MEdiating between Driver and Intelligent Automated Transport systems on Our Roads
Problem: Automated transport technology is developing rapidly for all transport modes, with huge safety potential. However, the transition to full automation brings new risks, such as misuse, overreliance, reduced situational awareness and mode confusion. The driving task changes to a more supervisory role, reducing the task load and potentially leading to degraded performance. Similarly, the automated system may not (yet) function in all situations; it must intelligently assess the strengths and weaknesses of both driver and system and select the best control mode according to the context.
Solution: MEDIATOR proposes an intelligent ‘mediating’ support system for road transport, enabling safe, real-time switching between human driver and system. It will constantly evaluate driving context, driver state and vehicle automation status, personalising its technology to the driver’s general competence.
Approach: MEDIATOR pursues a paradigm shift away from a view that prioritises either the driver or the automation, instead integrating the best of both. It will use state-of-the-art knowledge, including that from other transport modes, and develop new knowledge about human behaviour and human-machine interaction. It will apply the latest artificial intelligence technology to evaluate driver state, automation status and driving context in real time. It will produce several prototypes in the lab and in actual vehicles, for evaluation in simulation, simulator and on-road studies—as well as several tools for further exploitation.
Impact: MEDIATOR will optimise the safety potential of vehicle automation, especially during the transition to full automation. It will reduce future as well as current risks (such as inattention or fatigue). MEDIATOR will facilitate market exploitation by actively involving the automotive industry during the development process. Further, the involvement of experts from other transport modes will maximise the transfer of knowledge to these domains.