20 - Kinematic Analysis of a Tendon-Actuated Compliant Finger via Pseudo-Rigid-Body Modeling
Team Members:
Payton Cooper, Caden Page, Caleb Read, Pablo Diaz
Background:
Compliant mechanisms generate motion through material deformation rather than traditional rigid joints, enabling simpler, lightweight designs with fewer parts. With the rise of flexible 3D printing materials like TPU, these systems can now be fabricated as single, continuous structures with built-in flexibility.
In robotics, compliance improves adaptability, allowing grippers to handle objects of varying shapes without precise control. Tendon-driven actuation, inspired by biological systems, is commonly used to produce smooth, coordinated bending in such mechanisms.
To analyze these systems efficiently, the pseudo-rigid-body (PRB) model approximates flexible sections as rigid links with torsional springs, making it possible to apply classical kinematic methods while still capturing compliant behavior.
Summary/Overview:
This project focuses on designing and analyzing a tendon-actuated compliant robotic finger using the pseudo-rigid-body (PRB) modeling approach. Unlike traditional rigid mechanisms, the finger is made from flexible TPU with built-in flexure joints, allowing it to bend and adapt to objects. To make this complex deformation easier to analyze, the compliant structure is approximated as a system of rigid links connected by torsional springs, enabling the use of classical kinematics.
The goal is to understand how factors like tendon actuation, joint stiffness, and geometry influence the finger’s motion – particularly achieving sequential curling for effective grasping. The project involves developing mathematical models, building a 3D-printed prototype, and validating the predicted motion through experimental testing. Ultimately, the work aims to create a reliable method for predicting and designing compliant robotic fingers with controlled and efficient movement.
Problem Statement:
Designing a tendon-actuated compliant robotic finger with predictable and controllable motion is challenging because its behavior is governed by distributed material deformation rather than discrete joints. The interaction between tendon actuation, joint stiffness, and finger geometry creates coupled motion across the structure, making it difficult to anticipate how the finger will bend.
A key objective is to achieve sequential curling, where proximal joints close before distal joints to enable effective grasping. However, without a clear modeling framework, predicting and tuning this behavior typically requires iterative prototyping or complex simulations.
The problem addressed in this project is to develop a modeling approach that relates actuator input, flexure stiffness, and geometric parameters to the resulting finger motion, enabling accurate prediction and design of a tendon-driven compliant finger.
Table of Contents: