Abstract
In this study, we present a unified motion planner with low-level controller for continuous control of a differential drive mobile robot under variable payload values. Our deep reinforcement agent takes 11 dimensional state vector as input and calculates each wheel's torque value as a 2 dimensional output vector. These torque values are fed into the dynamic model of the robot, and lastly steering commands are gathered. In previous studies, intersection navigation solutions that uses deep - RL methods, have not been considered with variable payloads. Our study is focused specifically on service robotic applications where payload is subject to change. To the best of our knowledge, this is the first study in the literature that investigates intersection - navigation problem under variable payloads using deep-RL. In this paper, deep-RL based motion planning is performed by considering both kinematic and dynamic constraints. According to the simulations in a dynamic environment, the agent succesfully navigates to target with 98.2% success rate in test time with unseen payload masses during training. Another agent is also trained without payload randomization for comparison. Results show that our agent outperforms the other agent, that is not aware of its own payload, with more than 40% gap.
Original language | English |
---|---|
Title of host publication | Proceedings - 3rd IEEE International Conference on Robotic Computing, IRC 2019 |
Publisher | Institute of Electrical and Electronics Engineers Inc. |
Pages | 32-37 |
Number of pages | 6 |
ISBN (Electronic) | 9781538692455 |
DOIs | |
Publication status | Published - 26 Mar 2019 |
Event | 3rd IEEE International Conference on Robotic Computing, IRC 2019 - Naples, Italy Duration: 25 Feb 2019 → 27 Feb 2019 |
Publication series
Name | Proceedings - 3rd IEEE International Conference on Robotic Computing, IRC 2019 |
---|
Conference
Conference | 3rd IEEE International Conference on Robotic Computing, IRC 2019 |
---|---|
Country/Territory | Italy |
City | Naples |
Period | 25/02/19 → 27/02/19 |
Bibliographical note
Publisher Copyright:© 2019 IEEE.
Keywords
- Motion Planning
- Navigation
- Reinforcement Learning