In recent years, there has been an increasing demand for accurate measurement of the spatial dispersion of air temperature in air conditioning control for indoor environments. This spatial dispersion of temperature is caused by the unevenness of air conditioning equipment, local temperature gradients due to human activities, and local heat exchange with outdoor air, all of which hinder the provision of a comfortable environment. This issue requires resolution in various fields, including the agricultural sector, where it can negatively impact crop growth. To address this issue, it is necessary to measure air temperature at multiple points, although the number of sensors that can be installed is typically limited, and the temperature at a single point is typically used as a representation of the temperature in the space. Additionally, heat maps using infrared array sensors can acquire temperature distribution in a narrow area; however, they cannot directly determine air temperature as they only measure the surface temperature of an object. Therefore, we propose a deep learning-based method that utilizes both heat maps acquired from infrared array sensors and air temperature data obtained from environmental sensors to estimate air temperature. We evaluate its effectiveness by varying the experimental environment, the model structure, and the resolution and number of heat maps added as input. Furthermore, we assess the usefulness of utilizing a Graphics Processing Unit (GPU), which is a cost-effective solution that can be installed on low-resolution sensors and edge devices. The outcomes indicate that the proposed model can attain an estimation accuracy of under 1 mean squared error (MSE) in approximately 200 seconds, while also exhibiting practical feasibility for deployment on edge devices, including residential abodes and agricultural premises.

Short-term hot water consumption forecasting in an individual household is important to realize energy-efficient hot water management while meeting users' comfort. However, consumption traits in an individual household contain much irregularity, and that gives additional challenges to accurate forecasting. This paper proposes an Attention-PVS model to tackle this issue, which focuses on consumption values with similar past consumption trends. In this study, verification experiments were conducted on a real-household hot water consumption dataset collected by Electricity of France. The model was evaluated with its consumption forecasting error with MAE, MSE, and RMSE metrics and computational costs. As for the computational cost, experiments were conducted on NVIDIA Jetson Nano (Jetson) to validate the applicability to embedded systems. The results revealed that the proposed model performs consumption forecasting with comparable accuracy to other state-of-the-art models. Additionally, while LSTM scored lower error than the proposed model, Attention-PVS performed training and inference on Jetson in shorter times than LSTM.

In recent years, there have been many attempts in smart agriculture to increase efficiency and profitability, especially in horticultural agriculture, where profitability is high. One of the measures to achieve this goal is to realize uniform quality by equalizing temperatures in greenhouses which have a huge influence on a process of growth. The most common method for measuring temperatures in greenhouses is the use of temperature sensors. However, to measure continuous temperature distribution by scattering temperature sensors, a large number of temperature sensors must be installed, a method that should be avoided because of its high cost. Therefore, the goal of this paper is to estimate the temperature of substances such as crops and soil in greenhouses, which are secondarily affected by the atmospheric temperature, at a low cost instead of measuring atmospheric temperatures in a costly way. Temperature sensors for substances must be directly attached to the target object to measure its surface temperature, which can lead to the deterioration in quality. In contrast, infrared array sensors can measure the surface temperature of materials from a distance. They have been increasingly used in recent years due to growing demand, and they can be used to measure the surface temperature of a wide range of objects in a greenhouse. However, infrared array sensors also have many operational problems, such as dirty lenses, and the measurement error is larger than that of temperature sensors. Therefore, this paper proposes a machine learning model that predicts continuous temperature distribution in the form of a 16 ×18 pixels heat map from a limited number of temperature sensors. Evaluation results show that our approach is useful in different greenhouse environments, including different airconditioning systems. In addition, the model is computationally inexpensive enough to run in practical fields with limited computational resources; therefore, it can be run on relatively inexpensive embedded terminals. As for the accuracy, the average error of the heat map obtained by the proposed model is as small as 0.28 [°C/pixel].

In recent years, there has been an increase in the demand for three-dimensional data of geospatial information in various fields, such as automatic driving and disaster prevention. Beyond5G (above 5thgeneration network), which features wideband, low latency, reliable connection, allows for the easily collection of spatial information of many points and times from camera-equipped vehicles, such as self-driving cars and drones. Spatiotemporal information at any location and at any time can be provided depending on users’ requests. However, there are instances where the target point’s spatiotemporal information and the request’s target time have not been collected. Therefore, appropriate data completion is required to provide this service seamlessly. To create 3D models using photogrammetry, images are required; therefore, multiple videos that captured the streets near our university were prepared and split into frames. Then, those images were stored in our local storage. The information of the recording date, weather, file path to the image, and location (latitude and longitude) were saved to our database per frame. When users request spatiotemporal information, this system commences searching for images that meet the requirements. When the number of matched images are sufficient, the 3D model is created. When it is insufficient, the data are complemented appropriately.