By Kurtis Chien-Young
Amel Hassan is a Tufts junior researching Autonomous Intelligent Robots with Professor Jivko Sinapov. She is a recipient of the 2018 Laidlaw Scholarship, which funds Tufts students for research projects that span the course of two summers.
1. What does research with autonomous robots look like?
There are a variety of applications and research topics regarding autonomous robots. In the Autonomous Intelligent Robotics (AIR) Lab where I work, we focus on Autonomous Intelligent Service Robots – simply, robots that help out or interact with humans. Research with autonomous robots is collaborative – humans and robots working together. It requires a mutual understanding; the human has to understand what the robot is doing and the robot has to understand what the human wants it to do.
2. What do you specifically study through your research project?
During my summer research I worked alongside my peer and mentor to develop a platform to aid in human-robot interaction. We created an app that uses Augmented Reality to project what the robot is “thinking.” That is, the robot shares its data with the user through visual representations. The user is able to visualize the robot’s laser scan, path, costmap, and more. For instance, through an iPad’s camera, the user can see red particles overlaying a screen that represent the objects and edges that the robot detects with its camera. The project required setting up communication protocols between the robot and our app as well as data processing and visualizations.
This semester, we are using the platform we created over the summer in a research study. The research study explores how augmented reality can improve human navigation around robots. The study will take place in a tight hallway, with the research participant at one end and the robot at the other. As the participant and robot move towards each other, the goal is to avoid a collision. Our app will be used in the study as the variable that is predicated to improve human-robot navigation. It will take in the robot’s intended path and determine whether it is going to turn left or right. Blinkers in augmented reality will indicate to the user which way the robot is going. (The control trial will not be using our app.)
3. What are some of the challenges you face when researching? How do you go about solving the problems you come across?
The biggest challenge I face when researching is trying to troubleshoot pieces of the project that I do not understand fully. Since we use many external tools in our project, it is impossible to know and understand every detail of the inner workings which usually makes it hard to debug problems. I combat these challenges by researching, looking at source code, referencing examples online, and discussing with peers and mentors.
4. Why is your field of study important?
This field is becoming increasing important due to its wide variety of applications. For instance, autonomous robots can be used in dangerous situations such as search and rescue scenarios. They are also used in manufacturing and companies such as Amazon’s Kiva robots.
5. Do you have any advice for students who are interested in applying to become Laidlaw scholars? Any general advice for undergraduates who are looking to get into computer science / robotics research?
Prior to LaidLaw, I had little experience with research in Robotics. LaidLaw really gives you the opportunity to explore and grow in a field you are interested in but haven’t had a lot of exposure to.
Its normal to feel intimidated. In fields like robotics and computer science, there are new developments each day and it is nearly impossible to keep up, so it’s ok not to know every detail about technology. If you are looking into getting into computer science / robotics research, email or go to the office hours of a professor whose research interests you. Make sure you know what their field of interest is, preferably even projects they are working on, and even if you don’t have a lot of experience, be enthusiastic and demonstrate your willingness and eagerness to learn. If you don’t have a lot of experience, another piece of advice is to take a class with the professor you want to be your mentor for your LaidLaw project so that you can learn more about the concepts, software, and tools.