Final Reflection

 It has been two and a half months working as a research intern in the Collaborative Robotics Lab at Purdue University. Words cannot describe how grateful I am for the experience and knowledge gained during the internship. When I first started, I could not possibly imagine that I would be able to go this far in understanding, analyzing, and researching the hard concepts, structures, and algorithms in Artificial Intelligence and Robotics. As I worked mostly with AI, I have a lot to say about the progress that I have made as a researcher in this field. Starting off with the first few weeks in the lab, I can say that I felt overwhelmed and frustrated with all the new information and tools that I needed to comprehend and implement. I was hoping to work in a team of other research interns so that I would be able to collaborate and solve problems with other people. However, I had to work by myself as all the interns were assigned to different projects. At first, it scared me because I was not confident enough in my own abilities and skills. However, I persisted in persuading myself that I could do it, and that is exactly what happened. Such a shift in my personality changed me as a whole, and I became more confident in this professional field of study. Day after day, week after week, I was telling myself that I could work with no teammates and still progress as efficiently and even more. That worked magically because I was able to work on hand gesture recognition with absolutely no knowledge of what it is and how it works, and now I am knowledgeable enough to explain the concept to fellow interns or even my friends who are not CS-related. It is almost unbelievable to imagine that I worked on the project in machine learning, even though in college, I had not taken any related classes and had only experience with Python and algorithms, which were perhaps the only useful skills that I brought with me into this internship. In my project that was related to hand gesture recognition, I had to get some fundamental understanding of deep learning and, more specifically, reinforcement learning. Basically, reinforcement learning refers to the idea of a robot learning basic human skills from a human by either being trained or by learning from the algorithm. The idea sounded simple until I started getting more familiar with it. And that is the stage where I started having trouble with the project. I had to create an entire dataset of hand movements (moving my hand up, down, left, and right) using a 3D camera. The dataset should have included eighty videos overall, meaning twenty videos of each movement. I succeeded in this task; however, not from the first or even second trial because I had to fix a lot of issues related to YOLO. YOLO is a type of convolutional network that is used for real-time object detection. Basically, it detects the confidence level of the desired object to be what it is supposed to be, and if the desired object is on a certain confidence level or higher, it puts the object into the bounding box, where the computer recognizes the object. After countless trials and errors, I was able to create a suitable YOLO dataset that I used for my further progress. While working with the dataset, I was able to learn many techniques related to image and video frame manipulation and processing, which is the area that I am highly interested in but had never had prior experience with. I am glad I worked with it because now I got more experience with OpenCV and ROS. I have never taken any classes at Berea College or outside of it related to robotics or AI. So, I believe that it was very challenging for me to work on a project that includes computer vision, deep learning, machine learning, reinforcement learning, image processing, artificial intelligence, convolutional neural networks, and classification methods. I had to become more professional in the area of research because it was the only way to produce any results since all the areas that I studied here were novel to me. I believe researching is one of the most useful skills that I have developed during this internship because I need to be ready for explorations to achieve new results in the AI world. As I am planning to enter the world of academia right after I graduate from Berea College, I believe that this internship is crucial for my future decisions on a career path and my acceptance into graduate school. The things that I have been doing and experimenting with are mostly the concepts and ideas used by graduate and Ph.D. students. It is unbelievable for me now that I was able to stay on the same track and keep up with a similar pace with those intelligent people in the lab. Additionally, I overcame all the obstacles and successfully achieved my goals in training and testing the data for reinforcement learning. I got experience in extracting data, interpolating it, plotting it on the scatterplot, and classifying it for producing the trained and tested results for reinforcement learning. All of this work enhanced my knowledge of linear algebra, statistics, and probability, which are classes I have never taken before. Realizing that gave me so much inspiration to pursue the expansion of my intellectual, technical, analytical, creative, and problem-solving abilities. I can confidently call myself a scientist now because, for me, a scientist is not just a person with a higher degree. It is a person who strives to know more, learn, research, question, analyze, create, and most importantly, discover more. Even though I do not yet hold any qualifications for calling myself a scientist, I can definitely call myself a scientist in spirit because that is exactly what I have shaped myself into being.