Context-aware perception and manipulation strategies for PAPRAS dual-arm stand system in real-world scenarios

Dual-arm robotic systems hold immense potential for performing complex bimanual operations in human-centric environments. However, to fully leverage their capabilities, advanced perception and manipulation strategies tailored to address system constraints and dynamic conditions are crucial. This thesis presents the development and evaluation of context-specific perception and manipulation strategies for the Plug-And-Play-Robotic-Arm-System (PAPRAS) dual-arm stand system. The research is centered on enhancing the adaptability and effectiveness of the system in performing real-world interaction tasks, while considering the specific hardware constraints and requirements of the platform. To demonstrate the efficacy of the proposed approach, two key applications are explored: autonomous door opening and human-to-robot motion retargeting. These applications exemplify the need for robust perception, coordinated manipulation, and context-aware decision-making. The study integrates camera-based object recognition and localization, task-specific motion planning, and force feedback to address the challenges associated with each application. The research outcomes highlight the successful implementation of the proposed strategies, leading to improved overall performance and versatility of the PAPRAS dual-arm stand system in diverse and dynamic environments. By integrating unified perception and manipulation strategies with context-specific heuristics, this work contributes to enhancing the capabilities of dual-arm robotic systems for effective and safe human-robot interaction in various applications.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Electrical & Computer Engr

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2023

Author dc:creator

• Shin, Kazuki

Contributors dc:contributor

• Kim, Joohyung

• Dual-arm Robotics
• Autonomous Door-opening
• Context-specific Manipulation
• Motion Retargeting

Statement dc:rights

• Copyright 2023 Kazuki Shin

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/120461

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F120461