BriefGPT.xyz
Sep, 2022
机器人控制的DRL过程中的部分可观测性
Partial Observability during DRL for Robot Control
HTML
PDF
Lingheng Meng, Rob Gorbet, Dana Kulić
TL;DR
本文研究了Deep Reinforcement Learning在机器人控制任务中的应用，特别是在部分可观性条件下，比较了TD3、SAC和PPO算法的表现，并提出了改进部分可观性下TD3和SAC算法鲁棒性的多步版本算法。
Abstract
deep reinforcement learning
(DRL) has made tremendous advances in both simulated and real-world
robot control
tasks in recent years. Nevertheless, applying DRL to novel
→