~/projects/ascension

Ascension Robotics

> Computer vision developer on a combat-robotics student design team — building the perception and autonomy stack that finds, tracks, and targets opposing robots in real time.

// perception → actuation

camera feedlive frames on Linux
→ YOLO detectorarmor-panel detection
→ OpenCV post-processingcoordinates + angular orientation
→ Nav2 / Odin driversmotion planning
→ serial → MCUcustom ROS nodes drive the turret
Role
Computer Vision Developer
Since
Sept 2025
Platform
Jetson Orin Nano · ROS 2
Team
Student design team
Dataset
12,000+ images · 4× classes
OpenCVROS 2Nav2PythonC++Foxglove StudioLinuxSerial/MCU

Armor-panel detection

Real-time C++ detection system on a Jetson Orin Nano that automates aiming by tracking and predicting enemy robot motion — an OpenCV pipeline extracts precise target coordinates and angular orientation, with camera calibration tightened to cut RMS reprojection error by ~80%.

Dataset at scale

Processed over 12,000 images for the detection model, accelerating annotation by distilling labels from the team's older model and expanding the number of detected classes 4×.

ROS 2 motion planning

Implemented motion planning on the robot stack using Manifold-tech's Odin ROS drivers and Nav2, with Foxglove Studio for live introspection and debugging.

ROS → MCU bridge

Wrote custom ROS nodes that stream planner and targeting data over serial to the microcontroller, closing the loop between high-level autonomy and low-level actuation.