TY - GEN
T1 - Visual navigation of wheeled mobile robots using deep reinforcement learning
T2 - ASME 2020 Dynamic Systems and Control Conference, DSCC 2020
AU - Nwaonumah, Ezebuugo
AU - Samanta, Biswanath
N1 - Publisher Copyright:
Copyright © 2020 ASME
PY - 2020
Y1 - 2020
N2 - A study is presented on applying deep reinforcement learning (DRL) for visual navigation of wheeled mobile robots (WMR), both in simulation and real-time implementation under dynamic and unknown environments. The policy gradient based asynchronous advantage actor critic (A3C), has been considered. RGB (red, green and blue) and depth images have been used as inputs in implementation of A3C algorithm to generate control commands for autonomous navigation of WMR. The initial A3C network was generated and trained progressively in OpenAI Gym Gazebo based simulation environments within robot operating system (ROS) framework for a popular target WMR, Kobuki TurtleBot2. A pre-trained deep neural network ResNet50 was used after further training with regrouped objects commonly found in laboratory setting for target-driven visual navigation of Turlebot2 through DRL. The performance of A3C with multiple computation threads (4, 6, and 8) was simulated and compared in three simulation environments. The performance of A3C improved with number of threads. The trained model of A3C with 8 threads was implemented with online learning using Nvidia Jetson TX2 on-board Turtlebot2 for mapless navigation in different real-life environments. Details of the methodology, results of simulation and real-time implementation through transfer learning are presented along with recommendations for future work.
AB - A study is presented on applying deep reinforcement learning (DRL) for visual navigation of wheeled mobile robots (WMR), both in simulation and real-time implementation under dynamic and unknown environments. The policy gradient based asynchronous advantage actor critic (A3C), has been considered. RGB (red, green and blue) and depth images have been used as inputs in implementation of A3C algorithm to generate control commands for autonomous navigation of WMR. The initial A3C network was generated and trained progressively in OpenAI Gym Gazebo based simulation environments within robot operating system (ROS) framework for a popular target WMR, Kobuki TurtleBot2. A pre-trained deep neural network ResNet50 was used after further training with regrouped objects commonly found in laboratory setting for target-driven visual navigation of Turlebot2 through DRL. The performance of A3C with multiple computation threads (4, 6, and 8) was simulated and compared in three simulation environments. The performance of A3C improved with number of threads. The trained model of A3C with 8 threads was implemented with online learning using Nvidia Jetson TX2 on-board Turtlebot2 for mapless navigation in different real-life environments. Details of the methodology, results of simulation and real-time implementation through transfer learning are presented along with recommendations for future work.
KW - Asynchronous advantage actor-critic (A3C)
KW - Convolutional neural network (CNN)
KW - Deep neural network (DNN)
KW - Deep reinforcement learning (DRL)
KW - Edge computing
KW - Long short-term memory (LSTM)
KW - Machine learning (ML)
KW - Robot operating system (ROS)
KW - Transfer learning
KW - Visual navigation
UR - http://www.scopus.com/inward/record.url?scp=85101460837&partnerID=8YFLogxK
U2 - 10.1115/DSCC2020-3279
DO - 10.1115/DSCC2020-3279
M3 - Conference article
AN - SCOPUS:85101460837
T3 - ASME 2020 Dynamic Systems and Control Conference, DSCC 2020
BT - Adaptive/Intelligent Sys. Control; Driver Assistance/Autonomous Tech.; Control Design Methods; Nonlinear Control; Robotics; Assistive/Rehabilitation Devices; Biomedical/Neural Systems; Building Energy Systems; Connected Vehicle Systems; Control/Estimation of Energy Systems; Control Apps.; Smart Buildings/Microgrids; Education; Human-Robot Systems; Soft Mechatronics/Robotic Components/Systems; Energy/Power Systems; Energy Storage; Estimation/Identification; Vehicle Efficiency/Emissions
PB - American Society of Mechanical Engineers
Y2 - 5 October 2020 through 7 October 2020
ER -