混沌粒子群优化粒子滤波算法

针对基于粒子群优化算法的粒子滤波计算复杂度大,并且容易陷入局部最优,提出了一种新的基于混沌的粒子群优化粒子滤波算法。该算法在粒子群优化的基础上,引入混沌序列,利用混沌运动的遍历性、随机性等特点改善了初始样本的质量,同时利用混沌扰动避免搜索过程陷入局部最优,使算法具有更快的收敛速度和更好的全局搜索能力。最后利用UNGM模型将该算法与标准粒子滤波和粒子群粒子滤波进行仿真对比,并利用纯角度目标跟踪模型验证了算法的有效性。实验结果表明,该算法改善了粒子群优化算法的粒子滤波易陷入局部最优的现象,提高了粒子滤波的精度和速度,具有较高的应用价值。

A Chaos Particle Swarm Optimization Particle Filter Algorithm

Particle Filter based on Particle Swarm Optimization algorithm (PSO-PF) has high calculation complexity and is easily trapped in local optimum.To solve these problems a novel particle filter based on chaos particle swarm was proposed.On the basis of PSO this algorithm introduced chaos sequence by use of the ergodicity and randomness of chaos the quality of the original sample was improved.In the meanwhile chaos perturbation was utilized to avoid the search being trapped in local optimum.Therefore the algorithm had faster convergence speed and better global search capability.UNGM model was used for simulation to compare the algorithm with Particle Filter (PF) and PSO-PF and Bearings-only tracking (BOT)model was used to verify the effectiveness of the algorithm.The simulation results show that this algorithm not only reduces the local optimization but also improves the velocity and precision so it has a high application value.