DriveSafeAI
DriveSafeML Enhances Safety Validation for ML-Based Automated Driving Systems.
About the project
Project DriveSafeAI is a collaborative research partnership between Wayve and Warwick Manufacturing Group (WMG) at the University of Warwick, developing the evidence base for advanced AI safety assurance methods. The project addresses the need for independent safety validation to build confidence in the use of AI in automated vehicles (AV).
DriveSafeAI combines Wayve’s expertise in AI development with WMG’s world-leading expertise in safety verification and validation for self-driving technologies. This partnership will contribute to the evidence-base informing common assurance approaches to AI safety for the safe deployment of self-driving vehicles—unlocking a market projected to be worth £42bn to the UK economy by 2035.
The project aims to create a set of safety methods, tools, and datasets to ensure the safe deployment of AI software used in self-driving technology, paving the way for the commercial deployment of AV technology.
This includes the evidence-base for an AI safety assurance framework that is scalable to different Operational Design Domains (ODDs). Pioneering the novel OASISS (ODD-based AI Safety In Self-Driving Systems) approach, these methods can be easily adapted to new ODDs, ensuring straightforward verification and validation of AI-based software across different regions and deployment geographies.The innovations will benefit the entire self-driving industry, and are applicable to all system architectures that use AI, either in discrete components or fully end-to-end.
DriveSafeAI’s outputs will also help to shape emerging regulatory and standardisation frameworks at national and international levels by working closely with regulators and the UK Government through the CAVPASS programme to develop best practice and guidance on AI system safety validation.
Project Partners
- University of Warwick
- WAYVE TECHNOLOGIES LTD