Development of a smart glove

Project Title

Development of a smart glove

Project Area of Specialization Mechatronics EngineeringProject Summary

PROJECT SUMMARY:

The scope of this project is to design and manufacture a motor-actuated soft exo-glove that helps people suffering from muscular hand weakness and poor grip strength by collecting and integrating the data in the form of MMG/EMG signals. The model is trained to respond with the correct actuation commands through a tendon-based motor-driven mechanism that extends or flexes according to the intent of the user. The 3D model is designed to be light, flexible, portable, and power-efficient to enhance the load-bearing capabilities of the structure.

The dynamic model of this mechanical glove provides power assistance to the hand muscles to execute everyday tasks such as holding a glass of water, carrying a heavy object easily, etc.

This project delivers an essential advancement in the biomechanical area and fulfills a basic need for the unfortunates that suffer from muscular impairment. It is hoped that the developed prototype will be beneficial and play a vital role in bringing back the people to normal life.

Project Objectives

PROJECT OBJECTIVES:

Problem Statement:

Neuromuscular diseases such as stroke, spinal code injuries, multiple sclerosis, and weaknesses of the skeletal muscles may substantially affect the health and well-being of people, not only on the physical but also on emotional, cognitive, and behavioral levels. Further on, the ever-increasing population of the elderly and the proneness of the specified age group to acquire muscular impairments present a need for a piece of equipment to assist in routine tasks.

Project Aim:

The main objective is to provide people suffering from muscular impairment with a soft wearable exo-glove that could help enhance hand movement and improve grip strength. This could prove fundamental in rehabilitation by allowing patients to perform routine tasks independently, making them independent of any nursing assistance.

This can be achieved by designing and developing a motor-actuated mechanical glove that employs and integrates data collected from an MMG/EMG sensor for its operation.

The glove is conceptualized to be light, durable, and compact whilst being able to fulfill its primary function. This would allow it to be used in hospitals as well as on a day-to-day basis by patients outside medical premises.

Project Implementation Method

PROJECT IMPLEMENTATION:

The project implementation will be carried out in five phases:

1) The dynamic design of the glove

This stage comprises concept visualization and a listing of project constraints based on the human hand anatomy. The design is conceived in such a manner as to imitate human hand motion and provide the user with a natural degree of freedom; each finger can perform flexion/extension but adduction/abduction as well. This is best done using a soft glove. But the major issue with a soft wearable exo-glove is its limitation to power assistance. This is due to a weakness in the force display compared to the rigid exoskeletal robotic hand.

The objective is to optimize the design that is capable enough to provide adequate power while also being light, compact and flexible which is implemented using a tendon-based motor-driven mechanism. Tendon-driven systems require less space than linkage-based exoskeletons, as they utilize few, small anchoring structures. Therefore, they are well-suited for sensor implementation.

2) FEM Analysis

The CAD model is developed using the dimensionality of a team member’s hand. The FEM analysis is performed to understand and evaluate the stress and load-bearing capabilities of the structure as well as the maximum torsional limits. The results are used to further optimize the mechanism to increase its strength and reduce the overall weight and space occupied.

3) Data collection and integration from MMG/EMG sensors on a rapid prototype 

The data collection is done by attaching MMG/EMG electrodes to the flexor and extensor muscles of the impaired arm. The signals are received, amplified, optically isolated, and filtered using a signal conditioning circuit. These signals are then processed to actuate the mechanism to obtain the desired response.

4) Data Integration and Model Training

The model is trained to infer the user's intent from the collected data and respond with the appropriate actuation commands to the exo-tendon device. The control system is trained to recognize the changing patterns in different states of the hand. For example, if the model predicts that the user intends to open/close/relax the hand, the device is commanded to perform the corresponding action; retract or extend the tendons.

Benefits of the Project

Benefits:

1) Rehabilitation

Regular physiotherapy sessions using the exo-glove would help increase muscle strength and allow patients to gain enough gripping power to carry out routine tasks such as drinking water or writing with a pencil.

2) Self-Reliance

Being physically dependent on others to help with routine tasks could cause embarrassment and anxiety which might lead to severe mental havoc. This mechanical glove would allow them to become independent and perform their daily activities properly.

3) The feeling of inclusion:

Conditions leading to physical weakness can leave a person feeling left out and different from their peers. Most people suffering from the said diseases cannot compete in the athletic areas of life and are often segregated and perceived as an entirely different group. The use of this glove would allow them to overcome their deficiencies and feel more inclusive by being socially acceptable in society.

4) Improvement in self-confidence:

A poor self-image can put a person in a bad place mentally. The aesthetic design of the exo-glove will be imaged as cool and appealing, especially, to the younger generation. It would allow people to see this power glove as a unique technological piece of equipment that augments their physical strength and feel more powerful.

Technical Details of Final Deliverable

TECHNICAL DETAILS OF FINAL DELIVERABLE:

1) 3D Model and simulation of the design

The CAD of the exo-glove is made on SOLIDWORKS and FEM simulations are carried out on ANSYS. The 3-D model will be fully capable of depicting the dynamic movement of tendons that are powered by motors.

2)   Database

The relevant data would be acquired directly using clinical trials and testing as well as indirectly from the datasets available on the internet. We selected Cloud Service as our main database and storage point (subject to change). Cloud setup is completed and the dataset gathered is stored, be it in the form of images, videos, and any other format of data.

3)   ML Model Setup

The suitable machine learning model is selected and the dataset gathered is fed. The model is trained to classify different signal outputs from EMG/MMG electrodes. The ML model is then subjected to testing where different data outputs are compared and validated. Finally, the model is deployed where practical implementation of the model on the prototype is carried out to obtain desired actuation response.

4)   Prototype Setup

Material selection is performed as per the outlined constraints and a 3D model in line with these guidelines is made. The CAD is subjected to multiple simulations to determine the loading bearing capabilities of the mechanism as well as the torques and forces required to ensure desired actuation response from tendons.

The model is manufactured and the actuation setup is completed to check the DOF of each tendon. In the final phase, the manufactured glove is integrated with a machine learning model and subjected to further clinical testing

Final Deliverable of the Project HW/SW integrated systemCore Industry HealthOther Industries Manufacturing Core Technology RoboticsOther Technologies Wearables and ImplantablesSustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	79400
Muscle signal sensor EMG Sensor for Arduino/Raspberry	Equipment	3	4200	12600
Raspberry Pi 4B 1.5GHz Quad-Core 4 GB	Equipment	1	32000	32000
Actuators (servo/stepper motors)	Equipment	6	2800	16800
Batteries	Equipment	2	4000	8000
CNC Tendons	Miscellaneous	20	500	10000