This paper presents a probabilistic method for processing and analyzing residuals for the purpose of fault detection. The method incorporates residuals from multiple models using a hybrid dynamic Bayesian network in order to yield a low-cost, complete, diagnostic system. Continuous residuals are used as evidence directly in the network, and this paper discusses options for representing their probability distributions. The Bayesian network is used to model the temporal behavior of the faults, and the assumptions necessary to do this are analyzed. The diagnostic method is demonstrated on a car's handling system and experimental results are presented.

No abstract

No abstract

Passenger cars are increasingly available equipped with Autonomous Emergency Braking (AEB). AEB systems detect likely forward collisions and apply the vehicle’s brakes if the driver fails to do so, reducing vehicle speed in order to mitigate or potentially avoid a collision. The performance of these systems is experimentally evaluated in tests including those specified by the European New Car Assessment Program (Euro NCAP) and by the Insurance Institute for Highway Safety (IIHS). In both of these testing programs the subject vehicle is driven towards a Euro NCAP Vehicle Target, an inflatable device designed to have visual and radar reflective characteristics similar to the rear of a compact car. The results reported by Euro NCAP and the IIHS have revealed significant differences in the AEB test results achieved by various AEB-equipped vehicles. Such differences exist even between vehicles with similar sensing technologies, suggesting that the source of such disparities may be differences in sensor data processing methods or differences in collision mitigation and avoidance strategies. This paper details the performance of AEB as well as Forward Collision Warning (FCW) systems when tested with the Euro NCAP Vehicle Target. These results are analyzed, exploring the differences in the performance of these systems under the test conditions and discussing possible reasons for the observed disparities.