Real-time estimation and prediction of tire forces using digital map for driving risk assessment. (October 2019)

• Record Type: Journal Article
• Authors: Jiang, Kun; Yang, Diange; Xie, Shichao; Xiao, Zhongyang; Victorino, Alessandro Corrêa; Charara, Ali
• Is Part Of: Transportation research. Volume 107(2019)
• Page Start: 463
• Page End: 489
• Publication Date: 2019-10
• Subjects: Vehicle dynamics -- Nonlinear observer -- Tire forces estimation and prediction -- Driving risk assessment -- ADAS map; Transportation -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.011
• DOI: 10.1016/j.trc.2019.08.016
• Languages: English
• ISSNs: 0968-090X

Abstract

Highlights: We use tire forces to assess driving risk. All tire forces are estimated by using cascaded observers with only low-cost sensors. Estimation of tire forces does not need sideslip angle with the proposed models. Tire forces are predicted by getting upcoming road geometry from ADAS map. Abstract: This work aims to develop a driving risk warning system to enhance the road safety. Different from the existing lane departure warning system, speed limit warning system or collision warning system, the warning system proposed in this work focuses on the safety regarding vehicle's dynamics states. Many road accidents are caused by losing control of vehicle dynamics, such as the rollover, car drift and brake failure. First of all, the importance of monitoring vehicle dynamics states, especially the tire forces, is explained. Then the driving risk assessment criteria based on tire forces are developed in this work. The main contribution of this paper is the development of vehicle dynamics models and observers to estimate and predict individual tire forces using only low-cost sensors and ADAS (Advanced Driver Assistance Systems) map. The major new techniques developed in this study can be summarized in three aspects: (1) development of new vehicle dynamics models to estimate vertical, longitudinal, and lateral tire forces, (2) development of new nonlinear observers to minimize the estimation errors caused by sensor noises and model uncertainty, and (3) development of the tire forces prediction algorithm by taking advantage of digital map. The proposed warning system is validated by real vehicle experiments.