State and Parameters Estimation for Distributed Drive Electric Vehicle Based on Unscented Kalman Filter

Abstract

：In order to improve the control performance of distributed drive electric vehicle (EV) a more comprehensive vehicle state and parameter estimation method is proposed. For the problem that it is difficult to measure the vehicle states and parameters in real time, the distributed EV is considered as the object, and a method of vehicle states and parameters estimation based on unscented Kalman filter (UKF) is discussed. Firstly, the 7-DOF time-varying parameter vehicle model is established; secondly, taking the longitudinal acceleration, lateral acceleration, yaw angular velocity and wheel speed which are easy to be measured as the measured variables, a combined observer of vehicle states and system parameters is designed using the UKF algorithm to estimate longitudinal velocity, lateral velocity, lateral tire forces, vehicle mass, center position and inertia moment by extending the dimension of state variables. Finally, the combined observer is simulated and analyzed on Simulink/Carsim platform. The results show that the observer can estimate vehicle state and system parameters effectively, and has a good convergence effect.

Electric Vehicle State Parameter Estimation Based on DICI-GFCKF

“…Figures (5)(6) show that the error between the estimated and the actual values of both algorithms is small. Through local enlargement, the estimated value of the DICI-GFCKF algorithm is closer to the actual value than that of the GFCKF and the CKF.…”

Estimation of state parameters and road adhesion coefficients for distributed drive electric vehicles based on a strong tracking SCKF

“…At the same time, with the development of automotive electronics, an increasing number of automotive active safety systems are widely used, and the realization of these active safety system functions depends on the accurate acquisition of the car's driving state and road adhesion condition, so the accurate real-time estimation of a car's driving state parameters and road adhesion coefficient has become a trending topic in automotive control research. 1,2 1 School of Mechanical Engineering Anhui Polytechnic University, Wuhu, China 2 Currently, the estimation of vehicle driving state parameters can be divided into the following two types: the first is based on the kinematic model estimation method, 3 and its essence is to use the measurement data acquired by the sensor to estimate vehicle state parameters through kinematic relations. The kinematicbased estimation method has low model requirements and can improve the real-time performance of the algorithm, but it requires additional integration operations, and the cumulative error increases with time while requiring high sensor signal quality.…”

“…At the same time, with the development of automotive electronics, an increasing number of automotive active safety systems are widely used, and the realization of these active safety system functions depends on the accurate acquisition of the car’s driving state and road adhesion condition, so the accurate real-time estimation of a car’s driving state parameters and road adhesion coefficient has become a trending topic in automotive control research. 1,2…”

On co-estimation and validation of vehicle driving states by a UKF-based approach

“…When the system is strongly nonlinear, the EKF estimation results may lead to large errors or even diverge. In addition, when the estimated system is too complex, the computational load of the EKF algorithm will increase dramatically, which may cause no solution of the Jacobian matrix (Zhou et al, 2019;Strano and Terzo, 2018). Fortunately, compared with the EKF algorithm, the unscented Kalman filter (UKF) algorithm approximates linearization by sampling instead of calculating the Jacobian matrix, which can avoid the above-mentioned problems.…”