摘要
针对传统模糊控制隶属度函数主观性较强以及传统PID控制精度低、参数难以调整等缺点,提出一种基于模糊推理参数优化及神经网络的无人机自适应PID控制算法.(1)基于四旋翼无人机的运动学分析,建立了系统的动力学模型;(2)导出了基于神经网络控制的模糊推理参数优化方法;(3)采用带有全局极小动量项的梯度下降法修正网络学习权值,设计了模糊神经网络.实验结果表明,本文所提的算法相比其他传统算法具有更好的控制精度和鲁棒性.
In view of the strong subjectivity of membership function of traditional fuzzy control,the low precision of traditional PID control and the difficulty of parameter adjustment,this paper presented an adaptive PID control algorithm for UAV based on fuzzy inference parameter optimization and neural network.Firstly,the dynamics model of the system was established based on the kinematics analysis of the quadrotor UAV;then,the optimization method of fuzzy inference parameters based on neural network control was derived;finally,the gradient descent method with global minimum momentum term was used to modify the learning weight of the network,and the fuzzy neural network was designed.The results show that the proposed algorithm has better control accuracy and robustness than other traditional algorithms.
作者
张德华
张腾龙
ZHANG Dehua;ZHANG Tenglong(School of Artificial Intelligence/School of Computer and Information Engineering,Henan University,Henan Kaifeng475004,China)
出处
《河南大学学报（自然科学版）》
CAS
2021年第5期554-562,共9页
Journal of Henan University:Natural Science
基金
国家自然科学基金资助项目(61703142)
河南省高等学校重点科研项目(20A120005)
河南大学省级大学生创新创业训练计划项目(201910475084)
作者简介
张德华(1984-),男,河南新乡人,副教授,博士.研究方向:智能控制、自适应动态规划、非线性系统最优控制和神经网络控制等;通讯作者:张腾龙,E-mail:ztlsb666@163.com
