Precision agriculture uses robot gardeners for sustainable farming
In the middle of the Computer Science and AI Lab at MIT lies an indoor garden where robots are busy growing tomatoes.
The idea for a robot-tended garden came from work done in Professor Daniela Rus's Distributed Robotics Lab by postdoctoral assistant Nikolaus Correll. His ideas for agricultural robotics grew into a course in which students created robots capable of tending a small garden.
Each robot has an arm and a watering pump, and the plants are equipped with networked soil sensors. This allows plants to request water or nutrients, while the robots can minister to their charges, locate and pick a specific tomato, and even pollinate the plants.
A tool called LCM (Lightweight Communications Marshaller) allows the different robotic modules to communicate; the version used in the project came from the DARPA Grand Challenge Vehicle. The object recognition software is built on the back of image annotation tool LabelMe, and the robots are re-imagined versions of iRobot's Roomba.
Professor Rus calls the system precision agriculture, and suggests it could make farming more sustainable in two ways. First, monitoring of each plant's physical state means water, nutrients and care can be dispensed as-needed, rather than being broadcast indiscriminately. And second, mechanical harvesting removes the manual work from speciality crops such as fruits and vegetables.
In the long view, the researchers hope to develop a fully autonomous greenhouse, complete with robots, pots, and plants connected via computation, sensing and communication.
Looking past agriculture, Correll suggests similar systems could be applied automating services for older adults in residential care facilities, for example, or tending plants in greenhouses or hydroponic farms.