Skip to content
Monsson 2026 · CatRanger

TRACK EVERYMOVE

Detect a cat, hold its identity, and say exactly how far it is — in meters, with an honest confidence interval — from one ordinary camera. Then see where it will be, so you frame the shot before it lands.

±1 cm ground truthReal-timeMonocularBoT-SORT · ReID
detector approaches
2

detector approaches

every estimate
± CI

every estimate

FOV undistort
120°

FOV undistort

+0.5s → 1.74 mCAT #7 · 1.84 m±0.04 mclosing 0.21 m/sLIVERT-DETR · BoT-SORT28 FPS
Example readout
sample · track #7 · Mochi
1.84m
± 0.0495% CI
1.86 m
HC-SR04 truth
now
+0.5s
1.2 m2.4 m
where it will be: 1.74 m
now+1.0 s

Core capabilities

Three things, done honestly.

No magic numbers. Every distance carries its uncertainty, every prediction shows its lead time, and every robot move has a hard safety stop.

How far

Distance ± CI

Pinhole geometry and metric depth are fused into a confidence-weighted median, so you get meters with a real uncertainty band — not a single hopeful number.

Prediction

Where it will be

Per-track motion is extrapolated over an adjustable lead time. A ghost box shows the landing spot and its predicted distance, so the camera is ready before the cat is.

Control

Robot follow

An anti-oscillation controller drives the chassis to hold a safe distance and keep the cat centered — smooth, with a hard ultrasonic stop in firmware.

How it works

One frame, four stages.

Pretrained perception plus classical camera geometry — no training required to run. The same pipeline drives the overlay, the JSON, and the robot.

    01

    Detect

    YOLO11 or RT-DETR finds the cat (COCO class 15) on the undistorted frame.

    02

    Track

    BoT-SORT + ReID keeps a stable identity through full occlusion.

    03

    Distance

    Geometry ⊕ metric depth → meters ± a conformal confidence interval.

    04

    Predict & follow

    Extrapolate the next position, then drive the robot to hold its distance.

Hardware

Fully wireless. One laptop brain.

The laptop talks to the camera over Wi-Fi and the robot over Bluetooth — no gateway. Connect it all from the admin view in a few clicks.

Tapo C211

Eyes

Wi-Fi · RTSP

Live frames over the network, with optional pan/tilt.

Arduino Mega

Motion

Bluetooth · Serial

Drives the wheels and the camera-pan servo from the laptop brain.

HC-SR04

Ground truth

±1 cm ultrasonic

Confirms the model live: “1.84 m” estimated, “1.86 m” measured.

Point a camera. Know the distance.

Run the console locally, connect your camera and robot, and start a session. The pretrained baseline runs out of the box — no training, no calibration to get a first reading.