Human whole-body dynamic telelocomotion of humanoid robots

Humanoid robots have the potential to have a significant impact in a variety of challenging applications, including disaster response, healthcare, and manufacturing. However, achieving such an impact depends on these robots achieving human-level cognitive and physical intelligence. A promising approach is the teleoperation of these robots, which potentially bridges the gap between fully autonomous solutions. Through teleoperation, humanoid robots can in theory inherit human expert knowledge and sensorimotor skills to accomplish complex whole-body tasks. However, teleoperated humanoid robots have not yet reached this level of performance. This dissertation aims to address this problem at a fundamental level. Following the hypothesis that coordinated locomotion and manipulation teleoperation requires bilateral motor and mind synergy between humans and robots, we propose a novel bilateral feedback teleoperation framework. This framework enables direct control of the locomotion of a humanoid robot by a human pilot. Within this framework, a human pilot directly commands the robot's gait and receives haptic force feedback at the whole-body level to experience the robot's sense of balance and interaction forces with its environment. Named Telelocomotion in this dissertation, this type of teleoperation makes a significant stride toward creating robot avatars with human-level motor skills for physical interaction. Beyond designing the hardware-software architecture for this framework, the proposed work aims to explore the dynamics that arise between humans and humanoid robots during telelocomotion. In particular, understanding human motor adaptation to robot locomotion dynamics under various teleoperation re-targeting strategies and haptic feedback laws derived from reduced-order dynamical models for humanoid locomotion. This paves the way for a holistic bilateral feedback kinodynamic mapping that enables dynamic human-robot synchronized balance and locomotion that is robust to external disturbances. Our framework and associated methods are validated in real-time simulation and hardware experiments using our custom human-machine interface and the humanoid robot Tello. The work in this dissertation provides a fundamental step towards transferring human intelligence and reflexes to humanoid robots.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Mechanical Engineering

Grantor

University of Illinois Urbana-Champaign

Year dc:date

2025

Author dc:creator

• Colin Navarro, Guillermo

Contributors dc:contributor

• Ramos, Joao
• Dullerud, Geir
• Yim, Justin
• Bretl, Timothy
• Wensing, Patrick

• Humanoid Robots
• Bipedal Locomotion
• Teleoperation
• Controls

Statement dc:rights

• Copyright 2025 Guillermo Colin Navarro

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/129823

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F129823