Development and Verification of a Steering Safety Risk Assessment Algorithm for Highly Automated Vehicles Based on Indirect Kinematic Monitoring

The transition of highly automated vehicle (HAV) steering to a by-wire architecture (SbW) eliminates the backup mechanical link between the steering wheel and the road wheels and places a failure of the actuation subsystem in the highest safety integrity category, automotive safety integrity level D (ASIL D) under ISO 26262. Most known solutions achieve the required integrity level through duplication of steering angle sensors and electromechanical actuators, which increases cost and complicates certification. The aim of this work is the development and experimental validation of an independent algorithm for indirect kinematic verification of the steering actuation system that requires no additional sensors and relies exclusively on the standard wheel speed sensor signals of the anti-lock braking system/electronic stability program (ABS/ESP system). The computed steering angle of the road wheels is determined from the two-track Ackermann kinematic model based on the difference between the linear speeds of the non-steered axle wheels. An empirical speed-dependent correction is introduced to compensate for lateral tyre slip. The dynamic tolerance threshold for the angular disagreement is derived from the geometric condition that the vehicle must not depart from its lane within the reaction time. Risk accumulation is implemented by a hybrid algorithm combining cumulative sums with a weighted integrator whose growth aggressiveness depends on vehicle speed. The algorithm was implemented in MATLAB/Simulink (The MathWorks, Inc., Natick, MA, USA) and verified both on synthesized data and on telemetry from a real vehicle recorded via the Unified Diagnostic Services (UDS) diagnostic protocol. No false positives occurred throughout the entire experimental run. An artificially injected fault was reliably detected by the algorithm, with a subsequent correct decay of the risk indicator after the fault was removed. The proposed approach establishes a diagnostic channel that is physically independent of the steering actuation system and is based on already certified standard sensors, which reduces the implementation cost of the safety function without loss of sensitivity to the characteristic failure modes of SbW. The algorithm was validated on a production vehicle equipped with conventional electric power steering (EPS) using standard ABS/ESP telemetry; it is therefore presented as a candidate diagnostic channel, and extension to a dedicated SbW platform with architecture-specific fault injection—including a hardware-in-the-loop study—is identified as future work.