Bot-thoven: A Robot Musician

Plan

This week, I will construct more servo-mallet systems to use for testing the control of multiple servo movements with my PCA9685 PWM/Servo Driver from Adafruit.

Procedures

I took the stands which finished 3D printing from last week and attached the servo and mallet onto it to build three additional complete servo-mallet systems. I also had to solder all the pins onto the Servo Driver which arrived from Adafruit. 

Progress

Above is an image of the Servo Driver from Adafruit which arrived this week. The image shows the driver after I completed soldering on all of the pins. 

After completing putting together my new additional servo-mallet systems (shown above), I attached the servos to my Servo Driver shield. I spent Thursday taking the time to understand the PCA9685 library and running the example “Servo” code onto my board. Running the example code allowed me to verify that the Servo Driver works as intended in controlling multiple servo movements at the same time through just two analog pins. 

Problems

I will need to take what I have learned from familiarizing myself with the PCA9685 library and update the example code to have the servos move along my desired goal of appropriately striking a xylophone key. 

Plan

This week, I will continue printing more copies of the prototype of the servo-mallet system.

Procedures

I will continue 3D printing out more copies of the stand prototype and building more servo-mallet systems. This will allow me to test my code for varying servo speeds (for dynamics when hitting the xylophone key) and see if it will work when the microcontroller has to control several servos at once.

Progress

Above is an image from the CURA 3D printing software detailing the layout of my current print. The print is currently in progress (image below) and will be finished by Thursday evening.

I have also placed my order for the 16-Channel 12-bit PWM/Servo Driver from Adafruit this week, which will allow me to test if the shield is compatible with my microcontroller. 

Problems

Once the shield from Adafruit arrives, I will need to take all the servo-mallet systems that I made this week and test the shield to make sure it functions as intended: allowing a single microcontroller to control multiple servos at once. 

Plan

This week, I will print the prototype for the servo-mallet system I will be using. 

Procedures

Using the initial CAD I designed, I will use a 3D Printer to print out a stand to create my first prototype for the servo-mallet system. 

Progress

Below is the first physical prototype I created for the servo-mallet system.

To address the servo size problem I brought up in the past several weeks, David and I disassembled the xylophone into two separate parts to free up space around the xylophone frame. The wooden blocks show where the servo-mallet systems will be placed (all around the xylophone) to ensure we can fit all the servos.

Technical Specification

The microcontroller which I am using is the Arduino Leonardo. The technical specification, obtained from the Arduino website it shown below. 

All the parts I will be using for my project function and connect to the board as intended. A video of my HITEC Servo functioning when connected to the Leonardo microcontroller is available to view here. 

Problems

I will need to take the problems I found with my first prototype and create an improved design for the second prototype.

Plan

This week, I will test the prototype for the servo-mallet system I will be using. 

Procedures

Using the servo-mallet system I built last week,  I will continue testing out the prototype to find areas to fix and improve on. 

Progress

Below is an image of the servo-mallet system I have built as a first prototype. Currently, David and I are testing out the servo to move at different speeds, in order to achieve hitting the xylophone key with varying dynamics (loudness). 

In terms of fitting all 30 servos across the xylophone, I have decided to test the method of staggering mallet length. This week, I have built copies of the servo-mallet prototype. With these additional mallets, I will be able to physically test this method next week to see if it will work. If not, I will look into other options such as building a new, wider frame for the xylophone keys or using a smaller servo as a replacement. 

Problems

I will need to test the staggering mallet length method to see if I will be able to resolve my problem regarding how my servos are currently too big to fit 30 of these systems across all of the xylophone keys.

Plan

This week, I will create a prototype for the servo-mallet system I will be using. 

Procedures

Using the materials which arrived from Amazon last week, I built my first servo-mallet prototype to test it out on a xylophone. 

Progress

I first tested my code on varying the mallet speed by changing the servo’s rotation degrees per second. A video can be found here. 

I followed up by testing out the system onto the actual xylophone. A video can be found here. 

