Working Papers
Sorry, no publications matched your criteria.
Under Review
1.
Hao Tian, Ruozhou Wang, Feilong Zhou, Chaoyang Song, Jia Pan, Weihua Ou
Object Retrieval in Cluttered Scenes using Two-Stage Graph Neural Networks Online Forthcoming
Forthcoming, (Submitted to IEEE Robotics and Automation Letters).
Abstract | BibTeX | Tags: Authorship - Co-Author, Status - Under Review
@online{Tian2026ObjectRetrieval,
title = {Object Retrieval in Cluttered Scenes using Two-Stage Graph Neural Networks},
author = {Hao Tian and Ruozhou Wang and Feilong Zhou and Chaoyang Song and Jia Pan and Weihua Ou},
year = {2025},
date = {2025-11-24},
abstract = {In this study, we propose a novel object retrieval framework in
cluttered environments based on a two-stage Graph Neural Network (GNN)
architecture. The proposed method leverages a Traversability Graph
(T-graph) to represent the spatial layout of objects and their movable
relationships, enabling efficient planning for retrieving a target
object obstructed by multiple obstacles. In the first stage, an Edge
Prediction Network (EdgeNet) predicts feasible connections between
object pairs to construct the T-graph from spatial distributions. In
the second stage, a Path Prediction Network (PathNet) estimates the
optimal retrieval sequence on the predicted T-graph, indicating the
order of object removals required to access the target. To address
execution failures caused by collisions or kinematic constraints during
manipulation, we integrate the learned models with a heuristic Lifelong
Planning A* (LPA*) algorithm, facilitating efficient online replanning.
Extensive simulation and physical robot experiments demonstrate that
our approach significantly improves retrieval efficiency and success
rate compared with existing graph-based and search-based methods,
particularly in highly cluttered scenes with numerous objects. The
results verify that the proposed two-stage GNN framework provides a
scalable and data-driven solution for robotic object retrieval in
complex, confined environments.},
note = {Submitted to IEEE Robotics and Automation Letters},
keywords = {Authorship - Co-Author, Status - Under Review},
pubstate = {forthcoming},
tppubtype = {online}
}
In this study, we propose a novel object retrieval framework in
cluttered environments based on a two-stage Graph Neural Network (GNN)
architecture. The proposed method leverages a Traversability Graph
(T-graph) to represent the spatial layout of objects and their movable
relationships, enabling efficient planning for retrieving a target
object obstructed by multiple obstacles. In the first stage, an Edge
Prediction Network (EdgeNet) predicts feasible connections between
object pairs to construct the T-graph from spatial distributions. In
the second stage, a Path Prediction Network (PathNet) estimates the
optimal retrieval sequence on the predicted T-graph, indicating the
order of object removals required to access the target. To address
execution failures caused by collisions or kinematic constraints during
manipulation, we integrate the learned models with a heuristic Lifelong
Planning A* (LPA*) algorithm, facilitating efficient online replanning.
Extensive simulation and physical robot experiments demonstrate that
our approach significantly improves retrieval efficiency and success
rate compared with existing graph-based and search-based methods,
particularly in highly cluttered scenes with numerous objects. The
results verify that the proposed two-stage GNN framework provides a
scalable and data-driven solution for robotic object retrieval in
complex, confined environments.
cluttered environments based on a two-stage Graph Neural Network (GNN)
architecture. The proposed method leverages a Traversability Graph
(T-graph) to represent the spatial layout of objects and their movable
relationships, enabling efficient planning for retrieving a target
object obstructed by multiple obstacles. In the first stage, an Edge
Prediction Network (EdgeNet) predicts feasible connections between
object pairs to construct the T-graph from spatial distributions. In
the second stage, a Path Prediction Network (PathNet) estimates the
optimal retrieval sequence on the predicted T-graph, indicating the
order of object removals required to access the target. To address
execution failures caused by collisions or kinematic constraints during
manipulation, we integrate the learned models with a heuristic Lifelong
Planning A* (LPA*) algorithm, facilitating efficient online replanning.
Extensive simulation and physical robot experiments demonstrate that
our approach significantly improves retrieval efficiency and success
rate compared with existing graph-based and search-based methods,
particularly in highly cluttered scenes with numerous objects. The
results verify that the proposed two-stage GNN framework provides a
scalable and data-driven solution for robotic object retrieval in
complex, confined environments.
Journal Articles
16.
Shuqiao Zhong, Wanghongjie Qiu, Xudong Han, Chaoyang Song, Zhiyuan Zhou, Fang Wan, Jian Lin
Bio-inspired Rigid-Soft Interaction for Robust Picking Underwater Journal Article
In: Ocean Engineering, vol. 346, pp. 123774, 2026.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - Ocean Eng.
@article{Zhong2025Bioinspired,
title = {Bio-inspired Rigid-Soft Interaction for Robust Picking Underwater},
author = {Shuqiao Zhong and Wanghongjie Qiu and Xudong Han and Chaoyang Song and Zhiyuan Zhou and Fang Wan and Jian Lin},
doi = {10.1016/j.oceaneng.2025.123774},
year = {2026},
date = {2026-02-15},
urldate = {2026-02-15},
journal = {Ocean Engineering},
volume = {346},
pages = {123774},
abstract = {We present a Lobster-inspired Soft Touch Enhanced Rigid Grasp (LobSTER-Grasp) gripper for robust underwater manipulation, replacing conventional grippers on commercial underwater robotic arms by integrating passively adaptive, networked soft fingers with rigid fingers and a newly designed mounting structure. We demonstrate that the networked soft fingers exhibit significantly lower hydrodynamic drag than solid fingers, mitigating the disturbance often induced when approaching underwater targets. In addition, we systematically investigate the parameters of lobster-inspired serrations on the finger surfaces, identifying a design that achieves 261% higher frictional force compared with standard uniform serrations. We embed these serrations into rigid and soft fingers for enhanced grasp reliability. Through a user study, we show that the LobSTER-Grasp gripper improves 38.2% grasping efficiency and 77.9% task versatility compared to a commercial counterpart, and we further validate its effectiveness in a tank environment using a remotely operated vehicle. Our findings suggest that this bio-inspired, rigid–soft coupling design offers a simple yet superior solution for more robust and adaptive underwater operations with strong potential in marine applications such as resource exploitation, ecological monitoring, and deep-sea exploration.},
keywords = {Authorship - Co-Author, JCR Q1, Jour - Ocean Eng.},
pubstate = {published},
tppubtype = {article}
}
We present a Lobster-inspired Soft Touch Enhanced Rigid Grasp (LobSTER-Grasp) gripper for robust underwater manipulation, replacing conventional grippers on commercial underwater robotic arms by integrating passively adaptive, networked soft fingers with rigid fingers and a newly designed mounting structure. We demonstrate that the networked soft fingers exhibit significantly lower hydrodynamic drag than solid fingers, mitigating the disturbance often induced when approaching underwater targets. In addition, we systematically investigate the parameters of lobster-inspired serrations on the finger surfaces, identifying a design that achieves 261% higher frictional force compared with standard uniform serrations. We embed these serrations into rigid and soft fingers for enhanced grasp reliability. Through a user study, we show that the LobSTER-Grasp gripper improves 38.2% grasping efficiency and 77.9% task versatility compared to a commercial counterpart, and we further validate its effectiveness in a tank environment using a remotely operated vehicle. Our findings suggest that this bio-inspired, rigid–soft coupling design offers a simple yet superior solution for more robust and adaptive underwater operations with strong potential in marine applications such as resource exploitation, ecological monitoring, and deep-sea exploration.
15.
Chengxiao Dong, Yu Pan, Xuanyi Dai, Edmond Ho Nang, Chaoyang Song, Fang Wan
Enhancing Full-Arch Intraoral Measurement with Robotic Process Automation Journal Article
In: Journal of Bionic Engineering, vol. 0, iss. October, no. 0, pp. 0, 2025, (Accepted).
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - J. Bionic Eng. (JBE)
@article{Dong2024EnhancingFull,
title = {Enhancing Full-Arch Intraoral Measurement with Robotic Process Automation},
author = {Chengxiao Dong and Yu Pan and Xuanyi Dai and Edmond Ho Nang and Chaoyang Song and Fang Wan},
doi = {10.1007/s42235-025-00784-8},
year = {2025},
date = {2025-08-26},
urldate = {2025-08-26},
journal = {Journal of Bionic Engineering},
volume = {0},
number = {0},
issue = {October},
pages = {0},
abstract = {Intraoral scanning has become integral to digital workflows in dental implantology, offering a more efficient and comfortable alternative to conventional impression techniques. For complete edentulism, accurate scanning is crucial to successful full-arch dental implant rehabilitation. However, the absence of well-defined anatomical landmarks can lead to cumulative errors during merging sequential scans, often surpassing acceptable thresholds. Current mitigation strategies rely on manual adjustments in computer-aided design (CAD) software, a time-intensive process that depends heavily on the operator's expertise. This study presents a novel textit{segment-match-correct} robotic process automation (RPA) workflow to enhance full-arch intraoral scans' positioning accuracy and efficiency. By leveraging 3D registration algorithms, the proposed method improves implant positioning accuracy while significantly reducing manual labor. To assess the robustness of this workflow, we simulated four types of noise to evaluate their impact on scanning errors. Our findings demonstrate that the RPA workflow reduces dentist workload from 5-8 minutes per scan to less than 1 minute (about 57 seconds) while achieving a lower linear error of 45.16 $pm$ 23.76 unit{micrometer}, outperforming traditional scanning methods. We could replicate linear and angular deviations observed in real-world scans by simulating cumulative errors. This workflow improves the accuracy and efficiency of complete-arch implant rehabilitation and provides a practical solution to reduce cumulative scanning errors. Additionally, the noise simulations offer valuable insights into the origins of these errors, further optimizing intraoral scanner performance.},
note = {Accepted},
keywords = {Authorship - Co-Author, JCR Q1, Jour - J. Bionic Eng. (JBE)},
pubstate = {published},
tppubtype = {article}
}
Intraoral scanning has become integral to digital workflows in dental implantology, offering a more efficient and comfortable alternative to conventional impression techniques. For complete edentulism, accurate scanning is crucial to successful full-arch dental implant rehabilitation. However, the absence of well-defined anatomical landmarks can lead to cumulative errors during merging sequential scans, often surpassing acceptable thresholds. Current mitigation strategies rely on manual adjustments in computer-aided design (CAD) software, a time-intensive process that depends heavily on the operator's expertise. This study presents a novel textit{segment-match-correct} robotic process automation (RPA) workflow to enhance full-arch intraoral scans' positioning accuracy and efficiency. By leveraging 3D registration algorithms, the proposed method improves implant positioning accuracy while significantly reducing manual labor. To assess the robustness of this workflow, we simulated four types of noise to evaluate their impact on scanning errors. Our findings demonstrate that the RPA workflow reduces dentist workload from 5-8 minutes per scan to less than 1 minute (about 57 seconds) while achieving a lower linear error of 45.16 $pm$ 23.76 unit{micrometer}, outperforming traditional scanning methods. We could replicate linear and angular deviations observed in real-world scans by simulating cumulative errors. This workflow improves the accuracy and efficiency of complete-arch implant rehabilitation and provides a practical solution to reduce cumulative scanning errors. Additionally, the noise simulations offer valuable insights into the origins of these errors, further optimizing intraoral scanner performance.
