Research paper
Sequential convex optimization for collision-free motion
A trajectory optimizer combines sequential convex steps with increasing collision penalties and continuous-time collision constraints. Reported evaluations span arm and full-body planning, industrial box picking and a real PR2 whole-body example.
Reviewed 2026-11-05 · Source and bibliography checked
- Original title
- Finding Locally Optimal, Collision-Free Trajectories with Sequential Convex Optimization
- Publisher
- Robotics: Science and Systems Foundation
- Authors
- John Schulman, Jonathan Ho, Alex Lee, Ibrahim Awwal, Henry Bradlow, Pieter Abbeel
- DOI
- 10.15607/rss.2013.ix.031
- Proceedings publication date
- 2013-06-23
- Source language
- en
- Edition
- Robotics: Science and Systems IX
- Access
- Public bibliography and abstract
- Publication status
- Published
Source overview
A trajectory optimizer combines sequential convex steps with increasing collision penalties and continuous-time collision constraints. Reported evaluations span arm and full-body planning, industrial box picking and a real PR2 whole-body example.
Use and review
Local collision avoidance does not establish a globally feasible route. Geometry, sensing, moving-obstacle prediction and real-time execution assumptions need application-specific checks.
Sources & editions
Finding Locally Optimal, Collision-Free Trajectories with Sequential Convex Optimization ↗
Robotics: Science and Systems Foundation · Robotics: Science and Systems IX
Owning RSS2013 abstract/title/authors/DOI/BibTeX checked. Locally optimal trajectories are not a global completeness or injury guarantee.