Vote3Deep - Transferring tech from lab to vehicle for high performance, real-time object detection

Next-generation connected and autonomous vehicles (CAV) hold huge potential gains and benefits for the transport industry and are fundamental to realising smart mobility and cities of the future. It is predicted that the CAV market could deliver cumulative benefits to the UK of £51bn by 2030, but there is significant development to be completed before this can be realised. More accurate and efficient lidar solutions are needed for object perception and planning and will directly contribute to the substantial reductions of fatal accidents with an estimated 2500 lives saved by 2030. However these same lidar solutions can be used effectively before full autonomy to prevent accidents, both on cars and in other environments. To achieve a commercially viable solution, software must be able to run in real-time with extremely high accuracy, on low cost hardware. The Vote3Deep project intends to assess the feasibility of transferring a research solution that meets this brief from server-grade to portable systems ready for integration in CAV. The translation of this research solution into a commercially viable product will be enabled through a detailed test programme at RACE (UKAEA) Culham where it will be trialed extensively both statically and in-car in a secure environment.