During this semester, many advances were made. While shields were ordered for the RaspberryPi, smooth movement algorithms were tested using an Arduino. The algorithm can be found below. In order to achieve a smooth movement, the servo motors have to step one step at a time until it reaches its final position. These steps require a 20 ms delay to make the prosthetic move smoothly. All of the servos can move individually and make basic gestures. Issues arose when three servo motors were driven to make more complex gestures. The two wrist servos (HITEC HS-422) can be powered together to have an output of a waving motion (a video is attached below along with the program). Issues arose when a third servo was introduced to raise the elbow. The servo in the elbow is an Alturn ABDS-996HTG+HV. The Arduino powered these servos with different duty cycles, which leads me to believe that this causes the issue. The elbow motion video below is supposed to incorporate the waving motion as the elbow bends like a humanlike wave. Along with these achievements, the RaspberryPi was received with all of the required shields for this project. The Pi’s Linux operating system was installed and is now up and running.

The next steps for over break are to transfer the servo movement from the Arduino to the Pi along with correcting the duty cycle error so different servos can be driven at once. Then, EEG/EMG signals will be analyzed and mapped for certain movement patterns using PowerLab 8/35. Lastly, if the shoulder servo is received it will be installed on the prosthetic. 

To begin next semester, the Pi’s A/D converter shield will be experimented with. Once the hardware is understood, databases will be created for the EEG/EMG signals. This will set everything up for creating the full algorithm for the prosthetic.

Videos:

Waving Motion

Elbow Motion

Program:

SmoothServoProgram

Servo:

BephaMart High Voltage High Torque Servo

This is the last week of the semester. The servo motors on the arm had to be readjusted so the arm is actually a right arm and can move like one. Tests were done to check the proper angles of each servo to make the movements lifelike. These notes can be found below in the scanned notebook. Also during this week, basic smooth gestures using the arm were made (allowing the wrist and hand to wave). When applying more complex gestures issues arose. The two wrist servos and the one elbow servo wouldn’t run together. Therefore, the arm couldn’t lift at the elbow and wave. I believe it is due to the different duty cycles being received by the servos. Tests can be found in the notebook. Lastly, data was given to Karissa and I based on the brain signals. The results can be found in the scanned notebook below. It shows that the electrode placements on channels 3 and 5 produce the same signal. One final update will be made for the progress at the end of the semester and the next steps.

Scanned Notebook: Week-15