A Research Of Intelligent Vehicle System Based On FUZZY-PID Controller With Time Varying Modifying Weight Function

The intelligent vehicle, one kind of mobile robots, is an integrated system which integrates multiple functions such as environment awareness, planning decision and automatic drive. In order to improve the intelligent vehicle's performance in tracking and moving, an intelligent vehicle system based on FUZZY-PID controller with time varying modifying weight function is designed in this paper, which is researched from hardware systems, control strategy, control algorithm and optimization algorithm on the basis of intelligent vehicle principle.Firstly, after analyzing the performance of analytical description fuzzy control rules based on recursive function, an analytical description fuzzy control rules based on time varying modifying functionα(δ,γ) is designed in this paper. And the time varying modifying function has inertial characteristic, and can regulate the fuzzy control rules on-line based on errors of dynamic process.Secondly, aiming at the nonlinear and time-varying characteristics of DC motor speed control system, this paper, using time varying modifying weight function, introduces a FUZZY-PID control algorithm which combines the advantages of Fuzzy control, PID control algorithm and intelligent weight function. Therefore this algorithm, compared with integral separation PID and general Fuzzy control, is inferior in terms of its calculating time, but superior in terms of its performance.Then, by analyzing the principle of simulated annealing algorithm, an improved simulated annealing algorithm is presented in this paper. Taking system performance as optimized target, the Simulated Annealing-Golden Section algorithm is used to optimize the parameters of the FUZZY-PID controller with time varying modifying weight function.And finally, an intelligent vehicle hardware system, where the algorithm is applied in, is designed based on HCS12 MCU. Whereafter, the reliability of this system is tested, which proved that the hardware is simple and reliable; control algorithm can imitate the driving behavior of human to control the intelligent vehicle to track the path accurately and rapidly.