Research On Automatic Straight Line Control Of Tractor Based On Fuzzy Neural Network

Automatic navigation of agricultural vehicle use, can improve the productivity of agricultural machinery, operating precision and reliability; can greatly reduce the labor intensity of the driver to save labor and reduce production costs. Automatic tractor driving is generally in the conditions in the absence of controlled. And installed various sensors in the vehicle to obtain the characteristics of the external operating environment,thus obtaining the path information of the tractor.Then Simulation experienced drivers to make navigation decisions, To control the tractor to the independent movement.Tractor is a typical representative of farm vehicles,and is also an important driving force machinery in agricultural production. It can draw, hang and attach some farm machinery to complete some field work. Therefore,it has far-reaching significance to research tractor automatic navigation control technology.This article is researching an automatic navigation control system of the tractor.Under the existing experimental conditions,I select the Inertial navigation mode. It is used FOTON-4040 tractor as the test platform, developing a corresponding navigation system to complete a certain testing work. specific study are as follows:(1) The tractor navigation system is analyzed in this paper.According to the needs of this subject,selecting vehicle speed sensors, angle sensors, yaw sensors which are used in the tractor navigation system, and writing the corresponding serial procedure to collect the sensors'signal,then standardizing them.(2) The navigation control principle of the tractor is introduced in this paper;then steering control system of the tractor is alse introduced,and the steering system control design based on PID was designed;Next Neural network controller was designed to control tractor's automatic driving. Heading angle and front wheel angleare as the Input signal and Target front wheel angle is as the output signal. The performance of neural network controller was simulated on Matlab / Simulink simulation platform,and the results show that the control method is feasible.(3) The tractor's operating environment is very complexity and uncertainty. Using the feature which fuzzy control does not need to build the Controlled objects'precise mathematical model and neural network can realize online-learning automatically adjust agricultural vehicle control rules, the fuzzy neural controller is designed to control the vehicle automatic running.Based on the fuzzy neural network control, using the genetic algorithm to optimize controller.Finally some software program was developed in VC++ 6.0. Then neural network controller and fuzzy neural network controller optimized by genetic algorithm were tested. At 0.27m/s speeds,the Vehicle drives 50m, and maximum lateral deviation are respectively 12cm and 18cm. At 0.39 m/s speeds,using fuzzy neural network controller optimized by genetic algorithm, the Vehicle drives 50m, and maximum lateral deviation is 16cm.The result show that the designed controller is better control, and high stability, can basically meet the requirements of farm vehicle automatic driving.