A selection of ITRL's ongoing master thesis projects
Reinforcement Learning for Autonomous Driving
Reinforcement learning (RL) is widely used for games and robotic manipulators but applications to autonomous vehicles are still challenging due to high complexity of urban environments. ITRL is interested in evaluating two simulation environments: Carla (open source) and VTD (proprietary) for testing state of the art RL methods applied to auonomous vehicles as well as improving on current RL methods and solving novel problems with these simulation environments as tools.This thesis project will evaluate the two simulation environments and use them to solve an interesting problem in Autonomous Driving with Reinforcement Learning methods.
Innovation Diffusion Dynamics and Behavior of Actors in Road Freight Transportation
This study is going to investigate the potential dynamics and behavior of different actors for adoption self-driving trucks. Furthermore, the paper will present a model used to estimate the future adoption of autonomous trucks. By examining the market penetration rate of self-driving trucks, the hope is that the paper can serve as a facilitator for policymakers in infrastructure and legislation planning purposes.
One of the largest challenges that mankind is currently facing has been recognized to be climate change. It is estimated by the year 2050 that the consumption will be three times more than what is sustainable for planet earth. Moving towards climate neutrality and decoupling economic growth from resource use, and at the same time ensure long-term competitiveness becomes highly important.
Circular economy has in recent years become an area of interest for firms in their search to become more sustainable. The new type of business models allows end users to resell or return their goods with the intention to find a new user or purpose. Previous research has been done on certain services and applications, but little attention has been shown to how these new circular services will affect our transport demand. Therefore, this thesis aims to model the future demand of transportation related to circular services. Further, the study aims to evaluate the environmental impact of transportation in circular services in comparison to traditional linear supply chains.
Development of HMI for fault monitoring of driverless vehicles
The project involves the development of a Human Machine Interface for diagnosis and decision making of faulting driverless vehicles, in a control tower user case.
This thesis is a collaboration between Scania and ITRL.
