预警机制的鲸鱼算法及时间约束的航迹规划

Whale Optimization Algorithm Based on Early Warning Mechanism and Path Planning with Time Constraint

摘要:

针对高实时性要求的固定翼无人机航迹规划问题，在模型中引入算法的运行时间以契合实际工程要求，并提出基于预警机制的改进鲸鱼优化算法完成求解；该算法通过对适应度进行排序定义个体的预警概率，并借此控制个体更新机制的选取；随后引入与预警概率关联的权重系数控制螺旋更新机制的收缩扩张，同时使用莱维飞行改进随机游走机制加快收敛，达到平衡各机制开发与探索能力的目的；有效地改善算法收敛速度慢、精度低的缺陷；使用基准函数测试并验证算法的有效性，并在不同维度与距离的航迹规划对比仿真实验中量化改进算法在收敛精度与收敛速度的优越性；仿真实验表明，面对低维度航迹规划时，算法精度可提高8.0%；面对高纬度航迹规划时，算法收敛速度可提高50%；

Abstract:

For the fixed wing UAV path planning with high real-time requirements, time cost is introduced into the optimization model to fit the actual project, and the new improved whale optimization algorithm with early warning mechanism is proposed to complete the optimization solution. The algorithm establishes individual early warning probability by ranking fitness of the population, and uses it to control the selection of update mechanism. By introducing the weight coefficient associated with the early warning probability to control the expansion and contraction of the spiral update mechanism, and using Lévy flight to improve the random walk mechanism to accelerate the convergence, the goal of balancing the development and exploration capabilities of each mechanism is achieved, which is beneficial to alleviate the problems of whale optimization algorithm, such as slow convergence speed and low convergence accuracy. The simulation experiment uses benchmark function to prove the effectiveness of the algorithm, and the simulation of path planning in different dimensions and distances shows the superiority of the improved algorithm. The simulation results show that the algorithm accuracy can be improved by 8.0% when dealing with low dimension path planning and the convergence speed of algorithm can be improved by 50%.