Matschek, Janine ; Bethge, Johanna ; Findeisen, Rolf (2023)
Safe Machine-Learning-supported Model Predictive Force and Motion Control in Robotics.
doi: 10.48550/arXiv.2303.04569
Report, Bibliographie
 | Es ist eine neuere Version dieses Eintrags verfügbar. |
Kurzbeschreibung (Abstract)
Many robotic tasks, such as human-robot interactions or the handling of fragile objects, require tight control and limitation of appearing forces and moments alongside sensible motion control to achieve safe yet high-performance operation. We propose a learning-supported model predictive force and motion control scheme that provides stochastic safety guarantees while adapting to changing situations. Gaussian processes are used to learn the uncertain relations that map the robot's states to the forces and moments. The model predictive controller uses these Gaussian process models to achieve precise motion and force control under stochastic constraint satisfaction. As the uncertainty only occurs in the static model parts -- the output equations -- a computationally efficient stochastic MPC formulation is used. Analysis of recursive feasibility of the optimal control problem and convergence of the closed loop system for the static uncertainty case are given. Chance constraint formulation and back-offs are constructed based on the variance of the Gaussian process to guarantee safe operation. The approach is illustrated on a lightweight robot in simulations and experiments.
| Typ des Eintrags: | Report |
|---|---|
| Erschienen: | 2023 |
| Autor(en): | Matschek, Janine ; Bethge, Johanna ; Findeisen, Rolf |
| Art des Eintrags: | Bibliographie |
| Titel: | Safe Machine-Learning-supported Model Predictive Force and Motion Control in Robotics |
| Sprache: | Englisch |
| Publikationsjahr: | 8 März 2023 |
| Verlag: | arXiv |
| Reihe: | Robotics |
| Kollation: | 13 Seiten |
| DOI: | 10.48550/arXiv.2303.04569 |
| URL / URN: | https://arxiv.org/abs/2303.04569 |
| Zugehörige Links: | |
| Kurzbeschreibung (Abstract): | Many robotic tasks, such as human-robot interactions or the handling of fragile objects, require tight control and limitation of appearing forces and moments alongside sensible motion control to achieve safe yet high-performance operation. We propose a learning-supported model predictive force and motion control scheme that provides stochastic safety guarantees while adapting to changing situations. Gaussian processes are used to learn the uncertain relations that map the robot's states to the forces and moments. The model predictive controller uses these Gaussian process models to achieve precise motion and force control under stochastic constraint satisfaction. As the uncertainty only occurs in the static model parts -- the output equations -- a computationally efficient stochastic MPC formulation is used. Analysis of recursive feasibility of the optimal control problem and convergence of the closed loop system for the static uncertainty case are given. Chance constraint formulation and back-offs are constructed based on the variance of the Gaussian process to guarantee safe operation. The approach is illustrated on a lightweight robot in simulations and experiments. |
| Fachbereich(e)/-gebiet(e): | 18 Fachbereich Elektrotechnik und Informationstechnik 18 Fachbereich Elektrotechnik und Informationstechnik > Institut für Automatisierungstechnik und Mechatronik 18 Fachbereich Elektrotechnik und Informationstechnik > Institut für Automatisierungstechnik und Mechatronik > Control and Cyber-Physical Systems (CCPS) |
| Hinterlegungsdatum: | 22 Mär 2023 15:20 |
| Letzte Änderung: | 19 Dez 2024 11:27 |
| PPN: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
Verfügbare Versionen dieses Eintrags
- Safe Machine-Learning-supported Model Predictive Force and Motion Control in Robotics. (deposited 22 Mär 2023 15:20) [Gegenwärtig angezeigt]
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum