摘要
针对多飞行器协同目标搜索问题,提出了基于预测控制、探索与开发麻雀搜索算法的智能搜索策略。首先,将任务区域栅格化,利用目标存在概率图和信息确定度图建模任务区域;其次,借鉴模型预测控制的思想预测未来一段时间内飞行器协同搜索的航迹,利用目标存在概率和信息确定度对预测航迹进行量化,将多飞行器在线决策问题建模为优化问题;最后,利用麻雀搜索算法求解得到搜索决策;针对传统的麻雀搜索算法在处理复杂优化问题时全局最优性和收敛速度方面的缺陷,引入Tent混沌映射和精英反向传播策略,丰富初始种群多样性。利用黄金正弦策略更新生产者麻雀的位置,提高算法跳出局部极值的能力。结合强化学习探索与开发的思想更新追随者麻雀位置,利用余弦策略和贪心算法优化警戒麻雀数量并更新子代种群,加快算法收敛速度。通过仿真分析,验证了本文提出的算法可以提升协同搜索效率。
An intelligent search strategy based on predictive control and an exploration and exploitation sparrow search algorithm(EESSA)is proposed to address the cooperative search problem of multiple flight vehicles.First,the task area is gridded,and the target existence probability map and information certainty map are used to model the task area.Then,the idea of model predictive control(MPC)is adopted to predict the future flight paths of the vehicles for cooperative search over a certain period.The probability of target existence and the certainty of information are used to quantify the predicted flight paths,and the online decision-making problem of the multiple flight vehicles is modeled as an optimization problem.Finally,the sparrow search algorithm(SSA)is used to obtain the intelligent search decisions.To address the shortcomings of SSA in terms of global optimality and convergence speed when dealing with complex optimization problems,the Tent chaotic mapping and elite back propagation learning strategy are introduced to enhance the diversity of the initial population.The golden sine strategy is adopted to update the positions of the producer sparrows and improve the algorithm’s ability in escaping from local extrema.The positions of scroungers are updated by integrating the concept of exploration and exploitation.Additionally,the cosine strategy and greedy algorithm are utilized to optimize the number of scouter sparrows and update the offspring population,accelerating the convergence speed of the algorithm.Simulation analysis verifies that the proposed algorithm effectively improves cooperative search efficiency.
作者
陈韬
胥彪
李爽
宋勋
CHEN Tao;XU Biao;LI Shuang;SONG Xun(College of Astronautics,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,Jiangsu,China;Key Laboratory of Space Photoelectric Detection and Perception,Nanjing 211106,Jiangsu,China;Beijing Institute of Electronic System Engineering,Beijing 100854,China)
出处
《上海航天(中英文)》
2025年第3期117-127,共11页
Aerospace Shanghai(Chinese&English)
基金
空间光电探测与感知工业和信息化部重点实验室基金资助项目(NJ2022025-05)。
关键词
多飞行器
协同搜索
模型预测控制(MPC)
探索与开发
multiple flight vehicles
cooperative search
model predictive control(MPC)
exploration and exploitation
作者简介
陈韬(2000-),男,硕士生,主要研究方向为多无人机任务规划等;通信作者:胥彪(1986-),男,副研究员,博士,主要研究方向为飞行器精确制导、飞行器控制和非线性控制理论等。