Problems

After testing out the servo-mallet system on a xylophone, I realized that the size of the servos is currently too big to fit 30 of these systems and align it with all of the xylophone keys. For the upcoming weeks, I will need to think of a way to fit all 30 servo-mallet systems across my xylophone, whether it is using new servos, extending the xylophone frame, or modifying my mallet system.

Plan

This week, I will create CAD for the servo-mallet system prototypes I will be making.

Procedures

I used Fusion360 and the parts which arrived from Amazon to come up with an initial CAD design for the servo-mallet system. 

Progress

The mallets, zip ties, and xylophone arrived through Amazon as pictured above. 

I finished an initial CAD design showing how I will use the mallet, a servo, and zip tie to create a servo-mallet system to strike a xylophone key. 

Problems

I will need to begin testing out the prototype by 3D printing the body (in green). By testing out the initial design, I will be able to make adjustments and changes as necessary to improve the servo-mallet system.

Plan

This week, I need to find additional outside sources of support on my project and plan on contacting them for help. 

Procedures

I looked to find other researchers who are currently working to integrate the field of music with robotics. 

Progress

I found that Georgia Tech has a Center for Music Technology, where one of their main research topics is Robotic Musicianship. A possible source of support I found was Gil Weinberg, a professor at Georgia Tech who leads the Robotic Musicianship research group. He will be a great possible outside source to reach out to, due to his work and focus on developing artificial creativity and musical expression in robots.  

Problems

I will need to begin reaching out to outside sources whenever I am in need of help. Once my materials (xylophone, mallets, and servos) arrive, I will need to create a prototype of the servo-mallet system and test it on a xylophone key to see if the design is stable.

Plan

This week, I need to find outside sources of support on my project. In addition, I need to meet with David for our monthly check-up since it is now October.

Procedures

I looked into possible outside sources of support by finding other researchers working in the field of improving expressivity of instrument playing robots. I also met with David during 8th Period on Wednesday to check on his progress with mine. 

Progress

I found NIME - an International Conference on New Interfaces for Musical Expression. This annual conference gathers researchers from all around the world to share their work on new musical interface. The website contains hundreds of publications which have been published at the NIME conference, providing with numerous possible sources of contact. One possible source of contact I found is Steven Kemper - the Associate Professor of Music Technology and Composition at the Mason Gross School of the Arts at Rutgers, The State University of New Jersey.

When meeting with David, we were able to take his code (which reads in a MIDI file and creates an output file for the Arduino interface to read) and have LEDs play the tune of Yankee Doodle and Hot Cross Buns. 

Videos can be found here:

• Yankee Doodle 
• Hot Cross Buns

Problems

Once my materials (xylophone, mallets, and servos) arrive, I will need to create a prototype of the servo-mallet system and test it on a xylophone key to see if the design is stable.

Plan

This week, I finished my Project Proposal and created a preliminary design for the servo-mallet system I will be using to conduct my project. 

Procedures

I looked into examples of xylophone-playing robots made by other individuals to see how they were able to attach a mallet to a servo. 

Progress

My project proposal is available here. 

The image above shows the proposed design of my xylophone-playing robot. The mallet will be attached to a long servo horn and fastened using a zip tie. I decided to vary the rotation degrees per second on the servo motor to achieve different dynamics for my xylophone robot. 

Problems

I will need to create prototypes of the servo-mallet system and test it on a xylophone key to see if the design functions and is stable. 

Following my proposal presentation, I received feedback that I should find outside sources of support who I could reach out to for advice on my project. I will need to look into possible researchers in the field of robotic instruments.

Plan

This week, I finished my lab purchase sheet to decide the materials I will need to purchase for the project. 

Procedures

Found the servos, microcontroller, shield, and instrument I will need to carry out my project. 

Progress

The complete spreadsheet of all the materials can be found here:

https://docs.google.com/spreadsheets/d/1ogU2wCfnevOq0EUQLhm68Ls9j8SELpvh59MMGZaAmLA/edit?usp=sharing 

I also checked the Robotics Lab storage room to check which materials are already available for me to use (servos, microcontroller). 

Problems

For next week, I will need to complete my Project Proposal.