A vision-based approach to wheel camber angle and tyre loaded radius measurement

The accurate determination of a vehicle's lateral behaviour is necessary in the domain of vehicle dynamics to improve vehicle safety. Its interaction with the road surface must be understood thoroughly, which requires the integration of a force model, representing the tyre–ground interface, into the global vehicle model. The validity of such a model directly depends on its nature and more particularly on the accurate determination of its inputs and outputs. This paper focuses on one input: the wheel camber angle. Studies in collaboration with a car maker have shown that a measurement error below 0.1° is necessary to identify the tyre model parameters from real data accurately. At the same time, measurement performance must be insensitive to the test conditions, and more particularly to the road surface (tyre–road grip). For these reasons, a novel vision-based method for direct camber determination has been developed and a measurement prototype has been designed. Its vision-based principle is mainly aimed at giving accurate measurement. Its precision is robust to the road's unevenness and the vehicle's velocity; the tyre loaded radius is determined at the same time. The novel method's feasibility and the measurement performance of the designed prototype have been studied from real vehicle-on-track measurements with a test car.