Nitish Gupta

Autonomous systems, reinforcement learning, and safe control.

Doctoral research at NC State's CIIGAR Lab, five publications, fifteen built-and-shipped projects, and five courses taught — the research foundation underneath Robolabs AI.

EducationPublicationsProjectsTeachingLecturesMentoringHonorsCertificatesSkillsService

Education

  1. 2021 — 2023

    Doctoral Research, Computer Science

    North Carolina State University

    • Research: reinforcement learning & optimal control, real-time autonomous systems.
    • Advisors: Dr. David L. Roberts (current), Dr. Zhishan Guo (previous).
    • Began at the University of Central Florida before transferring to NC State.
    • Shifted focus to founding Robolabs AI in 2024.

  2. 2016 — 2018

    MS, Computer Engineering

    University of Central Florida

    • Research: machine learning, computer vision, deep learning, intelligent transportation systems, vehicular networks, ADAS.
    • Advisor: Prof. Yaser P. Fallah.
    • Thesis: Real-Time SIL Emulation Architecture for Cooperative Automated Vehicles.

  3. 2010 — 2014

    BE, Computer Engineering

    University of Mumbai

    • Ranked 1st of 120 students in the Electronics department.
    • Creative team head, "Pegasus" annual college festival.

Publications

  • Potential-Based Reward Shaping for Intrinsic Motivation

    AAMAS 2024, Auckland, NZ · May 2024 · Best Paper Award · with G. Forbes, L. V. Arias, D. L. Roberts, C. M. Potts, A. Jhala

  • Optimizing Real-Time Performances for Timed-Loop Racing under F1TENTH

    RTSS 2022, Industry Challenge · Dec 2022 · with Kurt Wilson, Zhishan Guo · also on arXiv

  • Real-Time Hardware-In-the-Loop Emulation Framework for DSRC-Based Connected Vehicle Applications

    CAVS 2019 · Sep 2019 · with Ghayoor Shah, Rodolfo Valiente, and 4 others

  • Real-Time SIL Emulation Architecture for Cooperative Automated Vehicles

    Electronic Theses and Dissertations, University of Central Florida · 2018

  • Real-World Modeling of Path Finding Agent Using Robot Operating System (ROS)

    FCRAR, vol. 30 · May 2017 · with S. J. A. Raza, G. R. Sukthankar, N. Chitalaya

Presenting the F1TENTH real-time performance paper at RTSS 2022

Selected Projects

  • 2023

    Red Wolf Racers — F1TENTH Autonomous Racing

    Led NC State's Red Wolf Racers F1TENTH team, building control algorithms including Adaptive Pure Pursuit and model-based approaches — ranked 1st in total laps and 3rd in time-trials at CPS-IoT Week 2023.

  • 2021

    Knights Auto Team (UCFKAT)

    Founded UCF's Knights Auto Team (UCFKAT), a student robotics club for autonomous racing, later featured in UCF news coverage.

  • 2021

    End-to-End Learning in the SVL Simulator ("Bumblebee")

    Trained an end-to-end driving policy in the SVL autonomous-vehicle simulator, mapping raw sensor input directly to steering and throttle commands for closed-loop evaluation.

  • 2020

    Intelligent Industrial Automation ("Panda")

    Built a factory-automation pipeline at NHK International: point-cloud acquisition, ROS integration, RANSAC/ICP-based segmentation and model fitting, and robot motion planning with MoveIt for dynamic industrial environments.

  • 2018

    Connected Automated Vehicle Emulator

    Designed a hardware-in-the-loop Remote Vehicle Emulator that simulates a network of virtual vehicles from a single V2X broadcaster, giving connected-vehicle safety applications a realistic, repeatable, real-time test environment.

  • 2017

    Vehicle Detection and Tracking

    Trained a linear SVM on HOG and color-histogram features to detect vehicles in dashcam footage (98.56% classifier accuracy), then tracked them frame-to-frame with a sliding-window search over normalized, randomized training data.

  • 2017

    Traffic Sign Classification

    Built and fine-tuned a convolutional neural network to classify traffic signs from a mounted camera, attaining 97% test accuracy on the German Traffic Sign dataset.

