论文部分内容阅读
针对不同类型威胁体存在的战场环境中无人车辆战术机动路径规划问题,提出了一种基于威胁代价地图的粒子群优化(Particle Swarm Optimization,PSO)方法。借助极坐标系中关键点的极角进行路径描述,并使用分段3次Hermite插值方法形成光滑路径,将路径规划问题转化为关键点极角的参数优化问题。针对基本PSO(BPSO)算法存在的早熟收敛和后期迭代效率低的缺陷,借鉴以群集方式生活的物种按照不同任务对种群进行分工的机制,提出了一种基于多任务子群协同的改进粒子群优化(Particle Swarm Optimization based on the Multi-tasking Subpopu-lation Cooperation,PSO-MSC)算法。借助该算法的快速收敛和全局寻优特性实现了最优路径规划。实验结果表明:该算法可以快速有效地实现战场环境下无人车辆的战术机动路径规划,且规划路径安全、平滑。
In order to solve the problem of maneuvering path planning for unmanned vehicle in battlefield environment with different threat types, a Particle Swarm Optimization (PSO) method based on threat cost map is proposed. The path description is made by means of the polar angle of the polar point in the polar coordinate system, and the parameter optimization problem of the path planning problem is transformed into the polar angle of the key point by using the cubic Hermitian interpolation method to form the smooth path. Aiming at the defects of premature convergence and late inefficiency of basic PSO (BPSO) algorithm, this paper proposes a new improved particle swarm based on multitasking subgroup coordination by referring to the mechanism of clustering live species according to different tasks. Particle Swarm Optimization based on Multi-tasking Subpopu-cooperation (PSO-MSC) algorithm. With the help of the fast convergence and global optimization of the algorithm, the optimal path planning is realized. The experimental results show that this algorithm can quickly and effectively realize the tactical maneuvering path planning of unmanned vehicles under the battlefield environment, and the proposed path is safe and smooth.