A Novel Comprehensive Scheme for Vehicle State Estimation Using Dual Extended H-Infinity Kalman Filter

Fengjiao Zhang,Yan Wang,Jingyu Hu,Dong Zhou,Song Chen,Guodong Yin,Hongdang Zhang

doi:10.3390/electronics10131526

Abstract

The performance of vehicle active safety systems relies on accurate vehicle state information. Estimation of vehicle state based on onboard sensors has been popular in research due to technical and cost constraints. Although many experts and scholars have made a lot of research efforts for vehicle state estimation, studies that simultaneously consider the effects of noise uncertainty and model parameter perturbation have rarely been reported. In this paper, a comprehensive scheme using dual Extended H-infinity Kalman Filter (EH∞KF) is proposed to estimate vehicle speed, yaw rate, and sideslip angle. A three-degree-of-freedom vehicle dynamics model is first established. Based on the model, the first EH∞KF estimator is used to identify the mass of the vehicle. Simultaneously, the second EH∞KF estimator uses the result of the first estimator to predict the vehicle speed, yaw rate, and sideslip angle. Finally, simulation tests are carried out to demonstrate the effectiveness of the proposed method. The test results indicate that the proposed method has higher estimation accuracy than the extended Kalman filter.

Highlights

IntroductionSome typical active safety systems include antilock braking systems [1], yaw stability systems [2], collision avoidance systems [3], and so on
Some typical active safety systems include antilock braking systems [1], yaw stability systems [2], collision avoidance systems [3], and so on. The prerequisite for these active safety systems to work accurately is the availability of accurate vehicle state information [4]
The second Extended H-infinity Kalman Filter (EH∞Kalman filter (KF)) estimator uses the result of the first estimator to predict the vehicle state

Summary

Introduction

Some typical active safety systems include antilock braking systems [1], yaw stability systems [2], collision avoidance systems [3], and so on. The prerequisite for these active safety systems to work accurately is the availability of accurate vehicle state information [4]. Existing onboard sensors are limited in the number of vehicle states they can measure. Some states that characterize vehicle stability, such as the sideslip angle, cannot be measured by onboard sensors. Some special sensors can measure the sideslip angle, they are expensive and require additional installation in mass production cars, which greatly limits the application of this measurement method. To obtain some key state information of the vehicle more economically, many scholars have proposed various interesting estimation methods

Methods

Results

Conclusion