Reinforcement Learning

Enabling deep reinforcement learning autonomous driving by 3D-LiDAR point clouds
SPIE, 14th International Conference on Digital Image Processing (ICDIP 2022)
Yuhan Chen, Rita Tse, Michael Bosello, Davide Aguiari, Su-Kit Tang, Giovanni Pau
PDF Cite Code Project DOI
Enabling deep reinforcement learning autonomous driving by 3D-LiDAR point clouds
Train in Austria, Race in Montecarlo: Generalized RL for Cross-Track F1tenth LIDAR-Based Races
IEEE, 19th Annual Consumer Communications & Networking Conference (CCNC 2022)
Michael Bosello, Rita Tse, Giovanni Pau
PDF Cite Code Project Slides Video DOI
Train in Austria, Race in Montecarlo: Generalized RL for Cross-Track F1tenth LIDAR-Based Races
F1tenth-RL
LIDAR data are used as input of a Deep Q-Network on a realistic 1/10 scale car prototype controlled by ROS. The robot-driver learns how to run in race tracks by exploiting the experience gained through a mechanism of rewards that allow the agent to learn without human supervision
PDF Code Slides Video DOI
F1tenth-RL
Integrating BDI and Reinforcement Learning: the Case Study of Autonomous Driving
University of Bologna, Master’s thesis
Michael Bosello
PDF Cite Code Slides Video DOI Seminar Project jason-rl Project f1tenth-rl
Integrating BDI and Reinforcement Learning: the Case Study of Autonomous Driving
From Programming Agents to Educating Agents – A Jason-Based Framework for Integrating Learning in the Development of Cognitive Agents
Springer, 7th International Workshop on Engineering Multi-Agent Systems (EMAS 2019)
Michael Bosello, Alessandro Ricci
PDF Cite Code Project Slides DOI
From Programming Agents to Educating Agents – A Jason-Based Framework for Integrating Learning in the Development of Cognitive Agents
Robot Drivers: Learning to Drive by Trial & Error
IEEE, 15th International Conference on Mobile Ad-Hoc and Sensor Networks (MSN 2019)
Michael Bosello, Rita Tse, Giovanni Pau
PDF Cite Code Slides Video DOI
Robot Drivers: Learning to Drive by Trial & Error
Jason-RL
Integrating Belief-Desire-Intention agents and RL, i.e., integrating planning and learning. The framework has been tested on a simulated urban scenario based on f1tenth, where BDI plans are used for high-level path planning and RL for low-level control.
PDF Code Slides DOI Seminar F1tenth-RL-BDI PDF F1tenth-RL-BDI Code F1tenth-RL-BDI Slides F1tenth-RL-BDI Video
Jason-RL