文献类型：期刊文献

中文题名：Magnetic Field-Based Reward Shaping for Goal-Conditioned Reinforcement Learning

作者：Hongyu Ding[1];Yuanze Tang[2];Qing Wu[2];Bo Wang[1,3];Chunlin Chen[1,3];Zhi Wang[1,3]

机构：[1]the Department of Control Science and Intelligence Engineering,School of Management and Engineering,Nanjing University,Nanjing 210093,China;[2]the Department of Power Engineering and Process Machinery,School of Mechanical and Power Engineering,East China University of Science and Technology,Shanghai 200237,China;[3]IEEE

年份：2023

卷号：10

期号：12

起止页码：2233

中文期刊名：IEEE/CAA Journal of Automatica Sinica

外文期刊名：自动化学报（英文版）

收录：CSTPCD;;Scopus;CSCD:【CSCD2023_2024】；

基金：supported in part by the National Natural Science Foundation of China(62006111,62073160);the Natural Science Foundation of Jiangsu Province of China(BK20200330)。

语种：英文

中文关键词：Dynamic environments;goal-conditioned reinforcement learning;magnetic field;reward shaping

摘要：Goal-conditioned reinforcement learning(RL)is an interesting extension of the traditional RL framework,where the dynamic environment and reward sparsity can cause conventional learning algorithms to fail.Reward shaping is a practical approach to improving sample efficiency by embedding human domain knowledge into the learning process.Existing reward shaping methods for goal-conditioned RL are typically built on distance metrics with a linear and isotropic distribution,which may fail to provide sufficient information about the ever-changing environment with high complexity.This paper proposes a novel magnetic field-based reward shaping(MFRS)method for goal-conditioned RL tasks with dynamic target and obstacles.Inspired by the physical properties of magnets,we consider the target and obstacles as permanent magnets and establish the reward function according to the intensity values of the magnetic field generated by these magnets.The nonlinear and anisotropic distribution of the magnetic field intensity can provide more accessible and conducive information about the optimization landscape,thus introducing a more sophisticated magnetic reward compared to the distance-based setting.Further,we transform our magnetic reward to the form of potential-based reward shaping by learning a secondary potential function concurrently to ensure the optimal policy invariance of our method.Experiments results in both simulated and real-world robotic manipulation tasks demonstrate that MFRS outperforms relevant existing methods and effectively improves the sample efficiency of RL algorithms in goal-conditioned tasks with various dynamics of the target and obstacles.