Experimental assessment of deep reinforcement learning for robot obstacle avoidance: a LPV control perspective

Sie sind hier: Homepage > Suche

Experimental assessment of deep reinforcement learning for robot obstacle avoidance: a LPV control perspective

Freier Zugriff

Incremona, Gian Paolo / Sacchi, Nikolas / Sangiovanni, Bianca / Ferrara, Antonella

This work presents the experimental assessment of a hybrid control scheme based on Deep Reinforcement. Learning (DRL) for obstacle avoidance in robot manipulators. More precisely, relying on an equivalent Linear Parameter Varying (LPV) state-space representation of the system, two operative modes, one based on both joint positions and velocities, one only based on velocity inputs, are activated depending on the measurement of the distance between the robot and the obstacle. Therefore, when the obstacle is close to the robot, a switching mechanism is introduced to enable the DRL algorithm instead of the basic motion planner, thus giving rise to a self-configuring architecture to cope with objects randomly moving in the workspace. The experimental tests of the DRL based collision avoidance hybrid strategy are carried out 011 a physical EPSON VT6 robot manipulator with satisfactory results.

Zugriff

Download

Exportieren, teilen und zitieren

Dokumentinformationen

Titel :

Experimental assessment of deep reinforcement learning for robot obstacle avoidance: a LPV control perspective

Beteiligte:

Incremona, Gian Paolo (Autor:in) / Sacchi, Nikolas (Autor:in) / Sangiovanni, Bianca (Autor:in) / Ferrara, Antonella (Autor:in) / Incremona, Gian Paolo / Sacchi, Nikola / Sangiovanni, Bianca / Ferrara, Antonella

Erscheinungsdatum :

2021-01-01

DOI :

https://doi.org/10.1016/j.ifacol.2021.08.586

Medientyp :

Aufsatz (Konferenz)

Format :

Elektronische Ressource

Sprache :

Englisch

Schlagwörter :

Collision avoidance , Deep reinforcement learning , LPV systems , Robot control