Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Aug 14;8(4):364.
doi: 10.3390/biomimetics8040364.

Bridging Locomotion and Manipulation Using Reconfigurable Robotic Limbs via Reinforcement Learning

Haoran Sun  1   2 , Linhan Yang  1   2 , Yuping Gu  1   2 , Jia Pan  2 , Fang Wan  3   4 , Chaoyang Song  1   5
Affiliations

Bridging Locomotion and Manipulation Using Reconfigurable Robotic Limbs via Reinforcement Learning

Haoran Sun et al. Biomimetics (Basel). .

Abstract

Locomotion and manipulation are two essential skills in robotics but are often divided or decoupled into two separate problems. It is widely accepted that the topological duality between multi-legged locomotion and multi-fingered manipulation shares an intrinsic model. However, a lack of research remains to identify the data-driven evidence for further research. This paper explores a unified formulation of the loco-manipulation problem using reinforcement learning (RL) by reconfiguring robotic limbs with an overconstrained design into multi-legged and multi-fingered robots. Such design reconfiguration allows for adopting a co-training architecture for reinforcement learning towards a unified loco-manipulation policy. As a result, we find data-driven evidence to support the transferability between locomotion and manipulation skills using a single RL policy with a multilayer perceptron or graph neural network. We also demonstrate the Sim2Real transfer of the learned loco-manipulation skills in a robotic prototype. This work expands the knowledge frontiers on loco-manipulation transferability with learning-based evidence applied in a novel platform with overconstrained robotic limbs.

Keywords: loco-manipulation; reconfigurable robot; reinforcement learning.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Locomotion as manipulation of terrain to unify the formulation of loco-manipulation problems, providing a unified observation as input to the GNN and MLP models.
Figure 2
Figure 2
Horizontal robot configuration (top) and vertical robot configuration (bottom). Locomotion tasks in the simulation are to control the rotation of the robot base; manipulation tasks are to control the rotation of the translucent plate. The frame icons on top of the robots indicate the current goal rotations of the robot base or the plate.
Figure 3
Figure 3
Performance of loco-manipulation transfer learning for the horizontal configuration (left), and the vertical configuration (right).
Figure 4
Figure 4
Performance of unified loco-manipulation learning (middle) for the horizontal configuration (left) and the vertical configuration (right).
Figure 5
Figure 5
Robot hardware setup for the sim-to-real test of the horizontal robot configuration.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Billeschou P., Bijma N.N., Larsen L.B., Gorb S.N., Larsen J.C., Manoonpong P. Framework for Developing Bio-Inspired Morphologies for Walking Robots. Appl. Sci. 2020;10:6986. doi: 10.3390/app10196986. - DOI
    1. Shi F., Homberger T., Lee J., Miki T., Zhao M., Farshidian F., Okada K., Inaba M., Hutter M. Circus ANYmal: A Quadruped Learning Dexterous Manipulation with its Limbs; Proceedings of the 2021 IEEE International Conference on Robotics and Automation (ICRA); Xi’an, China. 30 May–5 June 2021; pp. 2316–2323.
    1. Lynch K.M. Ph.D. Thesis. Carnegie Mellon University; Pittsburgh, PA, USA: 1996. Nonprehensile Robotic Manipulation: Controllability and Planning.
    1. Johnson A.M., Koditschek D.E. Legged Self-Manipulation. IEEE Access. 2013;1:310–334. doi: 10.1109/ACCESS.2013.2263192. - DOI
    1. Mason M.T. Toward Robotic Manipulation. Annu. Rev. Control. Robot. Auton. Syst. 2018;1:1–28. doi: 10.1146/annurev-control-060117-104848. - DOI