Developed reinforcement learning algorithms that dynamically adjust a robot’s level of autonomy based on a human collaborator’s cognitive load and situation awareness. Evaluated in a human-subjects study (n=119) against a heuristic baseline.
Fine-tuned, tested, and evaluated object detection and segmentation algorithms to detect spacecraft from thermal imagery. Deployed algorithms on a satellite that was launched in March 2025.
Coded on- and off-policy reinforcement learning algorithms from scratch for robot navigation by following a wall using LiDAR data. Implemented in ROS and Python, using Gazebo for simulation.
Developed code to compute inverse kinematics for a 6-DoF robot arm with both analytical and iterative methods. Implemented trajectory planning and control of the arm.