Projects
Planning Robot Actions That Reveal Human Objectives
Julia
Python
Optimization
ROS
Gazebo
LQR Control
Game Theory
How can robots understand human behavior without directly asking? We show that robots can act in ways that provoke humans to reveal their hidden objectives.
Zero to Autonomy in Real-Time
Python
ROS (Noetic)
PyTorch
Unity
Neural Networks
Model Predictive Control
Dynamics Modeling
How can robots adapt to terrains that evolve over time? We develop a way for robots to update their dynamics models on-the-fly as their terrain changes during autonomous missions.
Off-Road Terrain Adaptation for Robot Navigation
Python
C++
ROS (Noetic)
PyTorch
Unity
Neural Networks
Model Predictive Control
Dynamics Modeling
How can we estimate a robot's dynamics on unknown terrains at runtime? We developed a learned model that helps mobile robots quickly adapt to new terrain conditions, improving navigation accuracy and collision avoidance.
Vision-Based Autonomous Driving via Machine Learning
Python
PyTorch
Raspberry Pi
Supervised Learning
Convolutional Neural Networks
Electronics Integration
Our autonomous RC car navigates an obstacle course using a neural network that converts camera images into control actions.