Date of Award
12-2025
Culminating Project Type
Starred Paper
Styleguide
ieee
Degree Name
Computer Science: M.S.
Department
Computer Science and Information Technology
College
School of Science and Engineering
First Advisor
Adriano Cavalcanti
Second Advisor
Andrew A. Anda
Third Advisor
Jie H. Meichsner
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Keywords and Subject Headings
JAX, Random Forest, Brain-Computer Interface, Avatar, Robotics, AI, Machine Learning
Abstract
Neural control systems based on electroencephalography are changing the way that devices and brains talk to each other directly by making it possible to turn brainwave patterns into instructions that devices can use. This study investigates the utilization of noninvasive electroencephalographic-based Neural Control Interface (NCI) for the guidance of unmanned aerial vehicles and anthropomorphic robotic platforms. Using a JAX-based Random Forest ensemble classifier, we filtered, extracted features, and classified electroencephalographic data collected with an OpenBCI headset to identify mental command signatures corresponding to six directional commands: backward, forward, left, right, takeoff, and landing. The JAX implementation achieved 94.36% classification accuracy with training completed in under two minutes, demonstrating computational efficiency suitable for real-time BCI applications. The classified outputs were integrated with control modules on a SoftBank NAO6 humanoid robot and a DJI Tello aerial vehicle, enabling thought-driven multi-modal interaction including voice, vision, and motion control. The main contribution of this work is the integration of the JAX framework into the Avatar open-source BCI platform—the first implementation of this advanced computing library for brain-computer interface applications within this platform, establishing a foundation for future GPU-accelerated neural interface systems.
Recommended Citation
Patel, Yash J., "Neural Interface Avatar Systems for Humanoid and Drone Control" (2025). Culminating Projects in Computer Science and Information Technology. 64.
https://repository.stcloudstate.edu/csit_etds/64
Comments/Acknowledgements
I would like to extend my sincere gratitude to the faculty and staff of the Department of Computer Science and Information Technology at St. Cloud State University for their continuous support and encouragement throughout my academic journey. I am especially thankful to Dr. Adriano Cavalcanti for his invaluable mentorship, guidance, and for providing opportunities that have been truly valuable to me and will greatly benefit my future. I would also like to express my appreciation to Dr. Jie Meichsner and Dr. Andrew Anda, my committee members, for their thoughtful feedback and direction, which greatly contributed to the completion of this work.