Offroad-autonomy stack

Collaborating on a demonstration learning framework for robotic manipulation using implicit object representation and optimization. Evaluating performance with the addition of various forms of noise and occlusions in simulation. Also experimenting with alternative segmentation models and 3D reconstruction networks in the pipeline to improve accuracy.

Tools used: Python, ROS, PyBullet, Detectron2

Reminders App

Tools used: Objective-C, Swift, Xcode

Path planning, trajectory optimization for Indy Autonomous challenge and Roborace Season Beta.

Tools used: C++, ROS