Vehicle Motion Management - A model predictive control approach to realize holistic redundancy to enable actuator fail operational autonomous driving

Since the series introduction of ESC (1995), the number of vehicle dynamic control functions and ADAS has increased continuously. Today, there are many functions that control the vehicle motion in cooperation, so that the vehicle motion is often implemented suboptimally. Current megatrends (electrification, AD, new SW architectures) and new key technologies (X-by-Wire) enable the development of Integrated Chassis Control (ICC), which controls all motion-relevant components to optimize vehicle motion. This function, which is currently being developed by many OEMs, not only improves the driving safety and comfort but is also furthermore essential for AD. Vehicle Motion Management (VMM), the ICC solution by IAV GmbH, uses Model Predictive Control (MPC) to predict and optimize vehicle motion in real time. The VMM architecture already developed (Observer, Supervisor, MPC, Coordination) can be adapted to the powertrain topology. This article reports on the results of a simulative potential analysis regarding implicit redundancy. In this analysis, the steer-by-wire steering system fails during constant cornering. The VMM intervenes to maintain cornering even in the event of a steering system failure and to prevent the vehicle from leaving the lane. The curve radius is defined by the minimum curve radius of the standardized EKA 1A highway to map the corner case of the highway. This curve is driven through for two different, statistically relevant driving speeds that occur in real traffic. The potential analysis is repeated simulatively for three different powertrain topologies. In the first topology, the VMM can only set braking torques to maintain cornering. In the second and third topologies, additional electric drive torques and rear wheel steering angles can be set. To analyze the potential of additional motion-relevant components, the maximum track offset, the maximum yaw rate and the maximum side slip angle are evaluated as characteristic parameters.