Making autonomous vehicles safe requires a range of expertise and experience
The D-RISK project comprises four partners dRISK.ai, DG Cities, Claytex and Imperial College London.
The unique blend of skillsets and experience within the D-RISK consortium means we’re equipped to tackle the big barrier to making AVs a commercial reality: safety.
dRISK.ai is integrating a massively diverse set of data and expert curation into the ultimate data resource for AVs. Rather than enumerating each one of an infinite set of edge cases, we invented a way to “map” all edge cases into a comprehensive taxonomy - a knowledge graph of risk - which acts as both an edge case database and an evaluation space.
dRISK.ai’s patented technology trains, tests and validates AVs on comprehensive sequences of edge cases derived from our massive knowledge graph. The result is a step increase in AV responsiveness and safety.
DG Cities is an urban innovation consultancy, specialising in helping clients harness the power of technology and data to transform our towns and cities.
Our role within the D-RISK project is to ensure that public opinions are heard and shape the development of this new technology. We are inviting the public to share their views on what is ‘appropriate’ behaviour for automated vehicles through online surveys and focus groups.
Our work is also helping to extend the edge case library and ensure that real-life experiences are programmed - and appropriate responses generated - within AV software.
Consistently rated amongst the world’s best universities, Imperial College London is a science-based institution with an international reputation for excellence in teaching and research. The Transport Systems and Logistics Laboratory (TSL) at Imperial College London focuses on the study of transport networks, optimisation methods and multi-agent systems, as well as their applications in autonomous transport systems, urban infrastructure and logistics. The Personal Robotics Lab at Imperial focuses on the interaction between robotic devices and their users, learning from this interaction, as well as adapting the assistance provided in order to maximise their users’ physical, cognitive and social well-being.
Claytex develops simulation technology to explore the interaction of complex systems. We have developed a scenario-based AV simulation and testing solution that allows the AV to be immersed into the virtual environment.
Our solution supports open standards, including OpenSCENARIO and OpenDRIVE, to define the test case and uses digital twins of real-world locations to test the AV in their operating environment. Sensor realistic simulation is achieved for the AV sensor suite by using physics-based sensor models that respond to changes in material properties and environmental conditions.