Formula-E race strategy development using distributed policy gradient reinforcement learning

作者：

Highlights：

• Targeting the most popular race strategy problem in Formula-E races.

• Propose a novel race strategy approach using distributed reinforcement learning architecture.

• Reward shaping improves the reinforcement learning performance.

• Modification of actor neural network layout for hybrid-type action generation.

摘要

•Targeting the most popular race strategy problem in Formula-E races.•Propose a novel race strategy approach using distributed reinforcement learning architecture.•Reward shaping improves the reinforcement learning performance.•Modification of actor neural network layout for hybrid-type action generation.

论文关键词：Energy management,Formula-E race strategy,Deep deterministic policy gradient,Reinforcement leaning

论文评审过程：Received 21 September 2020, Revised 25 November 2020, Accepted 14 January 2021, Available online 20 January 2021, Version of Record 27 January 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.106781