What is it about?
It is a joint work with Franco Zambonelli and Giacomo Cabri with the ambition to shed light on the issue of coordination of autonomous vehicles to achieve cooperative driving, and the tradeoff between coordination enforcement and the degree of autonomy that vehicles have while participating in collective decision making. In brief, the paper first categorises traffic-related problems in a bi-dimensional taxonomy stemming from concurrency and coordination research. Then we describe such problems in view of typical concurrency and coordination terms, such as safety and liveness properties. We then shift attention to solutions to such problems, and categorise approaches on a “autonomy scale” ordering the degree of autonomy left to vehciles in decision making during the coordination process. Finally, we frame selected papers from the literature in the defined autonomy classes.
Featured Image
Photo by Cameron Venti on Unsplash
Why is it important?
Most of the current industrial and applied research in the area of autonomous vehicles concerns the methods and tools to enable individual vehicles to hit the road safely. However, it is getting increasingly recognized that, to take full advantage of autonomous vehicles so as to enable overall traffic and pollution reduction, a number of situations will compulsorily require coordinating the relative activities and movements of vehicles. This trend is witnessed by initiatives such as the Grand Cooperative Driving Challenge, which in its latest edition explicitly focused on cooperative automated driving. Examples of very diverse situations that require proper and careful coordination among groups of vehicles include: crossing intersections, entering a motorway, platooning, organizing urban deployment and rides for fleets of ride-sharing vehicles, trying to improve parking occupancy and reduce parking times. Supporting such coordination needs implies devising effective mechanisms and strategies to support coordination activities.
Perspectives
Our aim is not to provide quantitative comparisons among existing approaches, although we make some qualitative comparisons. Nevertheless, our framework can definitely help future research in this direction, by suggesting guidelines on the types of solutions that should be included in such quantitative comparisons. We also emphasize that this article is not a systematic literature review. From the literature, we tried to identify and discuss what we consider key proposals in the area, in order to give readers a sound overview of problems and solutions, each properly supported by relevant examples. To the best of our knowledge, this is the first survey that investigates the problem of coordinating autonomous vehicles in a uniform way, and the first that proposes a comprehensive taxonomy of coordination problems for autonomous vehicles and an organized list of potential solutions.
Stefano Mariani
Universita degli Studi di Modena e Reggio Emilia
Read the Original
This page is a summary of: Coordination of Autonomous Vehicles, ACM Computing Surveys, March 2021, ACM (Association for Computing Machinery),
DOI: 10.1145/3431231.
You can read the full text:
Contributors
The following have contributed to this page
