基于RBF神经网络的智能车速度控制系统研究

A Study on the Speed Control System of the Intelligent Vehicle Based on RBF Neural Network

针对传统PID控制算法在电磁导航智能车速度偏差处理中存在比例、积分、微分参数一经确定,不能在线调整、不具有自适应能力的缺点,提出了将RBF神经元网络控制器及其算法应用到智能车的调速系统中,对传统PID参数整定进行改进。RBF神经网络能够辨识智能车电机的数学模型,可以根据控制效果在线训练和学习,调整网络连接权重值,最终自适应地整定PID三个参数来实现智能车的速度控制。MATLAB仿真测试表明,与传统PID控制算法相比,RBF神经网络PID整定算法在智能车速度控制中具有响应快,超调量小、鲁棒性和适应性强的优点,大大提高了智能车电机控制系统的性能。

Considering that once proportional, integral and differential parameters are determined in speed deviation processing of the electromagnetically navigated intelligent vehicle according to the traditional PID control algorithm, they are not capable of online adjustment and do not have adaptive capability, this paper presents a scheme to apply the RBF neural cell network controller with its algorithm to the speed regulation system of the intelligent vehicle to improve the traditional PID parameter setting. The RBF neural network can identify the mathematical model of the intelligent car motor, conduct online training and learning according to the control effect, adjust the network connection weight and finally, adaptively adjust the three PID parameters to realize speed control over the intelligent vehicle. MATLAB simulation tests show that, compared with the traditional PID control algorithm, the PID setting algorithm of the RBF neural network has such advantages as quick response, small overshoot, robustness and strong adaptability in the speed control of the intelligent vehicle, thus greatly improving the performance of the intelligent vehicle motor control system.