Modeling Automatic Emergency Braking System Using Fuzzy Timed Petri Nets Based on the 3-Second Rule

Automatic Emergency Braking (AEB) systems play a critical role in enhancing road safety by assisting drivers in avoiding or mitigating collisions. This paper proposes a comprehensive method to model AEB systems using Fuzzy Timed Petri Nets tailored for the 3-second rule. The 3-second rule dictates that a vehicle should maintain a safe distance of at least 3 seconds from the advanced vehicle to minimize the risk of forward collisions. It is important to note that this paper's scope is limited to applying the three-second rule under specific conditions. We focus on scenarios where the weather is clear and ideal for driving, the road conditions are standard without any anomalies, and the vehicles involved are in optimal working conditions without mechanical faults. This narrowed focus allows for more in-depth analysis and understanding of the AEB system's performance under these parameters. By incorporating FTPN, which combines the benefits of Timed Petri Nets and fuzzy logic, we can effectively capture timing constraints and uncertainties associated with the AEB system. By integrating Artificial Intelligence (AI) technologies by combining the strengths of Petri Nets and fuzzy logic, FTPN allows us to accurately represent and infer real-time systems while effectively capturing the timing constraints and uncertainties inherent in reliable performance in critical scenarios for our system. Our model aimed at determining the decision of emergency Braking based on rules that consider the vehicle's speed, the distance to the vehicle in front, and the safety distance. An Expert System derives several rules from our database and presents them as Fuzzy Timed Production Rules (FTPR). The conversion from FTPR to FTPN is crucial in building this model since we cannot model this system without the correct formulation of the rules. The based model depicts a real-time system that responds to the surroundings and its speed to determine when it may start braking and to what extent it will. After obtaining the findings, we validated the model using a variety of scenarios given to several models that varied among themselves in terms of fuzzy intervals. Then, we compared the models with each other in terms of response speed and also the validity of the data. This work aims to advance road safety by enabling vehicles to make well-informed and timely emergency decisions.