Development of a Genetic Algorithm for Optimal Response Time Based on Full Scheduling Model of Electric Vehicle

Genetic algorithms (GAs) are frequently used in the design of electric vehicles to optimize various parameters, including battery capacity, motor size, and vehicle weight since they offer a powerful tool for improving their performance. In this paper, we focused our interest on the development of GA in the context of optimizing the response time based on the full scheduling model of electric vehicles applied to a modern vehicle of the Society of Automotive Engineers (SAE) Benchmark. The framework design is a set of many nodes connected through the Real-Time protocol FlexRay and the middleware Data Distribution Service (DDS). GA is implemented to find the optimal set of parameters that minimize the response time required for the static scheduling method applied to a SAE Benchmark application. This approach allows one to take advantage of FlexRay network high speed and to profit from DDS Quality-of-Service (QoS) management in the context of automotive electrical systems. Performance evaluations will be conducted to prove the efficiency, reliability, and robustness of GA proposed in this framework, and a comparison with other algorithms is discussed.