小型无人机气动参数辨识的新型HGAPSO算法

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。

HGAPSO: A new aerodynamic parameters identification algorithm for small unmanned aerial vehicles

In the development of small unmanned aerial vehicles (UAVs),aerodynamic parameter identification is needed for stability and control analysis and flight control law assessment.An improved hybrid genetic and particle swarm optimization algorithm (HGAPSO) is proposed for aerodynamic parameter identification.In this algorithm,the particle swarm optimization algorithm (PSO) is used as the main body,and the cross-over and mutation operation of genetic algorithm (GA) is included into the optimization of particle path to enhance the ability to jump out of the local optimal path.Kent mapping is also used to improve the initial distribution of the particle population,and to make the distribution more uniform and then the optimization more global.Based on the simulation results,the HGAPSO,PSO and GA algorithms are compared in terms of accuracy of identified aerodynamic parameters and success rate of identification.Monte Carlo simulations are further conducted to evaluate the effect of random noises in the measured signals.The results show that HGAPSO can provide both high efficiency and globalization in optimization,and has good resistance against measured noises.Flight testing data acquired from a small UAV are used to comprehensively evaluate the HGAPSO algorithm,and the HGAPSO shows satisfactory ability to identify aerodynamic parameters based on the flight data.