Search Results (27)

Indoor point clouds often present significant challenges due to the complexity and variety of structures and high object similarity. The local geometric structure helps the model learn the shape features of objects at the detail level, while the global context provides overall scene semantics and spatial relationship information between objects. To address these challenges, we propose a novel network architecture, PointMSGT, which includes a multi-scale geometric feature extraction (MSGFE) module and a global Transformer (GT) module. The MSGFE module consists of a geometric feature extraction (GFE) module and a multi-scale attention (MSA) module. The GFE module reconstructs triangles through each point’s two neighbors and extracts detailed local geometric relationships by the triangle’s centroid, normal vector, and plane constant. The MSA module extracts features through multi-scale convolutions and adaptively aggregates features, focusing on both local geometric details and global semantic information at different scale levels, enhancing the understanding of complex scenes. The global Transformer employs a self-attention mechanism to capture long-range dependencies across the entire point cloud. The proposed method demonstrates competitive performance in real-world indoor scenarios, with a mIoU of 68.6% in semantic segmentation on S3DIS and OA of 86.4% in classification on ScanObjectNN. Full article

This paper investigates a Local Strategy-Driven Multi-Agent Deep Deterministic Policy Gradient (LSD-MADDPG) method for demand-side energy management systems (EMS) in smart communities. LSD-MADDPG modifies the conventional MADDPG framework by limiting data sharing during centralized training to only discretized strategic information. During execution, it relies solely on local information, eliminating post-training data exchange. This approach addresses critical challenges commonly faced by EMS solutions serving dynamic, increasing-scale communities, such as communication delays, single-point failures, scalability, and nonstationary environments. By leveraging and sharing only strategic information among agents, LSD-MADDPG optimizes decision-making while enhancing training efficiency and safeguarding data privacy—a critical concern in the community EMS. The proposed LSD-MADDPG has proven to be capable of reducing energy costs and flattening the community demand curve by coordinating indoor temperature control and electric vehicle charging schedules across multiple buildings. Comparative case studies reveal that LSD-MADDPG excels in both cooperative and competitive settings by ensuring fair alignment between individual buildings’ energy management actions and community-wide goals, highlighting its potential for advancing future smart community energy management. Full article

UWB has been in existence for several years, but it was only a few years ago that it transitioned from a specialized niche to more mainstream applications. Recent market data indicate a rapid increase in the popularity of UWB in consumer products, such as smartphones and smart home devices, as well as automotive and industrial real-time location systems. The challenge of achieving accurate positioning in indoor environments arises from various factors such as distance, location, beacon density, dynamic surroundings, and the density and type of obstacles. This research used MFi-certified UWB beacon chipsets and integrated them with a mobile application dedicated to iOS by implementing the near interaction accessory protocol. The analysis covers both static and dynamic cases. Thanks to the acquisition of measurements, two main candidates for indoor localization infrastructure were analyzed and compared in terms of accuracy, namely UWB and LIDAR, with the latter used as a reference system. The problem of achieving accurate positioning in various applications and environments was analyzed, and future solutions were proposed. The results show that the achieved accuracy is sufficient for tracking individuals and may serve as guidelines for achievable accuracy or may provide a basis for further research into a complex sensor fusion-based navigation system. This research provides several findings. Firstly, in dynamic conditions, LIDAR measurements showed higher accuracy than UWB beacons. Secondly, integrating data from multiple sensors could enhance localization accuracy in non-line-of-sight scenarios. Lastly, advancements in UWB technology may expand the availability of competitive hardware, facilitating a thorough evaluation of its accuracy and effectiveness in practical systems. These insights may be particularly useful in designing navigation systems for blind individuals in buildings. Full article

In this paper, we present a novel localization scheme, location-aware ranging correction (LARC), to correct ranging estimates from ultra wideband (UWB) signals. Existing solutions to calculate ranging corrections rely solely on channel information features (e.g., signal energy, maximum amplitude, estimated range). We propose to incorporate a preliminary location estimate into a localization chain, such that location-based features can be calculated as inputs to a range-error prediction model. This way, we can add information to range-only measurements without relying on additional hardware such as an inertial measurement unit (IMU). This improves performance and reduces overfitting behavior. We demonstrate our LARC method using an open-access measurement dataset with distances up to 20 m, using a simple regression model that can run purely on the CPU in real-time. The inclusion of the proposed features for range-error mitigation decreases the ranging error 90th percentile (P90) by 58% to 15 cm (compared to the uncorrected range error), for an unseen trajectory. The 2D localization P90 error is improved by 21% to 18 cm. We show the robustness of our approach by comparing results to a changed environment, where metallic objects have been moved around the room. In this modified environment, we obtain a 56% better P90 ranging performance of 16 cm. The 2D localization P90 error improves as much as for the unchanged environment, by 17% to 18 cm, showing the robustness of our method. This method evolved from the first-ranking solution of the 2021 and 2022 International Conference on Indoor Position and Indoor Navigation (IPIN) Competition. Full article

