基于BP神经网络的四旋翼双闭环PID轨迹跟踪控制

Double Closed Loop PID Trajectory Tracking Control of Quadrotor Based on BP Neural Network

首先建立了四旋翼系统数学仿真模型,然后对飞行时轨迹跟踪和姿态稳定等控制问题进了分析,设计了基于BP神经网络优化的双闭环PID轨迹跟踪控制器。其中外环为空间位置环,输入为实际空间位置与期望轨迹路径差值,输出为姿态控制角;转换为角速度后作为内环姿态环输入,进行四旋翼姿态稳定控制。为增加系统鲁棒性,获得更优系统性能,内环PID参数采用BP神经网络优化调整得到。最后通过MATLAB平台仿真实验,可以看出基于BP神经网络优化的双闭环PID控制具有良好的轨迹跟踪性能。

This paper sets up the mathematical simulation model of the Quadrotor system,then analyzes the control prob- lems such as trajectory tracking and attitude stabilization,and designs a dual closed loop PID trajectory tracking controller based on BP neural network optimization.The outer loop is the space position loop,and the input is the difference between the actual space position and the desired path, and the output is the attitude control angle.converting control angle to the angular velocity, as the input of the inner loop attitude loop,for the attitude stability control of the Quadrotor.