Protecting bee populations with AI: recent inventions in the apiculture sector
Climate change is increasingly affecting the natural world; weather conditions are becoming more extreme, biodiversity has decreased, and the natural cycles of the planet are being disrupted. The agriculture industry is directly affected by these changing conditions, from issues with soil degradation and unpredictable crop yields to an increase in pest and diseases. In the face of these challenges, agritech is rapidly evolving. New technologies and methods of farming are being developed which focus on sustainability, with the aim of improving productivity in the current climate, without causing further damage for future generations.
A particular area of technological advancement is in livestock management, in particular animal husbandry. Animal husbandry is the practice of breeding and caring for livestock, focusing on health, welfare and productivity. Animal husbandry not only relates to traditional farm animals such as cows, sheep and pigs, but also, perhaps unexpectedly, to bees.
Bees are essential to agriculture, with more than a third of the world’s food crops relying on pollinators to reproduce. Wasps are also important, as they help to naturally control pest populations by feeding on insects which eat the crops. Wasps and bees are currently facing a number of challenges, from colony collapse disorder to the increase in populations of predatory hornets, all of which are depleting their numbers and posing a significant threat to the future of these vital pollinators.
This complex set of challenges and the complex nature of the hives themselves makes their management particularly difficult, but new developments in technology are looking towards mitigating the damage of these challenges to bee populations, from more effective monitoring of hives to actively deterring predators. In particular, AI is proving well suited to analysis of the noisy, multivariate, and sometimes indirect or incomplete data obtained from monitoring hives.
Recent inventions
As outlined in the European Patent Office’s (EPO’s) Insight Report on digital agriculture, a number of patents have published in the past five years which use artificial intelligence to assist with the management and protection of bee populations.
Some of these inventions relate to sensing and monitoring systems, which allow farmers to better understand and maintain the health of bee colonies. Patent application WO2021202065 was published in 2021 and relates to a system for beehive health and activity monitoring. The system incorporates a sensor bar into a honeybee frame that slides into a chamber of a beehive. The sensor bar includes a microphone that records sound within the hive and environmental sensors which measure conditions such as temperature and humidity. The sound recorded can be analysed using machine learning models to determine hive activity and other health issues such as the loss of the queen, which is then correlated to the health of the colony.
Similarly, patent application WO2022120496, published in 2022, relates to a system for monitoring and recording beehive status. The system combines sensor data from inside the beehive with human observation data of the beehive to identify health issues using machine learning. The resulting data set is used to train machine learning algorithms so that beehive health states can be identified using sensor data alone, allowing remote monitoring of hives.
Inventions using artificial intelligence to monitor pollination are also emerging. Monitoring pollination is beneficial not only for helping to maximise yield and quality of crops, but also for early identification of health issues within the hive. Patent application EP4417045 was published in 2024 and relates to an apparatus for identifying the presence and quantity of pollen on bees using a deep learning model and video footage of the bees. The model is trained on the characteristics of the morphology and colour of the bees and pollen. Patent application EP4276700, published in 2023, also relates to tracking and monitoring of pollination efficiency for crops. The invention uses machine learning to combine in-colony features (such as colony foraging activity, colony size, and brood status) with out-colony features (such as weather conditions, ambient temperature, crop variety, and crop yield) with ground truth labels which indicate pollination effectiveness.
The future
It is hoped that new technologies will continue to emerge in this field which allow farmers to maintain the health of their bee colonies and to expand bee populations, to reverse some of the damage being done by climate change. As the human population continues to grow and the climate becomes increasingly unstable, food security is a growing concern. Bees, although small in form, have a colossal impact on ecosystems. Their existence is essential, and there is an ever-growing need to find new ways to protect them and ensure their longevity for future generations.
The discussion here illustrates the growing trend amongst existing and new clients in using AI technologies in surprising and new fields. AI is opening the door to new technical functionality and commercial opportunities for clients. If you or your company would like IP advice relating to agritech inventions, or if you have any questions on this topic, please do not hesitate to get in touch with us.
Useful links
- EPO Insight Report, digital agriculture, September 2025 (PDF): dycip.com/epo-digital-agriculture
- WO2021202065A: dycip.com/beehive-monitoring
- WO2022120496A1: dycip.com/epo-id-record-beehive
- EP4417045A1: dycip.com/epo-pollen-learning
- EP4276700A1: dycip.com/epo-track-pollen
