This paper presents a method for indirect measurement of tire slip angles from chassis acceleration, yaw rate, and steer angle measurements. The chassis is assumed to be rigid so that acceleration data can be integrated to estimate velocities of the front and rear of the vehicle, from which slip angles can be predicted. The difference in front and rear slip angles is indicative of vehicle oversteer/understeer. Understeer data can then be correlated with position on the track to better understand vehicle handling behavior, aiding the tuning process.
The technique is presented, and shown to work well with simulated data, even when the data is corrupted with up to 20% noise. Therefore, the inversion process presented here is theoretically sound. However, when the technique is applied to measured data from race cars, it is shown to be inaccurate. One suspected problem is the difficulty of getting accurate yaw rate data. Bias errors in this data, integrated over time, corrupt the inversion process. Another problem is the inherent ill-conditioning of the inversion process, particularly when the yaw rate is small (as on a straight section of track). More accurate, yet still affordable yaw-rate sensors are needed to overcome these problems. There may also be improved data processing techniques that would help.