14.
Xi Xia, Xingxing Chen, Junli Shi, Zhibin Li, Bingfa Jiang, Kaixi Huang, Mengxue Guo, Zeyun Yang, Zelong Liao, Chaoyang Song, Chuanfei Guo
Micropillar-Enabled Tough Adhesion and Enhanced Sensing Journal Article
In: Matter, vol. 8, iss. October, no. 10, pp. 102221, 2025.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - Matter
@article{Xia2025MicropillarEnabled,
title = {Micropillar-Enabled Tough Adhesion and Enhanced Sensing},
author = {Xi Xia and Xingxing Chen and Junli Shi and Zhibin Li and Bingfa Jiang and Kaixi Huang and Mengxue Guo and Zeyun Yang and Zelong Liao and Chaoyang Song and Chuanfei Guo},
doi = {10.1016/j.matt.2025.102221},
year = {2025},
date = {2025-06-20},
urldate = {2025-06-20},
journal = {Matter},
volume = {8},
number = {10},
issue = {October},
pages = {102221},
abstract = {Skin-like soft sensors are a key technology for humanoid robots and wearables. Achieving both robust interfaces and promoted sensing performances in soft sensors may enable their applications in extreme mechanical conditions of high shear and large strain. However, introducing tough adhesion to the interfaces in a sensor often compromises its sensing properties. Here, we use micropillars of hyperbranched polyurethane with a diameter smaller than its length of flaw sensitivity serving as an adhesion layer for exceptional mechanical stability, and also as an adaptive spacer for enhanced sensing properties. We show a strong size effect of the structures to toughen the interface, with ultrahigh interfacial toughness up to 5095 J m-2 at a pillar diameter of 50 μm, which is one order of magnitude higher than the state-of-the-arts results. As a spacer, the micropillars provide enhanced sensitivity, adaptive limit of detection, rapid response to the acoustic range by decreasing the stiffness via elastic buckling. The sensors are ideal for the manipulation of heavy objects in humanoid robots and other applications. },
keywords = {Authorship - Co-Author, JCR Q1, Jour - Matter},
pubstate = {published},
tppubtype = {article}
}
Skin-like soft sensors are a key technology for humanoid robots and wearables. Achieving both robust interfaces and promoted sensing performances in soft sensors may enable their applications in extreme mechanical conditions of high shear and large strain. However, introducing tough adhesion to the interfaces in a sensor often compromises its sensing properties. Here, we use micropillars of hyperbranched polyurethane with a diameter smaller than its length of flaw sensitivity serving as an adhesion layer for exceptional mechanical stability, and also as an adaptive spacer for enhanced sensing properties. We show a strong size effect of the structures to toughen the interface, with ultrahigh interfacial toughness up to 5095 J m-2 at a pillar diameter of 50 μm, which is one order of magnitude higher than the state-of-the-arts results. As a spacer, the micropillars provide enhanced sensitivity, adaptive limit of detection, rapid response to the acoustic range by decreasing the stiffness via elastic buckling. The sensors are ideal for the manipulation of heavy objects in humanoid robots and other applications.
13.
Sen Li, Chengxiao Dong, Chaoyang Song, Fang Wan
ActiveSPN: Active Soft Polyhedral Networks with Pose Estimation for In-Finger Object Manipulation Journal Article
In: IEEE Robotics and Automation Letters, vol. 10, no. 8, pp. 8115 - 8122, 2025, (Accepted).
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - IEEE Robot. Autom. Lett. (RA-L)
@article{Li2025ActiveSPN,
title = {ActiveSPN: Active Soft Polyhedral Networks with Pose Estimation for In-Finger Object Manipulation},
author = {Sen Li and Chengxiao Dong and Chaoyang Song and Fang Wan},
url = {https://github.com/ancorasir/ActiveSPN},
doi = {10.1109/LRA.2025.3583616},
year = {2025},
date = {2025-06-16},
urldate = {2025-06-16},
journal = {IEEE Robotics and Automation Letters},
volume = {10},
number = {8},
pages = {8115 - 8122},
abstract = {Robotic grippers aim to replicate the remarkable functionalities of the human hand by providing advanced perception, adaptability, stability, and dexterity for complex tasks. Achieving these capabilities demands a sophisticated design hierarchy and robust perception mechanisms that ensure accurate manipulation. This paper introduces Active Soft Polyhedral Networks (ActiveSPN), a gripper design that leverages an active, non-biomimetic surface for precise in-hand manipulation. A vision system integrated directly into the fingers further facilitates accurate pose estimation of the in-finger object. The proposed system includes: (i) a soft polyhedral network featuring a transparent active belt to deliver complete three-dimensional adaptation and dexterous in-finger motion, and (ii) a generative learning-based pipeline for in-finger pose estimation. Experimental results demonstrate the ability of ActiveSPN to execute multi-degree-of-freedom in-finger manipulations, including two-axis rotation and one-axis translation. Moreover, the integrated vision-based pose estimation provides robust, real-time predictions, supporting consistent closed-loop control. Across diverse objects, the system achieves mean translational errors of 2.59 mm and rotational errors of 7 degrees, highlighting a promising paradigm for compact, efficient, and dexterous robotic manipulation. Codes are available at https://github.com/ancorasir/ActiveSPN.},
note = {Accepted},
keywords = {Authorship - Co-Author, JCR Q1, Jour - IEEE Robot. Autom. Lett. (RA-L)},
pubstate = {published},
tppubtype = {article}
}
Robotic grippers aim to replicate the remarkable functionalities of the human hand by providing advanced perception, adaptability, stability, and dexterity for complex tasks. Achieving these capabilities demands a sophisticated design hierarchy and robust perception mechanisms that ensure accurate manipulation. This paper introduces Active Soft Polyhedral Networks (ActiveSPN), a gripper design that leverages an active, non-biomimetic surface for precise in-hand manipulation. A vision system integrated directly into the fingers further facilitates accurate pose estimation of the in-finger object. The proposed system includes: (i) a soft polyhedral network featuring a transparent active belt to deliver complete three-dimensional adaptation and dexterous in-finger motion, and (ii) a generative learning-based pipeline for in-finger pose estimation. Experimental results demonstrate the ability of ActiveSPN to execute multi-degree-of-freedom in-finger manipulations, including two-axis rotation and one-axis translation. Moreover, the integrated vision-based pose estimation provides robust, real-time predictions, supporting consistent closed-loop control. Across diverse objects, the system achieves mean translational errors of 2.59 mm and rotational errors of 7 degrees, highlighting a promising paradigm for compact, efficient, and dexterous robotic manipulation. Codes are available at https://github.com/ancorasir/ActiveSPN.
12.
Rongzheng Zhang, Wanghongjie Qiu, Jianuo Qiu, Yuqin Guo, Chengxiao Dong, Tuo Zhang, Juan Yi, Chaoyang Song, Harry Asada, Fang Wan
Multimodal Intention Recognition Combining Head Motion and Throat Vibration for Underwater Superlimbs Journal Article
In: IEEE Transactions on Automation Science and Engineering, vol. 0, no. 0, pp. 0, 2025.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - IEEE Trans. Autom. Sci. Eng. (T-ASE)
@article{Zhang2024MultiModal,
title = {Multimodal Intention Recognition Combining Head Motion and Throat Vibration for Underwater Superlimbs},
author = {Rongzheng Zhang and Wanghongjie Qiu and Jianuo Qiu and Yuqin Guo and Chengxiao Dong and Tuo Zhang and Juan Yi and Chaoyang Song and Harry Asada and Fang Wan},
doi = {10.1109/TASE.2025.3554036},
year = {2025},
date = {2025-03-20},
urldate = {2025-03-20},
journal = {IEEE Transactions on Automation Science and Engineering},
volume = {0},
number = {0},
pages = {0},
abstract = {This paper presents a novel solution for underwater intention recognition that simultaneously detects head motion and throat vibration, enhancing multimodal human-robot interactions for underwater diving. The system pairs with an underwater supernumerary robotic limb (SuperLimb), providing propulsion assistance to reduce the diver's physical load and mental fatigue. An inertial measurement unit monitors head motion, while a throat microphone captures vocal vibrations. Learning algorithms process these signals to accurately interpret the diver's intentions and map them to the SuperLimb for posture management. The system features a compact design optimized for diving scenarios and includes a multimodal, real-time classification algorithm to distinguish various head motions and vocal signals. By collecting and analyzing underwater throat vibration data, the study demonstrates the feasibility of this approach, enabling continuous motion commands for enhanced diving assistance. The results show that the head motion recognition component of the system achieved a high classification accuracy of 95%, and throat vibration classification reached 86% accuracy on land and 89% underwater for various purposes.},
keywords = {Authorship - Co-Author, JCR Q1, Jour - IEEE Trans. Autom. Sci. Eng. (T-ASE)},
pubstate = {published},
tppubtype = {article}
}
This paper presents a novel solution for underwater intention recognition that simultaneously detects head motion and throat vibration, enhancing multimodal human-robot interactions for underwater diving. The system pairs with an underwater supernumerary robotic limb (SuperLimb), providing propulsion assistance to reduce the diver's physical load and mental fatigue. An inertial measurement unit monitors head motion, while a throat microphone captures vocal vibrations. Learning algorithms process these signals to accurately interpret the diver's intentions and map them to the SuperLimb for posture management. The system features a compact design optimized for diving scenarios and includes a multimodal, real-time classification algorithm to distinguish various head motions and vocal signals. By collecting and analyzing underwater throat vibration data, the study demonstrates the feasibility of this approach, enabling continuous motion commands for enhanced diving assistance. The results show that the head motion recognition component of the system achieved a high classification accuracy of 95%, and throat vibration classification reached 86% accuracy on land and 89% underwater for various purposes.
