Learning-based shared control for open world, dexterous robot teleoperation

Naughton, Patrick

Permalink

https://hdl.handle.net/2142/132495 Copy

Description

Title

Learning-based shared control for open world, dexterous robot teleoperation

Author(s)

Naughton, Patrick

Issue Date

2025-11-17

Director of Research (if dissertation) or Advisor (if thesis)

Hauser, Kris

Doctoral Committee Chair(s)

Hauser, Kris

Committee Member(s)

• Bretl, Timothy
• Lazebnik, Svetlana
• Pinto, Lerrel
• Srinivasa, Siddhartha

Department of Study

Siebel School Comp & Data Sci

Discipline

Computer Science

Degree Granting Institution

University of Illinois Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Teleoperation
• Shared Control
• Dexterous Manipulation
• Machine Learning

Abstract

Teleoperation allows people to extend their perception and interaction capabilities to remote locations in cases where distance or safety constraints prevent them from physically traveling. It is especially useful for operation in “open-world” environments, environments that cannot be accurately modeled ahead of time, where traditional automation techniques often prove unreliable. However, while humans are extremely adept at a wide variety of manipulation tasks, transferring these skills through a robot has remained an open challenge for over half a century. Given constraints on an operator’s total cognitive load, high degree-of-freedom systems face a tradeoff between flexibility and precision. Giving the operator direct control over every joint on the robot provides them with the maximum flexibility to complete many kinds of tasks, but the high cognitive load associated with such an interface makes it impossible to properly coordinate the joints for tasks requiring high precision. In contrast, shared control systems that let the operator control the robot at a higher level of abstraction reduce this cognitive load and can improve the operator’s precision, but simultaneously reduce the flexibility of the interface to accomplish tasks for which it was not explicitly designed. Towards the goal of creating truly telepresent systems, this thesis introduces several learning techniques to dynamically adapt robotic teleoperation interfaces to specific users and tasks, allowing operators to more naturally express their intent and enabling greater transfer of their manipulation skills. This approach has lead to improvements in mappings for retargeting operator inputs to robot actions, harnessing the operator’s existing manipulation skills to flexibly complete many kinds of tasks while maintaining a high degree of precision.

Graduation Semester

2025-12

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/132495

Copyright and License Information

Copyright 2025 Patrick Naughton

Owning Collections