您好,欢迎访问北京市农林科学院 机构知识库!

Research on scheduling path planning of multi-objective unmanned tractor based on reinforcement learning method

文献类型: 会议论文

第一作者: Wang Haichen

作者: Wang Haichen 1 ; Wu Huarui 1 ; Zhang Ning 1 ;

作者机构: 1.National Engineering Research Center for Information Technology in Agriculture|Research Center of Information Technology, Beijing Academy of Agriculture and Forestry Sciences|Key Laboratory of Digital Village Technology, Ministry of Agriculture and Rural Affairs, P.R.China

关键词: Processor scheduling;Decision making;Reinforcement learning;Agricultural machinery;Turning;Scheduling;Path planning

会议名称: IEEE International Conference on Cloud Computing and Intelligent Systems

主办单位:

页码: 426-431

摘要: In order to improve the efficiency of unmanned tractor ridge operation and save land costs, hence, a multi-objective optimization model is established in this paper, with the goal of minimizing the reserved turning distance and turning scheduling time at the headland. The model is solved by the improved reinforcement learning method according to the tractor’s turning action and motion state, and the optimal turning decision-making method that satisfies the multi-objective optimization conditions is obtained by using the TOPSIS method. On this basis, with the shortest global tractor turning time, the ant colony algorithm is used to plan the ridge operation path of the unmanned tractor. According to the experiment, the optimized unmanned tractor can save 17.8% of the turning time and 23.9% of the reserved length of the headland by operating in the shuttle operation mode; the total turning time of planning the global operation path combined with the ant colony algorithm can be saved by 48.22%.

分类号: tp3-53

  • 相关文献

[1]Development and Test of GNSS/IMU-Based Speed Measurement Device for Agricultural Machinery. Shuxia Guo,Junxiong Zhang,Shupeng Hu,Changhai Luo,You Li,Weiqiang Fu. 2019

作者其他论文 更多>>