Global chassis control based on inverse vehicle dynamics models

This work proposes to approach global chassis control (GCC) by means of model inversion-based feedforward with allocation directly on the actuator commands. The available degrees of freedom are used to execute the desired vehicle motion while minimizing the utilization of the tyre's grip potential. This is done by sampled constrained least-squares optimization of the linearized problem. To compensate for model errors and external disturbances, high-gain feedback is applied by means of an inverse disturbance observer. The presented method is applied in a comparison of eight vehicles with different actuator configurations for steer, drive, brake and load distribution. The approach shows a transparent and effective method to deal with the complex issue of GCC in a unitized way. It gives both a base for controller design and a structured way to compare different configurations. In practice, the transparency supports automatic on-board reconfiguration in the case of actuator hardware failure.