Connected On-Road Autonomous Mobility (CORAM)

Project 'Connected On-Road Autonomous Mobility' (CORAM) proposes to conduct a feasibility study to research and develop a system of capturing, processing and broadcasting fixed infrastructure sensing data via wireless network for providing non-line-of-sight perception or 'Look-Ahead' perception to autonomous vehicles much like the mirrors on blind corners, information displays on highway gantries and traffic signals on junctions assist human drivers. There is no reason why infrastructure based perception should not be available to a CAV. The central idea of CORAM is to bridge the gap between on-vehicle sensing capabilities that have a limited field of view and range and the wide field of view potentially available from fixed infrastructure sensing, by developing techniques to extract 'scene understanding' from a fixed vantage point possessing a long range and wide field of view. This scene understanding then will be delivered to the autonomous vehicle for use by on-vehicle perception in order to guide its local path planning decisions. This will result in enhanced robustness, improved safety and advanced capability of autonomous vehicles.