Robot motor skills can be acquired by deep reinforcement learning as neural networks to refect state–action mapping. The selection of states has been demonstrated to be crucial for successful robot motor learning. However, because of the complexity of neural networks, human insights and engineering eforts are often required to select appropriate states through qualitative approaches, such as ablation studies, without a quantitative analysis of the state importance. Here we present a systematic saliency analysis that quantitatively evaluates the relative importance of diferent feedback states for motor skills learned through deep reinforcement learning. Our approach provides a guideline to identify the most essential feedback states for robot motor learning. By using only the important states including joint positions, gravity vector and base linear and angular velocities, we demonstrate that a simulated quadruped robot can learn various robust locomotion skills. We fnd that locomotion skills learned only with important states can achieve task performance comparable to the performance of those with more states. This work provides quantitative insights into the impacts of state observations on specifc types of motor skills, enabling the learning of a wide range of motor skills with minimal sensing dependencies.
Identifying important sensory feedback for learning locomotion skills
2023-08-01
Nature Machine Intelligence , 5 (8) pp. 919-932. (2023)
Article (Journal)
Electronic Resource
English
DDC: | 629 |
Multi-sensory tactile feedback control for robot physical interaction and biped locomotion
TIBKAT | 2021
|Accessibility-Based Clustering for Efficient Learning of Locomotion Skills
BASE | 2022
|Locomotion control of a biomimetic robotic fish based on closed loop sensory feedback CPG model
Taylor & Francis Verlag | 2021
|Sensory-Motor Coordination in Crustacean Limbs during Locomotion
British Library Conference Proceedings | 1993
|