Developers of weeding robots are shifting away from machines which apply herbicides to mechanical and laser-based tools — and targeted improvements can see them break through to the farming mainstream.
In an assessment of progress in weeding robots over the last decade by Chris Lytridis and Theodore Pachidis from the HUMAIN-Lab in Kavala, Greece, they said that rising herbicide costs, environmental and health risks and plant resistance are driving the change in emphasis, which is aided by advancements in artificial intelligence.
They pointed to “significant progress” in the period, underlined by commercial robotic weeders now being available from companies such as Naïo Technologies and Farming Revolution. There are also a number of promising options currently going through field trials.
Computer vision, mechanical arms
The computer vision employed by weeding robots has improved, they said, due to more effective machine learning algorithms, increasing availability of data to train the models and better computational capability that can be integrated onboard the robots themselves. All of this adds up to better mechanical features to tackle weeds without the use of synthetic chemicals.
“Weeding tools themselves are evolving in accordance with the precision weed treatment capabilities provided by machine vision. This is especially the case with weeding tools attached, as end effectors, to robotic arms with enough degrees of freedom to guide the tool to precise locations,” they wrote in the journal AgriEngineering.
Autonomy is the aim
There is still considerable room to make further improvements which boost the effectiveness of the technologies for farmers, however. Lytridis and Pachidis highlighted three areas of focus: improved targeting of weeds through honing models and better taking into account environmental conditions in fields; standardisation of weeding tools; and a move towards full autonomy, both in terms of navigation and energy supply.
“These issues can be addressed by improved localisation and navigation algorithms, more accurate and effective detection methods and removal tools, and new, more efficient battery technologies and autonomous recharging techniques,” they added.
