![](https://cdn.mos.cms.futurecdn.net/sySV848uuCeXuCRfCGDyCj.jpg)
Using a Raspberry Pi to make a robotic dog is nothing new but putting it to work is practically unheard of. The team behind this project created a robotic dog, known as FREISA, that uses AI to determine when a plant needs watering and helps the thirsty plant by activating its onboard sprinkler system. This project was created by a team that goes by B-AROL-O with members Davide Macario, Andrea Podo, Gianluca Teti, Gianpaolo Macario, Orso Eric, and Pietro d’Agostino.
The robotic dog’s name, FREISA, is an acronym for "Four-legged Robot Ensuring Intelligent Sprinkler Automation” due to its design and function. It works by locomoting using its four limbs. A camera module is used to observe the surrounding environment. Plants are evaluated using AI and watered by FREISA if it looks too dry.
![](https://cdn.mos.cms.futurecdn.net/rmcx5jRVXMQTDabWiA92Nk.jpg)
![](https://cdn.mos.cms.futurecdn.net/HXhm9WN9EmMa6a569uJmEm.jpg)
![](https://cdn.mos.cms.futurecdn.net/mq26Qf7kbm3GJ3ZspeikBn.jpg)
![](https://cdn.mos.cms.futurecdn.net/YnvPNPMp3Qh6NYsp2JhMnn.jpg)
The team chose to start with a quadrupedal robot called the Mini Pupper 2. It’s been upgraded with a camera that can be used to capture images in real-time of plants as it treks through a garden. A Raspberry Pi is used to handle navigation functions for FREISA while the OAK-D Lite camera module handles the AI processing.
In addition to the Mini Pupper 2, some extra hardware is needed. Of course, the team is using a Raspberry Pi—in this case, they opted to use the Pi 4 B. The OAK-D Lite camera module is used for live video capturing as well as neural network inferencing. Some external peripherals are useful to have on hand, as well, like a laptop for programming hardware.
Instead of the popular Raspberry Pi OS, the team here is using Ubuntu on the Raspberry Pi. It’s working alongside Docker and OpenCV to help monitor the plants and determine when they need watering. The system was trained on models that enable to recognize healthy vs poor leaf conditions.
If you want to see this Raspberry Pi project in action, check out the project page at Hackster where you’ll find a detailed breakdown of its construction as well as a demo video.