Developing an open-source platform for the evaluation of intelligent traffic control algorithms

Intersection management is a key component of road transport systems. Envisaging a new age of road transport systems accommodating intelligent, connected, and autonomous vehicles, many novel intersection control algorithms have been proposed in the literature. These algorithms are often implemented using bespoke software and tested over custom built network models because of their complexity and the lack of freely accessible software tools. This in turn makes them difficult to evaluate and benchmark. To solve this issue, in this paper, we present the Traffic Control Test Bed project, the objective of which is to develop an open source microsimulation platform for the evaluation of intersection control algorithms. The platform provides a library of road network models together with an intuitive synthetic road network generator for user-defined layouts. It facilitates and streamlines the parallel execution of simulations. Outputs and performance indicators are monitored and visualised by the platform both during runtime and at post processing stage. We demonstrate the usage of the platform with a case study evaluating two simple signal optimisation methods. As well as being an arena for traffic control algorithms, the open source property of the platform also invites contributions from the wider research community to improve execution validity and efficiency of traffic control systems