具有自适应神经网络的自动驾驶车辆转向控制OA

This paper takes a permanent magnet synchronous motor(PMSM)-driven steer-by-wire(SBW)system as the research object.To address the external perturbation,parameter uncertainties,and unknown nonlinearity,it proposes an adaptive neural network control method combining preset performance and finite time.First,the dynamic effects of the PMSM are incorporated into the SBW system to build a new system dynamics model.Then,a virtual control direction method is employed to address the uncertainties of the control direction coefficient.Next,the unknown nonlinear part of the system is reconstructed by a new neural network and command filter design method to reduce the computational burden.A tracking differentiator is employed to estimate the angular velocity of the front wheels,reducing the number of sensors while effectively attenuating the measurement noise.The combination of finite-time and prescribed performance techniques ensures all system signals converge to the neighborhood of the origin in finite time.Finally,based on the finite-time Lyapunov stabilization criterion,the closed-loop system maintains stability in finite time.Simulation experiments further verify the effectiveness of the proposed algorithm.