Smart Vehicle Decision System for Emergencies Using AI and State Machines

What is it about?

The study proposes a decision-making method for autonomous vehicles in emergency conditions, integrating artificial potential fields (APFs) and finite state machines (FSMs). It models the vehicle's longitudinal and lateral potential energy fields to identify driving states and establish trigger conditions for path planning. The research designs state transition rules based on virtual forces and constructs a vehicle decision-making model using FSMs to maintain safety. A simulation model was developed on the MATLAB and CarSim platform, demonstrating the model's capability to accurately generate driving behaviors at various times. The study introduces a hierarchical vehicle state machine decision model to improve driving safety in emergencies. Mathematical models are created to determine state transition thresholds, leading to the formulation of rules for autonomous vehicles' state changes. The combined simulation model verified the feasibility of the proposed decision strategy.

Featured Image

Photo by NANDKUMAR PATEL on Unsplash

Photo by NANDKUMAR PATEL on Unsplash

Why is it important?

This study is important because it addresses the critical need for improved decision-making methods in autonomous vehicles (AVs) during emergency conditions, a vital aspect of ensuring passenger and pedestrian safety. By integrating artificial potential fields (APFs) and finite state machines (FSMs), the research provides a robust framework that enhances the vehicle’s ability to swiftly and effectively respond to unforeseen challenges on the road. This approach has significant implications for reducing accidents and improving the reliability of autonomous transportation systems, paving the way for safer and more efficient traffic management in real-world environments. Key Takeaways: 1. Hierarchical Decision Model: The study proposes a hierarchical vehicle state machine decision model that significantly enhances driving safety in emergency scenarios by allowing vehicles to transition smoothly between different states based on predefined rules. 2. Transition Thresholds: Mathematical models are developed to determine the transition thresholds for lateral and longitudinal vehicle states, facilitating accurate and timely state transitions that are crucial for effective emergency response. 3. Simulation Verification: Using a combined simulation model with the CarSim-MATLAB platform, the study validates the decision model's ability to accurately reflect vehicle driving behaviors across different time intervals, demonstrating its potential for practical applications in autonomous vehicle control systems.