top of page

Robotics

UMass Senior Design Project

Sensing and Navigation Lead

Silicon Synapse Lab

Deep Reinforcement Learning Multimodal Robot Controller

My current research at Northeastern University. I am working on using Deep Reinforcement Learning (DRL), a reward based learning scheme where good actions yield high reward, to develop a controller for a quadruped that can morph into a drone!

​

I am using IsaacLab to train using massive parallelism (4096 robots) in a game-like curriculum. The curriculum makes the terrain harder as the robots progress, so it's like a "walk before you run" training scheme. I use advanced techniques like Asymmetric Actor-Critic (A2C) with privileged observations to train the robots on more complex terrain. This allows the robot to train with more accurate value estimates and use the buffers to traverse the terrain.

UMass Senior Design Project

Sensing and Navigation Lead

UMass-Amherst Senior Capstone Project

Gathering Live Imagery, Mapping, and Pose Estimation (G.L.I.M.P.S.E.)

My senior capstone project at UMass. Working in an interdisciplinary team of computer and electrical engineers, we designed a fully autonomous wheeled robot, named Holly, capable of performing smooth dynamic obstacle navigation through partially crowded environments. The robot has a camera to track and follow scene landmarks called ArUco markers!

 

The full technical documentation for this project can be found in my GitHub repository linked on my CV

Sample ArUco Markers

IMG_1276_edited.jpg

UMass Senior Design Project

Sensing and Navigation Lead

I'm a paragraph. Click here to add your own text and edit me. It's easy.

UMass-Amherst Mechatronics and Robotics Research Lab

LiDAR Inertial Odometry and Navigation

In this research, I aimed to develop a semi-autonomous navigation algorithm using LiDAR and IMU for the Stretch RE1 medical assistant robot. I explored techniques in navigation and sequential decision making.

 

This project was in tandem with the Elaine-Marieb College of Nursing, where we worked with the nursing students to get feedback on how the robot should be used, and designed those applications. 

bottom of page