11.
Victor-Louis De Gusseme, Thomas Lips, Remko Proesmans, Julius Hietala, Giwan Lee, Jiyoung Choi, Jeongil Choi, Geon Kim, Phayuth Yonrith, Domen Tabernik, Andrej Gams, Peter Nimac, Matej Urbas, Jon Muhovic, Danijel Skocaj, Matija Mavsar, Hyojeong Yu, Minseo Kwon, Young J. Kim, Yang Cong, Ronghan Chen, Yu Ren, Supeng Diao, Jiawei Weng, Jiayue Liu, Haoran Sun, Linhan Yang, Zeqing Zhang, Ning Guo, Lei Yang, Fang Wan, Chaoyang Song, Jia Pan, Yixiang Jin, Yong A, Jun Shi, Dingzhe Li, Yong Yang, Kakeru Yamasaki, Takumi Kajiwara, Yuki Nakadera, Krati Saxena, Tomohiro Shibata, Chongkun Xia, Kai Mo, Yanzhao Yu, Qihao Lin, Binqiang Ma, Uihun Sagong, JungHyun Choi, JeongHyun Park, Dongwoo Lee, Yeongmin Kim, Myun Joong Hwang, Yusuke Kuribayashi, Naoki Hiratsuka, Daisuke Tanaka, Solvi Arnold, Kimitoshi Yamazaki, Carlos Mateo-Agullo, Andreas Verleysen, Francis wyffels
A Dataset and Benchmark for Robotic Cloth Unfolding Grasp Selection: The ICRA 2024 Cloth Competition Journal Article Forthcoming
In: The International Journal of Robotics Research, vol. 0, no. 0, pp. 0, Forthcoming, (Accepted).
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q2, Jour - Int. J. Robot. Res. (IJRR)
@article{DeGusseme2024BenchmarkingGrasp,
title = {A Dataset and Benchmark for Robotic Cloth Unfolding Grasp Selection: The ICRA 2024 Cloth Competition},
author = {Victor-Louis De Gusseme and Thomas Lips and Remko Proesmans and Julius Hietala and Giwan Lee and Jiyoung Choi and Jeongil Choi and Geon Kim and Phayuth Yonrith and Domen Tabernik and Andrej Gams and Peter Nimac and Matej Urbas and Jon Muhovic and Danijel Skocaj and Matija Mavsar and Hyojeong Yu and Minseo Kwon and Young J. Kim and Yang Cong and Ronghan Chen and Yu Ren and Supeng Diao and Jiawei Weng and Jiayue Liu and Haoran Sun and Linhan Yang and Zeqing Zhang and Ning Guo and Lei Yang and Fang Wan and Chaoyang Song and Jia Pan and Yixiang Jin and Yong A and Jun Shi and Dingzhe Li and Yong Yang and Kakeru Yamasaki and Takumi Kajiwara and Yuki Nakadera and Krati Saxena and Tomohiro Shibata and Chongkun Xia and Kai Mo and Yanzhao Yu and Qihao Lin and Binqiang Ma and Uihun Sagong and JungHyun Choi and JeongHyun Park and Dongwoo Lee and Yeongmin Kim and Myun Joong Hwang and Yusuke Kuribayashi and Naoki Hiratsuka and Daisuke Tanaka and Solvi Arnold and Kimitoshi Yamazaki and Carlos Mateo-Agullo and Andreas Verleysen and Francis wyffels},
url = {https://airo.ugent.be/cloth_competition/
https://doi.org/10.48550/arXiv.2508.16749},
doi = {10.1177/02783649251414885},
year = {2025},
date = {2025-01-10},
urldate = {2025-01-10},
journal = {The International Journal of Robotics Research},
volume = {0},
number = {0},
pages = {0},
abstract = {Robotic cloth manipulation suffers from a lack of standardized benchmarks and shared datasets for evaluating and comparing different approaches. To address this, we organized the ICRA 2024 Cloth Competition, a unique head-to-head evaluation focused on grasp pose selection for cloth unfolding. Eleven diverse teams competed with a shared dual-arm robot, utilizing our publicly released dataset of real-world robotic cloth unfolding attempts. We expanded this dataset with 176 live evaluation trials, which now encompasses 679 unfolding demonstrations across 34 garments. The competition established a key benchmark and reference for robotic cloth manipulation. Analysis revealed a significant discrepancy between competition performance and prior work, underscoring the importance of independent out-of-the-lab evaluation in robotic cloth manipulation. The resulting dataset, one of the most comprehensive collections of real-world robotic cloth manipulation data, is a valuable resource for developing and evaluating grasp selection methods, particularly for learning-based approaches. It can serve as a foundation for future benchmarks and drive further progress in data-driven robotic cloth manipulation.},
note = {Accepted},
keywords = {Authorship - Co-Author, JCR Q2, Jour - Int. J. Robot. Res. (IJRR)},
pubstate = {forthcoming},
tppubtype = {article}
}
Robotic cloth manipulation suffers from a lack of standardized benchmarks and shared datasets for evaluating and comparing different approaches. To address this, we organized the ICRA 2024 Cloth Competition, a unique head-to-head evaluation focused on grasp pose selection for cloth unfolding. Eleven diverse teams competed with a shared dual-arm robot, utilizing our publicly released dataset of real-world robotic cloth unfolding attempts. We expanded this dataset with 176 live evaluation trials, which now encompasses 679 unfolding demonstrations across 34 garments. The competition established a key benchmark and reference for robotic cloth manipulation. Analysis revealed a significant discrepancy between competition performance and prior work, underscoring the importance of independent out-of-the-lab evaluation in robotic cloth manipulation. The resulting dataset, one of the most comprehensive collections of real-world robotic cloth manipulation data, is a valuable resource for developing and evaluating grasp selection methods, particularly for learning-based approaches. It can serve as a foundation for future benchmarks and drive further progress in data-driven robotic cloth manipulation.
10.
Yu Pan, Xuanyi Dai, Fang Wan, Chaoyang Song, James KH Tsoi, Edmond HN Pow
A Novel Post-Processing Strategy to Improve the Accuracy of Complete-Arch Intraoral Scanning for Implants: An In Vitro Study Journal Article
In: Journal of Dentistry, vol. 139, iss. December, pp. 104761, 2023.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - J. Dent. (JoD)
@article{Pan2023ANovel,
title = {A Novel Post-Processing Strategy to Improve the Accuracy of Complete-Arch Intraoral Scanning for Implants: An In Vitro Study},
author = {Yu Pan and Xuanyi Dai and Fang Wan and Chaoyang Song and James KH Tsoi and Edmond HN Pow},
doi = {10.1016/j.jdent.2023.104761},
year = {2023},
date = {2023-12-01},
urldate = {2023-12-01},
journal = {Journal of Dentistry},
volume = {139},
issue = {December},
pages = {104761},
abstract = {[Objectives] To develop a new post-processing strategy that utilizes an auxiliary device to adjust intraoral scans and improve the accuracy of 3D models of complete-arch dental implants.
[Materials and methods] An edentulous resin model with 6 dental implants was prepared. An auxiliary device, consisting of an opaque base and artificial landmarks, was fabricated and mounted onto the resin model. Twenty intraoral scans (raw scans) were taken using this setup. A new post-processing strategy was proposed to adjust the raw scans using reverse engineering software (verified group). Additionally, ten conventional gypsum casts were duplicated and digitized using a laboratory scanner. The linear and angular trueness and precision of the models were evaluated and compared. The effect of the proposed strategy on the accuracy of complete-arch intraoral scans was analyzed using one-way ANOVA.
[Results] The linear trueness (29.7 µm) and precision (24.8 µm) of the verified group were significantly better than the raw scans (46.6 µm, 44.7 µm) and conventional casts (51.3 µm, 36.5 µm), particularly in cross-arch sites. However, the angular trueness (0.114°) and precision (0.085°) of the conventional casts were significantly better than both the verified models (0.298°, 0.168°) and the raw scans (0.288°, 0.202°).
[Conclusions] The novel post-processing strategy is effective in enhancing the linear accuracy of complete-arch implant IO scans, especially in cross-arch sites. However, further improvement is needed to eliminate the angular deviations.
[Clinical significance] Errors generated from intraoral scanning in complete edentulous arches exceed the clinical threshold. The elimination of stitching errors in the raw scans particularly in the cross-arch sites, through the proposed post-processing strategy would enhance the accuracy of complete-arch implant prostheses.},
keywords = {Authorship - Co-Author, JCR Q1, Jour - J. Dent. (JoD)},
pubstate = {published},
tppubtype = {article}
}
[Objectives] To develop a new post-processing strategy that utilizes an auxiliary device to adjust intraoral scans and improve the accuracy of 3D models of complete-arch dental implants.
[Materials and methods] An edentulous resin model with 6 dental implants was prepared. An auxiliary device, consisting of an opaque base and artificial landmarks, was fabricated and mounted onto the resin model. Twenty intraoral scans (raw scans) were taken using this setup. A new post-processing strategy was proposed to adjust the raw scans using reverse engineering software (verified group). Additionally, ten conventional gypsum casts were duplicated and digitized using a laboratory scanner. The linear and angular trueness and precision of the models were evaluated and compared. The effect of the proposed strategy on the accuracy of complete-arch intraoral scans was analyzed using one-way ANOVA.
