Automated Test Case Generation Method for Intersections Based on Cyber Physical System

This study analyzes the conflict dynamics between straight-moving and left-turning vehicles at intersections using a real-world dataset. By combining Markov chains, Monte Carlo simulations, and a newly proposed surrogate safety measure, Anticipated Collision Distance (ACD), we present a method for automatically generating test cases for left-turn and straight-moving vehicles. This approach addresses challenges related to low test case generation efficiency and insufficient coverage. Experimental validation shows that the generated trajectories closely match real-world data, with Frechet distance calculations confirming their authenticity. This method provides reliable benchmarks for testing systems such as Automatic Emergency Braking (AEB) and vehicle-cloud speed guidance algorithms within Intelligent Vehicle Cyber Physical System(IVCPS), ulti-mately enhancing intersection safety and passenger comfort.