Force-Driven Hook Disassembly via Constraint-Aware Object-Centric Reinforcement Learning
Force-Driven Hook Disassembly via Constraint-Aware Object-Centric Reinforcement Learning
This research proposes a reinforcement learning framework that uses contact forces to infer feasible disassembly directions for hook-type connections. The method is validated on real-world tasks, achieving success rates of 100%, 100%, 90%, and 75% with improved path efficiency, and significantly raises the success rate on an air-conditioning cover task compared to a baseline policy.
This paper introduces a constraint-aware reinforcement learning framework for force-driven disassembly tasks, demonstrating improved success rates and path efficiency in real-world scenarios compared to traditional methods.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
Reported success rate
100%
View original evidence
elief that provides the policy with temporally integrated directional information and reduces redundant exploratory motion. The approach is validated on four real-world tasks, from peg-in-hole pairs to the inner propeller and front cover of an end-of-life air-conditioning unit. The full method achieves success rates of 100%, 100%, 90%, and 75% with the highest path efficiency on every task, and raises the success rat
Open source S4task variety
4
View original evidence
elief that provides the policy with temporally integrated directional information and reduces redundant exploratory motion. The approach is validated on four real-world tasks, from peg-in-hole pairs to the inner propeller and front cover of an end-of-life air-conditioning unit. The full method achieves success rates of 100%, 100%, 90%, and 75% with the highest path efficiency on every task, and raises the success rat
Open source S4
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
What remains unknown
Not established in the collected evidence: Environment, Control, Data origin.
Reported performance applies to the described task. It does not establish general autonomy or deployment readiness.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
that infers feasible extraction directions online from contact forces. The policy is trained in a robot-agnostic, object-centric simulation on a simplified peg-in-hole pair and deployed zero-shot on a physical robot, where it transfers to real connections with unseen, multi-stage constraint structures. Because instantaneous contact forces are noisy and locally ambiguous, the policy is coupled with a particle filter b
Open source S3
elief that provides the policy with temporally integrated directional information and reduces redundant exploratory motion. The approach is validated on four real-world tasks, from peg-in-hole pairs to the inner propeller and front cover of an end-of-life air-conditioning unit. The full method achieves success rates of 100%, 100%, 90%, and 75% with the highest path efficiency on every task, and raises the success rat
Open source S4
e on the air-conditioning cover from 45% to 75% relative to a policy trained without the belief. These results indicate that force-driven reinforcement learning can enable disassembly without product-specific models or demonstrations.
Open source S5
Source：arXiv Robotics — research abstracts · arxiv.org