[Results] The linear trueness (29.7 µm) and precision (24.8 µm) of the verified group were significantly better than the raw scans (46.6 µm, 44.7 µm) and conventional casts (51.3 µm, 36.5 µm), particularly in cross-arch sites. However, the angular trueness (0.114°) and precision (0.085°) of the conventional casts were significantly better than both the verified models (0.298°, 0.168°) and the raw scans (0.288°, 0.202°).
[Conclusions] The novel post-processing strategy is effective in enhancing the linear accuracy of complete-arch implant IO scans, especially in cross-arch sites. However, further improvement is needed to eliminate the angular deviations.
[Clinical significance] Errors generated from intraoral scanning in complete edentulous arches exceed the clinical threshold. The elimination of stitching errors in the raw scans particularly in the cross-arch sites, through the proposed post-processing strategy would enhance the accuracy of complete-arch implant prostheses.
[Materials and methods] An edentulous resin model with 6 dental implants was prepared. An auxiliary device, consisting of an opaque base and artificial landmarks, was fabricated and mounted onto the resin model. Twenty intraoral scans (raw scans) were taken using this setup. A new post-processing strategy was proposed to adjust the raw scans using reverse engineering software (verified group). Additionally, ten conventional gypsum casts were duplicated and digitized using a laboratory scanner. The linear and angular trueness and precision of the models were evaluated and compared. The effect of the proposed strategy on the accuracy of complete-arch intraoral scans was analyzed using one-way ANOVA.
[Results] The linear trueness (29.7 µm) and precision (24.8 µm) of the verified group were significantly better than the raw scans (46.6 µm, 44.7 µm) and conventional casts (51.3 µm, 36.5 µm), particularly in cross-arch sites. However, the angular trueness (0.114°) and precision (0.085°) of the conventional casts were significantly better than both the verified models (0.298°, 0.168°) and the raw scans (0.288°, 0.202°).
[Conclusions] The novel post-processing strategy is effective in enhancing the linear accuracy of complete-arch implant IO scans, especially in cross-arch sites. However, further improvement is needed to eliminate the angular deviations.
[Clinical significance] Errors generated from intraoral scanning in complete edentulous arches exceed the clinical threshold. The elimination of stitching errors in the raw scans particularly in the cross-arch sites, through the proposed post-processing strategy would enhance the accuracy of complete-arch implant prostheses.
9.
Linhan Yang, Bidan Huang, Qingbiao Li, Ya-Yen Tsai, Wang Wei Lee, Chaoyang Song, Jia Pan
TacGNN: Learning Tactile-based In-hand Manipulation with a Blind Robot using Hierarchical Graph Neural Network Journal Article
In: IEEE Robotics and Automation Letters, vol. 8, iss. June, no. 6, pp. 3605-3612, 2023.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - IEEE Robot. Autom. Lett. (RA-L), Special - Dual-Track
@article{Yang2023TacGNN,
title = {TacGNN: Learning Tactile-based In-hand Manipulation with a Blind Robot using Hierarchical Graph Neural Network},
author = {Linhan Yang and Bidan Huang and Qingbiao Li and Ya-Yen Tsai and Wang Wei Lee and Chaoyang Song and Jia Pan},
doi = {10.1109/LRA.2023.3264759},
year = {2023},
date = {2023-04-05},
urldate = {2023-04-05},
journal = {IEEE Robotics and Automation Letters},
volume = {8},
number = {6},
issue = {June},
pages = {3605-3612},
abstract = {In this letter, we propose a novel framework for tactile-based dexterous manipulation learning with a blind anthropomorphic robotic hand, i.e. without visual sensing. First, object-related states were extracted from the raw tactile signals by a graph-based perception model - TacGNN. The resulting tactile features were then utilized in the policy learning of an in-hand manipulation task in the second stage. This method was examined by a Baoding ball task - simultaneously manipulating two spheres around each other by 180 degrees in hand. We conducted experiments on object states prediction and in-hand manipulation using a reinforcement learning algorithm (PPO). Results show that TacGNN is effective in predicting object-related states during manipulation by decreasing the RMSE of prediction to 0.096 cm comparing to other methods, such as MLP, CNN, and GCN. Finally, the robot hand could finish an in-hand manipulation task solely relying on the robotic own perception - tactile sensing and proprioception. In addition, our methods are tested on three tasks with different difficulty levels and transferred to the real robot without further training.},
keywords = {Authorship - Co-Author, JCR Q1, Jour - IEEE Robot. Autom. Lett. (RA-L), Special - Dual-Track},
pubstate = {published},
tppubtype = {article}
}
In this letter, we propose a novel framework for tactile-based dexterous manipulation learning with a blind anthropomorphic robotic hand, i.e. without visual sensing. First, object-related states were extracted from the raw tactile signals by a graph-based perception model - TacGNN. The resulting tactile features were then utilized in the policy learning of an in-hand manipulation task in the second stage. This method was examined by a Baoding ball task - simultaneously manipulating two spheres around each other by 180 degrees in hand. We conducted experiments on object states prediction and in-hand manipulation using a reinforcement learning algorithm (PPO). Results show that TacGNN is effective in predicting object-related states during manipulation by decreasing the RMSE of prediction to 0.096 cm comparing to other methods, such as MLP, CNN, and GCN. Finally, the robot hand could finish an in-hand manipulation task solely relying on the robotic own perception - tactile sensing and proprioception. In addition, our methods are tested on three tasks with different difficulty levels and transferred to the real robot without further training.
8.
Hao Tian, Chaoyang Song, Changbo Wang, Xinyu Zhang, Jia Pan
Sampling-Based Planning for Retrieving Near-Cylindrical Objects in Cluttered Scenes Using Hierarchical Graphs Journal Article
In: IEEE Transactions on Robotics, vol. 39, iss. February, no. 1, pp. 165-182, 2023.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - IEEE Trans. Robot. (T-RO)
@article{Tian2023SamplingBased,
title = {Sampling-Based Planning for Retrieving Near-Cylindrical Objects in Cluttered Scenes Using Hierarchical Graphs},
author = {Hao Tian and Chaoyang Song and Changbo Wang and Xinyu Zhang and Jia Pan},
doi = {10.1109/TRO.2022.3191596},
year = {2023},
date = {2023-02-01},
urldate = {2023-02-01},
journal = {IEEE Transactions on Robotics},
volume = {39},
number = {1},
issue = {February},
pages = {165-182},
abstract = {We present an incremental sampling-based task and motion planner for retrieving near-cylindrical objects, like bottle, in cluttered scenes, which computes a plan for removing obstacles to generate a collision-free motion of a robot to retrieve the target object. Our proposed planner uses a two-level hierarchy, including the first-level roadmap for the target object motion and the second-level retrieval graph for the entire robot motion, to aid in deciding the order and trajectory of object removal. We use an incremental expansion strategy to update the roadmap and retrieval graph from the collisions between the target object, the robot, and the obstacles, in order to optimize the object removal sequence. The performance of our method is highlighted in several benchmark scenes, including a fixed robotic arm in a cluttered scene with known obstacle locations and a scene, where locations of some objects or even the target object are unknown due to occlusions. Our method can also efficiently solve the high-dimensional planning problem of object retrieval using a mobile manipulator and be combined with the symbolic planner to plan complex multistep tasks. We deploy our method to a physical robot and integrate it with nonprehensile actions to improve operational efficiency. Compared to the state-of-the-art approaches, our method reduces task and motion planning time up to 24.6% with a higher success rate, and still provides a near-optimal plan.},
keywords = {Authorship - Co-Author, JCR Q1, Jour - IEEE Trans. Robot. (T-RO)},
pubstate = {published},
tppubtype = {article}
}
We present an incremental sampling-based task and motion planner for retrieving near-cylindrical objects, like bottle, in cluttered scenes, which computes a plan for removing obstacles to generate a collision-free motion of a robot to retrieve the target object. Our proposed planner uses a two-level hierarchy, including the first-level roadmap for the target object motion and the second-level retrieval graph for the entire robot motion, to aid in deciding the order and trajectory of object removal. We use an incremental expansion strategy to update the roadmap and retrieval graph from the collisions between the target object, the robot, and the obstacles, in order to optimize the object removal sequence. The performance of our method is highlighted in several benchmark scenes, including a fixed robotic arm in a cluttered scene with known obstacle locations and a scene, where locations of some objects or even the target object are unknown due to occlusions. Our method can also efficiently solve the high-dimensional planning problem of object retrieval using a mobile manipulator and be combined with the symbolic planner to plan complex multistep tasks. We deploy our method to a physical robot and integrate it with nonprehensile actions to improve operational efficiency. Compared to the state-of-the-art approaches, our method reduces task and motion planning time up to 24.6% with a higher success rate, and still provides a near-optimal plan.
7.
