Cooperative Mobile Manipulation using Learned Dynamical Systems
Organizations: Faculty of Electrical Engineering, Mathematics & Computer Science, TU Delft, The Netherlands
Abstract
Cooperative mobile manipulation requires robots to coordinate their motions while ensuring whole-body collision avoidance and preserving task-relevant behaviors. We propose CALDS (Cooperative Adaptation of Learned Dynamical Systems), a framework that enables mobile manipulators to adapt learned single-agent behaviors for coordinated execution. Building on individual dynamical systems (DSs) learned from demonstrations, CALDS enables cooperative adaptation through two key components: Cooperative Policy Selection, which identifies compatible combinations of learned policies, and Temporal Adaptation, which ensures collision-free cooperation while preserving the spatial structure and convergence properties of the original behaviors. We evaluate CALDS across three diverse scenarios: assistive dressing, restaurant cleaning and serving, and drawer opening and object placement. Quantitative results demonstrate its effectiveness in improving task success and enabling safe, coordinated simultaneous execution while maintaining fidelity to demonstrated behaviors.
Figures & tables
| Training/Fitting time [s] | Planning time [s] | Query time [ms] | |
|---|---|---|---|
| CALDS-CDSM (Ours) | 00 1.3 00 0.5 | 0.76 0.64 | 0.18 0.06 |
| CALDS-PUMA (Ours) | 652.0 234.2 | 1.49 1.73 | 0.19 0.21 |
| CBF-CDSM | 00 1.3 00 0.5 | 0.45 0.10 | 0.43 0.14 |
| CBF-PUMA | 652.0 234.2 | 0.67 0.17 | 0.45 0.23 |
| ONCol-DMP | 000000 - | 0.80 0.39 | 2.98 0.04 |
Appendix figures & tables2 assets
Supplementary material from the paper’s appendix.
Appendix
| Parameter | Value | Parameter | Value |
| Randomized start a | |||
| Base-position perturbation, [ ] | Min. robot–robot distance [ ] | 0.1 | |
| Base heading perturbation, [ ] | Min. distance to static objects [ ] | 1.0 | |
| Arm joint perturbation, [ ] | Runs per scenario | 200 | |
| Simulation | |||
| Time step [ ] | 0.1 | Horizon | 300 (30 ) |
Robotic Manipulation
Robotic manipulation studies how robots perceive, plan, and control physical interactions to grasp, move, assemble, and use objects.
See all 1,831 papers