Model predictive energy management with vehicle control for automated driving

With the introduction of automated driving new requirements arise for the functions involved in the automated transition to a safe state in case of a failure. The vehicle power net, which is responsible for the reliable power supply also in case of a failure, is of high importance for the functional safety of the automated driving systems. In the first part of this paper, new requirements for the power nets as well as a fail-operational power net topology fulfilling these requirements are presented. For the control of fail-operational power nets also new control strategies are required, providing a reliable power supply for normal and failure case operation. With the use of predictive and adaptive energy flow distribution, which can be integrated in the control of the powertrain, model predictive vehicle control with automated selection of the best scenario for the safe state transition is realized. The architecture of such a system as well as fault injection simulation results verifying the functionality are presented in this paper.