A vegetable-growing trial has pitted expert human gardeners against an AI-powered robot – both produced comparable crops, but the robot used about 40 per cent less water.
An automated robot gardener has shown its nurturing side by matching humans in trials of growing vegetables – and it did it with more efficient use of water.
 |
| The AlphaGarden robotic gardener is a cut above humans when it comes to water use Simeon Oluwafunmilore Adebola et al |
Simeon Adebola at the University of California, Berkeley, and his colleagues have developed an automated plot, called AlphaGarden, that contains a mix of plant types and tested whether it can perform as well as a team of six expert human horticulturalists, each with 10 years of gardening experience, on average.
Video Courtesy of NewScientist:
The robot and human-run plots both contained pairs of eight different common edible vegetables, such as kale, chard and radicchio. Where the humans were left to plant, water and trim the vegetation as they saw fit, AlphaGarden used a garden model and robotic pruning and watering controls to work out how best to grow the plants.
The competition was run twice, for 60 days at a time, and at the end of each period, the coverage, or size, of the plants, their diversity, or how many species fared well, and the water usage were compared. “The automated AlphaGarden performs comparably, in terms of coverage and diversity, to professional horticulturalists, but we find that AlphaGarden reduces water consumption by 40 per cent,” Adebola told the IEEE International Conference on Robotics and Automation in London on 30 May, where the work was presented ahead of publication.
Adebola and his colleagues also used data gathered from the robot garden during the trials to inform the computer simulation and then tweak the robot’s methods. For instance, the team used information to adjust an algorithm that describes how growth rates of plants can affect slower or faster-growing ones if they are all planted in one go, which led to the robot putting different species into the plot at different times to maximise growth.
The AI also used moisture sensors in the soil and a camera to work out when plants needed watering. This led to a marked reduction in the amount of water required for a similar growth rate. At the end of the two cycles, AlphaGarden used 37 and 44 per cent less water than the humans, respectively.
AlphaGarden did encounter some problems during the growing cycles and required human intervention for 14 irrigation and 24 pruning malfunctions. In future, the researchers hope to expand the amount of plant varieties that AlphaGarden can handle and see whether the system can be adapted to vertical farming, says Adebola.
Growing different species in the same plot is common in gardens, but less so in farming because it requires more manual labour. A system like AlphaGarden could help make multi-crop farming, which requires less pesticide and doesn’t deplete the soil of minerals so badly, cheaper, says Adebola.