A Data-Driven Framework for Tire Force Estimation of Distributed Electric-Drive Vehicles

In recent years, with the development of wheel-side motors and hub motors, distributed electric drive vehicles, gradually enter the electric vehicle market.Tire force are often derived from rule-based model in the past. However, distributed electric drive vehicles have a higher degree of freedom put forward new control requirements. This puts forward higher requirements for the accuracy of the tire force model. Rule-based model cannot meet the requirements quite well. Because of this, our study established a tire force residual correction framework for distributed electric drive vehicles. The framework consists of a neural network model (MLP, MLP-seq, and MLP-mixer) and a physical rule-based model. The framework was proved in the study to output a more accurate force estimation which will help dynamic modeling and control tasks.