Computer Engineering @ GWU · Class of '27

Andre
Arante

Focused on computer vision, reinforcement learning, and robotic systems — from simulation to real-world implementation.

View Projects GitHub →

About

Who I am

I'm an undergraduate computer engineering student at George Washington University with a deep focus on robotics and computer vision. My work spans the full stack — from training neural networks to wiring servos and electrical components.

Currently, I work in the Intelligent Aerospace Systems Lab (IASL) at GWU, where my research focuses on open-vocabulary obstruction detection for drone package delivery using YOLO-World deployed on a Jetson Xavier. Previously, I have worked on autonomous vehicle simulations such as CARLA/SUMO simulators, integrated with OpenCDA at IASL, and full-stack web development and building autonomous UGV systems as an intern at C-3 Comm Systems LLC. Outside of work, I'm currently building: a 6-DOF robotic arm, RL agents in Minecraft, and custom smart home systems.

ML / AI

Reinforcement Learning (SB3)PyTorch TensorFlowPPO

Robotics

ROS2MoveIt2 Jetson Orin / XavierMuJoCo Robotic ArmUGV

Languages & Tools

PythonC/C++ JavaJavaScript Flask / DjangoReact DockerLinux

Work

Projects

Research · CV

UAS Landing Zone Detection

Fine-tuned YOLO-World for open-vocabulary obstruction detection on drone landing pads. Zero-shot generalization to unseen obstacle classes.

YOLO-WorldFine-tuningJetson XavierVisDrone
RL · Simulation

Minecraft Parkour RL

A custom wrapper around Project Malmo, an open source Minecraft simulation enviornment for training RL agents.

PPOSB3MalmoReinforcement LearningImitation Learning
CV · Classification

Wildfire Early Detection

YOLO-based classification network for early wildfire detection in aerial imagery. Detects smoke and flame signatures across California.

YOLOv8ClassificationPython
Simulation · AV

OpenCDA Merging

Cooperative driving automation on CARLA + SUMO. Prototyping merging behaviors for autonomous vehicle fleets.

CARLASUMOOpenCDA
RL · Simulation

RL Gymnasium Agents

PPO agents trained across CarRacing-v3, LunarLander-v3, and Humanoid-v5. Custom reward shaping, domain randomization, and pose perturbation wrappers to reach competitive eval scores.

PPOGymnasiumStable-Baselines3PyTorch
IoT · Hardware

IoT Smart Home Network

Raspberry Pi smart home with voice activation via Vosk, Zigbee/MQTT for light control, and a local web interface. Fully offline — no cloud dependencies.

Raspberry PiMQTTZigbeeVoskPython
No projects match your search.

Experience

Timeline

Sep 2025
– Present

Undergraduate Research Assistant

IASL · George Washington University, Washington, DC

Deployed YOLO-World on Jetson Xavier for open-vocabulary real-time UAS obstruction detection and tracking, integrating with cloud-based VLMs for complex scene reasoning. Fine-tuned models on the VisDrone dataset for bird's-eye-view object detection, and built cooperative-driving test environments in CARLA + SUMO with OpenCDA to model multi-agent lane merging scenarios.

Feb 2024
– Feb 2025

Software Engineering Intern

C-3 Communication Systems, LLC · Bethesda, MD

Led end-to-end development of an autonomous unmanned ground vehicle (UGV) from concept through deployment. Re-architected the system to offload perception and planning to a Jetson Orin via a custom Python API, enabling YOLO, OpenCV, and ORB-SLAM at scale. Built a Flask-based command center and a full-stack OMS used by 8 employees to automate payroll and project analytics.

Jun – Aug 2023

Data Science Intern

Kaiser Permanente · San Jose, CA

Refactored a SAS pipeline to automate processing of millions of patient records. Developed Tableau dashboards with daily automatic updates to support hospital staffing decisions, and streamlined tracking of 80,000+ annual patient visits across five hospitals.

Let's talk

I'm always open to research collaborations, internship opportunities, and conversations about robotics, CV, or RL.