Research On Control Technology Of Vehicle Anti-Lock Braking Systems

As a kind of active security equipment in vehicle, Anti-lock Braking Systems( ABS) has been used all over the world. The road surface conditions vary from dry to ice. There are hysteresis in the actuator of ABS. All these factors result in difficulty of controller design course of ABS. The main difficulty arising in the design of Antilock Braking System (ABS) controller is due to the strong nonlinearity in the problem. The most important process parameters affecting the quality of control are the coefficient of friction between tire and road, the tire slip ratio, and the vertical force on the wheel. Robust theory must be adopted in the development of ABS controller.The main research fruits in this dissertation show below:At first, Three kinds of car model are established. A quarter car model is adopted in the design course of ABS controller. Half vehicle model is used for analysis of load transfer. Full vehicle model is used in special conditions such as steering and braking simultaneously, road and tire friction coefficient splitting in left and right side. Radial basis function (RBF) based neural network is applied to emulate tire force characteristics.Secondly, Fuzzy controllers have an inherently parallel structure , which allows the controller to respond immediately once a new situation has been identified. In this paper a genetic algorithm(GA) based fuzzy controller has been developed to solve the problem. The goal of the proposed controller is to maintain slip rate of the wheel within the vicinity of required value. The optimization process uses GA to determine the optimal location of membership function in the fuzzy controller. Simulations were conducted to verify the proposed strategy. The results illustrate how GA can be applied to produce feasible and effective fuzzy controller.Thirdly, In this paper a sliding mode based variable structure controller has been developed to solve the problem. The drawback of sliding mode control is chattering in its application when actuator dynamics were adopted in the controller designing process. In the paper a boundary layer is designed to eliminate chattering. With the idea simulations are done .In the simulations several situations such as braking in dry road,wet road and ice road are considered. The results illustrate that the idea can be applied to produce feasible and effective ABS controller.Fourthly, In this paper, H-inf control controllers which are robust to uncertain parameters of ABS are designed. The proposed controllers not only can satisfy the stability but also can reach some quadratic performance target.Next, A car model which takes into account lateral and longitudinal dynamics of car is established. In the model steer angle is regarded as control input and differential braking torque is treated as disturbance. Simulations are done to contrast response of car equipped with ABS with that of car equipped without ABS. The results show us that ABS can reduce the amplitudes of raw rate and lateral acceleration.In the end, A test system based on multi-dimension Wheel Force Transducer(WFT) is introduced. Some key sensors are introduced at detail. With the test system , road tests are done. Many experiments and tests are performed with different road surface and different braking beginning velocity. The results are analyzed and discussed. With the results, we can conclude that WFT based test system can be useful in the ABS performance test.