AI in multimodal transport management


With an increasing share of active and sustainable travel modes and emerging shared-mobility alternatives, including ride-hailing and ride-sharing services and electric and shared bike/scooter systems, our urban transport is bearing a shift in roadway space utilisation. The shared use of roadway space introduces inevitable and new conflicts among different modes of transport, as well as significant safety risks, traffic jams and delays for all road users. Future transport infrastructure should be capable of accommodating the multimodal travel demands efficiently.

In the current research within the Australian Integrated Multimodal Ecosystem (AIMES), the University of Melbourne has installed advanced mobility sensing devices at multiple intersections within a six square-kilometre zone north of the Melbourne CBD. These sensors include high-definition cameras connected to edge-computing devices, induction loops, LIDAR sensors, DSRC devices, mobile Wi-Fi detectors, and Bluetooth readers that collectively enable us to detect, classify and register vehicular traffic movements (including trams, buses, trucks) and vulnerable road users (pedestrian and cyclists) crossing or waiting at intersections with high precision in real-time.

This project aims to develop a comprehensive modelling framework, using artificial intelligence techniques and machine learning algorithms to simulate and optimise multi-objective and multimodal operation and control of smart intersections.


The expected project outcome is a suite of machine learning and advanced simulation and optimisation algorithms improve the multimodal performance of urban intersections from safety, efficiency and sustainability perspectives.


iMOVE Australia, RACQ and University of Melbourne.