Modelling Of The Pneumatic Brake System And Development Of The Driving State Estimation For The Commercial Vehicles

Nowadays,the dimensions and weights of the heavy-duty vehicles have been increased rapidly to fulfill the requirement of the freight services and logistic speed.To ensure the security of both the drivers and the passengers,most of the commercial vehicles are equipped with vehicle dynamics control(VDC)system except for the passive safety system,since the passive safety system such as the air bag or the life belt is not enough for ensuring the life security for both the drivers and the passengers in the heavy-duty vehicles.The typical VDC system needs the pre-knowledge of the vehicle driving states such as the side slip angle and the tyre normal force to operate well.Unfortunately,the sensors which can measure those states are very expensive and fragile.To overcome this problem,an alternative estimation method,which can accurately calculate the driving state based on the measurement from the low-cost sensors commonly integrated into the VDC system,should be taken into consideration.The estimation method can provide modularity and flexibility,therefore a good expansibility and maintainability can be expected.A computer-aided analytical dynamic model of a pneumatic brake system in commercial vehicles is first presented in this paper to investigate the accuracy of the proposed driving state estimation method in the MIL(Model in Loop)process.The brake system includes the brake pedal,treadle valve,quick release valve,load sensing proportional valve and brake chamber,and the simulation models for individual components of the brake system are established within the multi-domain physical modeling software-AMESim based on the logic structure.To obtain the structure parameters which cannot be directly measured in the load sensing proportional valve(LSPV),the mathematic model of LSPV is constructed in the form of state space equation.Then the estimation process is implemented relying on the experimental measurements.With the coefficients of the PC(Polynomial Chaos)expansion obtained by the numerical implementation,the output observation function can be transformed into a linear and time-invariant form.The uncertain parameter recursively update functions based on Newton method can therefore be derived fit for computer calculation.To improve the estimation accuracy and stability,the Newton method is modified by employing the acceptance probability to escape from the local minima during the estimation process.Then the bench test indicates that the proposed estimation method has a high accuracy with a good robustness against the measurement noise.This paper proposed a novel computational method which combines the Sliding-mode method and Kalman filter method for roll angle estimation in commercial vehicles.The estimation strategy is robust to the uncertainty of the tyre parameters and has no need for linearizing the dynamic system.The validation indicates that the proposed estimation method has a promising tracking performance with low computational cost and high convergence speed for the ECU unit.Then a 7 DOF planar dynamic model with the nonlinear Pacejka tyre model is constructed and a precise computational method based on Unscented Kalman Filter(UKF)is then proposed to estimate the velocities and the road friction.A novel quasi-stationary method is developed to predict normal tyre forces of heavy trucks.The novel quasi-stationary method for predicting normal tyre forces is able to characterize the typical chassis configuration of the heavy trucks.The validation is conducted through comparing the predicted results with those simulated by the TruckSim and it has a good agreement between these results without compromising the convergence speed and stability.In the last section of this paper,a detailed simulation system which includes both the pneumatic brake system and the commercial vehicle dynamic model is developed to validate the performance of the proposed driving state estimation method and the corresponding VDC control strategy.Then a HIL(Hardware in Loop)test bench is constructed to check the calculation efficiency and accuracy of the proposed estimation method in real-time condition.The test result indicates that the proposed estimation method has a promising tracking performance with low computational cost and high convergence speed.This approach enables a low-cost solution to provide the accuracy information for the VDC system to prevent the potential danger for the commercial vehicles.