Force translation and Gesture mimicking in Robotic hand using HD-sEMG

Project Title

Force translation and Gesture mimicking in Robotic hand using HD-sEMG

Project Area of Specialization Biomedical EngineeringProject Summary

The human hand is a powerful tool for sensing and operating in the environment. Hand loss can be perceived as devastating damage since it affects the level of autonomy, limiting the capability of performing working, social, and daily living activities. Amputation can be traumatic (due to an accident or injury) or surgical (due to any of multiple causes such as blood vessel disease, cancer, infection, excessive tissue damage, etc.) Upper limb amputations occur at a rate of 3.8 individuals per 100,000. Although the prosthetic devices used as a substitute for hands improves the quality of life but some upper limb prosthesis lack sensory feedback relating to the force exerted by the artificial hand on a grasped object and the degree of control is imprecise. Our project aims to develop a prosthetic hand that can evaluate a force feedback system considering design constraints, providing the user with closed-loop control. As an input signal, we will be using High-density Surface electromyography (HD-sEMG). HD-sEMG is used in rehabilitation research, providing spatial and temporal characteristics of myoelectric signal. We will be using data set for the HD-sEMG signals, then after analyzing the signals we will apply different signal processing techniques (filtering, amplifying, etc.) to extract the useful signal. The extracted signal will be given to the robotic hand (hardware prototype for testing the efficiency of data processing). The designed bionic hand will be able to recognize different gestures and also for the force translation. Being successfully employed in the industry, our project will provide a robust, easy to control, and capable of reliably grasping a large variety of objects by sensory feedback.

Project Objectives

Robotic hands are becoming more advanced, with major developments in control and functionality. However, up to 23% of prostheses are rejected, due to a lack of sensory feedback. This project presents the use of high-density surface EMG signals designed to study patterns of sEMG spatial distribution over upper limb muscles during voluntary isometric contractions. In this project, we are working on Sensory feedback and force estimation. Algorithms to improve neural interfacing signal extraction, force estimation, and gesture mimicking. The necessity for sensory input in prostheses is debatable; however numerous recent studies have indicated that prosthesis users want it in their hand prostheses. Using the right prosthesis with force translation and sensory feedback will improve their performance by addressing the needs of amputees, thus improving their quality of life and helping them perform their daily, economic and social activities and decrease their dependence on others.

Project Implementation Method

We will design a 3D printed robotic hand with five moveable digits, with the mechanical assembly using sensors and motors. We are using high-density surface EMG signals to study patterns of sEMG spatial distribution over upper limb muscles. After preprocessing the HD-EMG signals, different signal processing techniques, and Deep Learning Algorithms will be applied to obtain feature extraction and classification to develop and train an efficient AI model. Then it will be implemented on the hardware prototype by interfacing it with MATLAB software which will control the motors to provide the intended gesture. Then the external sensors will estimate the force applied and provide feedback to the prosthetic user. Therefore, the designed bionic hand will provide gesture recognition and force translation, an important milestone for restoring physical integrity as well as a sense of control.

Benefits of the Project

Amputation of a limb has an extensive effect on people’s lives, with people losing many physical functions and abilities. It has been found to impact people’s occupational status, leisure pursuits, social contacts, and activities of daily living. The loss of an upper limb potentially has a greater impact than lower limb amputation. This is because people’s hands and arms are not only particularly important functionally, allowing people to manipulate objects and carry out most of the activities of daily living, but socially as well. The approach of our project is thus the implementation of an effective and economic technique that provides sensory feedback and force estimation. Sensory feedback and Force estimation play an important role in the rehabilitation, biomechanics, sports, and assistive devices fields. It can also help in sports and physical activity for athletes with upper limb amputation.

We are using HD-sEMG signals that record motor unit action potentials (MUAP) over a muscle, using arrays of closely spaced electrodes. Unlike surface EMG, it accounts for a broader assessment of muscle electrophysiological activity. When applied to education, practical demonstration of this technique and its improvement in gesture recognition and force estimation can change the way students learn and ultimately create more knowledgeable and well-adjusted individuals. Not only would this help doctors but also educate individuals in different fields such as sports biomechanics, gait analysis, rehabilitation etc. They would be able to learn different applications such as signal decomposition (ie. classification of individual MUAP from the sEMG signal), the study of neuromuscular compartmentalization, the analysis of changes in the spatial distribution of MUAP, pattern recognition and force translation, among others. This would help students by making learning more engaging and collaborative. Rather than memorizing facts, they would learn by doing and through critical thinking. Therefore, our project would be essential in training and teaching by helping students improve their attention, design and working skills. The human hand demonstrates great resiliency and adaptability, which displays dependable capabilities with articulated mechanized hands. Upper limb amputees have no direct sense of grip force applied by a prosthetic hand therefore our aim is to strengthen institutional capabilities by the commercialization of our project. Thus our project will be one of the most mature technologies for our local industry. It would not only increase the capability of amputees but also it would reduce the burden on the economy and management.

Technical Details of Final Deliverable

The Technical details of the final product will be provided in the 3rd and 4th quarter of the project phase. The technical specification included a user manual, technical guide, and specification catalog. 

The final deliverable of the project is based on 4 quarters

In the 1st quarter, the Progress report along with the literature including the static analysis of project needs nationally and internationally. The data set and workflow of algorithm design will be presented.

In the 2nd quarter, the Initial design of the project (CAD design and dimensions) along with the list of components along with the technical details will be presented. The sensor calibration and schematic of the circuit layout will also be presented.

In the 3rd quarter, the Final design Software and hardware along with the algorithm design will be presented  

In the 4th Quarter, the Final product will be presented along with the technical details.

Final Deliverable of the Project Hardware SystemCore Industry MedicalOther Industries Manufacturing , Health Core Technology RoboticsOther Technologies Artificial Intelligence(AI), 3D/4D Printing, Shared EconomySustainable Development Goals No Poverty, Good Health and Well-Being for People, Affordable and Clean Energy, Decent Work and Economic Growth, Industry, Innovation and Infrastructure, Reduced InequalityRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	75850
Robotic Hand 3D printing	Equipment	1	45000	45000
Arduino Uno	Equipment	1	2250	2250
2.2” flex sensors	Equipment	3	2350	7050
Metal gear feedback servo motors	Equipment	5	400	2000
Force Sensitive Resistors 0.5’’	Equipment	5	1275	6375
Arduino Nano	Equipment	2	1250	2500
PCB board	Equipment	1	600	600
Base Structure	Equipment	1	4100	4100
Connecting wires	Equipment	40	3	120
Silicon glue	Miscellaneous	1	325	325
Latex gloves	Miscellaneous	3	150	450
Fishing lines	Miscellaneous	4	20	80
Publication	Miscellaneous	1	5000	5000