Created July 28, 2024

Neural network integrated Hydrogel-based exoskeleton

Meet the Hydrogel Exoskeleton: flexible, lightweight, and adaptive. Empowering those with low leg strength to walk independently and confide

Lucky Draw for Interview

Build2gether 2.0 — Inclusive Innovation Challenge

Neural network integrated Hydrogel-based exoskeleton

Things used in this project

Hardware components

Nordic Semiconductor nRF52840 Development Kit

Seeed Studio XIAO ESP32S3 Sense

Blues Notecard (Cellular)

Hand tools and fabrication machines

Soldering iron (generic)

Cable Cutter, 143mm

10 Pc. Jumper Wire Kit, 5 cm Long

Story

Background:

People living with mobility impairments face daily challenges that often go unseen by those who take their ability to walk for granted. Whether due to injury, illness, or other reasons, these individuals may find themselves unable to move freely, relying on mobility aids or the assistance of others to navigate the world around them. While some may have intact physical functionalities, the limitation lies in their minimal strength, rendering them unable to move effectively.

Despite advancements in assistive technology, there remains a significant gap in the market for devices specifically designed to address the needs of individuals with low leg strength. While wheelchairs and mobility scooters offer solutions for some, they may not fully meet the needs of those who desire greater independence and the ability to walk unassisted. In short, we need ways to have people with low leg strength to be able to walk independently, instead of relying on wheelchairs. On top of addressing the problem of having low leg strength, it is also worth noting that individuals with deteriorating leg strength also encounter other issues such as loss of balance or coordination of the body.

My aim is to develop a novel, neural network-assisted hydrogel exoskeleton, an advanced wearable device designed to assist individuals with low leg strength and coordination problems. This innovative exoskeleton aims to enable users to walk independently without relying on traditional mobility aids like wheelchairs, thereby enhancing their quality of life. Traditional solutions like wheelchairs and mobility scooters often fall short in meeting the needs of those who desire greater independence and the ability to walk unassisted. My goal is to help individuals, including those with conditions such as muscular dystrophy, maintain their mobility for as long as possible, potentially delaying the need for a wheelchair.

Device functionality:

The Hydrogel Exoskeleton leverages cutting-edge technology to provide comprehensive support and enhance mobility. Here’s how it works:

Materials: I chose a high-strength, highly stretchable photo-cross-llinkable PEG (polyethylene glycol) based hydrogel for the base material, as it is non-cytotoxic, durable and cost-efficient. This PEG-based hydrogel possesses water-swelling ability. By controlling the degree of swelling and the location of swelling, we are able to control precisely the location to provide mechanical support. I also embedded SEEED studio motion sensor inside the hydrogel to detect the motion. This motion sensor also sends one locational signal and one power signal to the control board. By knowing the location and the strength, the code will calculate the amount of force needed on the other leg or the other parts of the body to achieve a reasonable movement and force.

Neural Network and Feedback System: To address balance and coordination, the exoskeleton incorporates a sophisticated neural network and real-time feedback system. Embedded sensors throughout the device continuously monitor the user's posture, gait, and movements. This data is processed by the neural network, which dynamically adjusts the exoskeleton's support to optimize balance and coordination. For example, if the user begins to lose balance, the system can redistribute support to stabilize them, thereby enhancing safety and confidence.

To address balance and coordination, the exoskeleton incorporates a sophisticated neural network and real-time feedback system. Embedded sensors throughout the device continuously monitor the user's posture, gait, and movements. This data is processed by the neural network, which dynamically adjusts the exoskeleton's support to optimize balance and coordination. For example, if the user begins to lose balance, the system can redistribute support to stabilize them, thereby enhancing safety and confidence.

Ergonomic Design: The Hydrogel Exoskeleton is designed to be lightweight and ergonomic, ensuring maximum comfort and freedom of movement. It molds to the body’s contours, providing targeted support where it’s needed most without restricting the user's natural movements. This design also allows users to keep their hands free, further enhancing their independence.

