Author: Leonardo Georgescu, Email: georgl3@unlv.nevada.edu
Date Last Modified: 06/18/2018

Hi, my name is Leonardo Georgescu. I departed Las Vegas, Nevada on June 1, 2018 and I will be coming back on August 10, 2018. For the duration of the trip, I will be staying in Daejeon, South Korea, and I will study at the Unmanned Systems Lab under Professor David Shim at KAIST, a world renowned research university. This journal will not only show my research progress, but also my cultural experience. In each entry I will be discussing the following:

• 1. Photo of the Week
• 2. New Korean that I met
• 3. Cultural Insight
• 4. What did I learn about myself?
• 5. Research Status

First week in South Korea has been one of the most exciting weeks of my life. I not only focused on research, but I also had the honor to meet some Korean locals who showed me around Daejeon. The objective of this week was to get accustomed with the university, dorm, and Daejeon.

Even though I am not part of the Hubo Lab at KAIST, this is my photo of the week because I was extremely impressed upon arriving at their lab. Once we arrived at KAIST, our first point of contact was at the HUBO Lab. They were extremely nice, as many other Koreans, and helped us get settled into our dorms. Additionally, I caught the cold last week and needed to get some medicine from the pharmacy, but it was a holiday and the pharmacy on campus was closed. I reached out to them and they were quick to help me find a pharmacy around KAIST that was open. Also, later upon arrival they greeted us with some outdoor Korean barbecue. This was one of my first experiences with how friendly and helpful Koreans are. They were very quick to accept us in their group. It helps knowing that there is someone willing to help us if something does not work out. Their lab was also very impressive. Most of the things in their lab are manufactured there. They have someone who works only with the CNC machine. I was shocked by the number of Hubo robots they have. It was great being able to compare out lab in Las Vegas to the lab here at KAIST.

This is Professor Lee and I in front of the Donghaksa Buddhist Temple. Professor Lee is a professor at Woosong College teaching social issues. It was a great pleasure meeting him and being able to experience Korean culture with someone as knowledgeable as him. He is trying to practice his English while showing me around Korea. I cannot thank him enough for his kindness. I have learned so much about Korea from him.

Professor Lee took me to a Korean traditional restaurant. It was my first time experiencing this and I enjoyed every part of it. We had to take off our shoes and sit on the floor. The food was extremely delicious and the table was packed with a variety of vegetables and meat. The rice was served in a very interesting dish. This also let me firsthand explore true Korean food in the countryside.

After we finished eating, Professor Lee took me to the famous Donghaksa Buddhist Temple near Daejeon. I have always wanted to visit a Buddhist Temple, but I never had the opportunity. After parking the car, we walked up the mountain for 30 minutes and I arrived at this beautiful temple. The air was very clean and the atmosphere was peaceful. Professor Lee showed me some Buddhist prayers that I was able to partake in.

Professor Lee is very enthusiastic about connecting with foreigners and showing them Korean culture. I appreciate him taking the time out of his busy schedule to show me around.

Something very interesting about Korea that probably surprises many foreigners is that Koreans like to share. Whether I would go to a small seafood place in downtown Seoul or a traditional restaurant, Koreans like to share their food. In America we tend to order what we want from the menu and each person eats what is on his or her plate. On the contrary, in Korea everything is put on the table and shared between the diners. I think this is very important in creating a bond between the people. While eating in America feels more independent, in Korea it feels more group orientated. I enjoy going out and eating in Korea as it helps create better relationships with people.

Yu Hang and I went about our own ways and ended up meeting at a palace in Seoul. Before arriving in Korea, I was extremely worried about getting lost. In this picture, it shows both of us ending up where we wanted to be without getting lost. It's a little intimidating at first to be in a different place, but Yu Hang and I learned very fast. Our second day in Seoul we knew how to navigate around using the subway and bus. Obvious, the Naver app on the phone helped a lot. Even when we didn't speak the language, Koreans would do their best to help us around. I learned that I can survive even in a country that I don't speak the language.

I learned very much at the lab this past week. On the third day at the lab, the members took me for a test flight. A master's student it working on creating a deep learning program for collision avoidance with another aircraft. It surprised me how much equipment they have and how expensive it is. They flew the plane using high end military equipment and filmed it using a 4K DJI drone.

Of course, the purpose of my trip to Korea is to learn more about drones and RTK. One lab member designed a flight controller that they use for all the drones. Additionally, he programmed the software. I am surprised that they make their own products. I guess this is a very good way of having them really learn about what they're designing. I will be able to learn more about flight controllers from the people that made their own.

Last week I was able to learn how to use Autodesk Eagle to create transistors, op amps, and other electric components. Additionally, I was able to design a circuit board in Eagle. This is extremely useful to know since most of the things at this lab are made from scratch. Additionally, I would like to design my own circuit board and have it printed using the CNC machine once I arrive back at the DASL Lab.

I learned about flight controllers, and how they work. I learned the basic code programming for a drone using inertial measurements. First, I learned that a drone code has three main components to it. The first is for the controller to read sensor values, rotational rates and angles. Then we interpret what those mean by using time intervals. The second part is reading values from the remote control. The third part is the most difficult, and it involves processing the data to be able to create different flight modes such as stabilize, acro mode, etc.

Before coming to Korea, I was having issues with landing the RC fixed-wing. I spent a lot of time this week researching different methods to provide precise waypoints. Asking their advice, I was able to come up with a couple of plans for the skywriting project and how to get the best drawings on the sky.

One of my side projects is learning python and programming using DroneKit which allows an interface between a raspberry pi and pixhawk. I think this will be extremely useful in the skywriting project.

This week I will focus intensely on learning how to draw an image, interpret the image as GPS coordinates, and have them sent to the drone. Additionally, I will be working with a Master's student on getting the 50 drones to take off from one position instead of having them scattered around.

Being around these lab members helped motivate me to do more work. I am able to see their work ethic, as well as the interesting projects they are working on. I look forward to learning much more in the following weeks.

The focus of this week was integrating more into the lab, and working on a project. Surprisingly, getting used to living in the dormitory did not take very long.

I chose this as photo of the week because it accomplished part of my goal this last two weeks - integrating in the lab. The lab members are extremely welcoming and helpful, and they go out of their way to help me feel as I belong in the lab. The two lab members in the picture are Dasol Lee and Jamie. Dasol Lee is the lab member in charge of the drone swarm, as well as making his own flight controller and programming it. There were more lab members at the dinner, but I also wanted to get a picture of the delicious food as well. This was another traditional restaurant, and I had very good cold noodles. They may not sound good, but trust me, they were amazing. This dinner also helped us bond and get to know each other better.

Mr. Kokyungmo and his friend approached me on the street and asked me where I was from; they were very excited to find out I was from Las Vegas. I was surprised how well he spoke English. We exchanged business cards, and he wants to meet again to talk for longer. Unfortunately, I do not have a picture with him, but I will probably meet him this weekend. I plan on going out and meeting more people this weekend.

Towards the end is Inhwan Wee, a master's student in my lab who I am working with. The reason I chose to post this picture under cultural insight is to bring notice to the famous Korean chicken. I was so impressed how delicious it was, and how much Koreans care about the quality of it. I ate chicken at different places and it was the best chicken I have ever had. The chicken also comes in variety, so I could choose between spicy or with lots of cheese. I would totally recommend to anyone going to Korea that they try the Korean chicken.

I was impressed by the effort that the Korean government is putting into making the country more environmentally friendly. This is a parking lot to a park, and the cars are covered by solar panels. I would love to see more of these in Las Vegas because it would keep our cars cool, but also help promote renewable energy.

Again, I am impressed by how many parks there are, and how clean the Korean local government keeps them. This was a park next to the government center, and it was very well managed. I went to the park during the holiday, but it was still quite empty. Korea has many places to relax after a long day at work.

This was a forest resort south of Daejeon. It was very nice to sit down, eat some traditional Korean food, and listen to the quiet surroundings. The buildings in this picture are very traditional, as well as the design of the walkway.

Before coming to Korea, I was worried about getting used to the food. I had Korean food in Las Vegas, but I knew it was going to be different. I was shocked to see myself eat and enjoy Korean food. Actually, I want to always try different dishes. Even my Korean friends said they were surprised by how well I ate Korean food and enjoyed it. Apparently foreigners have a reputation for only eating burgers and chicken, so I am glad that I was able to show them that I like Korean food. I also learned that I enjoy trying different food, and look forward to trying different things.

The beginning of the week was mainly installing software - Linux, ROS, Python, Dronekit, APM Planner and OpenCV on my Macbook which proved to be quite problematic. First, I had to learn to navigate around Linux again because I have not worked with it in the past.

As I am progressing through this week, I keep thinking of new ideas for the skywriting paper. Dongbin, our lab manager, sent me an interesting paper title regarding path planning in wind fields. Obviously, to understand the paper I have to learn how an RC plane works. There is a book online that I found regarding fixed-wing aircrafts and the dynamics behind it. I spent a lot of time this week learning about the transformation from vehicle to body frame and MAV coordinate frames. These are very important concepts for the skywriting paper. It will also allow me to perform simulation in matlab.

My first project this week was learning some Python and getting an image into pixels so it can be transformed into GPS coordinates, which I will learn the next few days.

Here is my progress into the project. I am using Ihwan and Dasol's paper, “Button Filtering for Converting Image to Coordinates of Swarming Drones,” as reference.

This was the original picture I am using. The goal is to position the drone swarm to write UNLV in the sky. With this process I can take any picture and transform it into GPS coordinates.

The first step is using OpenCV to eliminate the background and only keep the letters that I am interested in.

After, I use the SLIC segmentation code to create the super pixels of the letter which will be the position of the drones.

The picture before is an SLIC image, and then I use the Canny Edge Detection to create better edges.

The next step will be to add the Button Filtering, which I forgot to do, and the pixels are not showing. There is a lot of future work involving this, such as developing a depth map and a color image to create an image similar to what Intel has created. This week my goal is to perfect the image drawing process, and learn how to precisely transfer these data points to the drone code. I believe that studying this can also help with the skywriting project since it's based on similar concepts, drawing in the sky. Instead of using the Mission Planner Waypoint Software, I would like to adapt this method for skywriting since it's more accurate.

More drones, approximately 300, will arrive in the next week, so Professor Shim asked if I could help in any way possible. I am more than excited to be able to learn from these lab members.