Five Computer Science students are working on a new capstone project conducting collaborative research with Dr. Bryan Donyanavard.
“The purpose of our capstone project is to measure the performance and efficiency of state-of-the-art perception and control algorithms executing on the latest commercial off-the-shelf embedded devices,” said Brycen Bissell, one of the five students.
They’re examining these algorithms by “simulating an autonomous vehicle moving and navigating in its virtual environment,” said Kathryn Shafer, another student working with Dr. Donyanavard.
The movements of the autonomous vehicle are controlled by a single board computer capable of supercomputer performance, the NVIDIA Jetson Navier NX, sending signals to a physics simulator called Ignition Gazebo. The simulator provides feedback about how efficiently the autonomous vehicle is navigating its environment, with information like speed, acceleration, and what the vehicle can see.
“After successfully integrating these components, we will evaluate the performance and efficiency of the perception and control algorithms with metrics such as precision and accuracy,” added Shafer.
About his experience on the project thus far, Eduardo Barraza said, “I feel very fortunate to work with four great individuals, each one bringing different skills and experiences to the team, who help me improve my skills in computer science and skills beneficial to a team project. This opportunity has been well worth the effort and I’m really thankful to Dr. Donyanavard for bringing me on to this capstone project he is overseeing.”
Anthony Mina emphasized how the capstone is preparing him for a career in industry, “I think the biggest takeaway from this project is to see what it’s really like to work in the field with a team on a massive project. It gives us hands-on experience with technologies that are sought after by so many companies.”
“I personally think that any form of research is crucial as it allows undergraduates to attain a better perspective in contrast to the traditional class teaching method as we are able to use the skills we learned in the workplace,” said Samir Hanan.
The students will continue to work on the project in the spring semester.