Robot Perception

This is a solid reference for anyone building vision systems on robots. It covers the practical stuff you actually need: camera calibration workflows with OpenCV, sensor comparisons (RealSense, ZED, OAK-D), and how to handle multi-sensor fusion. The code samples focus on intrinsic calibration with real warnings about coverage and reprojection error, which is helpful since bad calibration kills everything downstream. You'll want this if you're setting up perception pipelines, debugging why your depth maps look wrong, or figuring out which sensor to buy for manipulation versus navigation. It assumes you know Python and basic computer vision but walks through the robotics-specific parts like coordinate transforms and sensor sync.

Install to Claude Code

npx -y skills add arpitg1304/robotics-agent-skills --skill robot-perception --agent claude-code