A Nonlinear Decoupling Method for Wheel Force Transducer with High-Order Polynomial Model

Wheel force transducer (WFT) is an important device used for real-time monitoring of forces and moments acting on the wheels. However, the device faces a major challenge in vehicular applications—coupling problems between its multiaxis channels, leading to inaccurate measurements and limiting its important role in the study of vehicle dynamics, the assessment of vehicle safety performance and autonomous driving. The study of decoupled models with few parameters and high accuracy is important for low-cost devices with limited computational power. This paper presents a nonlinear decoupling method for WFT with high-order polynomial (HOP) model to effectively improve the measurement accuracy of WFT with eight-beam spoke elastomers. The method is based on a three-axis decoupling model of WFT designed with an eight-beam spoke structure, employing a multiple linear regression model to refine the physical model, resulting in a quadratic curve that enhances the nonlinear fitting capability and improves measurement accuracy. The experimental results show that the maximum decoupling error of the method is less than 44%, which proves the effectiveness and applicability of the method.