You Li, Zhuokang Huang, Xiaobo Liu, Yu Jie, Chaoyang Song, Chengzhi Hu
Calibrated Analytical Model for Magnetic Localization of Wireless Capsule Endoscope based on Onboard Sensing Journal Article
In: Robotica, vol. 41, no. 5, pp. 1500-1514, 2023.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q3, Jour - Robotica (ROBO)
@article{Li2023CalibratedAnalytical,
title = {Calibrated Analytical Model for Magnetic Localization of Wireless Capsule Endoscope based on Onboard Sensing},
author = {You Li and Zhuokang Huang and Xiaobo Liu and Yu Jie and Chaoyang Song and Chengzhi Hu},
doi = {10.1017/S0263574722001849},
year = {2023},
date = {2023-01-12},
urldate = {2023-01-12},
journal = {Robotica},
volume = {41},
number = {5},
pages = {1500-1514},
abstract = {We present an incremental sampling-based task and motion planner for retrieving near-cylindrical objects, like bottle, in cluttered scenes, which computes a plan for removing obstacles to generate a collision-free motion of a robot to retrieve the target object. Our proposed planner uses a two-level hierarchy, including the first-level roadmap for the target object motion and the second-level retrieval graph for the entire robot motion, to aid in deciding the order and trajectory of object removal. We use an incremental expansion strategy to update the roadmap and retrieval graph from the collisions between the target object, the robot, and the obstacles, in order to optimize the object removal sequence. The performance of our method is highlighted in several benchmark scenes, including a fixed robotic arm in a cluttered scene with known obstacle locations and a scene, where locations of some objects or even the target object are unknown due to occlusions. Our method can also efficiently solve the high-dimensional planning problem of object retrieval using a mobile manipulator and be combined with the symbolic planner to plan complex multistep tasks. We deploy our method to a physical robot and integrate it with nonprehensile actions to improve operational efficiency. Compared to the state-of-the-art approaches, our method reduces task and motion planning time up to 24.6% with a higher success rate, and still provides a near-optimal plan.},
keywords = {Authorship - Co-Author, JCR Q3, Jour - Robotica (ROBO)},
pubstate = {published},
tppubtype = {article}
}
We present an incremental sampling-based task and motion planner for retrieving near-cylindrical objects, like bottle, in cluttered scenes, which computes a plan for removing obstacles to generate a collision-free motion of a robot to retrieve the target object. Our proposed planner uses a two-level hierarchy, including the first-level roadmap for the target object motion and the second-level retrieval graph for the entire robot motion, to aid in deciding the order and trajectory of object removal. We use an incremental expansion strategy to update the roadmap and retrieval graph from the collisions between the target object, the robot, and the obstacles, in order to optimize the object removal sequence. The performance of our method is highlighted in several benchmark scenes, including a fixed robotic arm in a cluttered scene with known obstacle locations and a scene, where locations of some objects or even the target object are unknown due to occlusions. Our method can also efficiently solve the high-dimensional planning problem of object retrieval using a mobile manipulator and be combined with the symbolic planner to plan complex multistep tasks. We deploy our method to a physical robot and integrate it with nonprehensile actions to improve operational efficiency. Compared to the state-of-the-art approaches, our method reduces task and motion planning time up to 24.6% with a higher success rate, and still provides a near-optimal plan.
6.
Youcan Yan, Yajing Shen, Chaoyang Song, Jia Pan
Tactile Super-Resolution Model for Soft Magnetic Skin Journal Article
In: IEEE Robotics and Automation Letters, vol. 7, iss. April, no. 2, pp. 2589-2596, 2022.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - IEEE Robot. Autom. Lett. (RA-L), Special - Dual-Track
@article{Yan2022TactileSuper,
title = {Tactile Super-Resolution Model for Soft Magnetic Skin},
author = {Youcan Yan and Yajing Shen and Chaoyang Song and Jia Pan},
doi = {10.1109/LRA.2022.3141449},
year = {2022},
date = {2022-01-10},
urldate = {2022-01-10},
journal = {IEEE Robotics and Automation Letters},
volume = {7},
number = {2},
issue = {April},
pages = {2589-2596},
abstract = {Tactile sensors of high spatial resolution can provide rich contact information in terms of accurate contact location and force magnitude for robots. However, achieving a high spatial resolution normally requires a high density of tactile sensing cells (or taxels), which will inevitably lead to crowded wire connections, more data acquisition time and probably crosstalk between taxels. An alternative approach to improve the spatial resolution without introducing a high density of taxels is employing super-resolution technology. Here, we propose a novel tactile super-resolution method based on a sinusoidally magnetized soft magnetic skin, by which we have achieved a 15-fold improvement of localization accuracy (from 6 mm to 0.4 mm) as well as the ability to measure the force magnitude. Different from the existing super-resolution methods that rely on overlapping signals of neighbouring taxels, our model only relies on the local information from a single 3-axis taxel and thereby can detect multipoint contact applied on neighboring taxels and work properly even when some of the neighbouring taxels near the contact position are damaged (or unavailable). With this property, our method would be robust to damage and could potentially benefit robotic applications that require multipoint contact detection.},
keywords = {Authorship - Co-Author, JCR Q1, Jour - IEEE Robot. Autom. Lett. (RA-L), Special - Dual-Track},
pubstate = {published},
tppubtype = {article}
}
Tactile sensors of high spatial resolution can provide rich contact information in terms of accurate contact location and force magnitude for robots. However, achieving a high spatial resolution normally requires a high density of tactile sensing cells (or taxels), which will inevitably lead to crowded wire connections, more data acquisition time and probably crosstalk between taxels. An alternative approach to improve the spatial resolution without introducing a high density of taxels is employing super-resolution technology. Here, we propose a novel tactile super-resolution method based on a sinusoidally magnetized soft magnetic skin, by which we have achieved a 15-fold improvement of localization accuracy (from 6 mm to 0.4 mm) as well as the ability to measure the force magnitude. Different from the existing super-resolution methods that rely on overlapping signals of neighbouring taxels, our model only relies on the local information from a single 3-axis taxel and thereby can detect multipoint contact applied on neighboring taxels and work properly even when some of the neighbouring taxels near the contact position are damaged (or unavailable). With this property, our method would be robust to damage and could potentially benefit robotic applications that require multipoint contact detection.
5.
Sicong Liu, Yuming Zhu, Zicong Zhang, Zhonggui Fang, Jiyong Tan, Jing Peng, Chaoyang Song, Harry Asada, Zheng Wang
Otariidae-Inspired Soft-Robotic Supernumerary Flippers by Fabric Kirigami and Origami Journal Article
In: IEEE/ASME Transactions on Mechatronics, vol. 26, iss. October, no. 5, pp. 2747-2757, 2021.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - IEEE ASME Trans. Mechatron. (TMech)
@article{Liu2021OtariidaeInspired,
title = {Otariidae-Inspired Soft-Robotic Supernumerary Flippers by Fabric Kirigami and Origami},
author = {Sicong Liu and Yuming Zhu and Zicong Zhang and Zhonggui Fang and Jiyong Tan and Jing Peng and Chaoyang Song and Harry Asada and Zheng Wang},
doi = {10.1109/TMECH.2020.3045476},
year = {2021},
date = {2021-10-01},
urldate = {2021-10-01},
journal = {IEEE/ASME Transactions on Mechatronics},
volume = {26},
number = {5},
issue = {October},
pages = {2747-2757},
abstract = {Wearable robotic devices are receiving rapidly growing attentions for human-centered scenarios from medical, rehabilitation, to industrial applications. Supernumerary robotic limbs have been widely investigated for the augmentation of human limb functions, both as fingers and manipulator arms. Soft robotics offers an alternative approach to conventional motor-driven robot limbs toward safer and lighter systems, while pioneering soft supernumerary limbs are strongly limited in payload and dexterity by the soft robotic design approach, as well as the fabrication techniques. In this article, we proposed a wearable supernumerary soft robot for the human forearm, inspired by the fore flippers of otariids (eared seals). A flat flipper design was adopted, differing from the finger- or arm-shaped state-of-the-art works, with multiple soft actuators embedded as different joints for manipulation dexterity. The soft actuators were designed following origami (paper folding) patterns, reinforced by kirigami (paper cutting) fabrics. With this new approach, the proposed soft flipper incorporated eight independent muscles, achieving over 20 times payload to self-weight ratio, while weighing less than 500 g. The versatility, dexterity, and payload capability were experimentally demonstrated using a fabricated prototype with proprietary actuation and control. This article demonstrates the feasibility and unique advantages of origami + kirigami soft robots as a new approach to strong, dexterous, and yet safe and lightweight wearable robotic devices.},
keywords = {Authorship - Co-Author, JCR Q1, Jour - IEEE ASME Trans. Mechatron. (TMech)},
pubstate = {published},
tppubtype = {article}
}
Wearable robotic devices are receiving rapidly growing attentions for human-centered scenarios from medical, rehabilitation, to industrial applications. Supernumerary robotic limbs have been widely investigated for the augmentation of human limb functions, both as fingers and manipulator arms. Soft robotics offers an alternative approach to conventional motor-driven robot limbs toward safer and lighter systems, while pioneering soft supernumerary limbs are strongly limited in payload and dexterity by the soft robotic design approach, as well as the fabrication techniques. In this article, we proposed a wearable supernumerary soft robot for the human forearm, inspired by the fore flippers of otariids (eared seals). A flat flipper design was adopted, differing from the finger- or arm-shaped state-of-the-art works, with multiple soft actuators embedded as different joints for manipulation dexterity. The soft actuators were designed following origami (paper folding) patterns, reinforced by kirigami (paper cutting) fabrics. With this new approach, the proposed soft flipper incorporated eight independent muscles, achieving over 20 times payload to self-weight ratio, while weighing less than 500 g. The versatility, dexterity, and payload capability were experimentally demonstrated using a fabricated prototype with proprietary actuation and control. This article demonstrates the feasibility and unique advantages of origami + kirigami soft robots as a new approach to strong, dexterous, and yet safe and lightweight wearable robotic devices.
4.
Youcan Yan, Zhe Hu, Zhengbao Yang, Wenzhen Yuan, Chaoyang Song, Jia Pan, Yajing Shen
Soft Magnetic Skin for Super-Resolution Tactile Sensing with Force Self-Decoupling Journal Article
In: Science Robotics, vol. 6, no. 51, pp. eabc8801, 2021.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - Sci. Robot. (SciRob)
@article{Yan2021SoftMagnetic,
title = {Soft Magnetic Skin for Super-Resolution Tactile Sensing with Force Self-Decoupling},
author = {Youcan Yan and Zhe Hu and Zhengbao Yang and Wenzhen Yuan and Chaoyang Song and Jia Pan and Yajing Shen},
doi = {10.1126/scirobotics.abc8801},
year = {2021},
date = {2021-02-24},
urldate = {2021-02-24},
journal = {Science Robotics},
volume = {6},
number = {51},
pages = {eabc8801},
abstract = {Human skin can sense subtle changes of both normal and shear forces (i.e., self-decoupled) and perceive stimuli with finer resolution than the average spacing between mechanoreceptors (i.e., super-resolved). By contrast, existing tactile sensors for robotic applications are inferior, lacking accurate force decoupling and proper spatial resolution at the same time. Here, we present a soft tactile sensor with self-decoupling and super-resolution abilities by designing a sinusoidally magnetized flexible film (with the thickness ~0.5 millimeters), whose deformation can be detected by a Hall sensor according to the change of magnetic flux densities under external forces. The sensor can accurately measure the normal force and the shear force (demonstrated in one dimension) with a single unit and achieve a 60-fold super-resolved accuracy enhanced by deep learning. By mounting our sensor at the fingertip of a robotic gripper, we show that robots can accomplish challenging tasks such as stably grasping fragile objects under external disturbance and threading a needle via teleoperation. This research provides new insight into tactile sensor design and could be beneficial to various applications in robotics field, such as adaptive grasping, dexterous manipulation, and human-robot interaction.},
keywords = {Authorship - Co-Author, JCR Q1, Jour - Sci. Robot. (SciRob)},
pubstate = {published},
tppubtype = {article}
}
Human skin can sense subtle changes of both normal and shear forces (i.e., self-decoupled) and perceive stimuli with finer resolution than the average spacing between mechanoreceptors (i.e., super-resolved). By contrast, existing tactile sensors for robotic applications are inferior, lacking accurate force decoupling and proper spatial resolution at the same time. Here, we present a soft tactile sensor with self-decoupling and super-resolution abilities by designing a sinusoidally magnetized flexible film (with the thickness ~0.5 millimeters), whose deformation can be detected by a Hall sensor according to the change of magnetic flux densities under external forces. The sensor can accurately measure the normal force and the shear force (demonstrated in one dimension) with a single unit and achieve a 60-fold super-resolved accuracy enhanced by deep learning. By mounting our sensor at the fingertip of a robotic gripper, we show that robots can accomplish challenging tasks such as stably grasping fragile objects under external disturbance and threading a needle via teleoperation. This research provides new insight into tactile sensor design and could be beneficial to various applications in robotics field, such as adaptive grasping, dexterous manipulation, and human-robot interaction.
