Multiagent Meta-level Control for Adaptive Traffic Systems: A Case Study

Abstract

As cities across the globe continue to grow, traffic congestion has become globally ubiquitous with great economic and environmental costs associated with it. The increasing prevalence of self-driving vehicles creates an opportunity to build smart, responsive traffic infrastructure of the future. Such an infrastructure consisting of connected and autonomous vehicles and smart traffic lights would have the potential to cope with congestion, weather phenomena and accidents, while maintaining safety and ensuring privacy of information. This paper introduces an approach to address the challenge of dynamically adjusting traffic to the changes in the environment. We argue that multiagent meta-level control (MMLC) is an effective way to non-myopically determine how and when this adaptation should be done. The approach highlights the role of dynamic meta-reasoning in a platooning scenario, in which collaboration contributes to improved travel time for vehicles in the network as well as a positive environmental impact as related to fuel consumption and emissions. Specifically, for the case study described in the paper, our MMLC-based approach leads to approximately 44% decrease in travel time, 7% increase in average speed, a 32% decrease in fuel consumption and a 35% drop in emissions. We also see performance advantages for a scaled-up mixed traffic simulation environment.