Closed-Loop Object-Informed Control for Non-Prehensile Robot Manipulation
Closed-Loop Object-Informed Control for Non-Prehensile Robot Manipulation
This research presents a closed-loop object-informed (CLOI) method for non-prehensile robot manipulation, addressing challenges in long-horizon planning with discontinuous contact. By splitting the problem into object-level and robot-level planning solved via model predictive path integral control and coupled through consensus using ADMM, CLOI achieves a 35% success rate increase in simulation and 43% on hardware for planar pushing tasks with obstacles compared to standard MPPI under the same computational budget.
The paper introduces a closed-loop object-informed control method that splits planning into object-level and robot-level subproblems solved via model predictive path integral control, achieving a 35% success rate increase in simulation and 43% on hardware for planar pushing tasks with obstacles compared to standard MPPI.
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.
- Environment
- SimulationOpen source S4
Reported numbers
Reported success rate
35%
View original evidence
s revised toward robot-realizable object trajectories, while the robot plan is aligned with the object poses the task requires. In planar pushing tasks with obstacles using an xArm6 manipulator, CLOI increases the success rate by 35% in simulation and 43% on hardware, compared to standard MPPI given the same computational budget.
Open source S4Reported success rate
43%
View original evidence
s revised toward robot-realizable object trajectories, while the robot plan is aligned with the object poses the task requires. In planar pushing tasks with obstacles using an xArm6 manipulator, CLOI increases the success rate by 35% in simulation and 43% on hardware, compared to standard MPPI given the same computational budget.
Open source S4
- paper: not_reported
What remains unknown
Not established in the collected evidence: 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.
s revised toward robot-realizable object trajectories, while the robot plan is aligned with the object poses the task requires. In planar pushing tasks with obstacles using an xArm6 manipulator, CLOI increases the success rate by 35% in simulation and 43% on hardware, compared to standard MPPI given the same computational budget.
Open source S4
arXiv:2605.30778v3 Announce Type: replace Abstract: Non-prehensile robot manipulation is challenging due to discontinuous, long-horizon interactions between the robot and the objects it manipulates. Sampling-based model predictive control methods are effective with discontinuous contact but face challenges with finding promising trajectories in long-horizon planning. We propose a closed-loop object-informed (CLOI) me
Open source S2
Source：arXiv Robotics — research abstracts · arxiv.org