Soft Robot Gripper for Patients with Diabetic Neuropathy

This project is to create a gripper/soft robotic hand for people with disabilities to use. Some people are chronically impaired by diseases like arthritis and diabetes. Arthritis makes joints pain when used for basic functions. As a side effect of diabetes, patients have no sensation in their appendages like hands and feet. A gripper like this would help these two types of patients in dealing with their symptoms. 

The background section of this wiki will detail the research needed to make this product in its soft robotics aspect and the sensor aspect. For the soft robotics component, I am using silicone rubber, molded in such a way that when inflated it forms a gripper. For sensors, I am using a temperature and pressure sensor. Also, it will detail the creation of the idea.

The design section will show the design process for the robot and sensors, including the development of the idea. The idea is to help patients with Arthritis and Type II Diabetes, as stated above. 

The fabrication section will include the making of the robot and its prototypes, along with videos of the building process. It will also show issues with prototypes and how they were solved.

The testing section will include the final tests for the robot. These tests are for accuracy of the sensors and for the gripping power of the robot.

Our team is "Beri Good at Robotics." Our team consists of two students from Westford Academy in Westford, Massachusetts, Alekh and Serena Beri. Alekh is part of his local FRC robotics team, and Serena is part of her school's HOSA medical competition club. They combined their two passions to make a robot that also functions as a medical device.

 

                                 

soft_robotic_gripper.pdf405 KB

Background

My inspiration for this robot came from my grandmother, who currently has Type II diabetes and peripheral neuropathy, meaning that she can not sense things that she touches. Neuropathy occurs because of damage to the nerves in appendages like the hands and feet. Diabetes causes a lot of cases of neuropathy, but it can also be caused by other illnesses and injuries. I found that 20 million people in the US alone have this condition. I wanted to help these people, so I came up with the idea for this gripper. 

I found that soft robotics would be a great way to make this product, along with sensors to add to its usefulness. The small air channels in my gripper cause it to form a small hand-like structure when inflated. This can be used to pick up any object that it can get a hold of. The addition of specialized sensors like a temperature sensor, will allow users to find out if they can actually touch something. People with neuropathy can not sense temperature very well, and therefore could burn or seriously injure themselves without even knowing it.

Current treatments for this condition include prescription pain relieving drugs and antidepressants like oxycodone/oxycontin and amitriptyline. My product would not treat the pain relieving aspects of these drugs, but instead help patients perform everyday chores.

Design

My project started with an idea. I wanted to create a device to help people in need, and soft robotics seemed to be perfect for this. With its flexibility, soft robotics has the potential to do anything, and so I wanted to pair it with sensors. With the power of sensors, I can measure the temperature, pressure, and flex angle of whatever my soft robot touches. Using these two technologies, I came up with the idea to make a device to help people with neuropathy have functional hands again.

Clinical neuropathy is a condition that can occur with diabetes, which affects 29.1 million people in the US alone. It causes the patient to have severely decreased feeling in the hands and feet, which can be dangerous for the patient. For example, someone with this condition could be holding a very hot pan, but not feel a thing, burning their skin very badly. A gripper like this would fix this problem by it being able to touch something hot or under a lot of pressure before the human hand, almost like a safety or extra hand for the user. 

This idea is relatively simple to make too. A soft robot gripper is fairly simple to make with the right mold, and sensors can be made using an Arduino kit. The creation of the prototype will be outlined in the next section.

Fabrication

Fabrication of the soft gripper was done in 5 main steps:

  1.  Build the mold with 3D printing
  2.  Pour the Silicon mixture into mold and form the soft gripper structure 
  3.  Use an air source to power the soft robot gripper
  4.  Set up the Arduino sensors and microprocessor to collect data on the gripper 
  5.  Conduct testing and analyze data collected

Detailed fabrication information can be found in the subpages.

Bill of Materials

Materials Used for Soft Robotic Gripper

  • We used Ecoflex 00-30 (one "trial kit" is enough to make five to ten robots depending on size). You can also use materials like Elastosil M4601 A/B. We asked the manufacturer to provide samples as we did not need a huge quantity.
  • 1/16 inch ID, 1/8 inch OD polyethylene tubing (1 foot). Part number 5181K15 at McMaster-Carr.
  • 1/8 inch ID, 1/4; inch OD opaque silicone rubber tubing (1 inch). Part number 5236K83 at McMaster-Carr.
  • Squeeze bulb or bicycle pump as air source
  • 3D-printed mold: 3D models are provided in stl files from the website - some of them have to be modified for the application
  • Other household items used were tray, scissors, plastic cups, Popsicle sticks, disposable gloves and paper towels.

