Parking in busy locations eg at large-scale events (such as concerts and mass-participation sporting events) or in shopping centres at busy times (eg during ‘Christmas shopping’) often results in traffic congestion in surrounding areas (see, for example, [1], for an extreme example!). This may arise due to vehicles searching for, entering, and leaving parking spaces (even where adequate parking is available). Our hypothesis is that this is, at least in part, due to poor coordination between vehicles and that the situation could be improved if the actions of individual vehicles were coordinated. This might be achieved by providing appropriate (fine-grained) advice and/or directions to drivers to smooth the flow of traffic. Indeed, this is often attempted by the deployment of traffic police or stewards. This raises the question of how we can improve parking performance by coordinating and streamlining individual vehicles and their drivers?
As part of the Research Ireland funded ClearWay1 project, this project will explore the design of an ‘app’ or SatNav system to provide advice to drivers approaching a car park with the goal of smoothing the flow of traffic either to an appropriate (potentially reserved) parking space or away from the area in the case of through or departing traffic. Advice will be based on a model of vehicle coordination, developed in ClearWay, that we term ‘slot-based driving’. In slot-based driving, each vehicle is allocated a location-based time slot in which to travel for the duration of its journey. Machine learning (especially reinforcement learning) algorithms are expected to be used to optimize the flows of vehicles and thereby generate appropriate advice to drivers. The project is expected to prototype the design of the app and underlying control system and simulate its operation in a sample deployment.
[1] “Two-hour traffic delay spoils Dublin Half Marathon” https://www.irishtimes.com/news/ireland/irish-news/two-hour-traffic-delay-spoils-dublin-half-marathon-1.3231877
1 The ClearWay project is supported by the RI Frontiers for the Future Programme under award number 21/FFP-A/8957 from 2022 to 2027.