Neuroevolution Application to Collaborative and Heuristics-Based Connected and Autonomous Vehicle Cohort Simulation at Uncontrolled Intersection

Artificial intelligence is gaining tremendous attractiveness and showing great success in solving various problems, such as simplifying optimal control derivation. This work focuses on the application of Neuroevolution to Connected Autonomous Vehicle (CAV) cohorts operating at uncontrolled intersections. The proposed method implementation’s simplicity, thanks inclusion heuristics effective real-time performance are demonstrated. resulting architecture achieves nearly ideal conditions keeping average speeds close speed limit. It twice high mean throughput a controlled intersection, hence enabling lower travel time mitigating energy inefficiencies from stop-and-go vehicle dynamics. Low deviation road limit continuously sustained for most 50 m long. limitation can be mitigated with additional lanes that split into. concept also allows testing implementation fast-turning by simply replicating reconnecting each new crossing, scalability complex network analysis. controller successfully validated within high-fidelity dynamic environment, its potential driverless addition offering traffic simulation model future autonomous operation studies.