Dynamic Weight Sensing for a Cost-Effective Assistive Grip Device

Overview

Every day, over 20 million Americans struggle to perform simple vital tasks such as using a phone, writing, or even filling a glass of water, because of hand motor function degradation due to neurodegenerative diseases. Implementing inexpensive assistive strategies that work well has proven challenging. Assistive approaches can improve grip strength and limb mobility. Often, such devices are of limited use since grasping pressure is driven by a constant force regardless of fragility of an object. Such devices marginally improve function and assist with recovery, yet simple tasks such as filling a glass of water remain elusive. Workarounds include expensive tactile sensors, small cameras, and computer vision processing. Cheaper alternative approaches may greatly improve the effectiveness of the assistive device. This project assesses the feasibility of a low-cost assistive grip that dynamically adapts to changes in the objects manipulated by its user. It was hypothesized that this requires sensors capable to provide collocated normal and shear force measurements at fingers-object contact areas. The prototype grip assist uses actuators controlled by 3-dimensional sensors which measure shear and normal force at the same time. Users wear it like a glove that is minimally invasive. The controller calculates desired force to drive the actuator. The prototype assistive device uses 1 fingertip sensor along with a force tendon, servo actuator, Arduino circuit, and glove materials. The sensor’s static and kinematic friction coefficients are tested experimentally while simulating the effect of dynamically varying the grip force. Open-source video analysis software accurately evaluates the motion of objects as the gripping force tracks changes in the container weight. Overall, 13 experiments were conducted, each collecting an average of 45,000 data samples. The experimental results presented in the paper confirm the practical feasibility of the proposed approach. They demonstrate the gradual dynamic increase of the grip force in response to changing characteristics of the object that is being picked up, e.g. filling a cup with water. Results show that the assistive grip device prototype can rapidly respond to changing object characteristics (shape and weight), thus more accurately augmenting the existing capabilities of its wearer. A detailed error analysis is included and directions for future work were identified.

Competition history

• AJAS 2018 Category not listed

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• All Biomedical Engineering projects
• AJAS Biomedical Engineering projects
• Every AJAS 2018 project