© Copyright 2025 Mobirise - All Rights Reserved

HELLO


I'm Varundev and I am a Computer Engineer

I work on software for robots and other
autonomous cyber-physical systems.
Welcome to my internet home!

Formal Education

I've received the following degrees.

Doctor of Philosophy

Computer Engineering
Class of 2023
University of Virginia

Master of Science

Computer Engineering
Class of 2018
University of Virginia

Bachelor of Science

Electrical Engineering
Class of 2014
Visvesvaraya Tech. University

Professional Experience

I've worked in the industry as a software engineer and as a research engineer, In this section, you'll find my professional experience listed chronologically.

HYPR (Hyprlabs, Inc.)

Research Engineer

I worked on a contract to create the company's first prototype of a two-wheeled self-balancing side-walk delivery robot. I used rapid prototyping techniques (CNC, FDM, PCB/A, etc.) to build the prototype and I integrated the controllers, actuators, and sensors to the robot. I also created a software stack that could localize the robot and move it to target positions.

Nimble Robotics

Software Engineer II

I worked on a customized grid-based robotic system that would travel on top of a densely packed storage cluster in a warehouse. My work involved creating a control stack that would allow the robot to accurately start and stop motion profiles. I was able to accomplish this task using a model-based approach by considering the dynamics of the robot along the generated trajectory.

Leidos & U.S. DOT

Graduate Research Engineer

I worked on this contract project from the U.S. DOT administered by Leidos. The CARMA is a software stack the is built using functional nodes of Autoware to satisfy the requirements of the U.S. DOT. My work involved porting CARMA to the F1/10 testbed and demonstrate that the F1/10 can be used as a CARMA testbed called CARMA 1-TENTH.

Project Portfolio

I am a computer engineer by training, but I have a deep passion for all things autonomous. I've worked on several different projects where software based autonomy was central to the mission.

Head-to-Head Autonomous Racing

PhD Research

My dissertation is a modular framework that I created that could handle autonomous overtakes and position defense in a head-to-head racing scenario. I theorized how a multi-agent race can be decomposed into several simultaneous head-to-head races, and I used graphs and finite-state machines to further break-down tasks associated with the racing maneuvers.

Cavalier Racing @ Indy Autonomous Challenge

Technical Lead for Motion Planning and Controls

I was the technical lead for motion planning and controls for the Cavalier team at the 2021 IMS and 2022 LVMS race seasons. During the race at IMS Speedway Oval, my work on an adaptive-lookahead pure-pursuit and EKF based GNSS filtering helped my team achieve speeds over 125mph (the fastest for an American team at that time) safely and reliably.

F1tenth.dev: Autonomous Racing Simulator

A ROS and Gazebo based open-source simulator

I created an autonomous racing simulator designed to mirror the behavior and performance of the F1/10 racecar platform. This simulator can be used to verify the performance of autonomous racing algorithms before testing on the real F1/10 racecar. The simulation platform supports ROS SLAM and navigation packages with tutorials on how to get started available in the GitHub link below.

F1/10 Racecar and Teaching Platform

1/10th the scale, 10 times the fun!

I was part of the core team that developed the F1/10: a platform for researchers to rapidly explore automotive cyber-physical systems by providing a platform for real-world experimentation. F1/10’s biggest value is in taking care of the most tedious aspects of putting together an autonomous vehicle testbed so that the user can focus directly on the research and learning goals.

Autonomous Heavy-Lift Multirotor

Technical Collaborator

I helped build a drone system based on 960-class hexa-rotor frame with an all-up weight capacity of 12 kgs and a 12Ah battery which provides about 20-30 minutes of flight time. Using the PX4 flight stack on a Pixhawk 3 Pro with a ZED-F9P RTK (real-time kinematics) GPS, I was able to build a autonomous UAV navigation system with a 1-cm 3D localization accuracy.

Teaching

I've led instruction in the following college-level classes.

ABCs of Electronics    

Helped with lecture and lab sessions for an undergraduate class on topics including signals, transforms, electrical devices, embedded electronics, etc. Graded homework and exams.

Cyber-physical systems    

Used MATLAB and other tools (MLE, TensorFlow etc.) to model energy, medical and automotive cyber physical systems. Topics include building energy modeling, pacemaker design, etc.

Software for autonomy    

Built, maintained and supported a fleet of F1/10 racecars for undergraduate course on software for vehicular autonomy. Provided lectures, instruction materials, and led lab sessions.

Updated: 2023, Varundev Sukhil