Collaborative sensing and control in large-scale transportation systems

Transportation is the circulatory system of our economy. Yet many of our traditional transportation systems are inadequate to serve the needs of the 21st century. Thus, Intelligent Transportation Systems (ITS) has been proposed as a promising direction to provide innovative services in various modes of transport and traffic management. Researchers have accumulated abundant knowledge for designing ITS systems based on surveys and feedbacks from users and operators. However, the data collected from these manually conducted methods are often incomplete, inaccurate and out-of-date. Thus, based on these data, the applications, correlations and interactions among different forms of transportation are under-exploited [1]. This inefficiency calls for a new architecture, which collaboratively integrates sensing and control aspects in the data processing chain of ITS systems, i.e., data acquisition, data analysis, and data utilization, from multimodal transit systems, e.g., taxicab, bus, and subway, by fully automatic realtime methods. Because of the limited understanding on how to collaboratively interconnect different transit systems for realworld applications, we face an urgent and challenging task to investigate the theory and practice in order to coordinate the sensing and control aspects efficiently and collaboratively. To accomplish this task, this research aims at (i) addressing a fundamental challenge that uniquely defines sensing and control aspects in ITS - heterogeneity, (ii) proposing a design for a multi-level architecture to collaboratively handle different procedures of ITS datasets, and (iii) testing our architecture in one of the largest transportation systems in the world as reference implementation in real-world scenarios.