Autonomous Vehicle and Risk Assessment

Abstract: Accidents with vehicles happen more often than they should, whether it is by road traffic accidents or maritime accidents out in the sea. One big contributing factor is human error. Distraction, influence of alcohol/drugs, carelessness and speeding are examples of factors that can lead to severe injuries for the passengers, the surrounding people and to infrastructure. By transferring human maneuvers to a more autonomous operation, it is believed that safety will increase. This is true for all vehicle types, such as cars, ships, drones and trains – which are the vehicles the thesis is focusing on. Autonomous vehicles are developing fast, and it is also the case of their risk assessment methods. Research in this field is new and a general applicable method that works on most systems and situations does not exist. Not even the industries have a clear image of how risk assessment should be done on such vehicles. At the same time, new security risk emerges. When a technology becomes less human reliant and have fewer manual functions, malicious attackers find new areas to strike. The risk assessment method must thus cover both safety and security perspectives. The thesis’ aim is to increase knowledge about risk assessment of autonomous vehicle and to analyze the found information in order to assemble a holistic risk assessment framework. A Scoping Study and consultative interviews were conducted to investigate the current knowledge about assessment methods and the results formed the RAAV framework. The RAAV framework is based on a customized S&S model.