Study Of Control Strategy For Vehicle Active Suspension Based On Kalman Filter

The car has become an important part in our production and lives, Especially, with the development of related technologies, the overall performance of the car is constantly required to improve, especially for the automobile riding comfort and the handling stability performance requirements. Suspension system plays an important role in improving overall vehicle performance, especially for the riding comfort and handling stability,and it is general concerned by domestic and foreign scholars and automobile manufacturers. the vehicle active suspension system which can achieve optimal overall performance has been widespread concerned, and have received significant development. Active suspension systems are mainly focused on control strategies and actuator. The active suspension system using a single control strategy can not meet the requirements of the people, therefore, the hybrid control strategy of two or more control strategies will be a developing trend of vehicle active suspension systems.Based on previous research, the main purpose of this dissertation is to improve vehicle riding comfort, the most important work is to do some research on control strategy for vehicle active suspension system, and do computer simulation through MATLAB, RECURDYN on active suspension system. This specific research work is as follows.In this dissertation, the dynamic models of active and passive vehicle suspension in the whole car with seven degrees of freedom were built, at the same time, their coefficient matrix and control matrix were derived. And use RECURDYN software to create three-dimensional solid model of the whole vehicle. The speed sensor and acceleration sensor are complex, cost too much, can not be measured directly, because of the shortcoming, in this dissertation, only four displacement sensors were installed between the chassis and the arm (or oblique arm), in this way, not only can it easily distinguish the relative displacement signal, but also it can obtain the corresponding relative displacement and relative velocity through first-order differential, second-order differential respectively.when the vehicles are moving, some state variables are not easily obtained by measuring, therefore, in this dissertation the Kalman filter state estimator was designed to estimate the vehicle state variables, an accurate estimate of the state variables of the vehicle was achieved. The compound control strategies based on fuzzy control and LQG optimal control was proposed, the inputs of fuzzy control were the relative velocity and relative acceleration of the suspension, the output was the road excitation grades, and in MATLAB verified the accuracy of the fuzzy controller. The road excitation would be further divided into12grades, obtained optimal feedback gain by adjusting the weights on the different levels of pavement, achieved optimal control on the road of any grade in order to achieve the overall performance improvement. And through the joint simulation of MATLAB/Simulink and RECURDYN verified the effectiveness and reliability of this compound control strategy.