Autonomous Vehicles (AVs) will need to perform the driving function safely within a test parameter space of almost infinite complexity and variability. Different road geometries, combinations of other road user movements and combinations of environmental conditions all contribute to this complexity. The software and sensors that constitute the Automated Driving System (ADS) must therefore be highly sophisticated, able to perceive the world around themselves and make safe, predictable decisions. Demonstration of confidence in such a system is a major step towards the introduction of AVs to our roads.
It is unlikely to be viable to achieve appropriate confidence solely from physical ‘real-world’ testing. This physical testing is time consuming, has limited control of test parameters, requires a huge number of miles (even with a targeted focus on the difficult use cases) and requires repeating every time the system is changed. Testing in a virtual environment has the potential to address these issues and when combined with physical testing should enable rigorous, controlled and timely evaluation of ADS systems.
This study seeks to explore simulation capabilities from the perspective of developing and testing an ADS, and also suggests how these tools could potentially be used to support a regulatory objective.
The Transport Systems Catapult were instructed by The Centre for Connected and Autonomous Vehicles (CCAV) and the DfT’s International Vehicle Standards division (IVS) to deliver a study investigating information that Government could ask an organisation seeking to apply for an exemption to a Construction and Use Regulation, to enable testing of automated vehicles on UK public roads in a way which could be interpreted as inconsistent with UK law, to help inform a decision by Ministers on whether to provide approval for that test.
The report details evidence needed to demonstrate that the test will be carried out safely and responsibly, and considerations Government officials need to bear in mind when assessing information.
The report also summarises the international regulatory framework and investigates evidence of public attitudes towards automated vehicles.
The primary focus of this white paper, produced by Catherine Menon of Hertfordshire University, is on the intersection of safety and ethics for automated vehicles. The material for this white paper has its origins in the research performed and the workshops organised by the Transport Systems Catapult on behalf of the Department for Transport.
Autonomous systems (a category which includes automated vehicles) have been proposed for use in multiple domains, with examples including nuclear containment, defence systems, health and transport. This paper discusses the ethical landscape surrounding the introduction and operation of autonomous vehicles as a form of transport on the public road network. Use of autonomous vehicles in other domains (e.g. as a defence capability or a form of medical transport) is likely to impose ethical requirements which go beyond the scope of this document, and for which this paper refers the reader to existing literature.
This report summarises the output from Work Package 7 (titled ‘Autonomous and Connected Vehicles Impact’); a deliverable of the Station Innovation 2 project which has been undertaken by the Transport Systems Catapult (TSC) on behalf of the Department for Transport (DfT).
This work package considers how Connected and Automated Vehicles (CAVs) may develop over the next decade and how these could be integrated with future station designs and operations in order to provide an efficient interchange in a multi-modal transport system.
This project investigates abnormal driving situations, classifies them into a taxonomy and goes on to consider potential solutions for how to handle some of the more challenging automated driving scenarios.
The majority of the driving task is relatively routine, but occasionally situations demand the driver to take action which is out of the ordinary or requires the driver to make an interpretation of the situation and act in a considered manner (common sense driving). Such situations could present challenges to Automated Vehicles (AVs) and their developers. AVs will need to adhere to rules governing their behaviour. If the rules and regulations governing vehicle behaviour within abnormal situations are not clear, then this could lead to unexpected or undesirable behaviour amongst AVs. Indeed, AVs may behave differently to the same abnormal situation depending on the AV manufacturer and the software algorithms that have been deployed.
This study investigates how planning, designing, appraisal, implementation and operation of road infrastructure could adapt as a result of the introduction of CAVs.
Highways authorities, public bodies, developers and other organisations rely on planning and guidance material to guide future transport provision and investment priorities. Connected and Automated Vehicles (CAVs) have the potential to revolutionise transport, but many planning and guidance documents remain silent on the issue. In some cases, this is because the research that contributed to these documents pre-dates the technological progress that has been made in recent years in relation to CAVs. In other cases, there may be a reluctance to comment on a future which can appear to be unclear and rapidly changing. What is certain, however, is that the more we discuss the potential opportunities and issues that CAVs present, and the more strategies that are developed for maximising the benefits of them, the more likely it is that a positive outcome from their implementation can be achieved.
This award winning paper by TSC Technologist Carl Goves presents the results of applying an artificial neural network to estimate traffic conditions 15 minutes into the future on a section of motorway within the UK.