HORIBA MIRA, a member of CAM Testbed UK, is a global provider of pioneering engineering consultancy, research, and testing services, and has been involved in the development of connected and automated mobility (CAM) solutions for over 15 years as part of an overall vision of making journeys safer, cleaner and smarter. This guest blog post was written by Richard Hillman, Principal Engineer within the Connected and Autonomous Vehicles (CAV) team at HORIBA MIRA.
Safety assurance for autonomous vehicles
Whilst the prospect of CAM is seen by many as an opportunity to improve mobility and drive economic growth, there are many who have understandable concerns about new and unproven technology; history has shown that the introduction of revolutionary transport technology inevitably leads to new and previously unforeseen hazards. It is therefore imperative that robust test methodologies are developed and applied from the start.
Providing assurance of safety is made especially challenging by the sheer complexity of both the systems themselves and their ‘Operational Design Domains’ (i.e. the environment they operate within). Every driver will be used to encountering unusual and unexpected occurrences, such as confusing road markings or erratic behaviour of other road users, but whereas humans can adapt to novel situations, machines cannot. Therefore, the engineering and testing of the vehicle must cover every conceivable permutation.
In addition to the occasional rare and unusual event, normal driving on public roads also involves a large proportion of repetitive, uneventful driving, making for an inefficient test method. What is therefore needed is an approach that allows the automated vehicle to be deliberately exposed to a wide range of different scenario permutations as part of a designed test programme, eliminating the excessive repetition whilst ensuring broad coverage of the challenges that the system will face in the real world. We refer to this approach as ‘Scenario-Based Testing’.
This requires the ability to control what happens in the scenario, including the behaviour of other road users. There are two main ways to do this:
- Simulation testing, where the parameters in the scene (e.g. light level, speed of an oncoming vehicle) can be fully controlled. This allows a vast range of scenarios to be tested relatively quickly and cheaply.
- Controlled environment (proving ground) testing, allowing some (but not all) parameters to be controlled. The use of robotic soft target actors, such as those shown in the image, enables precise control and allows challenging scenarios to be conducted safely.
The role of CAM Testbed UK
Zenzic is driving the creation of a world-leading testing capability through their funding and support for CAM Testbed UK, a network of test facilities providing a diverse range of environments. HORIBA MIRA is contributing to this through:
- The construction of ‘Assured CAV Highway’, a large and fully configurable test area that will allow maximum flexibility in the range of scenarios possible;
- The construction of ‘Assured CAV Parking’, which incorporates a multi-story car park and various parking bay permutations to allow comprehensive testing of automated parking systems;
- The modelling of various urban and country road scenes in the virtual domain within ‘Assured CAV Routes’, created as part of the Midlands Future Mobility project. This will allow a wide variety of scenarios to be reproduced within a realistic simulation environment.
The UK Connected and Automated Mobility Roadmap 2030: CAM Creators Update
The Roadmap is a tool that shows a single vision of interdependencies and addresses developments needed to achieve connected and automated mobility at scale by 2030. This brand-new release provides key insights into specific progress within the UK, highlighting details of 117 CAM Creators; the organisations who are actively making advancements in the UK. HORIBA MIRA has contributed through providing feedback to inform the Roadmap updates and involvement in a range of projects that are helping UK industry realise the vision.