HumanDrive Digital Twin Case Study

This paper covers how a digital twin has been created, the tools used, techniques developed and their application to the HumanDrive project (https://humandrive.co.uk/) (an Innovate UK/CCAV office sponsored Connected and Autonomous Vehicle (CAV) project). Much of the focus of this paper is on that of data used, data processing and its ultimate exploitation in creating an accurate real-world digital twin of public roads and the Multi User Environment for Autonomous Vehicle Innovation (MUEAVI) test track (https://www.cranfield.ac.uk/facilities/mueavi).

Digital Twin technology is the concept surrounding the creation of a digital “twin” or replica of a physical asset. It can provide valuable, actionable insight into operations and in this case have been used to analyse human driving behaviour and support safety work being undertaken by the Connected Places Catapult in the HumanDrive project.

The UK automotive industry is in prime position to assert a global leadership position in the demonstration, testing, development and deployment of CAV technologies. Advancements in computing power, sensor technology, Artificial Intelligence (AI) and Machine Learning have enabled organisations to bring these together in a way that has the potential to make driving safer, easier, quicker, cleaner and more accessible.

The UK features a world-class physical test environment ecosystem however, to further support organisations working on the development of CAVs, the use of virtual test environments for simulation and testing has been identified to strengthen CAV development. A virtual environment, or a digital twin of a real-world location or scenario, can be used to test for example, a product within a geospatially accurate replica of a physical environment. Virtual test environments allow models to be built up, which require fewer resources and preparation time than that of its physical test environment counterpart. Further to this, production aspects such as testing characteristics, vehicle paraments and environmental features are easily adapted and modified.

Sign up to our newsletter

Subscribe