Reliable CPS Design for Mitigating Semiconductor and Battery Aging in Electric Vehicles

Reliability and performance of cyber-physical systems (CPS) in electric vehicles (EVs) are influenced by three design aspects: (i) controller design, (ii) battery usage, i.e., Battery rate capacity and aging effects, (iii) processor aging of the in-vehicle embedded platform. In this paper, we present a two-phase design optimization framework involving control algorithm development, battery modeling and processor aging analysis. First, before processor aging, a Pareto front between quality of control (QoC) and battery usage is obtained. Second, as the processor ages, the controller is re-optimized considering the prolonged sampling period to avoid QoC deterioration of safety-critical applications with a compromise on battery usage and comfort-related applications. An outlook following this line of research on reliable CPS design in EVs is discussed, that we believe, will be pursued by researchers in the CPS community.