An Investigation into Multi-Core Architectures to Improve a Processing Performance of the Unified Chassis Control Algorithms 2010-01-0662

This paper describes an investigation into multi-core processing architecture for implementation of a Unified Chassis Control (UCC) algorithm. The multi-core architecture is suggested to reduce the operating load and maximization of the reliability to improve of the UCC system performance. For the purpose of this study, the proposed multi-core architecture supports distributed control with analytical and physical redundancy capabilities. In this paper, the UCC algorithm embedded in electronic control unit (ECU) is comprised of three parts; a supervisor, a main controller, and fault detection/ isolation/ tolerance control (FDI/FTC). An ECU is configured by three processors, and a control area network (CAN) is also implemented for hardware-in-the-loop (HILS) evaluation. Two types of multi-core architectures such as distributed processing, and triple voting are implemented to investigate the performance and reliability. A vehicle simulator and brake HILS are used to evaluate the proposed multi-core architectures. From the test results, it is shown that all of the proposed multi-core systems have better performance and improved reliability compared with the single-core system.