The Hydrogel Exoskeleton is designed to be lightweight and ergonomic, ensuring maximum comfort and freedom of movement. It molds to the body’s contours, providing targeted support where it’s needed most without restricting the user's natural movements. This design also allows users to keep their hands free, further enhancing their independence.

Adaptive and Versatile: The exoskeleton is highly adaptable, making it suitable for a wide range of users, including those with varying degrees of leg strength and coordination issues. Whether assisting individuals with mobility impairments, enhancing athletic performance, or aiding in rehabilitation, the Hydrogel Exoskeleton offers unmatched versatility.

The exoskeleton is highly adaptable, making it suitable for a wide range of users, including those with varying degrees of leg strength and coordination issues. Whether assisting individuals with mobility impairments, enhancing athletic performance, or aiding in rehabilitation, the Hydrogel Exoskeleton offers unmatched versatility.

In summary, the hydrogel exoskeleton is a revolutionary device that combines advanced materials and intelligent systems to provide comprehensive support for individuals with low leg strength and coordination challenges, enabling them to walk independently and confidently.

Code

Code for location and force detection

import time
import board
import busio
import adafruit_mpu6050
import numpy as np

# Initialize I2C communication
i2c = busio.I2C(board.SCL, board.SDA)
sensor = adafruit_mpu6050.MPU6050(i2c)

# Placeholder function for neural network prediction
def neural_network_predict(data):
    # Simulate complex data processing with a neural network
    # Generate mock predictions for demonstration purposes
    location = np.random.uniform(0, 1)  # Mock location between 0 and 1
    swelling_degree = np.random.uniform(0, 1)  # Mock swelling degree between 0 and 1
    force = np.random.uniform(0, 10)  # Mock force between 0 and 10 Newtons
    return location, swelling_degree, force

# Placeholder function to control the exoskeleton based on neural network predictions
def control_exoskeleton(location, swelling_degree, force):
    # Implement the control logic for the exoskeleton here
    # Adjust the support provided by the hydrogel based on the predictions
    print(f"Adjusting exoskeleton: Location={location}, Swelling Degree={swelling_degree}, Force={force}")
    # Example control logic (to be replaced with actual implementation):
    # If swelling_degree is high, reduce water absorption rate
    # If force is low, increase support in the corresponding location
    pass

# Main loop to continuously read sensor data and adjust the exoskeleton
while True:
    # Read motion sensor data
    acceleration = sensor.acceleration
    gyro = sensor.gyro
    
    # Preprocess sensor data for neural network input
    data = np.array([
        acceleration[0], acceleration[1], acceleration[2],
        gyro[0], gyro[1], gyro[2]
    ])
    
    # Normalize data (example normalization, adjust as needed)
    data = data / np.linalg.norm(data)
    
    # Predict swelling location, degree, and generated force
    location, swelling_degree, force = neural_network_predict(data)
    
    # Output the results
    print(f"Swelling Location: {location:.2f}")
    print(f"Swelling Degree: {swelling_degree:.2f}")
    print(f"Generated Force: {force:.2f} N")
    
    # Control exoskeleton based on the neural network output
    control_exoskeleton(location, swelling_degree, force)
    
    # Wait before next reading
    time.sleep(0.1)

Credits

Mushroom Chen

1 project • 1 follower

Comments

Awards

Lucky Draw for Interview

Build2gether 2.0 — Inclusive Innovation Challenge

Neural network integrated Hydrogel-based exoskeleton

Things used in this project

Hardware components

Hand tools and fabrication machines

Story

Schematics

Setup

Code

Code for location and force detection

Credits

Mushroom Chen

Comments

Awards

Embed the widget on your own site

Neural network integrated Hydrogel-based exoskeleton

Neural network integrated Hydrogel-based exoskeleton

Things used in this project

Hardware components

Hand tools and fabrication machines

Story

Schematics

Setup

Code

Code for location and force detection

Credits

Mushroom Chen

Comments

Awards