Cooperative Game Based Traffic Control Scheme for the Dual-channel CAN Platform

Abstract

In recent years, automotive manufactures have installed dozens of electronic controller units (ECUs) for various functions related to safety and convenience of drivers. Each ECU controls specific functions of a vehicle, and exchanges information by using the controller area network (CAN). It is characterized by low cost, efficient communication, and high flexibility. In this work, we aim to study the property of CAN bus, and design a new traffic control algorithm to get an excellent CAN system performance. By employing the concepts of two different cooperative game solutions, our approach explores the impact of ECUs' mutual-interaction relationship and devises a novel two-step game model to take the full advantages of CAN resource sharing process. To adaptively handle different data service requirements, control decisions in our two-step game are mutually dependent each other, and each individual ECU acts cooperatively in the proper collaboration manner. This interactive coordinated process continues until a desirable solution is obtained; it is a practical and suitable approach in real world CAN system operations. Finally, simulation testbed is constructed and the numerical analysis is conducted to demonstrate the performance improvement of our proposed method. In addition, several research challenges are discussed and open issues are also outlined.