Robot uses tactile sensitivity to identify plants

A team of researchers has developed technology for plant identification: a robot equipped with a "tactile sensory system" capable of distinguishing plant species and monitoring their growth stages simply by touching the leaves.

The robot has the ability to measure features such as texture and water content, factors that visual methods such as image analysis cannot capture with the same accuracy.

How does the robot "feel" the plants?

Inspired by the structure of human tactile receptors, the device uses an "iontronic" interface that measures properties such as the foil's capacitance and resistance to touch.

By applying a small electrical voltage, the system can determine how much charge a leaf can store and how much resistance it offers to the flow of current. This data, combined with machine learning algorithms, allows the robot to accurately identify both the plant species and its stage of development.

In tests, the device achieved an average of 97,7% accuracy in identifying ten plant species and achieved 100% accuracy for bauínia leaves at different stages of growth.

Potential applications in agriculture

With the advancement of this technology, the robot can become an essential tool for precision agriculture.

According to researcher Zhongqian Song from Shandong First Medical University, this device could in the future be used by farmers and researchers to monitor crop health, adjust irrigation and fertilizer levels, and even carry out more efficient pest control.

Song believes that innovation can not only improve agricultural productivity, but also facilitate early detection of diseases, which is essential for food security.

Overcoming limitations

Traditional plant monitoring methods, such as cameras and visual sensors, face limitations imposed by weather conditions, light variations and interference in the background of images.

In contrast, the developed touch system utilizes foil as a natural conductive material, creating an ion exchange interface that is less susceptible to these external variables.

This approach could open doors to a new way of studying plant physiology under field conditions, where plants are exposed to different types of environmental stress.

Limitations and next steps

Despite its initial success, the robot still has limitations. It has difficulty identifying plants with very rigid leaves or complex structures, such as thorns or needles.

According to Song, these limitations could be addressed in the future with improvements to the device's electrode design. In addition, the researchers plan to expand the plant species database, allowing the robot to recognize a larger number of plants with the same accuracy.

Another goal is to integrate the device with sensors that can display results in real time, even without the need for an external power source.

More information can be found at doi.org/10.1016/j.device.2024.100615