引用本文:熊志刚,刘忠,罗亚松.未知环境下基于虚拟角速度跟踪的避障策略[J].控制理论与应用,2023,40(6):1122~1128.[点击复制]
XIONG Zhi-gang,LiIU zhong,LUO Ya-song.Obstacle avoidance strategy in unknown environments with virtual angular rate tracking[J].Control Theory and Technology,2023,40(6):1122~1128.[点击复制]
未知环境下基于虚拟角速度跟踪的避障策略
Obstacle avoidance strategy in unknown environments with virtual angular rate tracking
摘要点击 1420  全文点击 414  投稿时间:2021-12-26  修订日期:2023-06-14
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2022.11273
  2023,40(6):1122-1128
中文关键词  智能体  未知环境  避障  执行器饱和  李雅普诺夫方法
英文关键词  intelligent agents  unknown environments  obstacle avoidance  actuator saturation  Lyapunov method
基金项目  国家自然科学基金项目(51679247), 湖北省自然科学基金项目(2017CFB377, 2018CFC865)
作者单位E-mail
熊志刚 海军工程大学 xiongzgzm2@163.com 
刘忠 海军工程大学  
罗亚松* 海军工程大学 yours-baggio@sina.com 
中文摘要
      当智能体自主执行任务时, 局部障碍物可测的未知环境增加了局部极值和执行器饱和发生的概率. 对此, 本文提出了虚拟角速度跟踪的避障策略. 首先, 基于简易障碍物的几何模型构造虚拟的避障引导角, 并利用李雅普 诺夫方法设计角速度控制律, 通过受限制的虚拟角速度跟踪来实现避障控制. 然后, 引入方位因子改进距离型权值 分配器, 强化轨迹附近障碍物的影响以降低局部极值发生的概率. 最后, 对于不完全可测的复杂障碍物, 根据历史探 测信息建立以边界点为中心的简易障碍物模型. 仿真结果表明, 该策略能够避让低速动态障碍物及U型复杂障碍 物, 并且可实现抗饱和控制.
英文摘要
      Unknown environments in which obstacles can be measured partially increases the probability of local extremum and actuator saturation for intelligent agents performing tasks autonomously. Therefore, an obstacle avoidance strategy with virtual angular velocity tracking is proposed in this paper. First, the virtual heading is constructed for obstacle avoidance based on the geometric model of simple obstacles, and the Lyapunov method is applied to design the control law of angular rate, with which obstacle avoidance is realized by tracking the bounded virtual angular rate. Then, the azimuth factor is introduced to improve the distance weight distributor to reduce the probability of local extremum via enhancing the influence of obstacles near the trajectory. For avoiding complex obstacles that are not completely measurable, a simple obstacle model centered on the boundary point is established based on the historical detected information. Simulations show that the proposed strategy can help avoid low-speed dynamic obstacles and U-shaped complex obstacles and realize anti-saturation control.