基于多无人机协同的巡检路径规划研究  

Study on the Inspection Path Planning Based on Multi-UAV Cooperation

作　　者：刘修康 李龙[1] LIU Xiu-kang;LI Long(Fudan University,Shanghai 200000,China)

机构地区：[1]复旦大学,上海200000

出　　处：《航空计算技术》2025年第2期65-70,共6页Aeronautical Computing Technique

基　　金：复旦大学引进人才项目资助(JIH2126062Y)。

摘　　要：工业生产、港口物流等行业需要定期对仓库内的储备货物进行检查,以确保库存及进出数据的准确性。传统基于单无人机的仓库巡检和货物盘点面临耗时长、效率低等问题,不利于仓储数据的实时动态跟踪。为了更好地服务实际需求,亟需开展多无人机协同巡检的路径规划研究。在保证无人机不发生碰撞前提下,充分考虑了检查仓库内货物的完成率需求和能耗成本需求,提出了针对射频识别点的多无人机协同巡检路径规划模型。此外,针对传统灰狼算法收敛性不足的问题,提出一种基于自适应变异机制策略的改进型灰狼算法。并对某仓库典型环境建模,验证了算法在多机协同巡检路径规划中的有效性,并且AMGWO的收敛性相较另一种基于非线性控制因子策略和随机权重策略改进的灰狼算法(IGWO)提高了40.1%。Industrial production,ports,and other industries require regular inspections of reserve goods in warehouses to ensure the accuracy of inventory and transaction data.Traditional warehouse inspections and inventory management using a single UAV face challenges such as lengthy duration and low efficiency,which are not conducive to real-time dynamic tracking of storage data.To better address these practical issues,studies on path planning for multi-UAV cooperative inspection are urgently needed.This paper proposes a multi-UAV cooperative inspection path planning model for radio frequency identification points(RFID)by considering the requirements for UAV inspection completion rates and UAV energy consumption costs within the warehouse while ensuring that the UAVs do not collide with obstacles.Additionally,to address the issue of insufficient convergence in the Grey Wolf Optimizer(GWO),an improved algorithm based on an adaptive mutation mechanism strategy,referred to as the Adaptive Mutation Grey Wolf Optimizer(AMGWO)is proposed.A typical enterprise warehouse environment is modeled to verify the effectiveness of the algorithm in multi-UAV cooperative inspection path planning and the convergence of the AMGWO is enhanced by 40.1%compared to the IGWO.

关 键 词：无人机 路径规划 仓库巡检 多机协同 智能优化算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]