Biosensing Final Project: the muscle-driven Hexbot

Team: Angus Kitchell, Charles Elmer, Ross Mansfield

Project Abstract:

With this project, we aim to create an interactive experience that is both fun and thought provoking. Muscle impulses from the drummer inspire the hexbot to move forward, while hits registered from the sticks influence the direction it travels in. People are moved in many ways by music, whether physically or emotionally, or both. To view the hexbots moved by the biosensed impulses of a drummer’s arm and see it dance to the rhythm is to reimagine our understanding of music. We can become the influencer and the observer- and yet still partake in the experience.

This technology could also be viewed as a way to create a multimedia art piece. While we chose to have our muscle imposes control the movement hexbot, these impulses could be transcoded into a wide array of other formats to create a visual output. So, with muscle sensors hooked up to a drummer (as they are in this demonstration), the muscle impulses involved in the drumming process could be converted into an accompanying visual output that is intrinsically tied to the audio output. Thus, a dynamic multimedia piece is born, creating a bi-sensory experience for the audience. By presenting the muscle impulses of a drummer in a visual format, the audience can suddenly experience and appreciate the skill of the drummer (typically evaluated as an auditory output) in a whole new medium, potentially expanding their enjoyment and understanding of the performance. In addition, this concept could expand far beyond drumming. For example, this technology could be used to convert the muscle impulses from a soccer player’s calf, a weightlifter’s biceps, or an orchestra conductor’s forearm into a musical number, a visual display, or a tactile experience.


Project Description:

The key biological component of this project consisted of electromyography, which measures the electric potential associated with muscle activity. We began by testing a muscle sensor (Muscle Sensor v3, available on Sparkfun:, adjusting the positioning of the sensor pads and the sensitivity of the board until we settled on a combination that produced a consistent and controlled output. Given that our demonstration utilized drumming as the means to control the hexbot, we experimented with multiple muscle groups that are active during a drummer’s motion, ultimately targeting the Brachioradialis muscle on the inner forearm.

To convert the muscle impulses into directions for the hexbot, we downloaded an example Arduino code and hacked it to fit the needs of our project. This involved setting a threshold voltage to initiate the command for forward movement and adjusting the regularity of the signal so that the hexbot would respond to muscle activity with minimal lag between the signal and the action. Thus, the regularity of the signal had to be frequent enough that the hexbot could respond to multiple commands made in quick succession (or stop moving shortly after the last command was issued), but with enough space between signals for the mechanical aspects of the hexbot to register and respond to each signal.

Signals were transmitted to the hexbot via the hexbot remote, which we wired to the output ports of the Arduino board. This required a wires to be soldered to the contacts for each command on the remote, allowing the outgoing signals from the Arduino board to be transmitted through the remote’s LED transmitter. Thus, while the remote was still functional, we simply used it as a “transmitting tower.” When the incoming signals from the muscle sensor reached the required threshold, an outgoing signal was sent through the remote to command to hexbot to move forward.

Side-to-side motion was controlled by the vibrations created by drumming. In this case, we placed a piezo element ( between two notebooks (acting as our drum pad), and wired it to the Arduino board. Vibrations registered by the piezo element were transcoded into a signal that commanded the the hexbot to turn right. While we had hoped to have one piezo element for right turns and another for left turns, we were only able to use one successfully, resulting in a hexbot that only turned right. Again, the outgoing signal created by the Arduino board was wired directly to the “right” command on the hexbot remote, such that any vibrations that passed the command threshold caused to hexbot to turn right.

We used a breadboard to facilitate the wiring involved in this project, connecting the muscle sensor to its batteries, as well as the piezo element and the Arduino board. A basic summary of the wiring for the muscle sensor can be seen in the figure below.sensordiagram

Personal Contribution: 

This was definitely more of a learning project than a leading project for me. As someone who has little experience in coding and wiring, my ability to help on a technical level was fairly limited. However, our partner Ross is a very capable technician who found time to explain some details to me as he was working. As a result, I gained a basic conception of the processes required to implement a project of this nature, as well as a better understanding of how to actually perform the technical aspects. For example, I was introduced to coding when Ross walked me through the sample code for the Arduino board that he had taken from and hacked to suit the needs of our project. He then showed me how to upload a program from a computer to an Arduino board, which is the crucial step in transcoding incoming biological signals (muscle impulses) into outgoing digital signals. Given that this required multiple pieces of technology to work in unison with each other, I was also introduced to some basic wiring and circuity, including the value of a breadboard for connecting multiple pieces of technology into a single circuit. I was able to absorb some information as I watched Ross disassemble the Hexbot remote and solder new wires onto the command contacts. Since my ability to help with technical aspects of this project was fairly limited, I volunteered to act as a testing dummy for the muscle sensor, helping Ross to troubleshoot the system when we came up against early difficulties with the equipment’s sensitivity (at first un-responsive, then overly sensitive). Again, I experimented with multiple muscle locations before settling on the forearm as an effective muscle group for our activity (drumming). At the end of the project, I took responsibility for writing the project description. This served as a good way to recap all that I had learned through this project, and present the multiple steps in a cohesive format.
Overall, my greatest contribution was in the conceptual phase of this project. My background in ecology has had a profound impact on my worldview, and I enjoyed the opportunity to incorporate biological systems or properties with technology. This project was a great example of the value of multiple persepectives/backgrounds in a group project, and was one of most rewarding experiences I have had with an interdisciplinary course. During the “initial ideas” phase of this project, I came up with a variety of ways in which I wanted to see the biological world integrated with the digital world, and Ross and Charles were able to take my ideas and explain how they could be implemented with technology. For example, I expressed an interest in creating a robot that would interact with its environment in a biological manner, responding to stimuli and acting without the need for a human to issue commands. In response, Ross and Charles suggested that we could utilize the infrared signaling capabilities of the Hexbot by placing it within a network of infrared “pillars” or “signal towers” with which it could interact, sending and receiving signals and changing its movement in response to the position of other objects in its environment. As an alternative, Ross suggested that we could attach a sensor to the bottom of the Hexbot that would allow it to register differences in light/color, allowing it to follow (or avoid) paths or markings that we created on its walking surface. For example, using a sharpie to draw a dark black path (or obstacles) on a large sheet of white paper, we could create an “environment” for the hexbot to interact with. Thus, this bio-sensing project was almost like a meta-project for me: as I developed ideas with my biology-minded brain, Ross and Charles “transcoded” those ideas into technical versions, allowing them to be implemented in class.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s