OpenCV Spatial AI Competition 3rd prize winner project "Automatic Lawn Mower Navigation" aiming to utilize OAK-D to enable multi-modal obstacle and limit detection using object detection and depth estimation and classification simultaneously. For detailed documentation check out the corresponding page.
Simply clone or download the DepthAI repository and follow the installation tutorial on the Luxonis website. Copy the files contained in this repository into the depthai repository and if you have been able to run the depthai_demo.py OAKMower should run as well. However, so far I have only performed tests on MacOS.