In order to design a robust, fault tolerant HAD system, the automated vehicle must also handle subsystem faults. The subsystems for which a fault could occur might e.g. be a sensor as part of the AD system or a classical mechanical subsystem.
The focus of this work was to investigate how such a fault case can be investigated and which extensions and adaption for the models and the architecture have to be applied. Furthermore, additional signals have to be introduced. On the one hand a fault injection signal, which is used to trigger the faulty behavior. On the other hand a fault detection signal, as commonly used in vehicles.
To demonstrate the feasability of the proposed approach an exemplary fault scenario was chosen. In this scenario a fault in the steering system is assumed, which results in a limitation of maximum steering velocity.
The HAD function is extended to react on the changed steering system status in the fault case. In the traffic scenario of a small urban crossing, the ego vehicle performs a right turn with other road users nearby.
Simulations of this scenario were carried out with additional fault injection and both with and without proper reaction of the HAD function on this fault. Simulation results show distinct different behavior.