Materials Used for Sensor Circuits

  • Arduino Uno Microprocessor
  • Sparkfun Breadboard
  • Sparkfun Jumper Wires
  • Tekscan A401 Pressure Sensor
  • TMP 36 Temperature Sensor
  • Flex Sensor 2.2"
  • 10K Ohm Resistor
  • Computer with Arduino Software (Can be Mac or PC)

Creating a Soft Robotic Gripper

For the soft robot we followed instructions from the website for Ecoflex as described in website. Details of the websites used are listed in our bibliography. The video shows the step by step fabrication process. 

 

Works Cited

Finio, Ben. “Squishy Robots: Build an Air-Powered Soft Robotic Gripper.” Science Buddies. N.p., n.d. Web. 29 June 2016.

“SIK Experiment Guide for Arduino.” Sparkfun. N.p., n.d. Web. 29 June 2016.

“2016 Soft Robotics Competition.” Soft Robotics Toolkit. Harvard University, n.d. Web. 30 June 2016.

 

Creating Sensor Circuits

Below are the circuit diagrams from the sparkfun website:

  1.  Flex Sensor
    1. Create circuit using fritzing diagram below and materials listed: 
  2. Temperature Sensor
    1. Create circuit using fritzing diagram below and materials listed:
  3. Pressure Sensor
    1. Create circuit using fritzing diagram below and materials listed:

Testing

This section will outline the testing of the prototype done by us. The accuracy of the sensors will be shown, along with how the soft robot gripper compares with that of the human hand. An analysis will be conducted on the linearity/consistency of the soft robot gripper vs the human hand vs the sensor alone, which will serve as the control group. This analysis will happen in three major senses used in gripping: flex, temperature, and pressure. This will show how well the gripper will work with a patient with neuropathy.

For these tests, there are 3 steps:

  1. Evaluate linearity of sensors individually
    1. Temperature sensor between 12 - 42 degrees C - sensitivity between .05 to .2 degress
    2. Pressure Sensor up to 25 psi
    3. Flex angle 180 degrees with accuracy to 1 degree
  2. Create a working model
    1. Add sensors to microprocessor to build prototype circuit
    2. Test soft robot in environment and changes in parameters
  3. Evaluate Data
    1. Compare soft robot data to that of human hand

Next steps:

  1. Develop prototype with wireless sensors 
  2. Evaluate Communication Connectivity - make an app or way for feedback for the user
  3. Keep developing prototype so that it is more consistent to make and use 

Flex Sensor

The graph above shows the linearity of the flex sensor in a normal circuit with a servo motor. In this circuit, the sensor is alone, showing that it works properly and can be used in the gripper. 

This graph shows flex position over time, with the blue line being for no manipulation (control), the red line being for the contours of the gripper, and the yellow line being for the contours of the human hand. Without manipulation, the flex resistance does not change at all, which is expected. The gripper and hand both fluctuate wildly, but they both cover the same range. This means that the gripper can "flex" to any position the human hand will normally, showing that the flex sensor will work properly with the gripper for patients with neuropathy.

Temperature Sensor

The graph above shows the linearity of the temperature sensor, showing that it can be used in the soft robotic gripper with accuracy.

This graph shows the temperature profiles of room temperature, the gripper, and the human hand over time. The blue line is for room temperature, the red line is for the soft robotic gripper, and the yellow line is for the human hand. The room temperature stays within 1 degree, which is not unusual. The gripper and hand have temperatures about 3 degrees away from one another, but this is not enough to make a significant impact on the results of temperature sensing in the gripper. Overall, there is no convincing evidence to suggest that the temperature sensor will not work in this soft robotic gripper.

Pressure Sensor

This graph comes from the Tekscan website, where I got the pressure sensor. The one I am using corresponds to the red 25lb force range line, which is relatively linear. This shows that the pressure sensor can be used in the soft robotic gripper. 

This graph shows the pressure of the gripper and the human hand over time. The left graph is for the soft robot gripper, while the right graph is for the hand pressure. Both were measured when pressing as hard as possible, which is not necessarily needed to hold an object. The soft robot's pressure only went up to .75 psi at the most, which is barely enough to hold something, but the gripper should be fine to hold things depending on the air source. A weak air source may produce too little pressure, while a strong one may produce too much. The hardest hand pressure is about 2.75 psi, which is definitely enough to hold something, but harm something fragile. Although these pressures are very different, they should not affect the overall function of the gripper too much, therefore showing that the gripper will work properly in the aspect of pressure.