Mi3 Lab
Machine Intelligence, Interaction, and Imagination
"How do we ascribe meaning to objects and actions in the open world to make safe, goal-aligned decisions?"
About Our Lab
The Mi3 Lab at UC Merced researches machine intelligence, interaction, and imagination to develop safer autonomous vehicles, robots, and human-interactive intelligent systems. Our primary focus is on safe autonomous driving and intelligent transportation systems, where we work to ensure that self-driving vehicles can safely interact with all road users—including first responders, pedestrians, cyclists, and vulnerable populations.
We address critical challenges such as how autonomous vehicles should respond to emergency vehicles, how to make mobility accessible for all populations, and how intelligent systems can understand both verbal and non-verbal human communication. Our research spans artificial intelligence, computer vision, multimodal machine learning, and human-robot interaction, with a strong emphasis on safety validation, explainability, and real-world deployment.
Research Areas
Safe Autonomous Driving & Intelligent Transportation
Perception, prediction, and decision-making systems for autonomous vehicles that safely interact with other road users, including first responders and vulnerable populations.
Robotic Perception & Planning
Multimodal machine learning that enables robots to perceive, plan, and act in unstructured real-world settings.
Human-Machine Communication & Co-Creativity
Interfaces and models that understand verbal and non-verbal communication, enabling natural collaboration between humans and intelligent systems.
Vision-Language Representations
Models that bridge visual perception and natural language understanding for more interpretable AI systems.
Safety, Explainability & Trust
AI systems that are safe, reliable, and can explain their decisions in ways humans can understand and trust.
Adaptability & Generalization
Systems that adapt to new environments and generalize learned knowledge to novel situations.
Recognition & Funding
NVIDIA Academic Research Grant (2026)
Edge-Deployed Multimodal Safety Reasoning for Autonomous Vehicles
ASCENDr Seed Grant (2026 · $200,000)
Naturalistic Multimodal Human Instruction for Resilient Human–Machine Teaming in Autonomous Defense Systems
California DMV Sponsored Research
Autonomous Vehicles Safety Research
SAE / GM Civic Progress Challenge Grant for Accessible Mobility
First Prize — NHTSA Enhanced Safety of Vehicles Safety Design International Competition (2026)
Vision-Language Work Zone Intelligence for Safety-Critical Speed Regulation of Mixed-Autonomy Vehicles in Dynamic Environments