A masters thesis about development of a framework of infrastructure requirements to analyse routes for use in autonomous bus systems.
Author: Victor Berndtsson and Oskar Vamling
Supervisor and examiner: Gunnar Stefansson
Department of Technology Management and Economics
Chalmers University of Technology
New technology, and especially technology related to autonomous vehicles (AV), will create new demands on infrastructure and the infrastructure owners. The purpose of the report is to create a framework of general characteristics that a route should possess in order to make it feasible to implement AVs on routes, and to analyse a few routes on roads that is owned by the Swedish Transportation Administration, using the framework for evaluation of the suitability of AV operations. In addition to the infrastructure requirements, future predictions when the technology is expected to be implemented, and which application that is suitable to use for autonomous buses is also included. The report mainly focuses on passenger transportation, but the framework is a general guideline that could in practice be used for other applications as well.
To formulate the requirements a literature study was conducted in conjunction with interviews with representatives of different companies and organisations that are involved in the development and operations of AVs in relation to passenger transportation. The framework was then formulated using the data collected in the literature review and using the answers from the interviews the different categories was classified either low, medium, and high depending on how many of the interviewees covered the subject. The routes to be analysed was selected from four different road categories; highway, four-lane road, separated 2-lane road, and 2-lane road. One route of each category was selected depending on the combined traffic volume of cars and public transit passengers, where it was found that there were the most potential users of the service, if implemented. For the future predictions, mainly data from the interviews was used to create a time-line but it was also supplemented by data from the literature study. As for the potential applications of AVs in public transport, both literature and interviews is used.
The study found that the eight areas of road layout, signs, intersections, lane markings, maintenance, digital maps, V2I/I2V, and connectivity should be included in the framework, the first five classified as physical infrastructure and the remaining three under digital infrastructure. The most important factors, that was mentioned the most in the interviews, is the road layout and connectivity, due to their direct and heavy impact on the feasibility in AV operations. The category that is deemed the least important is the lane markings, mostly due to existing technologies that is able to function without them even though AV operations become easier if they are present due to less difficult to read situations. The rest of the categories is classified as having roughly the same importance, somewhere in-between the previously mentioned characteristics. Each of the categories was each further divided into subcategories that each had either positive or negative effects on the evaluation of the routes, depending on the factors are present or not. The analysed routes had different prerequisites in almost all categories making the recommendation of a specific route, or route type, hard to do because of all the factors involved that is going to vary from case to case. The result is therefore to not generally recommend any specific type of road without analysing all context for the specific application in detail first. The general predictions made for when AVs will become available was a span between 2025 to 2045, with most interviewees and literature agreeing on that between 2030 and 2035 was the most likely for AVs with limited capabilities. The potential application of AVs in public transit is mainly concentrated to two areas, first- and last-mile transit as well as hub-to-hub transport. The implementation of AVs in public transit would likely bring the benefits of either operating current systems at lower cost, or to be able to expand the system without increasing the cost as the operating cost of AVs are lower than conventional vehicles. Overall, it can be said that different AVs are suitable for different situations and a deeper analysis of the intended application is necessary to determine the suitability.