Master's Spotlight: Dan Stoianovici and the Creation of Nyku
Authored by Dan Stoianovici
From the start of my Master’s program at DU, I already knew what my thesis was going to be on. Having already worked on Dr. Mahoor’s Dream Face team and having developed a passion for robotics in undergrad, I was tasked with building a robot to meet needs of children with Autism Spectrum Disorders (ASD), which we called Nyku. Of course when I started I had no idea what this would entail. Before this project most of the requirements I had been designing to meet were hard engineering numbers—dig the hole this deep, lift this specific object, move the arm at 60 deg/second. However, with this robot the design was based much more in aesthetics and other hard to quantify metrics which in the end, I think that made it a much more fun project to work on.
I started the project as any engineer would with an analysis of the needs of ASD children. I spent the first year of my Mechatronics Systems Engineering degree researching Autism Spectrum Disorders. Coming from a technical background, this was a large but very welcomed shift in the type of work I was doing. During this period, I also volunteered as a caretaker with a few ASD children in order to better understand their needs. Although I read a lot about how ASDs present themselves, gaining the experience firsthand helped me develop a new level of empathy for both the children and their caregivers, which helped motivate me in the trickier times of the process.
Having gotten started so early, I had to opportunity to go through a lot of failed design iterations. One of which resulted in a few broken monitors and a bloodied chest when I exploded a flywheel in the DUCV lab. Luckily, Dr. Mahoor kept faith and let me work through the design process for Nyku. From the start, we knew that Nyku had to be appealing to children and not be intimidating, while being able to pack a serious punch when it came to therapeutic potential and computing power. So Nyku was designed to look like a penguin and be capable of posture mimicry using computer vision techniques among other things.
In the end I was able to develop Nyku into a full robot, complete with posture mimicry. Sadly, due to COVID-19 we were not able to run trials with ASD children using Nyku, so my thesis was presented on the design of the social robot. The original intention was to have the children play video games with Nyku, using the positive and negative rewards given by the game as teaching moments to describe what postures relate to which emotions, as well as which emotions/behaviors are appropriate for what stimuli. Although Nyku is ready for these trials, the system which runs them must still be orchestrated and the research plan must have IRB approval. Despite these hiccups, a lot was accomplished. First, because of the design of the base which allows for posture mimicry, a novel solution to spherical manipulator using three omni wheels was developed which we intend on publishing in ICRA later this year. In addition to this, DU has decided to fund a non-provisional patent for Nyku, which has been very rewarding for me as a young inventor.
Being a robot, the scope of the project was equal parts mechanical, electrical, and computer engineering. And as a result I had to ask for help from most of the faculty. When it came time to machine a part or validate a mechanism, it was amazing to be able to go down into the Machine Shop and ask Justin Huff for his advice. Or walk up to the fourth floor and ask Dr. Davidson if I was applying dynamics concepts correctly. When I needed a circuit validated, I could always ask Dr. Martins. When my code needed checking or simply needing a second pair of eyes there was always a lending hand in my lab from Rohola, Ali, or Soroush. As I was defending my thesis during COVID-19, I was able to invite all the people who have helped me along to the way to participate in my Zoom presentation, which was a small silver lining to presenting to an empty room. The defense was by far the best part of the thesis process and it was an honor to be able to share my work with the academic community, my parents, and my friends. After all the work is done and the writing finished, you are so comfortable with your thesis that presenting is fun and you can finally look back at all your work and see it in its entirety. Though I did underestimate how long it would take me to make the presentation, so I pulled one last all-nighter in my academic career.
When I first started at DU as an undergrad, I did not really understand what the purpose of a university is, it was my privilege that made it the next step in my education. Sure, they produce research and help advance science, but that seemed very detached from the intro courses we take. During my time spent at the Ritchie School, I came to realize I wasn’t just there to earn a degree. I was there to spend time with the smartest people I’ve ever met and develop personal relationships with people from every walk of life, learning much more than engineering along the way. Over these last months in quarantine I have sorely missed the Ritchie School cohort and it has made saying goodbye very difficult. So I’d like to say thank you for the opportunity to serve as a TA and be a part of the community.
*The title has been updated to correctly reflect Dan Stoianovici's degree of a master's degree rather than a PhD.