@ARTICLE{Koptev:RAL:2021,
  author={M. {Koptev} and N. {Figueroa} and A. {Billard}},
  journal={IEEE Robotics and Automation Letters}, 
  title={Real-Time Self-Collision Avoidance in Joint Space for Humanoid Robots}, 
  note={[Selected for presentation in ICRA 2021]},
  year={2021},
  volume={6},
  number={2},
  pages={1240-1247},
  doi={10.1109/LRA.2021.3057024},
  url={https://ieeexplore.ieee.org/document/9345975}, 
  abstract={In this letter, we propose a real-time self-collision avoidance approach for whole-body humanoid robot control. To achieve this, we learn the feasible regions of control in the humanoid's joint space as smooth self-collision boundary functions. Collision-free motions are generated online by treating the learned boundary functions as constraints in a Quadratic Program based Inverse Kinematic solver. As the geometrical complexity of a humanoid robot joint space grows with the number of degrees-of-freedom (DoF), learning computationally efficient and accurate boundary functions is challenging. We address this by partitioning the robot model into multiple lower-dimensional submodels. We compare performance of several state-of-the-art machine learning techniques to learn such boundary functions. Our approach is validated on the 29-DoF iCub humanoid robot, demonstrating highly accurate real-time self-collision avoidance.}}


%% 2020 %%

{"_id":"uoqyrPgNGeQR3dY5X","bibbaseid":"koptev-figueroa-billard-realtimeselfcollisionavoidanceinjointspaceforhumanoidrobots-2021","authorIDs":["6weNhrK95SmkxE6z5"],"author_short":["Koptev, M.","Figueroa, N.","Billard, A."],"bibdata":{"bibtype":"article","type":"article","author":[{"firstnames":["M."],"propositions":[],"lastnames":["Koptev"],"suffixes":[]},{"firstnames":["N."],"propositions":[],"lastnames":["Figueroa"],"suffixes":[]},{"firstnames":["A."],"propositions":[],"lastnames":["Billard"],"suffixes":[]}],"journal":"IEEE Robotics and Automation Letters","title":"Real-Time Self-Collision Avoidance in Joint Space for Humanoid Robots","note":"[Selected for presentation in ICRA 2021]","year":"2021","volume":"6","number":"2","pages":"1240-1247","doi":"10.1109/LRA.2021.3057024","url":"https://ieeexplore.ieee.org/document/9345975","abstract":"In this letter, we propose a real-time self-collision avoidance approach for whole-body humanoid robot control. To achieve this, we learn the feasible regions of control in the humanoid's joint space as smooth self-collision boundary functions. Collision-free motions are generated online by treating the learned boundary functions as constraints in a Quadratic Program based Inverse Kinematic solver. As the geometrical complexity of a humanoid robot joint space grows with the number of degrees-of-freedom (DoF), learning computationally efficient and accurate boundary functions is challenging. We address this by partitioning the robot model into multiple lower-dimensional submodels. We compare performance of several state-of-the-art machine learning techniques to learn such boundary functions. Our approach is validated on the 29-DoF iCub humanoid robot, demonstrating highly accurate real-time self-collision avoidance.","bibtex":"@ARTICLE{Koptev:RAL:2021,\n author={M. {Koptev} and N. {Figueroa} and A. {Billard}},\n journal={IEEE Robotics and Automation Letters}, \n title={Real-Time Self-Collision Avoidance in Joint Space for Humanoid Robots}, \n note={[Selected for presentation in ICRA 2021]},\n year={2021},\n volume={6},\n number={2},\n pages={1240-1247},\n doi={10.1109/LRA.2021.3057024},\n url={https://ieeexplore.ieee.org/document/9345975}, \n abstract={In this letter, we propose a real-time self-collision avoidance approach for whole-body humanoid robot control. To achieve this, we learn the feasible regions of control in the humanoid's joint space as smooth self-collision boundary functions. Collision-free motions are generated online by treating the learned boundary functions as constraints in a Quadratic Program based Inverse Kinematic solver. As the geometrical complexity of a humanoid robot joint space grows with the number of degrees-of-freedom (DoF), learning computationally efficient and accurate boundary functions is challenging. We address this by partitioning the robot model into multiple lower-dimensional submodels. We compare performance of several state-of-the-art machine learning techniques to learn such boundary functions. Our approach is validated on the 29-DoF iCub humanoid robot, demonstrating highly accurate real-time self-collision avoidance.}}\n\n\n%% 2020 %%\n","author_short":["Koptev, M.","Figueroa, N.","Billard, A."],"key":"Koptev:RAL:2021","id":"Koptev:RAL:2021","bibbaseid":"koptev-figueroa-billard-realtimeselfcollisionavoidanceinjointspaceforhumanoidrobots-2021","role":"author","urls":{"Paper":"https://ieeexplore.ieee.org/document/9345975"},"metadata":{"authorlinks":{"figueroa, n":"https://nbfigueroa.github.io/"}},"downloads":10},"bibtype":"article","biburl":"https://www.dropbox.com/s/x2bon6vqlc79ekc/FigueroaPublications.bib?dl=1","creationDate":"2021-02-03T22:06:39.534Z","downloads":10,"keywords":[],"search_terms":["real","time","self","collision","avoidance","joint","space","humanoid","robots","koptev","figueroa","billard"],"title":"Real-Time Self-Collision Avoidance in Joint Space for Humanoid Robots","year":2021,"dataSources":["QscCmn2DGDfW7oiMm","kJNNQTBYdRiDNqEoj","Myci9twJQBkzgwjJt","CaGiEo4vZFJvt7iqr"]}