Intelligent Vehicle Control Based On Fuzzy Neural Network

At present,under the rapid development of driverless technology,research on intelligent vehicle control systems has a very important significance.For the two main characteristics of smart cars,one is the nonlinear steady system,and the other is the situation that it is susceptible to external interference.The main research content of this paper is to establish a control system to achieve the control of the smart car.The main task of this paper is to realize the recognition of the runway by the car and the simulation of the running of the car on straight and curved roads through a combination of hardware and software design.And verify the correctness of the design through simulation.The completed work is as follows:1.In terms of hardware,the selection of each part of the chip was first determined,and a suitable image sensor was selected by comparison.At the same time,the design of the relevant circuit was completed,including four-way motor drive circuit,WiFi transmission circuit,etc.,and the main control was completed.Chip PCB design.2.In terms of software,the image processing on the target runway is first of all,image deformity correction,to achieve the maximum degree of restoration of the runway;secondly,image binarization is performed to obtain local adaptive binarization.The noise effect is better.Finally,the edge detection algorithm of the neural network is compared with the algorithm of the Canny operator,and the result shows that the edge detection of the BP neural network has the best effect.3.This paper builds a model of fuzzy sliding mode controller,which uses a combination of sliding mode controller and fuzzy control to control the car.Under the Matlab simulation,it shows that the fuzzy sliding mode controller has less oscillation and is relatively stable.The actual verification can ensure that the car runs normally along the track line.It is proved that the fuzzy sliding mode controller can effectively control the trolley.