In the context of precision horticulture, decision support tools play a significant role in providing fruit growers with insights into orchard conditions, facilitating informed decisions regarding orchard management practices. This study presents the development of an autonomous yield estimation system designed to provide decision support to small commercial orchards. Autonomous yield estimation is based on the application of UAVs and AI. AI is used to identify and quantify fruitlets and fruits in photographs collected by UAV. In this article, we present our prototype of an autonomous yield estimation system. The adapted “4+1” architecture was applied to design a system with a holistic approach analyzing software, hardware, and ecosystem requirements. Six datasets are presented, which contain the images of fruitlets and fruits of apples, pears, and cherries. Three CNN models were trained: YOLOv8m, YOLOv9m, and YOLOv10m. The experiment showed that the most accurate was YOLOv9m, which achieved mean accuracies of 0.896 mAP@50 and 0.510 mAP@50:95 for all datasets.

Orchard management can benefit greatly from the use of modern technology to reach higher yields, decrease costs and achieve more sustainable farming. Implementation or such a smart farming approach into orchard management can be realised via application of unmanned aerial vehicles (UAV) for data collection and artificial intelligence (AI) for yield estimation and forecasting. On top of that, a digital twin of the orchard can be implemented to represent the physical system of the orchard in the digital format allowing implement modern data-driven decision-making based on fruit-growing automation.The aim of this study is to present a digital twin based on application of UAV and AI for orchard management that is being developed as part of a research project lzp-2021/1-0134. At this moment, we are developing a user-centred design which is oriented to satisfy horticulture specialists’ needs for an autonomous monitoring system and to help them in decision-making. Within the framework of this study an enterprise model of orchard management is designed, which supports the digital twin concept and provides autonomous orchard monitoring. The study is scoped with subjects: apples, pears and cherries, and yield management based on orchard monitoring using UAV.

This article provides a systematic review of innovations in smart fruit-growing. The research aims to highlight the technological gap and define the optimal studies in the near future moving toward smart fruit-growing based on a systematic review of literature for the period 2021–2022. The research object is the technological gap until the smart fruit-growing. The research question of the systematic review was related to understanding the current application of vehicles, IoT, satellites, artificial intelligence, and digital twins, as well as active studies in these directions. The authors used the PRISMA 2020 approach to select and synthesise the relevant literature. The Scopus database was applied as an information source for the systematic review, completed from 10 May to 14 August 2022. Forty-three scientific articles were included in the study. As a result, the technology gap analysis was completed to highlight the current studies and the research trends in the near future moving toward smart fruit-growing. The proposed material will be useful background information for leaders and researchers working in smart agriculture and horticulture to make their strategic decisions considering future challenges and to optimise orchard management or study directions. Considering the current challenges, authors advise paying attention to decision-making, expert, and recommendation systems through the digital twin paradigm. This study will help the scientific community plan future studies optimising research to accelerate the transfer to new smart fruit-growing technologies as it is not sufficient to develop an innovation, but it must be done at the appropriate time.

The smart farming solutions are mainly based on the application of convolutional neural networks for object detection tasks. The number of open datasets is restricted in the agricultural domain. Therefore, it is required to find the answer to the question: how big a dataset must be collected to train a convolutional neural network for object detection tasks? To solve this task, the YOLOv8 framework was selected for the experiment. Three datasets were prepared: MinneApples, PFruitlets640 and mosaic dataset using both previously named datasets. 100 images were selected for testing. Other images were used to create training datasets, which had the size from 100 until 1000 images with step 100 images. Training was repeated 10 times with each size of dataset. The experiment showed that the increase of dataset from 100 to 500 images provides an accuracy growth up to 15.48% mAP@0.5, but from 600 to 1000 images - only 2.98% mAP@0.5. This study experimentally proves that the dataset size equal to 500 images is the most efficient. Meanwhile, the experiment with the mosaic dataset shows constant accuracy improvement. Therefore, it is more advisable to collect different classes with 500 images than one large dataset. This study will be interesting not only for smart farming experts as well as for all machine learning experts.

Translating knowledge into formal representation for the purpose of building an expert system is a daunting task for domain experts and requires information technology (IT) competence and software developer support. The availability of open and robust expert system shells is a way to solve this task. A new architecture of a rule-based expert system combining the digital twin paradigm and a capability-driven approach is presented in this study. The aim of the architecture is to provide a user-friendly framework for domain experts to build upon without the need to delve into technical aspects. To support this architecture, an open-source no-coding shell RecGen has been developed (Python and Django framework). RecGen was validated on a use case of an expert system for providing recommendations to reduce plate waste in schools. In addition, the article presents experiments with large language models (LLMs) by implementing a question-answering functionality in an attempt to improve the user experience while working with large expert system knowledge bases. A mean classification accuracy of 74.1% was achieved experimentally using the injection method with language prefixes. The ablation test was applied in order to investigate the effect of augmentation, injection, a linear layer size, and lowercase text on LLM accuracy. However, the analysis of the results showed that clustering algorithms would be a more suitable solution for future improvements of the expert system shell RecGen.

Food waste indicates ineffective and irresponsible consumption of resources, particularly during the food consumption stage. The aim of our research study is to optimize the catering management process at Latvian schools by reducing the amount of plate waste. The experts developed a set of recommendations aimed at improving the catering management process at schools. The recommendations developed were supported by measurable parameters, which must be monitored by school staff. The capability-driven development approach was applied to model the recommendation system. A plate waste predictive module and a large language model classifier were integrated into the system to support sustainable decision-making. The large language model classifier was trained to filter questions and recommendations. Three training methods were compared: training from scratch and finetuning by using datasets DBPedia and News Category Dataset. As a result, we present the list of recommendations based on the literature review, and the prototype of the knowledge-based recommendation system was developed to audit the school catering management process and promote sustainable school management and decision-making. The recommendation system aims to reduce plate waste due to deficiencies in the implementation of the catering process and to promote responsible food consumption at schools.