Enabling mobile robot perception and shared-control with vision-based sensor fusion systems

Mobile robots with various locomotion mechanisms have been developed to adapt to different environmental conditions and mission requirements. The advancement in mobile robot development has raised the requirements for robot perception to ensure the safe and smooth execution of robot missions. In the past decade, the emergence of a novel self-balancing mobile robot system, known as the ballbot, has opened up new avenues of research in the field of mobile robots. While offering enhanced maneuverability, the under-actuated dynamics of the ballbot invalidate most existing ground robot perception solutions. Moreover, when the ballbot is designed as a riding mobility device, it imposes additional demands and constraints on robot perception and human-robot shared motion control. Specifically, the ballbot’s omnidirectional capability requires operators to possess significant proficiency and experience in riding and controlling the platform effectively. To address these challenges, there is a critical need for advanced and innovative robot perception techniques and higher-level control systems that can reduce the operator’s cognitive load and ensure system safety by preventing unwanted collisions. This dissertation aims to investigate robot perception and safeguarding challenges by leveraging vision-based sensor fusion techniques and commonly available commercial sensors. The research specifically focuses on developing robot perception and shared-control systems that can be deployed on a physical ballbot prototype, adhering to strict Size-Weight-and-Power Constraints (SWaP-C), while also being generalizable to mobile robots with simpler or similar dynamic characteristics as ballbots.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Mechanical Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2023

Author dc:creator

• Chen, Yu

Contributors dc:contributor

• Norris, William R
• Hsiao-Wecksler, Elizabeth T
• Ramos, João
• Sreenivas, Ramavarapu S
• Driggs-Campbell, Katie
• Wang, YuXiong

• Deep Learning For Visual Perception
• Rgb-d Perception
• Vision-based Navigation
• Self-balancing Robot
• Robot Perception
• Vision Sensor
• Sensor Fusion
• Shared-control

Statement dc:rights

• Copyright 2023 Yu Chen

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/122017

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F122017