3.
Juan Yi, Xiaojiao Chen, Chaoyang Song, Jianshu Zhou, Yujia Liu, Sicong Liu, Zheng Wang
Customizable Three-Dimensional-Printed Origami Soft Robotic Joint with Effective Behavior Shaping for Safe Interactions Journal Article
In: IEEE Transactions on Robotics, vol. 35, iss. February, no. 1, pp. 114-123, 2019.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - IEEE Trans. Robot. (T-RO)
@article{Yi2019Customizable3D,
title = {Customizable Three-Dimensional-Printed Origami Soft Robotic Joint with Effective Behavior Shaping for Safe Interactions},
author = {Juan Yi and Xiaojiao Chen and Chaoyang Song and Jianshu Zhou and Yujia Liu and Sicong Liu and Zheng Wang},
doi = {10.1109/tro.2018.2871440},
year = {2019},
date = {2019-02-01},
urldate = {2019-02-01},
journal = {IEEE Transactions on Robotics},
volume = {35},
number = {1},
issue = {February},
pages = {114-123},
abstract = {Fast-growing interests in safe and effective robot–environment interactions stimulated global investigations on soft robotics. The inherent compliance of soft robots ensures promising safety features but drastically reduces force capability, thereby complicating system modeling and control. To tackle these limitations, a soft robotic joint with enhanced strength, servo performance, and impact behavior shaping is proposed in this paper, based on novel three-dimensional-printed soft origami rotary actuators. The complete workflow is presented from the concept of origami design and analytical modeling, joint design, fabrication, control, and validation experiments. The proposed approach facilitates a fully customizable joint design towards the desired force capability and motion range. Validation results from models and experiments using multiple fabricated prototypes proved the excellent performance linearity and superior force capability, with 18.5-N·m maximum torque under 180 kPa, and 300-g self-weight. The behavior shaping capability is achieved by a low-level joint-angle servo and a high-level variable-stiffness regulation; this significantly reduces the impact torque by 53% and ensures powerful and safe interactions. The comprehensive guidelines provide insightful references for soft robotic design for wider robotic applications.},
keywords = {Authorship - Co-Author, JCR Q1, Jour - IEEE Trans. Robot. (T-RO)},
pubstate = {published},
tppubtype = {article}
}
Fast-growing interests in safe and effective robot–environment interactions stimulated global investigations on soft robotics. The inherent compliance of soft robots ensures promising safety features but drastically reduces force capability, thereby complicating system modeling and control. To tackle these limitations, a soft robotic joint with enhanced strength, servo performance, and impact behavior shaping is proposed in this paper, based on novel three-dimensional-printed soft origami rotary actuators. The complete workflow is presented from the concept of origami design and analytical modeling, joint design, fabrication, control, and validation experiments. The proposed approach facilitates a fully customizable joint design towards the desired force capability and motion range. Validation results from models and experiments using multiple fabricated prototypes proved the excellent performance linearity and superior force capability, with 18.5-N·m maximum torque under 180 kPa, and 300-g self-weight. The behavior shaping capability is achieved by a low-level joint-angle servo and a high-level variable-stiffness regulation; this significantly reduces the impact torque by 53% and ensures powerful and safe interactions. The comprehensive guidelines provide insightful references for soft robotic design for wider robotic applications.
2.
Katja Hölttä-Otto, Kevin Otto, Chaoyang Song, Jianxi Luo, Timothy Li, Carolyn C. Seepersad, Warren Seering
The Characteristics of Innovative, Mechanical Products—10 Years Later Journal Article
In: Journal of Mechanical Design, vol. 140, iss. August, no. 8, pp. 084501, 2018.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q2, Jour - J. Mech. Des. (JMD)
@article{HolttaOtto2018TheCharacteristics,
title = {The Characteristics of Innovative, Mechanical Products—10 Years Later},
author = {Katja Hölttä-Otto and Kevin Otto and Chaoyang Song and Jianxi Luo and Timothy Li and Carolyn C. Seepersad and Warren Seering},
doi = {10.1115/1.4039851},
year = {2018},
date = {2018-08-01},
urldate = {2018-08-01},
journal = {Journal of Mechanical Design},
volume = {140},
number = {8},
issue = {August},
pages = {084501},
abstract = {Ten years prior to this paper, innovative mechanical products were analyzed and found to embody multiple innovation characteristics—an average of two more than competing products in the marketplace. At the time, it was not known whether these products would be successful over time and whether the number or type of innovation characteristics would be related with success. In this work, products from the previous study were categorized into well- and under-adopted products. Also, each product was categorized according to the type of firm that launched it: a new venture or an established firm. The innovative products enjoyed a success rate of 77% on average. The success was not dependent on the number or type of innovation characteristics embodied by the product. However, products developed in new ventures embody, on average, one more innovation characteristic and enjoy a slightly higher success rate than those launched by established firms.},
keywords = {Authorship - Co-Author, JCR Q2, Jour - J. Mech. Des. (JMD)},
pubstate = {published},
tppubtype = {article}
}
Ten years prior to this paper, innovative mechanical products were analyzed and found to embody multiple innovation characteristics—an average of two more than competing products in the marketplace. At the time, it was not known whether these products would be successful over time and whether the number or type of innovation characteristics would be related with success. In this work, products from the previous study were categorized into well- and under-adopted products. Also, each product was categorized according to the type of firm that launched it: a new venture or an established firm. The innovative products enjoyed a success rate of 77% on average. The success was not dependent on the number or type of innovation characteristics embodied by the product. However, products developed in new ventures embody, on average, one more innovation characteristic and enjoy a slightly higher success rate than those launched by established firms.
1.
Juan Yi, Xiaojiao Chen, Chaoyang Song, Zheng Wang
Fiber-Reinforced Origamic Robotic Actuator Journal Article
In: Soft Robotics, vol. 5, no. 1, pp. 81-92, 2018.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, JCR Q1, Jour - Soft Robot. (SORO)
@article{Yi2018FiberReinforced,
title = {Fiber-Reinforced Origamic Robotic Actuator},
author = {Juan Yi and Xiaojiao Chen and Chaoyang Song and Zheng Wang},
doi = {10.1089/soro.2016.0079},
year = {2018},
date = {2018-02-01},
urldate = {2018-02-01},
journal = {Soft Robotics},
volume = {5},
number = {1},
pages = {81-92},
abstract = {A novel pneumatic soft linear actuator Fiber-reinforced Origamic Robotic Actuator (FORA) is proposed with significant improvements on the popular McKibben-type actuators, offering nearly doubled motion range, substantially improved force profile, and significantly lower actuation pressure. The desirable feature set is made possible by a novel soft origamic chamber that expands radially while contracts axially when pressurized. Combining this new origamic chamber with a reinforcing fiber mesh, FORA generates very high traction force (over 150N) and very large contractile motion (over 50%) at very low input pressure (100 kPa). We developed quasi-static analytical models both to characterize the motion and forces and as guidelines for actuator design. Fabrication of FORA mostly involves consumer-grade three-dimensional (3D) printing. We provide a detailed list of materials and dimensions. Fabricated FORAs were tested on a dedicated platform against commercially available pneumatic artificial muscles from Shadow and Festo to showcase its superior performances and validate the analytical models with very good agreements. Finally, a robotic joint was developed driven by two antagonistic FORAs, to showcase the benefits of the performance improvements. With its simple structure, fully characterized mechanism, easy fabrication procedure, and highly desirable performance, FORA could be easily customized to application requirements and fabricated by anyone with access to a 3D printer. This will pave the way to the wider adaptation and application of soft robotic systems.},
keywords = {Authorship - Co-Author, JCR Q1, Jour - Soft Robot. (SORO)},
pubstate = {published},
tppubtype = {article}
}
A novel pneumatic soft linear actuator Fiber-reinforced Origamic Robotic Actuator (FORA) is proposed with significant improvements on the popular McKibben-type actuators, offering nearly doubled motion range, substantially improved force profile, and significantly lower actuation pressure. The desirable feature set is made possible by a novel soft origamic chamber that expands radially while contracts axially when pressurized. Combining this new origamic chamber with a reinforcing fiber mesh, FORA generates very high traction force (over 150N) and very large contractile motion (over 50%) at very low input pressure (100 kPa). We developed quasi-static analytical models both to characterize the motion and forces and as guidelines for actuator design. Fabrication of FORA mostly involves consumer-grade three-dimensional (3D) printing. We provide a detailed list of materials and dimensions. Fabricated FORAs were tested on a dedicated platform against commercially available pneumatic artificial muscles from Shadow and Festo to showcase its superior performances and validate the analytical models with very good agreements. Finally, a robotic joint was developed driven by two antagonistic FORAs, to showcase the benefits of the performance improvements. With its simple structure, fully characterized mechanism, easy fabrication procedure, and highly desirable performance, FORA could be easily customized to application requirements and fabricated by anyone with access to a 3D printer. This will pave the way to the wider adaptation and application of soft robotic systems.
