Reproduce “Motion Planner Augmented Reinforcement Learning for Robot Manipulation in Obstructed Environments” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jun Yamada, Youngwoon Lee, Gautam Salhotra, Karl Pertsch, Max Pflueger, Gaurav S. Sukhatme, Joseph J. Lim, Peter Englert
MoPA-RL combines model-free reinforcement learning with motion planning for simulated contact-rich manipulation. Its policy proposes joint-space moves: large displacements invoke a collision-free planner, while small ones are executed directly, preserving fine control. MoPA-SAC learned all four obstructed tasks and showed earlier-rising success curves than alternatives, but evidence is limited to simulation, not physical-robot deployment.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.