Proposal Summary and Objectives

Freight logistics are critical to quality of life and economies. However, freight mobility, especially along major freight corridors in urban areas, rarely get special consideration in signal timing. The advent of the Internet of Things (IoT) makes vast information collection a reality. The rich data environment, combined with the boost in computational power, has brought unprecedented opportunities closer to reality than ever before for real-time, information-driven intersection traffic control under variants of traffic scenarios. The rich information collected through sensors and through inter-vehicle communication has enabled large scale application of machine leaning, a proven powerful tool for efficient and responsive decision making. This proposal will build a coordinated traffic control mechanism based on machine learning in the context of big data with a specific objective of improving freight mobility along corridors. More specifically, this research will focus on developing a new traffic responsive network signal control in general, but with freight traffic considered in particular, and provide new measures for optimal switching points for network signal control by directly translating general delay minimization into maximization of intersection throughput, and thus provide a solid theoretical basis for the subsequent reward design of the reinforcement learning. Finally, combine transport theory with reinforcement learning methods to design highly efficient network control algorithms. Numerical test via simulation will be conducted to show the benefits of the developed model and algorithms under different scales of truck traffic.

Funding Amount:
Status: Active
Duration: June 1, 2021 – May 31, 2022