An autonomous vehicular system based on muli-agents control: Architecture and behavior simulation

Since the 21
^st
century, vehicles have attracted a great interest due to their potential applications for transportation of people and goods. The initial concerns of industries and researchers were that radio-equipped vehicles are able to keep the drivers informed about risks and road conditions. However, recent researches focuses more on providing the drivers with more comfort an less effort. For instance, air-conditioning, automatic features, GPS, etc, insure the quality of service (QoS) for the users. Moreover, the researches are still in advance to promote more autonomy and dynamic to the vehicle. In this paper, we present an autonomous vehicular system based on multi-agents to reduce the complexity of the autonomous system. In fact, instead of executing several tasks by one process, we split them between different agents. Therefore, this complexity reduction reduces the execution time and provide a quick intervention in complex scenarios. Furthermore, our multi-agent system can be applied for all vehicular brands. In contradictory to the existing systems that were dedicated to specific brands.