AI Comes To Bee Rescue

New-generation technology is transforming agricultural practices worldwide, meeting the challenge of productivity, labor shortages, and sustainability.

Satellite technology and the internet of things applications are developing rapidly and artificial intelligence is being increasingly harnessed in the cause of feeding the world’s population as the environment is challenged by climate change.

As a major agricultural producer and exporter, Australia has a significant stake in future farming. Right now, there are more than 20 AI startups in the agricultural space, many of them using applications developed in universities and then spun off commercially.

One company is SwarmFarm Robotics, which has developed autonomous robots for agricultural applications. Their robots automatically spray weeds, prune selectively and learn to co-ordinate themselves to know which areas have been sprayed, if they need refilling or charging, and how to navigate a farm landscape physically. 

Another startup, Hillridge, is more in the financial services area, offering an AI-based crop insurance platform for farmers. Using weather data and machine learning applied to a particular farmer’s property, it provides insurance and self-executing contracts based on blockchain. Ultimately the goal is to help farmers mitigate the financial impact of poor weather on crops or livestock. 

Habitats under pressure

Another AI application is still in the research phase but may have the most powerful impact on agriculture because it addresses a critical issue of biodiversity. 

The plight of bee species and the implications for ecosystems worldwide has been well documented. One encouraging response to the problem has come from a group of Australian researchers harnessing the power of AI.

Pollination by bees is a critical link in many ecosystems. Their contribution to the pollination of plants plays a role in biodiversity and food security by maintaining crop quality. Human activity, however, has destroyed many bee habitats and put them at risk, along with the role they play for plant species.

“Knowing the extent to which a crop has been pollinated allows growers to alter hive locations and numbers to boost pollination levels”

Enter AI with a solution. At Monash University in Melbourne, researchers have developed a new monitoring system that uses AI to understand the movement of bees and help improve pollination and crop yield. 

The research has recently been published in the International Journal of Computer Vision and details how the team recorded pollinators like honey bees, hoverflies, moths, butterflies and wasps to build a database of over 2000 insect tracks at a commercial strawberry farm in Victoria.

The recordings were then analyzed using AI to track individual movements of individual insects, count them and monitor their flower visits. This enabled farmers and researchers to understand the contribution of different species to pollination.

Same day data

Traditional methods of insect monitoring are time-consuming, labor-intensive and liable to inaccuracy. The AI monitoring system can generate same-day data on crop pollination levels and provide farmers with the evidence they need to make decisions.

The researchers developed customized software from the monitoring system to analyze the enormous volume of data and reliably track how individual insects fly through complex and often dense foliage.

“Knowing the extent to which a crop has been pollinated allows growers to alter hive locations and numbers to boost pollination levels,” said Associate Professor Alan Dorin, lab director of the NativeBee+Tech Facility at Monash.

“Farmers might also open or close greenhouse sidewalls to encourage or discourage insect visits from particular directions. They may decide to add flowers to entice insects to explore crop regions that have not been pollinated adequately.”

The software developed for the system combines AI-based object-detection capabilities with separate foreground detection algorithms to identify the precise position of insects and flowers.

Lead researcher Dr. Malika Ratnayake said the software also includes features to make data processing more efficient and save on computer power.

“We have opted to keep this software open source, so it is accessible to anyone who wants to build similar monitoring systems or other applications to optimize and analyze different data points captured through videos,” Ratnayake said.

For the future, the plan is to use the monitoring system to study the long-term impact and results of precise pollination techniques and how they change the quality of food production and yield over several crop cycles.

From the initial project at the strawberry farm, a blackberry farm is next, and collaborations are underway with institutions in Germany and Italy.

From small things, big things grow. AI might yet have a role in protecting bees and pollination and contributing to more efficient farming and a more sustainable planet. 

Lachlan Colquhoun is the Australia and New Zealand correspondent for CDOTrends and the NextGenConnectivity editor. He remains fascinated with how businesses reinvent themselves through digital technology to solve existing issues and change their entire business models. You can reach him at [email protected].

Image credit: iStockphoto/Firn