Using Capacitive Sensors for Limb Posture Analysis

Presenting my work at the 2024 IEEE Green Technologies poster session

Author: Abhinav Komanduri | Major: Electrical Engineering | Semester: Spring 2024

I first heard about Dr. Nelson when he won the National Science Foundation CAREER award, given to rising research faculty for high-impact work. When I started in the lab, we had a meeting where he just tried to gauge my interests. My first task was to read through his grant and pick out areas where I thought I could be of service. I quickly identified that my area of interest was in embedded systems and printed circuit board (PCB) designs. Dr. Nelson’s underlying NSF grant was investigating different technologies that can be developed to be used in capacitive sensing to help rehabilitate patients of strokes and ischemic attacks. I was very excited to begin working when I realized that my work could go toward helping other people in need.

Capacitance is the ability to hold an electric charge. A gap between two objects with different electric potentials causes an electric field to form. A new and ubiquitous technology is Capacitive sensors that use this principle and can detect human touch. When we, as conductors, touch a capacitor we begin to shunt the charge away from it. From this, Capacitive Sensor Arrays (CSAs) are derived which are grids of capacitive sensors that are mutually coupled with one another. One useful application of CSAs is in the pose estimation of human or robotic limbs. CSAs can provide valuable information about joint position, movement speed, and much more. It would also be feasible, cost-effective, and low power, meaning patients can afford it and receive high-quality care while having control over all the data generated.

In the spring grant period, my work took these CSAs and found a way to extend their resolution. Currently, the CSAs are powered by a Texas Instrument microcontroller (MCU) which can only support up to 64 sensors. I decided to use the method of daisy-chaining to connect multiple of the MCUs together to support a higher number of sensors. Similar to how a monitor with more pixels has better resolution, more CSA sensors will translate to better data. Figure 1 shows a diagram representation of how I implemented my idea. To connect the boards together, a serial protocol known as the Inter-Integrated Circuit (I2C) protocol and data could be transmitted between boards. The results I got were that daisy-chaining 128 sensors together could transmit data in 630 microseconds with no data loss, proving that it was a fast and reliable method to expand sensors.

I had the opportunity to present this work at the 2024 IEEE Green Technologies Conference and publish the paper in the conference proceedings. I was also able to present this work at the 2024 UARK Honors Research Symposium. This work is still ongoing, and my next steps are to simplify the design I created into a singular printed circuit board.

In my time with this project and in the AESIR lab, I acquired a great many skills and interests. I was able to refine my skills in electronics and get a much better understanding of the low-level hardware working of the microcontrollers. I also got to refine one of the most important skills in research – communication. With the grant proposals, papers, posters, and presentations I was able to greatly improve how I communicate my results and convey the message of why my work is important.