IMU-Enabled Robotic Finger Force Estimation With Modular Soft Actuators and Virtual Linkages

• Published in: IEEE transactions on magnetics Vol. 60; no. 9; pp. 1 - 5
• Main Authors: Lin, Sheng-Guan, Chang, Hsien-Ting, Chang, Jen-Yuan
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.09.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accelerometers; Actuators; Couplings; End effectors; Estimation; Fingers; Force estimating; Force measurement; Humanoid; inertial measurement units (IMUs); Inertial platforms; Jacobi matrix method; Jacobian matrices; Jacobian matrix; Linkages; Magnetometers; modular soft actuators (MSAs); Modular units; Position measurement; Position sensing; Robot control; Robotics; Sensor arrays; Shaft encoders; Soft robotics; Task complexity; Torque
• ISSN: 0018-9464; 1941-0069
• DOI: 10.1109/TMAG.2024.3427630

This research presents a novel force estimating method without utilizing force gauges for soft robotic fingers that leverage inertial measurement units (IMUs) to detect finger joint angles and to calculate virtual linkages and torques. Motivated by the limitations of soft actuators in service-oriented humanoid robots, especially in tasks requiring precise fingertip positioning and gripping, this study aims to enhance position control and force estimating. Unlike conventional methods that rely on rotary encoders and flex sensors, IMU arrays and modular soft actuator (MSA) are proposed and examined in this study to overcome the inherent challenges of soft actuators, for their inability to accurately control fingertip positions and force directions. The proposed methodology involves transforming curvature-bending soft actuator fingers into a conventional rotary axis framework, allowing for precise control over external forces applied to objects through the calculation of virtual torques and the employment of the Jacobian matrix. Experimental validations confirm the feasibility of this approach, showcasing its potential to significantly improve the functionality and applicability of soft robotic fingers in complex manipulation tasks.