// research
Research
Exploring the intersection of robotics, perception, and human interaction. Current focus on multimodal AI for autonomous characters.
// primary focus
Efficient Multimodal Tokenization for Real-Time Social Perception in Autonomous Robotic Characters
Investigating how autonomous robots can perceive and respond to human social cues in real time. The core challenge is building a tokenization scheme that fuses vision, audio, and proprioceptive data streams efficiently enough to run on edge hardware like the Jetson Orin, while maintaining the latency requirements needed for natural human-robot interaction.
// ballbot build log
Build Progress
Documenting the mechanical design, control systems, and behavior layer development of a self-balancing inverted pendulum robot with character-driven social interaction.
// papers & writeups
Multimodal tokenization for real-time social perception
Edge-optimized transformer architectures for robotic perception
Character behavior modeling in autonomous social robots