Research Domain
Physical AI & Embedded Systems
Connecting intelligence to the physical world: custom electronics, sensor integration, embedded computers, and edge devices that let AI observe and act on real environments.
Research Question
What does it take for AI to perceive, measure, and respond to the physical world — not just text?
D-CSIL is run by an engineer who prototypes hardware as readily as software. Devices built so far include an RV leveling system and a machine-health monitor that tracks the lab computer's vibration and temperatures — AI-adjacent instrumentation designed and built in-house.
The lab's embedded fleet is already real: Bailey (a Raspberry Pi 5) serves the lab's memory, Pi nodes host the MQTT agent hive, SensorTile firmware and BLE telemetry dashboards handle sensing, and Piper runs neural TTS on edge hardware. The direction is physical AI: models that consume live sensor data and act on it.