‘EDGE CASES’ — these are the countless risky vehicle scenarios which are individually unlikely, but together make up all the risk. For Autonomous Vehicles to become a commercial reality, developers, regulators and insurers will have to validate their performance on a huge range of edge cases.
But no one has known what all the edge cases are… until now.
D-RISK is assembling a huge volume and heterogeny of edge cases based on large proprietary data sources, including:
-CCTV data from the world’s most sensorized cities
-Full text accident reporting from global insurers
-Front-facing camera data
We also have a patented technology for fusing these data and every major publicly available data source and scenario library into a single, uniformly queriable taxonomy – a knowledge graph of every source of vehicular risk. Out of this knowledge graph come exhaustive test sequences that can be applied in simulation, or a recommendation on the next best test in real life.
D-RISK’s customers and partners include:
who are all preparing for the proliferation of AV systems, reducing risk and safety for passengers and public.
Timeline & Funding
D-RISK is funded via the a UK Government’s Centre for Connected and Autonomous Vehicles CCAV. The project is overseen and managed by Innovate UK as part of the CAVSIM portfolio of projects. The project commenced November 2019 and is due to end in January 2022 and has a total grant funding value of £3M.