The vehicle uses AI-based reasoning to plan its future trajectory based on the desired behaviour in its current context and the risk of its actions. This ‘situation awareness’ allows the vehicle to demonstrate safe behaviour in complex situations in automated driving. Decision making is one of the most challenging tasks for an automated vehicle. State-of-the-art solutions are either provably safe but unable to scale up to complex situations, or difficult to assess in terms of safety but handling complex situations well. The Hybrid-AI approach in this demonstrator is able to traverse complex situations while at the same time explicitly taking into account the risks involved in the actions of the vehicle.
In the demonstrator, a factory default vehicle, retrofitted with automated driving hard- and software, will be autonomously driving in its lane. At some point, the lane will be blocked by an obstacle. The vehicle will have to decide whether it is safe to plan and execute a trajectory around the obstacle, or wait behind it.
Watch the video.