Contact task execution by robot with non-rigid fixation
Start Date: 1st March 2024
Expected End Date: 1st September 2027
This PhD project will research the following problem: how to use robots with a non-rigidly fixed chassis to perform contact tasks on a much larger workpiece that has been placed arbitrarily on the floor of a factory. Contact tasks, such as grinding, polishing, or welding, typically involve high-precision, high-contact-force hybrid force/position control. To perform contact tasks on large workpieces, one feasible approach is to use mobile robots. The mobile chassis extends the robot’s workspace, enabling it to cover large workpieces. However, the introduction of a mobile chassis also results in non-rigid fixation of the robot to the ground, leading to reduced positioning accuracy, and decreased overall rigidity and stability of the robot. My research will focus on exploring and addressing these issues.
Research challenges include precise localization between the mobile robot and the workpiece, optimization of hybrid force/position control, and specific challenges posed by certain contact tasks (e.g., vibrations in grinding tasks).
Recent Updates
- Collaboration with Vaulta on robotic grinding to remove oxide layers from electrical contact points.
- Exploration of microphone-based vibration sensing during grinding as an additional modality to improve reliability.
- Development of a method using microphone feedback to control the robotic grinding process, replacing costly force sensors and providing a low-cost, easy-to-install alternative.
Outcomes
The future goals of this project are:
- Application of the microphone feedback grinding control method in the collaboration project with Vaulta.
- Use of multimodal sensing to improve stability in robotic grinding tasks under non-rigid fixation conditions.
- Exploration of the use of humanoid robots for performing grinding tasks.
Publications
- Research paper “Acoustic Feedback for Closed-Loop Force Control in Robotic Grinding” submitted to ICRA 2026.
Supervisory Team
- Principal Supervisors: Professor Jonathan Roberts,
- Associate Supervisors: Professor Will Browne and Dr Chris Lehnert
Associated Researchers