Reproduce “Keep it Upright: Model Predictive Control for Nonprehensile Object Transportation with Obstacle Avoidance on a Mobile Manipulator” 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
Adam Heins, Angela P. Schoellig
Researchers at the Technical University of Munich and the University of Toronto developed a whole-body model predictive control (MPC) approach for a mobile manipulator that enables nonprehensile object transportation while actively avoiding static and dynamic obstacles. The controller successfully balanced various objects, including stacked configurations, and reacted to a thrown projectile at end effector speeds up to 2.0 m/s and accelerations up to 7.9 m/s^2, utilizing minimum statically-feasible friction coefficients for robustness and computational efficiency.
Key insights
- Planning with minimum statically-feasible friction coefficients enhances robustness against frictional uncertainty and vibrations while significantly reducing MPC computation time by simplifying force parameterization.
- A whole-body MPC can achieve high-speed, reactive nonprehensile manipulation on mobile manipulators by directly optimizing joint-space trajectories based on task-space goals and constraints.
- Softening object dynamics constraints within the MPC formulation provides flexibility, allowing the use of tangential contact forces when required for acceleration while maintaining computational benefits.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.