Abstract
This paper presents the design, development, and testing of an autonomous collision avoidance system for a model multirotor by team Elite Ball Knowledge (EBK) at San Diego State University (SDSU). This project was developed as part of the Baseball Avoidance Multirotor (BAM) challenge from NASA Langley Research Center (LaRC). The objective of this project was to detect, predict, and avoid a baseball in real time while minimizing deviation from the drone's original trajectory. The system is divided into three main components: perception, prediction, and planning. The entire system was implemented entirely in MATLAB/Simulink. The final program implementation successfully avoided the baseball; however, it was not able to return to its original path autonomously on any tested trajectories.
My Role
As the Project Lead, I was responsible for the overall project management, system design, and integration. I led the team in developing the perception, prediction, and planning algorithms. I also coordinated the testing and validation of the system, ensuring that all components worked together seamlessly. Additionally, I contributed to the implementation of the MATLAB/Simulink code and the preparation of the final paper and presentation materials.
Pictures
Project Poster
Here's the project poster that we presented at SDSU Senior Design Day at the end of the semester.
Videos
Here's two videos of the project.
The first video is our class presentation:
The second video shows the drone in action:
GitHub Repository
All sharable code can be found in the GitHub repository linked below: