The Lateral Control And Longitude Speed Adaptive Control Of The Small-Scale Autonomous Vehicles

The research achievements of the small-scale autonomous vehicle can be used for driver assistant system to enhance safety and comfort while driving,they can also help conducting experiments of driverless car to develop driving strategy in a more economical and securer way.The bottom control which includes longitude speed adaptive control and lateral turning control for path tracking is mainly discussed in this work.The main studies are as follows:(1)In order to get the fixed parameters of the speed controller,the longitudinal velocity output model is established according to the standard operating parameters of the longitudinal drive motor and used as simulation object.The kinematics model and the geometric model provide a reference for pose estimation and the basis for the lateral control law design.(2)The longitude speed controller is designed based on single neuron adaptive PID to achieve the goal of one controller adapts to the control situation that the small-scale autonomous vehicles with different structural parameters work in different conditions.While designing the controller,the quadratic learning algorithm is used to optimize the control increment and reduce the output error.This work presents a structure that the single neuron adaptive PID is combined with a hyperbolic function to further constrain the control increment,as a result,the ability of anti-interference is enhanced.The approach “Open Loop Identification & Control Simulation” offers the original value of the changeable parameters in the controller,with witch the time of adjusting parameters is reduced;this improves the rapidity of the system.Comparison experiments under different operating conditions reveal that the controller has strong adaptability.(3)In this work,the indoor ultrasonic positioning system is installed and tested.The special program has been developed for positioning data extracting and transmitting.According to the positioning data and the geometric location relationship of installed tags,the localization and pose of the small-scale autonomous vehicle are calculated.In order to estimate the current state,the average delay time with which the time delay of positioning data is compensated and the kinematic model are combined.The kalman filter is utilized to improve the accuracy of the positioning system.(4)The lateral control law is designed based on path tracking algorithm to control the steering.Through straight path tracking experiments,the gain value K for determining look-ahead distance is obtained.To solve the problem that there are more than one target computed,this work proposes the downward search strategy.In the research,the position error is added as a basis for the control of steering angle output,and this weakens the influence of positioning error on lateral control.The method of quantitative compensation is used to reduce the error of steering output due to the clearance of steering mechanism.The path tracking experiments on typical paths show that the method can meet the requirements of the system.In summary,this work has completed the design of basic controllers for small-scale autonomous vehicles,and the relevant experiments are conducted,the experiments show that longitude and lateral controllers perform well.