Feature Level Implementation of System Health Monitoring concept using Adaptive AUTOSAR platform

Abstract

Electronic Control Unit (ECU) level health monitoring mechanisms have been developed and widely used in automotive industry since last decade. Existing health monitoring mechanisms from AUTOSAR architecture do not provide a standardized way of exchanging health related information between ECUs having different architectures. As the automotive sector moves towards higher degrees of automation with adaptive environments and adaptive behaviour of applications running on ECUs, it is imperative that analyzing the health of different ECUs at runtime would be of high importance. Thus, a system wide co-ordination of health related information of different ECUs would be necessary to enhance the existing safety functionalities implemented in subsystems in the vehicle. The work carried out in this thesis showcases an approach of realization of the system wide health monitoring concept for an Advanced Driver Assistance Systems (ADAS) feature. The subsystems associated with the ADAS feature are analyzed and health related information of the subsystems is used to determine the ADAS feature level health indications. The health indications are used as a means of indication of reliability and/or performance to enhance the existing safety functionality provided by the ADAS feature. The results of this thesis show that the realization of system wide health monitoring concept for ADAS feature is viable. The proposed architecture for the realization of system wide health monitoring can be used as an additional safety implementation on top of the existing functionality provided by the ADAS feature.