A robot capable of identifying plants species at different stages of growth by "touching" their leaves with an electrode has been developed by a team of researchers from China’s Sun Yat-sen University.

According to its developers, the robot can accomplish this by measuring properties including surface texture and water content — properties that could not be previously determined using visual approaches.

The team explained that when the robot’s electrode makes contact with a leaf, it learns about the plant by measuring many different properties — such as the amount of charge stored at a given voltage, the difficulty electrical current encounters when moving through the leaf and the contact force as the robot grips the leaf.

This data is then processed using machine learning, which classifies the plant because different values for each measure correspond with different plant species and different stages of growth.

During trials of the robot, the researchers discovered that it could identify 10 different plant species with a reported average rate of accuracy of roughly 97.7%. Further, the robot identified leaves of the flowering bauhinia plant with an estimated 100% rate of accuracy at various growth stages.

The researchers suggest that the robot could one day be used to monitor the health and growth of crops and make decisions about how much water and fertilizer to give their plants and how to address pest control.

An article detailing the robot, “Iontronic tactile sensory system for plants species and growth stage classification,” appears in the journal Device.