Conference Papers
7.
Nuofan Qiu, Chaoyang Song, Fang Wan
Describing Robots from Design to Learning: Towards an Interactive Lifecycle Representation of Robots Conference
IEEE International Conference on Advanced Robotics and Mechatronics (ICARM2024), Tokyo, Japan, 2024.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, Conf - ICARM
@conference{Qiu2024DescribingRobots,
title = {Describing Robots from Design to Learning: Towards an Interactive Lifecycle Representation of Robots},
author = {Nuofan Qiu and Chaoyang Song and Fang Wan},
url = {https://github.com/bionicdl-sustech/ACDC4Robot
https://apps.autodesk.com/FUSION/en/Detail/Index?id=5028052292896011577},
doi = {10.1109/ICARM62033.2024.10715835},
year = {2024},
date = {2024-06-08},
urldate = {2024-06-08},
booktitle = {IEEE International Conference on Advanced Robotics and Mechatronics (ICARM2024)},
address = {Tokyo, Japan},
abstract = {The robot development process is divided into several stages, which create barriers to the exchange of information between these different stages. We advocate for an interactive lifecycle representation, extending from robot morphology design to learning, and introduce the role of robot description formats in facilitating information transfer throughout this pipeline. We analyzed the relationship between design and simulation, enabling us to employ robot process automation methods for transferring information from the design phase to the learning phase in simulation. As part of this effort, we have developed an open-source plugin called ACDC4Robot for Fusion 360, which automates this process and transforms Fusion 360 into a user-friendly graphical interface for creating and editing robot description formats. Additionally, we offer an out-of-the-box robot model library to streamline and reduce repetitive tasks. All codes are hosted open-source. (https://github.com/bionicdl-sustech/ACDC4Robot)},
keywords = {Authorship - Co-Author, Conf - ICARM},
pubstate = {published},
tppubtype = {conference}
}
The robot development process is divided into several stages, which create barriers to the exchange of information between these different stages. We advocate for an interactive lifecycle representation, extending from robot morphology design to learning, and introduce the role of robot description formats in facilitating information transfer throughout this pipeline. We analyzed the relationship between design and simulation, enabling us to employ robot process automation methods for transferring information from the design phase to the learning phase in simulation. As part of this effort, we have developed an open-source plugin called ACDC4Robot for Fusion 360, which automates this process and transforms Fusion 360 into a user-friendly graphical interface for creating and editing robot description formats. Additionally, we offer an out-of-the-box robot model library to streamline and reduce repetitive tasks. All codes are hosted open-source. (https://github.com/bionicdl-sustech/ACDC4Robot)
6.
Linhan Yang, Bidan Huang, Qingbiao Li, Ya-Yen Tsai, Wang Wei Lee, Chaoyang Song, Jia Pan
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS2023), Huntington Place, Detroit, Michigan, USA, 2023, (Dual-track Submission with RAL: https://doi.org/10.1109/LRA.2023.3264759).
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, Conf - IROS, Special - Dual-Track
@conference{Yang2023TacGNN-IROS,
title = {TacGNN: Learning Tactile-Based In-Hand Manipulation with a Blind Robot Using Hierarchical Graph Neural Network},
author = {Linhan Yang and Bidan Huang and Qingbiao Li and Ya-Yen Tsai and Wang Wei Lee and Chaoyang Song and Jia Pan},
url = {https://ieee-iros.org/},
year = {2023},
date = {2023-10-01},
urldate = {2023-10-01},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS2023)},
address = {Huntington Place, Detroit, Michigan, USA},
abstract = {In this letter, we propose a novel framework for tactile-based dexterous manipulation learning with a blind anthropomorphic robotic hand, i.e. without visual sensing. First, object-related states were extracted from the raw tactile signals by a graph-based perception model - TacGNN. The resulting tactile features were then utilized in the policy learning of an in-hand manipulation task in the second stage. This method was examined by a Baoding ball task - simultaneously manipulating two spheres around each other by 180 degrees in hand. We conducted experiments on object states prediction and in-hand manipulation using a reinforcement learning algorithm (PPO). Results show that TacGNN is effective in predicting object-related states during manipulation by decreasing the RMSE of prediction to 0.096 cm comparing to other methods, such as MLP, CNN, and GCN. Finally, the robot hand could finish an in-hand manipulation task solely relying on the robotic own perception - tactile sensing and proprioception. In addition, our methods are tested on three tasks with different difficulty levels and transferred to the real robot without further training.},
note = {Dual-track Submission with RAL: https://doi.org/10.1109/LRA.2023.3264759},
keywords = {Authorship - Co-Author, Conf - IROS, Special - Dual-Track},
pubstate = {published},
tppubtype = {conference}
}
In this letter, we propose a novel framework for tactile-based dexterous manipulation learning with a blind anthropomorphic robotic hand, i.e. without visual sensing. First, object-related states were extracted from the raw tactile signals by a graph-based perception model - TacGNN. The resulting tactile features were then utilized in the policy learning of an in-hand manipulation task in the second stage. This method was examined by a Baoding ball task - simultaneously manipulating two spheres around each other by 180 degrees in hand. We conducted experiments on object states prediction and in-hand manipulation using a reinforcement learning algorithm (PPO). Results show that TacGNN is effective in predicting object-related states during manipulation by decreasing the RMSE of prediction to 0.096 cm comparing to other methods, such as MLP, CNN, and GCN. Finally, the robot hand could finish an in-hand manipulation task solely relying on the robotic own perception - tactile sensing and proprioception. In addition, our methods are tested on three tasks with different difficulty levels and transferred to the real robot without further training.
5.
Juan Yi, Xiaojiao Chen, Zhonggui Fang, Yujia Liu, Dehao Duanmu, Yinyin Su, Chaoyang Song, Sicong Liu, Zheng Wang
A Soft Wearable Elbow Skeleton for Safe Motion Assistance by Variable Stiffness Conference
ASME International Design Engineering Technical Conferences and Computers and Information in Engineering Conference (DETC/CIE2022), Cleveland, Ohio, USA, 2022.
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, Conf - DETC/CIE
@conference{Yi2022ASoft,
title = {A Soft Wearable Elbow Skeleton for Safe Motion Assistance by Variable Stiffness},
author = {Juan Yi and Xiaojiao Chen and Zhonggui Fang and Yujia Liu and Dehao Duanmu and Yinyin Su and Chaoyang Song and Sicong Liu and Zheng Wang},
doi = {10.1115/DETC2022-90320},
year = {2022},
date = {2022-08-14},
urldate = {2022-08-14},
booktitle = {ASME International Design Engineering Technical Conferences and Computers and Information in Engineering Conference (DETC/CIE2022)},
address = {Cleveland, Ohio, USA},
abstract = {Wearable robots could provide external physical assist, contributing to the well-being of elderly or disabled users in accomplishing tasks or therapeutic procedures. However, closely integrating robot dynamics into human activity demands for safety, pleasance and effectiveness simultaneously, requiring both strength and compliance. Soft robotics is generally regarded as a suitable alternative to rigid motor-based actuators for their light weight and passive compliance. However, existing approaches either have predetermined and/or limited passive compliance, or have moderate payload or motion range. While proven very successful in hand actuation in terms of various robotic gloves, their fundamental limitations restrict further expansion to driving other human body parts with higher demands in payload and speed. Previously we have developed an origami soft robotic joint with high torque, customizable dimensions and motion range. In this paper, we report the most recent results on controller development and wearable system integration of the proposed soft actuator in achieving excellent variable-stiffness compliant performance with inherent safety. Using the newly proposed controller, we demonstrate that a higher stiffness leads to quicker passive recovery than human’s normal reaction, suitable in stabilizing common object; while a more submissive configuration, represented by a lower stiffness, could delay the submissive interaction time allowing for a more delicate control of the dynamics of the interaction port with further involvement of human commands, suitable for objects requiring smoother dynamics like preventing a cup of coffee from sloshing. The proposed control strategy and framework has been implemented in a 3D-scanned, 3D-printed wearable robot with elbow actuation, to demonstrate the advantages and new features through experimental results.},
keywords = {Authorship - Co-Author, Conf - DETC/CIE},
pubstate = {published},
tppubtype = {conference}
}
Wearable robots could provide external physical assist, contributing to the well-being of elderly or disabled users in accomplishing tasks or therapeutic procedures. However, closely integrating robot dynamics into human activity demands for safety, pleasance and effectiveness simultaneously, requiring both strength and compliance. Soft robotics is generally regarded as a suitable alternative to rigid motor-based actuators for their light weight and passive compliance. However, existing approaches either have predetermined and/or limited passive compliance, or have moderate payload or motion range. While proven very successful in hand actuation in terms of various robotic gloves, their fundamental limitations restrict further expansion to driving other human body parts with higher demands in payload and speed. Previously we have developed an origami soft robotic joint with high torque, customizable dimensions and motion range. In this paper, we report the most recent results on controller development and wearable system integration of the proposed soft actuator in achieving excellent variable-stiffness compliant performance with inherent safety. Using the newly proposed controller, we demonstrate that a higher stiffness leads to quicker passive recovery than human’s normal reaction, suitable in stabilizing common object; while a more submissive configuration, represented by a lower stiffness, could delay the submissive interaction time allowing for a more delicate control of the dynamics of the interaction port with further involvement of human commands, suitable for objects requiring smoother dynamics like preventing a cup of coffee from sloshing. The proposed control strategy and framework has been implemented in a 3D-scanned, 3D-printed wearable robot with elbow actuation, to demonstrate the advantages and new features through experimental results.
4.
Youcan Yan, Yajing Shen, Chaoyang Song, Jia Pan
Tactile Super-Resolution Model for Soft Magnetic Skin Conference
IEEE International Conference on Robotics and Automation (ICRA2022), Philadelphia (PA), USA, 2022, (Dual-track Submission with RAL: https://doi.org/10.1109/LRA.2022.3141449).
