论文部分内容阅读
针对小型无人机(UAVs)研制中操稳特性和飞行控制律设计评估对气动参数辨识的需求,提出了一种混合遗传粒子群优化算法(HGAPSO)。该算法以粒子群优化算法(PSO)为主体,在粒子优化路径中,引入遗传算法(GA)的交叉变异操作,增强粒子群跳出局部最优的能力;同时采用Kent映射改进粒子种群的初始化,使初始种群在可行解空间内分布更加均匀,增强全局优化能力。基于仿真结果,依据辨识准度及辨识成功率,对比了HGAPSO、常规PSO和GA优化算法气动参数辨识的结果,然后用蒙特卡洛仿真测试随机观测噪声的影响,结果表明该算法兼备PSO算法高的搜索效率和GA算法的全局优化能力,对随机观测噪声不敏感。最后,通过设计小型UAV试飞示例进行综合应用评价,结果表明:HGAPSO算法基于真实试飞数据进行气动参数辨识取得了满意结果,具有良好的实用性。
Aim To meet the need of aerodynamic parameter identification in handling stability design and flight control law design evaluation of small UAVs, a hybrid genetic particle swarm optimization (HGAPSO) algorithm was proposed. Particle Swarm Optimization (PSO) is the main part of this algorithm. Genetic algorithm (GA) crossover mutation is introduced in the particle optimization path to enhance the ability of particle swarm to jump out of local optimum. At the same time, Kent mapping is used to improve the particle population initialization, So that the initial population is more evenly distributed in the feasible solution space and enhances the global optimization ability. Based on the simulation results, the aerodynamic parameter identification results of the HGAPSO, conventional PSO and GA optimization algorithms are compared based on the recognition accuracy and the recognition success rate. The Monte Carlo simulation is used to test the influence of random observation noise. The results show that the proposed algorithm combines PSO algorithm The search efficiency and the global optimization ability of GA algorithm are not sensitive to the random observation noise. Finally, a comprehensive UAV test sample design application evaluation, the results show that: HGAPSO algorithm based on real test data for aerodynamic parameter identification obtained satisfactory results, with good practicability.