Coming soon...

Modern mobile robots need to navigate complex environments with boundless types of obstacles and terrain. Currently, most traversability prediction methods rely on heuristics, human demonstration, or pretraining on a specific set of object classes. At the Robotic Systems Lab, I investigated applications of large pretrained vision language models to traversability prediction.

This project involved implementing existing traversability analysis methods in simulation, coming up with a novel approach to traversability analysis using visual semantics, developing a ROS package to run the method interactively on an Anymal D robot dog, and designing experiments to ablation test the method. We demonstrated long horizon unguided exploration using  zero shot traversability prediction. Additionally, we investigated how traversability knowledge can act as a prior for imitation learning. We presented a poster at the 2023 Cornell Rawlings Presidential Research Symposium, and an initial writeup of our findings can be found here.

Shown is the Anymal Robot navigating a large construction site, a randomly selected traversability mask (red is untraversable, blue is traversable), and the results after projecting the traversability prediction onto a robot-centric elevation map. 

A major problem in mobile robotics is predicting what parts of a given terrain will be traversable for a robot, and planning to avoid untraversable areas. My most recent project with the Greenberg Lab is SuperEdge, a self-supervised learning method for traversability analysis from partial environment observations. As part of my final project for CS 4756: Robot Learning at Cornell, I wrote up some of my findings.

To support this project, I developed a python library containing end-to-end motion planning algorithms that interface with legged robots in Nvidia Isaac Sim. It includes traversability estimation, environment graph generation, graph search, and path tracking. The code can be found here.

I also implemented and benchmarked a variety of supervised and reinforcement learning methods for a quadruped to perform obstacle avoidance tasks in Isaac Sim. These included behavioral cloning, DAgger, q-learning, and actor-critic policy gradient. 

As a research intern at NASA IV&V, I was tasked with applying combinatorial testing methods to HLS Starship flight control software. In collaboration with verification researchers at NIST, systems engineers at NASA, and software engineers at SpaceX, I surveyed the current methods for system-level verification, and proposed a workflow that would integrate combinatorial testing.

In order to improve the efficiency of this workflow and make it viable for use by engineers, I developed a desktop application that would automatically generate unit tests with a specified degree of combinatorial coverage. This required me to design a graph optimization algorithm to find the most efficient ordering of tests in state based systems. Ultimately, the application was able to suggest tests that would improve coverage, and it significantly increased the efficiency of combinatorial coverage analysis.

Atomic Force Microscopy (AFM) allows for measurements of certain properties of materials at nanometer scale. Nanomachining by Regressively Actuated Setpoint (NRASP) is a novel method that uses an AFM to precisely machine any pattern into the surface of the material. I designed and implemented the NRASP algorithm, which uses linear regression and gaussian smoothing to calculate and apply an optimal voltage to the material at different locations. In the process of interfacing with the AFM, I found a bug in its source code that had existed for over 10 years. The below images show how the surface of the material starts with a noisy topography and ends as a dinosaur with nanometer depth and lateral precision.

Publication coming soon!