中国邮电高校学报(英文) ›› 2022, Vol. 29 ›› Issue (5): 10-20.doi: 10.19682/j.cnki.1005-8885.2022.0025

所属专题: Special Topic on Artificial Intelligence of Things

• Special Topic: Artificial Intelligence of Things • 上一篇    下一篇

Vehicle-following system based on deep reinforcement learning in marine scene

张新,娄皓然,蒋励,肖前浩,蔡著文   

  1. 西安邮电大学
  • 收稿日期:2022-04-19 修回日期:2022-09-06 出版日期:2022-10-31 发布日期:2022-10-28
  • 通讯作者: 娄皓然 E-mail:871025829@qq.com
  • 作者简介:2022-05-31
  • 基金资助:
    《动力轴系的振动参数和能效指标的远程分析与监控系统技术研发》;《扭矩动态测量及准确性验证技术研究》

Vehicle-following system based on deep reinforcement learning in marine scene

Zhang Xin, Lou Haoran, Jiang Li, Xiao Qianhao, Cai Zhuwen   

  • Received:2022-04-19 Revised:2022-09-06 Online:2022-10-31 Published:2022-10-28
  • Contact: Hao-Ran Lou E-mail:871025829@qq.com
  • Supported by:
    Remote analysis and monitoring system technology development of vibration parameters and energy efficiency index of power shafting;Research on Torque Dynamic Measurement and Accuracy Verification Technology

摘要:

In order to solve the problems that the feature data type are not rich enough in the data collection process about the vehicle-following task in marine scene which results in a long model convergence time and high training difficulty, a two-stage vehicle-following system was proposed. Firstly, semantic segmentation model predicts the number of pixels of the followed target, then the number of pixels of the followed target is mapped to the position feature. Secondly, deep reinforcement learning algorithm enables the control equipment to make decision action, to ensure that two moving objects remain within the safe distance. The experimental results show that the two-stage vehicle-following system has a 40% faster convergence rate than the model without position feature, and the following stability is significantly improved by adding the position feature.

关键词: vehicle-following| semantic segmentation| reinforcement learning| double deep Q-network (DDQN)

Abstract:

In order to solve the problems that the feature data type are not rich enough in the data collection process about the vehicle-following task in marine scene which results in a long model convergence time and high training difficulty, a two-stage vehicle-following system was proposed. Firstly, semantic segmentation model predicts the number of pixels of the followed target, then the number of pixels of the followed target is mapped to the position feature. Secondly, deep reinforcement learning algorithm enables the control equipment to make decision action, to ensure that two moving objects remain within the safe distance. The experimental results show that the two-stage vehicle-following system has a 40% faster convergence rate than the model without position feature, and the following stability is significantly improved by adding the position feature.

Key words: vehicle-following| semantic segmentation| reinforcement learning| double deep Q-network (DDQN)