Design and control of an origami-enabled soft crawling autonomous robot (OSCAR)

Soft mobile robots offer unique benefits as they are highly adaptable to the terrain of travel and safe for interaction with humans. However, the lack of autonomy currently limits their practical applications. Autonomous navigation has been well studied for conventional rigid-bodied robots; however, it is underrepresented in the soft mobile robot research community. Its implementation in soft robots comes with multiple challenges. However, the major challenge is the significant motion uncertainties due to the robot compliance, ground interactions, and limited available sensing. These uncertainties prevent high-level control implementation, such as autonomous navigation, to be performed successfully. Therefore, soft robots require robust design methods, as well as path following and path planning algorithms, to mitigate these uncertainties and enable autonomy. This dissertation develops and implements autonomous navigation for a novel origami-enabled soft crawling autonomous robot called OSCAR. In order to implement autonomous navigation, it first mitigates the OSCAR’s motion uncertainties by a multi-step iterative design process. Analysis has shown that OSCAR’s motion uncertainties are the result of: (i) the ground-feet interaction, (ii) effectiveness of low-level closed-loop control and, (iii) variability in the manufacturing assembly process. The iterative control-oriented design allows a robust and reliable OSCAR performance and enables high-level path following control implementation. To design and implement path following control, this research presents an idealized kinematic model and introduces an empirically based correction to make the model predictions match the experimental data. The dissertation investigates two separate path-following controllers: a model-based pure pursuit and a feedback controller. The controllers are investigated in both simulation and experiment and the need for feedback is clearly demonstrated. Finally, this research presents the path planning in order to complete OSCAR’s autonomous navigation. The simulation and experimental results show that OSCAR can accurately navigate in a 2D environment while avoiding static obstacles. Lastly, the coupled locomotion of multiple OSCARs demonstrates an extension of functionality and expands the potential design and operation space for this promising type of soft robot.

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

2021

Author dc:creator

• Angatkina, Oyuna

Contributors dc:contributor

• Alleyne, Andrew
• Wissa, Aimy
• Tawfick, Sameh
• Hsiao-Wecksler, Elizabeth
• Chowdhary, Girish

• soft robot
• origami
• path following
• path planning
• autonomy
• kinematic model
• pure pursuit
• feedback controller
• mobile robots
• crawling
• modular robot

Statement dc:rights

• © 2021 Oyuna Angatkina

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/110414

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/110414