Date on Master's Thesis/Doctoral Dissertation

12-2023

Document Type

Doctoral Dissertation

Degree Name

Ph. D.

Department

Electrical and Computer Engineering

Degree Program

Electrical Engineering, PhD

Committee Chair

Popa, Dan

Committee Member

Walsh, Kevin

Committee Member

Roussel, Tommy

Committee Member

Harnett, Cindy

Author's Keywords

Robot skin sensor; additive manufacturing; robotics; piezoresistive organic polymer; cleanroom fabrication

Abstract

Enhancing physical human-robot interaction in modern robotics relies on refining the tactile perception of robot skin sensors. This research focuses on crucial aspects of the development process, including fabrication techniques, miniaturization, and integration for a more efficient collaborative human-robot interface. The fabrication process of robot skin sensors, designed to mimic human skin, is explored both within and outside cleanroom environments. An enhanced technique is presented to increase fabrication yield and create more miniaturized sensor designs with feature sizes in the tens of microns. These sensors function as piezoresistive arrays using organic polymers like PEDOT: PSS as the pressure-sensing medium. Various deposition techniques, such as cleanroom spin coating and direct-write inkjet printing with Aerosol inkjet printers, are discussed. A NeXus microfabrication platform is introduced to eliminate errors, simplify the cleanroom process, and reduce production time for sensor arrays. This platform is employed for the prototyping of tactile strain gauges, integrating an Aerosol jet printer station for patterning sensor electrodes on flexible substrates and a piezo-electric fluid dispenser for PEDOT:PSS deposition, bypassing cleanroom photolithography. The post-processing phase is detailed, highlighting the sintering of patterned silver traces using an oven or intense pulse light (IPL). The curing process determines the resistance and conductivity of printed samples, with IPL offering flexibility and efficiency compared to traditional ovens. Cured samples undergo testing on a specialized testbench equipped with an indenter, force feedback control, motorized stage, and computer vision functionality. LabVIEW Programs synchronize testing components, producing tangible results for each tactile sensor test. Test quality influences the integration of tactile sensors with a robotic arm. A novel tactile fingerprint design, realizable in the NeXus, is proposed and characterized based on performance and reliability. Sensitivity, indentation cycles, and spatial resolution studies contribute to a comprehensive understanding of the proposed design. The research's ultimate goal is to integrate tactile sensors, including commercially available options like Flexiforce sensors and robot skin sensor patches, with a robot to enhance direct interaction. The effective use of the Robot Operating System (ROS) and local area connectivity to implement the robot's response to physical touch on the skin sensors marks a significant stride in advancing human-robot interaction. The abstract encompasses the critical elements of improved fabrication, miniaturization, and integration, making strides toward more effective and adaptable physical human-robot collaboration.

Download

Share

COinS