Research into the infrastructure needs of autonomous vehicles and how their sensors interact with the road environment.
Through surveys and workshops with experts in the field, we will assign a weighting to each aspect of road infrastructure being assessed to produce an overall score for a road.
Data for a study area in Leeds, UK will be collected and fed into our Road Scoring Index to produce spatial insights and CAV routes through the city.
This project builds on an idea suggested by Dr Zia Wadud for development by Lawrence Penn as his master's dissertation project at the Institute for Transport Studies, University of Leeds. Whilst the idea is still being shaped, it is hoped that this project will provide a better understanding of how the application of the index score can provide insights into CAV readiness across our test bed area. This complementary approach allows us to share our findings in aid of existing work for further development in the future.
1. Discussion on the need for a Road Scoring Index
Our approach will begin by making the case for the introduction of CAVs. This will cover the state of the art of CAVs and the potential societal, economic and environmental benefits, such as improvements to road safety, optimising lane capacity and providing access for those with disabilities. We will then highlight the many concerns over the safety of CAVs in the initial stages of implementation and how a specifically developed road scoring index could help to identify and address these issues.
2. Understanding how CAVs uniquely perceive the road environment
The second stage of our approach will investigate the suite of sensors, artificial intelligence and machine learning processes, and the subsequent infrastructure requirements of CAVs to set ideal standards for a safe operating environment. This understanding will differentiate the needs of CAVs from human drivers and inform which aspects of the physical and digital road environment should be considered, as well as how they are graded in terms of quality.
3. Feasibility study, parameter and data review
Academic literature and expert consultation will be used to create a scoring taxonomy and methodology for each aspect of the road environment identified in the previous stage which reflects a CAVs unique requirements. The efficacy of the index is only as good as the quality of the inputted data and its ability to accurately capture the nature of the road environment. A data review will be undertaken to explore potential data sources that could feed into the index parameters.
The weightings of each parameter will be informed by expert consultation. Surveys will be sent out to academics and professionals in the fields of CAVs and road safety which will ask respondents to comment on and weight the parameters in terms of their relative importance. The results of this investigation will be presented and discussed in expert workshops where deeper discussion will be encouraged, and a consensus reached.
4. Data collection
Data will be collected from the multiple sources identified in the previous stage. The index will use a combination of primary and secondary datasets dependent on availability. The index test bed will be located locally in Leeds to keep the cost of primary data collection low.
5. Index application
This stage will involve a GIS exercise to apply the parameter scorings to a shapefile layer of the road network to map the index parameters and resultant compound index score.
6. Analysis and discussion of results
A report will be produced discussing the results of the index to provide insights and analysis of our test bed area, including measures to mitigate any safety issues arising during CAV deployment. A series of scenarios will be developed to investigate the impact of changing parameters such as different weather conditions and the maturity of V2I technology.
A framework to assess the suitability of the road network for CAVs that considers physical and digital infrastructure needs.
A report on the suitability of a test bed area of Leeds for CAVs. This will show which areas and routes are viable.
Discussion of next steps and wider application for public benefit.
GIS analysis will be undertaken to show the current readiness of the road network. This will enable key insights into suggested routes for autonomous vehicles, as well as a network-level assessment of CAV readiness across a given region.