Abstract | Links | BibTeX | Tags: Authorship - Co-Author, Conf - ICRA, Special - Dual-Track
@conference{Yan2022TactileSuper-ICRA,
title = {Tactile Super-Resolution Model for Soft Magnetic Skin},
author = {Youcan Yan and Yajing Shen and Chaoyang Song and Jia Pan},
doi = {10.1109/LRA.2022.3141449},
year = {2022},
date = {2022-01-10},
urldate = {2022-01-10},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA2022)},
issue = {2},
address = {Philadelphia (PA), USA},
abstract = {Tactile sensors of high spatial resolution can provide rich contact information in terms of accurate contact location and force magnitude for robots. However, achieving a high spatial resolution normally requires a high density of tactile sensing cells (or taxels), which will inevitably lead to crowded wire connections, more data acquisition time and probably crosstalk between taxels. An alternative approach to improve the spatial resolution without introducing a high density of taxels is employing super-resolution technology. Here, we propose a novel tactile super-resolution method based on a sinusoidally magnetized soft magnetic skin, by which we have achieved a 15-fold improvement of localization accuracy (from 6 mm to 0.4 mm) as well as the ability to measure the force magnitude. Different from the existing super-resolution methods that rely on overlapping signals of neighbouring taxels, our model only relies on the local information from a single 3-axis taxel and thereby can detect multipoint contact applied on neighboring taxels and work properly even when some of the neighbouring taxels near the contact position are damaged (or unavailable). With this property, our method would be robust to damage and could potentially benefit robotic applications that require multipoint contact detection.},
note = {Dual-track Submission with RAL: https://doi.org/10.1109/LRA.2022.3141449},
keywords = {Authorship - Co-Author, Conf - ICRA, Special - Dual-Track},
pubstate = {published},
tppubtype = {conference}
}
Tactile sensors of high spatial resolution can provide rich contact information in terms of accurate contact location and force magnitude for robots. However, achieving a high spatial resolution normally requires a high density of tactile sensing cells (or taxels), which will inevitably lead to crowded wire connections, more data acquisition time and probably crosstalk between taxels. An alternative approach to improve the spatial resolution without introducing a high density of taxels is employing super-resolution technology. Here, we propose a novel tactile super-resolution method based on a sinusoidally magnetized soft magnetic skin, by which we have achieved a 15-fold improvement of localization accuracy (from 6 mm to 0.4 mm) as well as the ability to measure the force magnitude. Different from the existing super-resolution methods that rely on overlapping signals of neighbouring taxels, our model only relies on the local information from a single 3-axis taxel and thereby can detect multipoint contact applied on neighboring taxels and work properly even when some of the neighbouring taxels near the contact position are damaged (or unavailable). With this property, our method would be robust to damage and could potentially benefit robotic applications that require multipoint contact detection.
3.
Xinyao Hu, Chuang Luo, Hao Li, Liyao Jia, Chaoyang Song, Zheng Wang, Xingda Qu
An Ankle Based Soft Active Orthotic Device Powered by Pneumatic Artificial Muscle Conference
IEEE International Conference on Real-time Computing and Robotics (RCAR2019), Irkutsk, Russia, 2019.
Links | BibTeX | Tags: Authorship - Co-Author, Conf - RCAR
@conference{Hu2019AnAnkle,
title = {An Ankle Based Soft Active Orthotic Device Powered by Pneumatic Artificial Muscle},
author = {Xinyao Hu and Chuang Luo and Hao Li and Liyao Jia and Chaoyang Song and Zheng Wang and Xingda Qu},
doi = {10.1109/RCAR47638.2019.9043948},
year = {2019},
date = {2019-08-04},
urldate = {2019-08-04},
booktitle = {IEEE International Conference on Real-time Computing and Robotics (RCAR2019)},
address = {Irkutsk, Russia},
keywords = {Authorship - Co-Author, Conf - RCAR},
pubstate = {published},
tppubtype = {conference}
}
2.
Xiaojiao Chen, Tommy Hu, Chaoyang Song, Zheng Wang
Analytical Solution to Global Dynamic Balance Control of the Acrobot Conference
IEEE International Conference on Real-time Computing and Robotics (RCAR2018), Kandima, Maldives, 2018.
Links | BibTeX | Tags: Authorship - Co-Author, Conf - RCAR
@conference{Chen2018AnalyticalSolution,
title = {Analytical Solution to Global Dynamic Balance Control of the Acrobot},
author = {Xiaojiao Chen and Tommy Hu and Chaoyang Song and Zheng Wang},
doi = {10.1109/rcar.2018.8621827},
year = {2018},
date = {2018-08-01},
urldate = {2018-08-01},
booktitle = {IEEE International Conference on Real-time Computing and Robotics (RCAR2018)},
address = {Kandima, Maldives},
keywords = {Authorship - Co-Author, Conf - RCAR},
pubstate = {published},
tppubtype = {conference}
}
1.
Juan Yi, Zhong Shen, Chaoyang Song, Zheng Wang
A Soft Robotic Glove for Hand Motion Assistance Honorable Mention Conference
IEEE International Conference on Real-time Computing and Robotics (RCAR2016), Angkor Wat, Cambodia, 2016.
Links | BibTeX | Tags: Authorship - Co-Author, Award - Best Conference Paper Finalist, Award - Paper, Conf - RCAR
@conference{Yi2016ASoft,
title = {A Soft Robotic Glove for Hand Motion Assistance},
author = {Juan Yi and Zhong Shen and Chaoyang Song and Zheng Wang},
doi = {10.1109/rcar.2016.7784010},
year = {2016},
date = {2016-06-06},
urldate = {2016-06-06},
booktitle = {IEEE International Conference on Real-time Computing and Robotics (RCAR2016)},
address = {Angkor Wat, Cambodia},
keywords = {Authorship - Co-Author, Award - Best Conference Paper Finalist, Award - Paper, Conf - RCAR},
pubstate = {published},
tppubtype = {conference}
}
Conference Workshops & Extended Abstracts
2.
Tianyu Wu, Xudong Han, Haoran Sun, Zishang Zhang, Bangchao Huang, Chaoyang Song, Fang Wan
2025, (Short Paper accepted to CoRL 2025 Data Workshop on Making Sense of Data in Robotics: Composition, Curation, and Interpretability at Scale & Demo Presentation at CoRL 2025).
@workshop{Wu2025MagiClaw,
title = {MagiClaw: A Dual-Use, Vision-Based Soft Gripper for Bridging the Human Demonstration to Robotic Deployment Gap},
author = {Tianyu Wu and Xudong Han and Haoran Sun and Zishang Zhang and Bangchao Huang and Chaoyang Song and Fang Wan},
doi = {10.48550/arXiv.2509.19169},
year = {2025},
date = {2025-09-28},
urldate = {2025-09-28},
abstract = {The transfer of manipulation skills from human demonstration to robotic execution is often hindered by a "domain gap" in sensing and morphology. This paper introduces MagiClaw, a versatile two-finger end-effector designed to bridge this gap. MagiClaw functions interchangeably as both a handheld tool for intuitive data collection and a robotic end-effector for policy deployment, ensuring hardware consistency and reliability. Each finger incorporates a Soft Polyhedral Network (SPN) with an embedded camera, enabling vision-based estimation of 6-DoF forces and contact deformation. This proprioceptive data is fused with exteroceptive environmental sensing from an integrated iPhone, which provides 6D pose, RGB video, and LiDAR-based depth maps. Through a custom iOS application, MagiClaw streams synchronized, multi-modal data for real-time teleoperation, offline policy learning, and immersive control via mixed-reality interfaces. We demonstrate how this unified system architecture lowers the barrier to collecting high-fidelity, contact-rich datasets and accelerates the development of generalizable manipulation policies. Please refer to the iOS app at https://apps.apple.com/cn/app/magiclaw/id6661033548 for further details.},
note = {Short Paper accepted to CoRL 2025 Data Workshop on Making Sense of Data in Robotics: Composition, Curation, and Interpretability at Scale & Demo Presentation at CoRL 2025},
keywords = {},
pubstate = {published},
tppubtype = {workshop}
}
The transfer of manipulation skills from human demonstration to robotic execution is often hindered by a "domain gap" in sensing and morphology. This paper introduces MagiClaw, a versatile two-finger end-effector designed to bridge this gap. MagiClaw functions interchangeably as both a handheld tool for intuitive data collection and a robotic end-effector for policy deployment, ensuring hardware consistency and reliability. Each finger incorporates a Soft Polyhedral Network (SPN) with an embedded camera, enabling vision-based estimation of 6-DoF forces and contact deformation. This proprioceptive data is fused with exteroceptive environmental sensing from an integrated iPhone, which provides 6D pose, RGB video, and LiDAR-based depth maps. Through a custom iOS application, MagiClaw streams synchronized, multi-modal data for real-time teleoperation, offline policy learning, and immersive control via mixed-reality interfaces. We demonstrate how this unified system architecture lowers the barrier to collecting high-fidelity, contact-rich datasets and accelerates the development of generalizable manipulation policies. Please refer to the iOS app at https://apps.apple.com/cn/app/magiclaw/id6661033548 for further details.
1.
Jianwen Luo, Sicong Liu, Chengyu Lin, Yong Zhou, Zixuan Fan, Zheng Wang, Chaoyang Song, Harry Asada, Chenglong Fu
Mapping Human Muscle Force to Supernumerary Robotics Device for Overhead Task Assistance Workshop
2020, (Extended Abstract accepted to the IEEE/ASME AIM 2020 Workshop on Supernumerary Robotic Devices).
@workshop{Luo2020MappingHuman,
title = {Mapping Human Muscle Force to Supernumerary Robotics Device for Overhead Task Assistance},
author = {Jianwen Luo and Sicong Liu and Chengyu Lin and Yong Zhou and Zixuan Fan and Zheng Wang and Chaoyang Song and Harry Asada and Chenglong Fu},
url = {https://aim2020srd.wixsite.com/aim2020srd},
doi = {10.48550/arXiv.2107.13799},
year = {2020},
date = {2020-11-01},
urldate = {2020-11-01},
note = {Extended Abstract accepted to the IEEE/ASME AIM 2020 Workshop on Supernumerary Robotic Devices},
keywords = {},
pubstate = {published},
tppubtype = {workshop}
}
Doctoral Thesis
Sorry, no publications matched your criteria.