Cost-aware reliability task scheduling of automotive cyber-physical systems

New generation of automotives has become a typical Cyber-Physical System, including various sensors, complex communication networks and distributed computing devices. Manufacturers pay great attention to functional safety and cost control of automotive. High-end computing devices such as electronic control units (ECU) that can perform multiple tasks bring higher reliability and also lead to soaring costs. In this paper, we propose an efficient scheduling algorithm that satisfies cost and functional reliability constraints under multifunctional mixed-criticality task replication execution scenarios. Using the proposed fault-tolerant hardware cost-aware multifunctional safety assurance (FT_HCMSA) algorithm can reduce the deadline miss ratio (DMR) of high-criticality function significantly while meeting the autumotive functional safety and hardware cost constraints. FT_HCMSA uses a fine-grained ECU replacement strategy to optimize task allocation. Our experiments validate that FT_HCMSA can reduce the DMR of high-criticality level functions while meet the requirements of safety and cost.