Tuna Gül

Systems

Two-camera distance to targets Archive

Distance to the reflective FRC targets from two cheap cameras, without a full stereo calibration. Blobs are paired across the two frames, and distance is fitted against 1/disparity from real measurements.

Role
Main author
Period
Jan–Mar 2023
Team
Team NF (FRC 9029)
Stack
Python · OpenCV · NumPy
Impact
A straight-line fit alone gave r ≈ 0.99

Why

Our FRC robot needed the distance to the reflective targets on the field. My teammates and I had two cheap cameras, and instead of a full stereo calibration we measured.

What I did

I wrote the vision code by hand, in Python with OpenCV:

  • Finding targets: HSV thresholding, erode and dilate, then contours, keeping the blobs that fill enough of their rotated bounding box.
  • Pairing: a blob in the left frame is paired with the blob in the right frame whose contour area is most similar, if the smaller area is at least 0.8 of the larger.
  • Distance: the disparity is the distance between the two paired centroids. Distance is a third-degree polynomial in 1/disparity, fitted to distances we measured.

Results

A straight line in 1/disparity already fitted with r = 0.9934, which is what the pinhole model (distance = focal length × baseline / disparity) predicts. The fit was checked with R² on a second data set taken at an angle.

The code is in private team repositories, so there is no link.

Related

Illustration of the method with simulated readings: six Hall sensors on a ring, a distance circle around each, the pairwise intersections nearest the magnet's path, and their average marking the magnet and the shaft angle.

Hall-sensor absolute encoder Archive

An attempt at a cheap absolute encoder for our FRC robot. A ring of Hall sensors around a magnet, and the circle geometry to find the magnet, worked out by hand in C.
Building cross-section with four elevator shafts, the cars at their floors and passengers queued at the landings, fed by a small floor-versus-time trace of the last minute, from a small dispatch simulation.

Elevator dispatch simulator Won

A building and elevator simulator in Rust, written in a hackathon week with one teammate, with cascaded PID control for every car. Our team took 1st place at Boğaziçi AI UpForAChallenge.
Illustration of a streaming JSON decoder: a small nested list drawn as an indented tree on a white card, its deeper part boxed in amber as a lazy generator below the max_depth cut, leaving the card through a port beside the root bracket as a stream of token chips on a winding track, with chevrons showing the direction, that runs off to the right.

DBEX Archive

A JSON-compatible encoder and decoder in Python built on generators, so it can stop at a chosen depth and hand back the rest lazily. On PyPI since 11th grade.