The optimal design of wheeled robot tracking system

One single neuron PID control algorithm based on Elman neural network identification to optimize design of wheeled robot tracking system is proposed in this paper. The error between predetermined path and actual running track of the system is employed as the input of controller for single neuron PID control algorithm based on Elman neural network identification. Neuronal synaptic weights are constantly adjusted online by controller, which outputs a signal to control the motor drive module to adjust the steering and speed of servo motor. The control algorithm proposed in this paper is combined with the L298N chip for the first time. The function of robot self-steering is realized by differential steering of two servo motor. The algorithm is tested by Matlab software, the results suggests that the control effect in is obviously superior to the traditional PID control schemes.