• Google工程师：从零开始学习无人驾驶技术 ——端到端无人驾驶
• ROS Robotics Projects（8）自动驾驶汽车
• 在ROS Kinetic和Gazebo 8中使用智能汽车仿真演示

demo

• A 4 servo arm control program learn to perform reaching task by deep reinforcement learning.
• This module allows to carry out a two-finger stable grasp by means of a controller developed using reinforcement learning.
• Solutions to various tasks using reinforcement learning and optimal control (trajectory optimization)

ROS-RL-Kinetic

• Factory_RL_Gazebo
• gym-gazebo-turtlebot-RL
• Deep RL Arm Manipulation
• QLearning-NAO_plays_Agario
• Open Ai Gym extension for robotics
• Reinforcement learning for ROS simulation
• a turtlebot reinforcement learning in Gazebo
• Reinforcement Learning framework for Robotics
• Reinforcement Learning Framework and Repository
• Autonomous Drifting using Reinforcement Learning
• Reinforcement Learning with Gazebo & Rospy & Keras
• Deep (Reinforcement?) Learning for stage simulated robots
• An OpenAI gym extension for using Gazebo known as gym-gazebo
• ROS/Gazebo simulations for reinforcement learning exercises
• Overcoming Exploration in Reinforcement Learning with Demonstrations
• Deep reinforcement learning for UAV in Gazebo simulation environment
• A reinforcement learning environment using Gazebo, ROS and openAI gym
• ROS implementation of multi-task reinforcement learning for Turtlebot 2
• SLAM-Safe Planner: Preventing Monocular SLAM Failure using Reinforcement Learning
• A drone control system based on deep reinforcement learning with Tensorflow and ROS
• Implement DQN in Stage simulator in ROS for Deep Reinforcement Learning using Google DeepMind
• S-RL Toolbox: Reinforcement Learning (RL) and State Representation Learning (SRL) for Robotics
• RosLoveGym: for the great RL Dream

论文复现

• A Robotic Auto-Focus System based on Deep Reinforcement Learning

ROS-RL-Indigo

• ROS package for reinforcement learning on Baxter robot
• Training Neural Networks for Event-Based End-to-End Robot Control
• Packages related to autonomous mobile robots using Deep Learning and Reinforcement algorithms
• This project is a simulation set- up for ROS Indigo upon which Reinforcement LEarning Algorithms can be used and applied
• Reinforcement for state of the art visual slam algorithms with Deep Learning based solutions. Part of my master thesis at University of Freiburg
• Using a Unity3D package to connect to ROS topics, robots can be simulated in the game-engine with communication to a Reinforcement Learning Algorithm

Self Driving

• Self-Driving Truck in Euro Truck Simulator 2, trained via Reinforcement Learning