D3PG: Decomposed Deep Deterministic Policy Gradient for Continuous Control

Dong, Yinzhao; Yu, Chao; Ge, Hongwei

doi:10.1007/978-3-030-64096-5_4

Cited by 1 publication

(1 citation statement)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Decaying the exploration rate and coordinatingly shrinking the action space can mitigate these issues. To coordinate the agents in the system to jointly explore the environment, Dong et al (2020b) uses the state information of other agents to enhance the state information of the individual learning agent or Dong et al (2020a) decomposes the global critic into the weighted sum of multi-task local critics. We use the decentralized reinforcement learning method to learn the robot motion control policy.…”

Section: Decentralized Reinforcement Learningmentioning

confidence: 99%

Decentralized Policy Learning for Quadruped Robots Multi-Gait Control

Zhang

2023

Preprint

View full text Add to dashboard Cite

This paper proposes the Decentralized Multi-Critic Actor Learning (DMultiCriticAL) method for quadruped robots to overcome the multi-task gradient conflict in multi-gait learning. In this method, each leg joint angle of quadruped robots is served as the action space of decentralized learning agents rather than learning directly to control all robot joint angles. Because of the high similarity in quadruped robot local action space, the DMultiCriticAL method can avoid the multi-task gradient conflict in multi-gait learning. Compared with the conventional multi-task learning method, we exploit the decentralized learning systems to limit the useless exploration of environments by agent, not design complex reward functions to track the multi-gait reference motion trajectory. This also means we can avoid the suboptimal reference motion trajectory design. Our experiment results show that the multi-gait motion controller based on the DMultiCriticAL method can yield more stable locomotion without complex reward function designs.

show abstract

Section: Decentralized Reinforcement Learningmentioning

confidence: 99%

Decentralized Policy Learning for Quadruped Robots Multi-Gait Control

Zhang

2023

Preprint

View full text Add to dashboard Cite

show abstract

D3PG: Decomposed Deep Deterministic Policy Gradient for Continuous Control

Cited by 1 publication

References 12 publications

Decentralized Policy Learning for Quadruped Robots Multi-Gait Control

Decentralized Policy Learning for Quadruped Robots Multi-Gait Control

Contact Info

Product

Resources

About