It is established that perimter control is beneficial compared to fixed time, but is it better or worse than adaptive signal control? Should we use perimeter control, adaptive control or apply both concurrently? Those were the motivating questions of our study.
In this study, we cast the route and signal control controller into a distributed approach. We use the augmented Lagrangian to model the constraints into the objective and decompose the model into a set of objective-coupled subproblems, one for each intersection, enabling the solution to be computed by a distributed-gradient projection algorithm. We show that the approach is feasible to be deployed in real-time.
The automation aspect of connected and automated vehicles, namely the possibility of allowing short headways and therefore higher capacity, is extensively investigated in the CAV realm compared to the connectivity aspect. What can we do with connectivity information while vehicle are still human-driven? In this study, we propose a method for concurrent routing and adaptive signal using a multi-commodity approach.
Research at Universidade Federal de Santa Catarina concerning the deployment of an adaptive signal control system.