基于约束多目标优化的多区域无人机路径规划

Multi-Region UAV Path Planning Based on Cons trained Multi-Objective Optimization

无人机路径规划常以约束多目标优化问题形式建立数学模型,且这些模型几乎只考虑三维建模空间中两点之间的路径规划;然而缺少无人机从起点出发,经过指定的多个作业区域并到达终点的路径规划问题研究。针对上述问题,基于不可行解引导种群进化提出带有地理信息指导搜索策略的约束多目标进化算法(DW-LS)。首先,根据无人机飞行所受环境约束、性能约束和访问多区域任务建立新的约束多目标优化模型;其次,基于不可行解利用机制,设计地理信息指导搜索策略优化不可行解以进一步协助种群搜索最优进化方向;最后,通过仿真对比了DW-LS与具有代表性的三个约束多目标进化算法,实验结果表明DW-LS所得可行解具有更好的收敛性和多样性。

Drone path planning often establis hes mathematical models in the form of constr ained multi-objective optimization problems,and these models almo st only consider path planning between two points in the three-dimensional modeling space;However,there is a lack of research on the path planning problem of unmanned aerial vehicles starting from the starting point,passing thr ough multiple designated work areas,and reac hing the endpoint.To remedy this issue,this paper proposes a constrained multi-objective evolutionary algorithm with a geographic informationguided search strategy based on infeasible solution-guided population evolution(DW-LS).First ly,a new constrained multi-objective UAV path planning problem was built by combining environmental constraints,performance constraints,and multiple region visit tasks of UAV flight.Secondly,by having the utilizati on mechanism of the infeasible solution,a geographic information-guided se arch strategy was designed to optimize the in feasible solution to further assist the population in searching for the optimal evolutionary direction.Finally,DW-LS was compared with three representative constrained multi-objective evolutionary algorithms,and the experimental results show that the performance of our algorithm is superior in terms of convergence and diversity of the obtained feasible solutions.