  • 2017

    Driver Behavior Cloning

    Trained an end-to-end convolutional network to clone human driving behavior, predicting steering angle directly from front-facing camera images in a simulated driving environment.

  • 2017

    Lane-Line Segmentation

    Built a classical computer-vision pipeline for lane detection: camera calibration and distortion correction, color/gradient thresholding, a perspective transform, and lane-boundary fitting.

  • 2017

    Autonomous Bot Using ROS

    Built an autonomous mobile robot on ROS, with the approach and results documented in an associated publication.

  • 2017

    Q-Learning in Grid World

    Implemented tabular Q-learning in a custom grid-world environment, with an interactive GUI for visualizing the learned policy and value function.

  • 2017

    Robot Path Planning with A*

    Implemented the A* search algorithm for optimal path planning in a grid-world environment, comparing paths against alternative heuristics.

  • 2017

    Autonomous Rescue Robot

    Built a 4-wheeled autonomous car for search-and-rescue in disaster-affected areas: ROS nodes gathering odometry and Kinect scans into a Raspberry Pi for 2D occupancy mapping, a particle filter for localization, and path planning to goals over offline-built maps. Led to a publication (FCRAR 2017).

  • 2016

    Concurrent Physics Engine

    Built a physics engine simulating circles colliding under random velocities, then implemented a concurrent, lock-free version of Sweep-and-Prune and hash-grid collision detection in C++.

  • 2014

    Surveillance Based on Tracking and Targeting

    Led a team of three to build a MATLAB-based real-time object detection and tracking system controlling a camera-laser-mounted robotic arm, aimed at continuously tracking and targeting a suspect.

Teaching

  1. Jan 2024 — May 2024

    CSC 584 — Building Game AI

    North Carolina State University

  2. Aug 2023 — Dec 2023

    CSC 454 — Human-Computer Interaction

    North Carolina State University

  3. Aug 2022 — Dec 2022

    CSC 216 — Software Development Fundamentals Lab

    North Carolina State University

  4. Aug 2022 — Dec 2022

    CSC 714 — Real-Time Computer Systems (Lectures)

    North Carolina State University

  5. Jan 2021 — May 2021

    EECS 140 — Introduction to Digital Logic Design Lab

    University of Kansas

Special Lectures

Mentoring

North Carolina State University

  • Jason Wong
  • Cassidy Petrykowski
  • Arvin Kushwaha
  • Anthony Brown
  • Ashok Kumar Selvam
  • Jubitta John
  • Andres Ayala-Lagunas
  • Anupam Mijar
  • Tej Joshi
Mentoring students testing autonomous RC vehicles for the F1TENTH program

Honors & Awards

  • 2022Selected for the Frank M. Hubbard Engineering Scholarship (declined, moved from UCF to NC State).
  • 2021Graduate Dean's Fellowship.
  • 2021Hoglund Fellowship.
  • 2014Dean's Award for Excellence in Academics.

Certificates

  • Machine Learning — Stanford UniversityJan 2017
  • Self-Driving Car Nanodegree — Udacity
  • Robotics Engineer Nanodegree — Udacity
  • Transformer-Based Natural Language Processing Applications — NVIDIA Deep Learning InstituteJun 2022
  • Autonomous Robots: Model Predictive Control — UdemyAug 2021

Technical Skills

Languages

Python, C/C++, MATLAB

Libraries

PyTorch, TensorFlow/Keras, OpenAI Gym, Stable Baselines, OpenCV, Scikit-Learn, Pandas, NumPy, Matplotlib, PCL, VTK, Eigen

Tools

ROS1/ROS2, Gazebo, MoveIt, Git, Docker, Linux, SUMO, NS3, CloudCompare, Qt, VS Code

Hardware

LIDAR, Depth cameras, NVIDIA Jetson, Raspberry Pi, Arduino, FPGA, IMU

Service

Peer reviewer: RTSS, RTAS, DAC, TNNLS, TCRTS, TIST, RTNS, HDIS, ICIST, RTCSA, ICICIP, ICESS, MetroCAD, Field Robotics.

  • Program Committee, 2nd Workshop on Opportunities and Challenges with Autonomous Racing (ICRA 2022)