The public defence of Shahriar Hasan's doctoral thesis
The public defence of Shahriar Hasan's doctoral thesis in Computer Science will take place at Mälardalen University, room Paros (Västerås Campus) at 09:30 on 5 October 2023 and via Zoom.
The title of the thesis is On Transient Communication Outages among Collaborating Connected and Automated Vehicles.
The faculty examiner is Professor Javier Manuel Gozalvez Sempere, Miguel Hernández University (Spain).
The examining committee consists of Docent Mikael Asplund, Linköping University (Sweden); Professor Alexey Vinel, Halmstad University (Sweden); Professor Maria Kihl, Lund University (Sweden).
Reserve is Professor Jakob Axelsson, Mälardalen University (Sweden).
The doctoral thesis has serial number 384.
Summary
The advances in wireless communication technology over the recent years enable connecting automated vehicles and thus the possibility of creating safer, more environmentally friendly and intelligent transportation systems. In this context, the concept of platooning has received significant research attention. Platooning refers to a group of wirelessly connected automated vehicles forming a vehicle train that cooperates by exchanging information, such as speed, acceleration, steering angle, warnings, etc. Thanks to the vehicles communicating wirelessly and being able to use advanced sensor fusion technology, platooning can improve traffic safety, fuel efficiency, road capacity, and traffic flow significantly.
Despite these societal, environmental and economic opportunities, platooning, for safety reasons, has not yet been introduced on public roads. This is because each of the connected autonomous vehicles is a safety-critical system and the consequences of a possible failure could, in the worst case, endanger human life or cause extensive damage to property or the environment. The challenges are further aggravated due to the fact that the vehicles are traveling at high speed and with small gaps between the vehicles. In addition, there may be temporary outages in the wireless communication.
In this doctoral thesis, a new approach is proposed for modeling and handling temporary communication outages. This enables automatic handling of the errors that may occur when there is an outage in wireless communication. In addition, several strategies are proposed for emergency braking of the vehicle train to avoid collisions in case of hazards on the road. With the help of the strategies proposed in the thesis, a vehicle train can continue to travel forward with maintained speed and safety even when communication problems arise. Instead of completely discontinuing platooning, the distance between vehicles may be temporarily increased, possibly with a temporary decrease in fuel efficiency as a result. The thesis also uses Artificial Intelligence (AI) to enable the cooperating autonomous vehicles to predict if and when delays or outages in the wireless communication occur, as well as whether collisions between the vehicles could occur if one tries to brake with a certain force in a certain emergency situation. The information gathered with the help of AI is also used to understand the causes of a possible collision and to prevent it. The thesis thus investigates the complexity of the technologies that enable platooning, addresses the problem of temporary outages in the wireless communication and uses AI to improve the safety, autonomy and efficiency of collaborating connected autonomous vehicles and vehicle trains.