Abstract for presentation (Poster or Podium)
AI in Transportation
Paul Slone, PE, PTOE
Director of Traffic Operations & ITS
Cincar Consulting Group, LLC
Woodstock, GA, United States
Paul Slone, PE, PTOE
Cincar Consulting Group, LLC
Woodstock, Georgia, United States
Traditional methods for traffic signal coordination have relied on fixed-cycle timing plans, occasionally incorporating technological advancements such as traffic-responsive operation or an adaptive system. Adaptive signal systems marked a pivotal shift, promising substantial improvements in traffic flow, but came with increased costs of maintenance and technical expertise for transportation agencies. Despite this progress, these adaptive systems still operate within predetermined timing plans crafted by system operators. Their efficacy falters when faced with unusual traffic patterns, such as non-recurring congestion or large-scale special events, posing significant challenges. The pivotal question emerges: can artificial intelligence enable real-time control of traffic signal systems? The answer is yes; AI for coordinated traffic signal systems is currently being studied and will soon be achievable. With existing computational power, scalability, and robust communication infrastructure, the realization of AI-driven real-time traffic signal control is not a matter of if, but when. This presentation will summarize research and trials to date.