Abstract
Learning from demonstration (LfD) has emerged as a promising approach enabling robots to acquire complex tasks directly from human demonstrations. However, tasks involving surface interactions on freeform 3D surfaces present unique challenges in modeling and execution, especially when geometric variations exist between demonstrations and robot execution. This paper proposes a novel framework called probabilistic surface interaction primitives (ProSIP), which systematically incorporates the surface path and the local surface features into the learning procedure. An instrumented tool allows seamless recording and execution of human demonstrations. By design, ProSIPs are independent of time, invariant to rigid-body displacements, and apply to any robotic platform with a Cartesian controller. The framework is employed for an edge-cleaning task of bathroom sinks. The generalization capability to various object geometries and significantly distorted objects is demonstrated. Simulations and an experimental setup with a 9-degrees-of-freedom robotic platform confirm the performance.
Originalsprache | Englisch |
---|---|
Titel | Proceedings of the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) |
Seiten | 5956-5963 |
DOIs | |
Publikationsstatus | Veröffentlicht - 25 Dez. 2024 |
Veranstaltung | 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) - Abu Dhabi, Vereinigte Arabische Emirate Dauer: 14 Okt. 2024 → 18 Okt. 2024 |
Konferenz
Konferenz | 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) |
---|---|
Land/Gebiet | Vereinigte Arabische Emirate |
Stadt | Abu Dhabi |
Zeitraum | 14/10/24 → 18/10/24 |
Research Field
- Complex Dynamical Systems
Fingerprint
Untersuchen Sie die Forschungsthemen von „ProSIP: Probabilistic Surface Interaction Primitives for Learning of Robotic Cleaning of Edges“. Zusammen bilden sie einen einzigartigen Fingerprint.Auszeichnungen
-
IROS Best Application Paper Award
Unger, C. (Empfänger/-in), Hartl-Nesic, C. (Empfänger/-in), Vu, M. N. (Empfänger/-in) & Kugi, A. (Empfänger/-in), 17 Okt. 2024
Auszeichnung: Best paper award für Beitrag in einer Fachzeitschrift/ auf einer Konferenz