双延迟深度确定性策略的AUV路径规划控制  

AUV Path Planning Control with Double Delay Depth Deterministic Strategy

作　　者：吕金锐[1] 付燕[2] LYU Jinrui;FU Yan(Taiyuan City Vocational College,Taiyuan 030027,China;Xi’an University of Science and Technology,Xi’an 710054,China)

机构地区：[1]太原城市职业技术学院,太原030027 [2]西安科技大学,西安710054

出　　处：《火力与指挥控制》2024年第10期49-60,共12页Fire Control & Command Control

基　　金：山西省重点研发计划基金(201903D121171);来晋优秀博士科研基金资助项目(2022LJ042)。

摘　　要：为了在复杂未知环境中提升水下机器人的实时控制精度,提出一种基于双延迟深度确定性策略的水下机器人路径规划控制方法。在没有环境先验知识的情况下,通过在线测量使用双延迟深度确定性策略算法计算水下无人机的连续控制输入。考虑洋流随机高斯模型,引入一种综合奖励函数,并且提出一种基于阶跃强化学习的六自由度动力学水下无人机自适应运动规划和避障技术。仿真结果表明提出方法能够有效实现位置环境下的高精度实时控制,并且具有一定的泛化能力。In order to improve the real-time control accuracy of underwater robots in complex unknown environments,a path planning control method for underwater robots based on double delay depth deterministic strategy is proposed.Firstly,without prior knowledge of the environment,the continuous control input of the underwater unmanned aerial vehicle is calculated with a dual delay depth deterministic strategy algorithm through online measurement.Then the stochastic Gaussian model of ocean current is considered,a comprehensive reward function is introduced,and an adaptive motion planning and obstacle avoidance technology for six degrees of freedom dynamic underwater UAV based on step reinforcement learning is proposed.The final simulation results show that the proposed method can effectively achieve high-precision real-time control in a positional environment and has a certain generalization ability.

关 键 词：水下机器人 路径规划 实时控制 阶跃强化学习 

分 类 号：V249.1[航空宇航科学与技术—飞行器设计] TP183[自动化与计算机技术—控制理论与控制工程]