TR2023-121
Constrained Dynamic Movement Primitives for Collision Avoidance in Novel Environments
-
- "Constrained Dynamic Movement Primitives for Collision Avoidance in Novel Environments", IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), September 2023.BibTeX TR2023-121 PDF Video
- @inproceedings{Shaw2023sep,
- author = {Shaw, Seiji and Jha, Devesh K. and Raghunathan, Arvind and Corcodel, Radu and Romeres, Diego and Konidaris, George and Nikovski, Daniel},
- title = {Constrained Dynamic Movement Primitives for Collision Avoidance in Novel Environments},
- booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
- year = 2023,
- month = sep,
- url = {https://www.merl.com/publications/TR2023-121}
- }
,
- "Constrained Dynamic Movement Primitives for Collision Avoidance in Novel Environments", IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), September 2023.
-
MERL Contacts:
-
Research Area:
Abstract:
Dynamic movement primitives are widely used for learning skills that can be demonstrated to a robot by a skilled human or controller. While their generalization capabilities and simple formulation make them very appealing to use, they possess no strong guarantees to satisfy operational safety constraints for a task. We present constrained dynamic movement primitives (CDMPs), which can allow for positional constraint satisfaction in the robot workspace. Our method solves a non-linear optimization to perturb an existing DMP’s forcing weights to admit a Zeroing Barrier Function (ZBF), which certifies positional workspace constraint satisfaction. We demonstrate our approach under different positional constraints on the end-effector movement on multiple physical robots, such as obstacle avoidance and workspace limitations.