Notebook Category
Poster for Design Day
Apr. 16, 2019—The poster that will be presented in our final presentation on 04/22/2019 is attached below and was used for consideration in design meetings on 04/10/19 and 04/17/19. BME_5 Design Day Poster
PID Final Validation in Servo Construct
Apr. 16, 2019—Because of the differences in state seen between the prosthesis and the servo motor used as a proof of concept, two distinct PIDs were constructed for clinical use and for servo validation. The intial PID was formed based on pre-existing arduino libraries, but the analog output, did not translate as smoothly as our matlab model...
Validation Data for Software
Apr. 16, 2019—Final device validation, performed with software and hardware constructs with servo motor actuation model. The first graph illustrates the ability for the software to accurately determine the user intended action for future actuation. The final graph shows the distribution of action across the three main elements of the software: mean absolute value of differential EMG...
Servo Actuation- Proof of Concept
Apr. 9, 2019—Servo Motor Actuation with a Continuous Force-Based Control System The below video serves as a partial proof of concept demonstrating the effectiveness of our EMG acquisition, pre-processing, processing, and actuation output in accurately controlling and maintaining the posture of a servo.
Weekly Meeting 15 (4/08/19)
Apr. 8, 2019—Letter of support and grade needed Will clarify info about the grade Reprinted housing for components Worked on validating servo actuation Lots of flutter Threshold too low Using 3 Hz low pass filter to smooth signal Removes noise Can try implementing function based on strength of contraction for threshold PID would help but geared towards...
Safety Design Considerations
Apr. 3, 2019—A risk analysis was performed previously during Grant development and is shown above. Probability of Safety Event: 1 = extremely unlikely, 10 = guaranteed Severity of Safety Event: 1 = no harm to the user, 10 = fatal
3D Models of Cage Designs
Apr. 3, 2019—Original cage design: Modified cage design: Lid for modified cage:
Logistics Classifier Machine Learning Training
Apr. 3, 2019—Logistics Classifier Machine Learning Training Collect pre-processed training data across subjects with varying gender, musculature, and EMG data Calculate MAV of extensor and flexor using a 200 ms sliding window every 50 ms Calculate the differential between the extensor and flexor MAV for each subject Average all subjects’ MAV values and varying contraction levels Fit...
Coding Repository
Apr. 3, 2019—Github Repository To integrate version control and robust development practices, Git and Github were used throughout the software development process. Here is a link to our GitHub page: https://github.com/christian-stano/Force-Based-Myoelectric-Controller
Description of FDA Category
Apr. 3, 2019—Because there are a variety of commercially-available powered and mechanical prosthetics, it is believed that this device should not require pre-market approval (PMA). Since this is a joint effort between our team and Case Western, it is understood that they will be responsible for bringing the product or certain components of this project to market....
Gesture Classifier Possible Fits
Apr. 3, 2019—Preprocessed emg data from the current hardware landscape with varied percent contractions across three subjects was taken and used to train a gesture classifier for software implementation. The three fits shown below were considered for use in the current scheme. The third order polynomial presented too much variation at the extremes of the data set...
Initial Prototype of Servo with Implemented Algorithm and Actuation
Apr. 3, 2019—Calibration Step Actuation
SNR Recalculations
Apr. 3, 2019—SNR values were previously recalculated given changes in baseline considerations from previous studies, as well as, with outputted data from the software preprocessing system to assess are ability to achieve similar SNR in real time, to that seen in previously manually filtered data. We found the values shown below which remained consistent with previous tests,...
Comparison Teensy Processed Data to Raw Signal
Apr. 3, 2019—Software Development Progress with inputted Signals On the left above is the output from the Teensy v3.6 analog pin following intake of raw signals from the EMG microcontrollers, shown on the right, and processing using a Tustin Approximation in the developed software system. Varying contraction levels are evident throughout the approximation as well as varied...
Progress Report 8 (4/03/19)
Apr. 2, 2019—Previous Goals Last week, we wanted to continue working on the algorithm, implementing the buffer management system and beginning to implement the differential algorithm. We also sought to determine the relationship between pulse widths and servo angles and whether or not the servo returns to a “set point” when no pulses are sent. In terms...
Weekly meeting 14 (4/01/19)
Apr. 1, 2019—Classifier Leaning towards using linear piecewise function (other two options are polyfit and piecewise polyfit) Want to include calibration for MVC of each person The x-value will scale Can try plotting each person individually which will show the variability between people Servos Tried altering pulse-width with analog, but could only control frequency Found a library...
Weekly Meeting 13 (3/25/19)
Mar. 25, 2019—SNR Recalculated baseline without the co-contractions and this raised all of the values Initial contraction classifier Recorded raw EMG, Tested using: pure extensor test pure flexor test partial extensor/flexor Tustin approximation is working well Includes a discrete integral in the calculation allowing us to smooth as we go Helps with the flow and speed of...
Pre-processing Results at Varying Contraction Levels
Mar. 19, 2019—Testing of Tustin Approximation Low Pass filtering at 3 Hz. Varying levels of contraction clearly visible.
Progress Report 7 (3/20/19)
Mar. 19, 2019—Previous Goals Last week, we sought to test and finalize the acquisition and preprocessing code, troubleshooting the infinite loop error. We also strived to implement and test the EMG differential and buffering code and finish SNR analysis (determine a primary subject and find the ideal electrode placement). In terms of hardware, we wanted to incorporate...
Example Manually Processed Data
Mar. 19, 2019—Example of manually processed data: The average of the baseline for each muscle (flexor and extensor) was calculated and then subtracted from the entire data set. The absolute value was taken, and the result was graphed.
Weekly Meeting 12 (3/18/19)
Mar. 18, 2019—Emg Calculated SNR Should consider co-contractions as signal, not noise Should increase SNR values Servo motor Control angle We should think about how you can compare servo motor to prosthetic Prosthetic uses variable speed, servo does not Ottobock They will send us an address to send the arm this week Will get appraised for how...
Weekly Meeting 11 (3/11/19)
Mar. 11, 2019—Letter of Support We mentioned last semester We will check in with Dr. Walker EMG We have tested everyone and calculated SNR Currently Christian has best SNR Will retest Ally and Rebecca Matlab processing We will do correlations and matlab preprocessing to correlate MAV to hand angle Import equation into arduino code Two bipolar pairs...
Actuation: Ottobock Communications and Proof of Concept
Feb. 26, 2019—When we started our project, we were given a donated Ottobock prosthetic arm. It was unclear whether or not the prosthesis would work, and after many troubleshooting attempts, we found that the arm was not functional. Upon mechanical isolation between the motor and gears, it was found that no jamming was occurring. We also increased...
Progress Report 6 (2/27/19)
Feb. 26, 2019—Previous Goals We had previously strived to continue hardware addition and construction with battery pack incorporation and the identification of myoware sensor processing issues. We also aimed for translation and effective implementation of PID controller in arduino IDE. EMG testing was begun and we hoped to continue the identification of effective phantom users and become...