top of page

Focus on First Open Call winners: AI-at-the-Edge for advanced functionality of a beehive monitoring system and optimization of its maintenance


AI at the Edge for advanced functionality of a beehive monitoring system and optimization of its maintenance (BeeEdgeAI) experiment is aimed toward implementing AI-at-the-Edge within our product – beehive monitoring system (hereinafter the system). The current version of the system measures beehive mass, inhive and ambient temperature, relative humidity, and pressure, and detects rainfall. It sends measured data to the cloud via NarrowBand Internet of Things protocol every 10 minutes, where the data are stored and available for further analysis. This functionality is far from being optimized when it comes to energy consumption as well as it does not offer any unique value proposition. The main objective is to apply AI-at-the-Edge at the system to evaluate the extent of the foraging activities of honeybees and then send the relevant information only when necessary. By reaching the described main technical objective we will be able to optimize the system by means of increasing its battery life by 30% (the most energy consuming components NB-IoT module) and reducing the data volume being sent from the system to server by 30% (less amount of data will be sent less often). Both these objectives are relevant from the aspect of product maintenance, since they reduce the needed number of battery replacements during the product lifetime as well as the costs associated with topping up the SIM card once data volume is depleted. The second technical objective is to develop an email-based notification system, which will benefit both, beekeepers (users of the system) and us (the manufacturer). The former will get information on what is going on in beehive based on the information about the assessed foraging activity provided by AI-at-the-Edge, while the latter the information on battery life and data volume available on SIM cards, which will allow us to implement a predictive maintenance strategy. With our experiment we are targeting manufacturing (predictive maintenance, optimization of the energy consumption of the product) and agriculture (precision beekeeping) sectors. The key technologies to be used will be NB-IoT, distributed machine learning (related to AI-at-the-Edge), RESTful API, serverless computing, and email communication service.



Technological impact. With this experiment we aim to independently develop AI-at-the-Edge solution for the first time based on our knowledge of embedded software engineering, expertise in beekeeping and knowledge on AI obtained through knowledge transfer arising from collaboration with AI experts in the past.

Economic impact. Implementation of AI-at-the-Edge will reduce the number of times data is being sent to the server as well as the actual amount of data being sent, which will reduce total data volume by 30%. This will impact the costs associated with topping up NB-IoT SIM cards installed in the system. In addition, optimization of the performance of the system will also reduce the costs associated with the battery replacement due to being able to increase the battery life by 30%. The last impact is related to the cost associated with the cloud. Since we are relying on a serverless solution and a consumption plan, the costs occur only after an application is triggered. Since the amount of application triggers will be reduced, the costs associated with cloud provider will reduce as well.

Commercial impact. The proposed experiment will make our entire product more relevant by possessing a unique value proposition for the end-users.


Senso4s d.o.o.



  • link-icon-svg-2
bottom of page