Digital Modelling of Millbrook Proving Ground Advances Autonomous System Development

Millbrook virtual proving ground - Belgian Pavé test track render

Millbrook is investing in the creation of a Digital Model of its Proving Ground in the UK. The model, created by UK software specialist rFpro, will enable vehicle manufacturers to significantly improve the development of ADAS and autonomous systems.

The development of connected and autonomous vehicles necessitates a virtual test environment. Testing of future systems needs to begin today, so we are ensuring that our customers have the tools and capabilities required to take full advantage of our facilities as they develop the vehicles of tomorrow.Alex Burns, President of Millbrook

Millbrook Proving Ground provides a safe and secure environment for controlled testing of autonomous experiments and a natural, real-world, extension to its customers’ software engineering processes. Any corner cases identified during real-car testing can then be incorporated in the customer’s regression testing for further training and validation.

The success of this validation depends on the correlation between the software and the real world, and hinges on precise measurement and analysis. rFpro is an industry leader in this field, by utilising phase-based laser scanning survey data it is able to create models with an accuracy of around 1mm in Z (height) and in X and Y (position). The Millbrook model contains over 50km of various types of road, including ADAS and autonomous facilities, wet and dry handling, ride and durability routes.

The software for ADAS and autonomous systems need to be trained and tested on a digital copy of the proving ground before they are ready for real world testing. By investing in a digital model, Millbrook’s proving ground becomes a simulated part of its customers’ Continuous Integration software development tool-chain. This significantly reduces the development and validation time, and therefore cost, of such systems.Chris Hoyle, Technical Director at rFpro.

The virtual world created by rFpro can be populated by ego vehicles (the customer’s vehicles) as well as by semi-intelligent Swarm traffic and Programmed traffic. Vehicular and pedestrian traffic can share the road network correctly with perfectly synchronised traffic and pedestrian signals, following the rules of the road, while also allowing ad-hoc behaviour, such as pedestrians stepping into the road, to provoke an emergency. This allows digital experiments to precisely mirror the real-world tests conducted on the physical proving ground at Millbrook.

Human test drivers can interact with this virtual world in full-scale driving simulators, or at desktop workstations with basic steering and pedal controls. This allows drivers to test cars with ADAS systems, to be passengers in a car under the control of a fully autonomous system, or to simply drive around the virtual world to either subjectively assess the behaviour of autonomous vehicles, or to provoke emergency scenarios and evaluate the response.