Register now After registration you will be able to apply for this opportunity online.
DynaGELLO: A Low Cost "Puppet" Teleoperation System for the DynaArm
The GELLO system proposed in [1] is a low-cost “puppet” robot arm that is used to teleoperate a larger, main robot arm. This project aims to adapt this open source design to enable teleoperation of the DynaArm, which is a robot manipulator arm custom designed by the Robotic Systems Lab to be mounted on the ANYmal quadruped platform. Such a system provides a simplification over the existing DynaArm teleoperation interface consisting of a second identical DynaArm used purely as a human interface device [2], which may be an unnecessarily expensive and cumbersome solution. The system developed may have applications in remote teleoperation for industrial inspection or disaster response scenarios, as well as providing an interface for training imitation learning models, which may optionally be explored as time permits.
[1] Wu, Philip et al. "GELLO: A General, Low-Cost, and Intuitive Teleoperation Framework for Robot Manipulators". arXiv preprint (2024)
[2] Fuchioka, Yuni et al. AIRA Challenge: Teleoperated Mobile Manipulation for Industrial Inspection. Youtube Video. (2024)
Teleoperation systems for robot arms have wide-ranging applications in industrial inspection, search-and-rescue, and more recently as a popular approach for collecting datasets for imitation learning. Human teleoperation interfaces have ranged from game controllers, 6DOF spacemouse devices, haptic devices, exoskeletons, hand pose tracking with computer vision (such as those mounted on virtual reality goggles), and a second robot arm. In particular, [2] and [3] present systems where the human operator backdrives a robot arm to specify commands to the main teleoperated arm, which was shown to enable simple controller implementation and easy operation due to the clear relationships between leader and follower device motions.
At the Robotics Systems Lab, there exists a teleoperation system where the DynaArm, which is a robot arm custom designed in-house in the lab, is teleoperated by another identical DynaArm [2]. However, it can be expensive and cumbersome to have a second DynaArm dedicated purely as a human interface device. In contrast, [1] presents a system where a small “puppet” arm is used to teleoperate the main arm, leveraging its identical kinematic structure despite its low cost.
In this project, we would like to create a GELLO-like system that can be used to teleoperate the DynaArm. This involves both the design and construction of the hardware device, as well its software integration with the DynaArm robots. Although the minimal project requirements are well defined, there are several open-ended extension directions depending on the aptitude and interests of the student–-especially with regards to whether the project is carried out as a Bachelor’s, Semester, or Thesis project.
[1] Wu, Philip et al. "GELLO: A General, Low-Cost, and Intuitive Teleoperation Framework for Robot Manipulators". arXiv preprint (2024)
[2] Fuchioka, Yuni et al. "AIRA Challenge: Teleoperated Mobile Manipulation for Industrial Inspection". Youtube Video. (2024)
[3] Zhao, Tony et al. “Learning fine-grained bimanual manipulation with low-cost hardware”. RSS (2022)
Teleoperation systems for robot arms have wide-ranging applications in industrial inspection, search-and-rescue, and more recently as a popular approach for collecting datasets for imitation learning. Human teleoperation interfaces have ranged from game controllers, 6DOF spacemouse devices, haptic devices, exoskeletons, hand pose tracking with computer vision (such as those mounted on virtual reality goggles), and a second robot arm. In particular, [2] and [3] present systems where the human operator backdrives a robot arm to specify commands to the main teleoperated arm, which was shown to enable simple controller implementation and easy operation due to the clear relationships between leader and follower device motions.
At the Robotics Systems Lab, there exists a teleoperation system where the DynaArm, which is a robot arm custom designed in-house in the lab, is teleoperated by another identical DynaArm [2]. However, it can be expensive and cumbersome to have a second DynaArm dedicated purely as a human interface device. In contrast, [1] presents a system where a small “puppet” arm is used to teleoperate the main arm, leveraging its identical kinematic structure despite its low cost.
In this project, we would like to create a GELLO-like system that can be used to teleoperate the DynaArm. This involves both the design and construction of the hardware device, as well its software integration with the DynaArm robots. Although the minimal project requirements are well defined, there are several open-ended extension directions depending on the aptitude and interests of the student–-especially with regards to whether the project is carried out as a Bachelor’s, Semester, or Thesis project.
[1] Wu, Philip et al. "GELLO: A General, Low-Cost, and Intuitive Teleoperation Framework for Robot Manipulators". arXiv preprint (2024)
[2] Fuchioka, Yuni et al. "AIRA Challenge: Teleoperated Mobile Manipulation for Industrial Inspection". Youtube Video. (2024)
[3] Zhao, Tony et al. “Learning fine-grained bimanual manipulation with low-cost hardware”. RSS (2022)
- Hardware design and build for a “puppet” robot arm, adapted from the open source GELLO design [1].
- Software implementing a ROS-based interface between the designed arm and the DynaArm.
- (Optional) Implement active gravity compensation.
- (Optional) Explore options for bilateral force feedback control.
- (Optional) Devise a method to enable locomotion control, to enable teleoperated locomanipulation.
- (Optional) Deploy the system on an imitation learning pipeline.
- Hardware design and build for a “puppet” robot arm, adapted from the open source GELLO design [1]. - Software implementing a ROS-based interface between the designed arm and the DynaArm. - (Optional) Implement active gravity compensation. - (Optional) Explore options for bilateral force feedback control. - (Optional) Devise a method to enable locomotion control, to enable teleoperated locomanipulation. - (Optional) Deploy the system on an imitation learning pipeline.
Please send a mail to yfuchioka@student.ethz.ch with the Subject: "Application - Your Name - DynaGELLO" with:
- BS/MS Transcript of Records
- CV
- Short motivation for your interest in the project
Please send a mail to yfuchioka@student.ethz.ch with the Subject: "Application - Your Name - DynaGELLO" with:
- BS/MS Transcript of Records - CV - Short motivation for your interest in the project
