Print Email Facebook Twitter Multi-Agent Deep Reinforcement Learning for Automated Highway Driving Title Multi-Agent Deep Reinforcement Learning for Automated Highway Driving Author Bakker, Lou (TU Delft Mechanical, Maritime and Materials Engineering) Contributor Grammatico, Sergio (mentor) Degree granting institution Delft University of Technology Programme Mechanical Engineering | Systems and Control Date 2019-06-08 Abstract Recent advances in Deep Reinforcement Learning have sparked new interest in many different research topics, including Automated Highway Driving where agents model autonomous vehicles. The main advantage of Deep Reinforcement Learning is that the training algorithm is adaptable to its environment. In highway driving, researchers often simplify the framework of an agent by using lower level controllers and observers. However, agent observations do not yet include lane change intentions of surrounding vehicles. Resulting agents were able to drive on a maximum of three lanes and unfit to drive in lane changing environments. We aim to simplify the current state-of-the-art agent frameworks even further to improve performance. We also believe that observing other vehicles lane-change intent, or blinker status, is essential for collision avoidance in a highway driving environment. In this paper, we try to implement multi-agent Deep Reinforcement Learning on a six-lane highway, including lane changes. After training, agents are able to avoid collisions while reaching destination lanes. Moreover, a lane-selection strategy according to desired speed evolved from open freeway training. Subject Deep Reinforcement LearningMulti-Agent Systemautomated driving systemsAutomated driving To reference this document use: http://resolver.tudelft.nl/uuid:9b825cbf-0d95-4cd7-b531-337885b89892 Part of collection Student theses Document type master thesis Rights © 2019 Lou Bakker Files PDF mscThesis_LJB_Final.pdf 1.53 MB Close viewer /islandora/object/uuid:9b825cbf-0d95-4cd7-b531-337885b89892/datastream/OBJ/view