Google has launched an autonomous robot which uses artificial intelligence to monitor crops.
Powered by solar panels, the buggy travels through crops using cameras and sensors to predict agricultural yields and help farmers decide how much pesticide and fertiliser to use. It has been launched by X, Google's "moonshot factory" which develops experimental technologies.
Tested on soybean fields in Illinois and strawberry farms in California, the buggy is one of the first products developed by Mineral, the team focused on agriculture, which the secretive X unit first began discussing last year.
Earlier this year the team met US Department of Agriculture officials in Washington to discuss their work.
Its larger aim is to identify new plant species that could be developed for human consumption and might be resistant to climate change.
"Intensively growing just a few varieties of plants makes our food supply vulnerable to pests, disease, and a changing climate.
"Over time, it also depletes the soil of nutrients and minerals, reduces the diversity of the soil’s microbiome, and diminishes the soil’s ability to store carbon," the team said on its website.
The buggy is designed to help farmers understand the needs of individual plants so that growing conditions can be more closely tailored, and also predict their yields more accurately by counting individual plants and buds.
This is supposed to reduce the use of unnecessary water and chemicals, making farming more efficient, as well as limiting the need for extra labour.
"By combining data collected from the field, like plant height, leaf area and fruit size, with environmental factors like soil health and the weather, Mineral’s software tools can help breeders understand and predict how different varieties of plants respond to their environments," the group added.