Details
| Original language | English |
|---|---|
| Pages (from-to) | 56-67 |
| Number of pages | 12 |
| Journal | International Conference on Agents and Artificial Intelligence |
| Volume | 1 |
| Publication status | Published - 23 Feb 2025 |
| Event | 17th International Conference on Agents and Artificial Intelligence, ICAART 2025 - Porto, Portugal Duration: 23 Feb 2025 → 25 Feb 2025 |
Abstract
The field of mobile robotics has undergone a transformation in recent years due to advances in manipulation arms. One notable development is the integration of a 7-degree robotic arm into mobile platforms, which has greatly enhanced their ability to autonomously navigate while simultaneously executing complex manipulation tasks. As such, the key success of these systems heavily relies on continuous path planning and precise con trol of arm movements. In this paper, we evaluate a whole-body control framework that tackles the dynamic instabilities associated with the floating base of mobile platforms in a simulation closely modeling real-world configurations and parameters. Moreover, we employ reinforcement learning to enhance the controller’s per formance. We provide results from a detailed ablation study that shows the overall performance of various RL algorithms when optimized for task-specific behaviors over time. Our experimental results demonstrate the feasibility of achieving real-time control of the mobile robotic platform through this hybrid control frame work.
Keywords
- Reinforcement Learning, Simulated Mobile Robot Platform, Whole-Body Control
ASJC Scopus subject areas
- Computer Science(all)
- Artificial Intelligence
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: International Conference on Agents and Artificial Intelligence, Vol. 1, 23.02.2025, p. 56-67.
Research output: Contribution to journal › Conference article › Research › peer review
}
TY - JOUR
T1 - MIRSim-RL
T2 - 17th International Conference on Agents and Artificial Intelligence, ICAART 2025
AU - Li, Qingkai
AU - Ma, Zijian
AU - Li, Chenxing
AU - Liu, Yinlong
AU - Recker, Tobias
AU - Brauchle, Daniel
AU - Seyler, Jan
AU - Zhao, Mingguo
AU - Eivazi, Shahram
N1 - Publisher Copyright: © 2025 by SCITEPRESS– Science and Technology Publications, Lda.
PY - 2025/2/23
Y1 - 2025/2/23
N2 - The field of mobile robotics has undergone a transformation in recent years due to advances in manipulation arms. One notable development is the integration of a 7-degree robotic arm into mobile platforms, which has greatly enhanced their ability to autonomously navigate while simultaneously executing complex manipulation tasks. As such, the key success of these systems heavily relies on continuous path planning and precise con trol of arm movements. In this paper, we evaluate a whole-body control framework that tackles the dynamic instabilities associated with the floating base of mobile platforms in a simulation closely modeling real-world configurations and parameters. Moreover, we employ reinforcement learning to enhance the controller’s per formance. We provide results from a detailed ablation study that shows the overall performance of various RL algorithms when optimized for task-specific behaviors over time. Our experimental results demonstrate the feasibility of achieving real-time control of the mobile robotic platform through this hybrid control frame work.
AB - The field of mobile robotics has undergone a transformation in recent years due to advances in manipulation arms. One notable development is the integration of a 7-degree robotic arm into mobile platforms, which has greatly enhanced their ability to autonomously navigate while simultaneously executing complex manipulation tasks. As such, the key success of these systems heavily relies on continuous path planning and precise con trol of arm movements. In this paper, we evaluate a whole-body control framework that tackles the dynamic instabilities associated with the floating base of mobile platforms in a simulation closely modeling real-world configurations and parameters. Moreover, we employ reinforcement learning to enhance the controller’s per formance. We provide results from a detailed ablation study that shows the overall performance of various RL algorithms when optimized for task-specific behaviors over time. Our experimental results demonstrate the feasibility of achieving real-time control of the mobile robotic platform through this hybrid control frame work.
KW - Reinforcement Learning
KW - Simulated Mobile Robot Platform
KW - Whole-Body Control
UR - http://www.scopus.com/inward/record.url?scp=105001689090&partnerID=8YFLogxK
U2 - 10.5220/0013159400003890
DO - 10.5220/0013159400003890
M3 - Conference article
AN - SCOPUS:105001689090
VL - 1
SP - 56
EP - 67
JO - International Conference on Agents and Artificial Intelligence
JF - International Conference on Agents and Artificial Intelligence
SN - 2184-3589
Y2 - 23 February 2025 through 25 February 2025
ER -