A5052963938- Reinforcement Learning in Robotics
- Optimization and Search Problems
- Robotic Locomotion and Control
- Prosthetics and Rehabilitation Robotics
- Robotic Path Planning Algorithms
Harbin Institute of Technology
2020-2021
This article deals with the problem of mobile robot path planning in an unknown environment that contains both static and dynamic obstacles, utilizing a reinforcement learning approach. We propose improved Dyna- <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${Q}$ </tex-math></inline-formula> algorithm, which incorporates heuristic search strategies, simulated annealing mechanism, reactive navigation...
This paper is concerned with locomotion problems for the quadruped robot in unknown and unstructured terrains, utilizing emerging deep reinforcement learning technique. The state-of-the-art deterministic policy gradient (DDPG) algorithm leveraged to acquire gait by from interactions environment. framework of system presented based on actor-critic architecture, additional domain-specific knowledge exploited shaping reward function enhance efficiency during training. Moreover, a simulation...