Base On Neural Network Research Control System Of Vehicle Four-Wheel Steering

Apart from front-wheel steering, rear-wheel steering steering also attend, it is called Four-wheel steering. Because of four-wheel steering vehicle has the flexibility at low speed and can reduce the vehicle turning radius obviously; at high speed with good lateral stability. Four-wheel steering can improve initiative vehicle safety and handling stability, this paper characteristics are studied.At present, it has been set up the model of two degrees of freedom, linear and nonlinear three degrees of freedom. This paper analyzed and compared to determine the three degrees of freedom nonlinear as a main model study. Because of this model has strong non-linear factors,it is very difficult to simulation visually, the paper design a BP neural network with feedback loops to instead of the mathematical model. The neural network was trained by MATLAB and converted it into multi-input and multi-output ports, it has the form of kinetic behavior of SIMULINK. For the neural network model and design a PID neural network controller. Through training the PID neural network, the controller model has been set up. Compare with PID neural network control and four-wheel steering proportional control, feedback control and the front wheels steering,after input parameters and simulation , the results show that the neural network control effects better than the other three control methods .In order to verify the design right or not, design the handware which connect with PC of the systerm test car(including schematic and PCB circuit board). selected as the CC1110 RF chip, MAX232 and the PC to the CC1110 Electrical Level Translators, TPS7333 as a voltage converter, a CC1110 RF circuit design and software programming of the preliminary work.