Rose flowers have been cultivated for their fragrance and their garden value since ancient times. Very ancient cultivars became famous locally for their specific use, and competitive horticultural activities progressively established, leading, with time, to landraces with limited polymorphism. The most famous examples are the oil-bearing Damask roses from Iran and the Yueyue Fen garden strain from China. In 1817, a new rose, allegedly a hybrid from the two previous lineages, was discovered at Reunion. From this plant, as early as the 1820s, a new founder group, the Bourbon roses, was developed in France, which immediately stirred up deep passions among botanists and skilled enthusiasts. Today, more than 30,000 named cultivars have been raised either as garden and landscape plants for the cut rose market or as indoor pot plants. The market handles billions of euros a year, making the rose the most economically important crop worldwide. Following the inheritance of SSR DNA markers, we here propose a reconstitution of the very early lineage of Bourbon roses, clarifying one of the major steps, if not the major one, that links these very ancient heritage roses to modern roses. Full article

Personalization of workloads is essential for optimizing training processes and minimizing the risk of injuries in sports. Precise knowledge of the external load demands borne by basketball players during competition is necessary for this purpose. The objective of this research was to determine the objective external load demands of five variables during a basketball competition, three kinematic (speed, accelerations, and decelerations) and two neuromuscular variables (impacts/min and Player Load/min), and subsequently establish workload ranges. Six official matches from preparatory tournaments involving professional basketball players from the Spanish first division, Liga ACB, were analyzed. Inertial devices and an UWB system were used for variable localization and recording within indoor spaces. Two methods, two-step and k-means clustering, were employed for workload range classification. The results revealed different workload thresholds clusters based on the data analysis technique used. The following speed ranges were identified in professional basketball players: Standing, <2.95 km/h; Walking, 2.96 to 7.58 km/h; Jogging, 7.59 to 12.71 km/h; Running, 12.72 to 17.50 km/h; and Sprinting, >17.51 km/h. The center of cluster 5 was found to determine the concept of a sprint (>19 km/h) as well as high-speed running (>17.50 km/h). Acceleration and deceleration ranges displayed few cases but with considerably high values, which must be considered when designing injury prevention tasks. The distribution of impacts showed a normal pattern, with identified periods during which players withstood significant G-forces (14%). Finally, the Player Load value at which an activity is considered to be very high, 1.95 au/min, was identified. Considering the obtained results, basketball is proposed as a sport with a high neuromuscular load. Coaches should choose the classification method that best suits their needs. These reference values are the first of their kind for this population of top-level professional players and should aid in adjusting training processes to match competition demands. Full article

Sports centers are significant energy consumers. This article outlines the engineering design for a comprehensive energy performance upgrade of the indoor sports hall in Arkalochori, Greece, and presents the projected results. The indoor sports hall constitutes a major sport facility on the mainland of Crete, hosting a broad cluster of sport municipal activities and the official basketball games of the local team in the 2nd national category. Having been constructed in the mid-1990s, the facility exhibits very low thermal performance, with considerably high U-factors for all constructive elements (from 4 to 5 W/m²∙K), still use of diesel oil for indoor space heating and domestic heat water production, and ineffective old lamps and luminaries covering the lighting needs of the facility. The energy performance upgrade of the indoor sports hall was studied, and the following passive and active measures were considered: Opaque-surfaces’ thermal insulation and openings’ replacement, stone wool panels, installation of heat pumps for indoor space conditioning, removal of diesel oil for any end use, production of domestic hot water from a novel solar-combi system, upgrade of lighting equipment, installation of solar tubes on the main sports hall roof for natural lighting as well as of a photovoltaic system for covering the remaining electricity consumption. With the proposed interventions, the studied building becomes a zero-energy facility. The payback period of the investment was calculated at 26 years on the basis of the avoided energy cost. This work was funded by the “NESOI” Horizon 2020 project and received the public award “Islands Gamechanger” competition of the NESOI project and the Clean Energy for EU Islands initiative. Full article

Semantic segmentation of large-scale indoor 3D point cloud scenes is crucial for scene understanding but faces challenges in effectively modeling long-range dependencies and multi-scale features. In this paper, we present RegionPVT, a novel Regional-to-Local Point-Voxel Transformer that synergistically integrates voxel-based regional self-attention and window-based point-voxel self-attention for concurrent coarse-grained and fine-grained feature learning. The voxel-based regional branch focuses on capturing regional context and facilitating inter-window communication. The window-based point-voxel branch concentrates on local feature learning while integrating voxel-level information within each window. This unique design enables the model to jointly extract local details and regional structures efficiently and provides an effective and efficient solution for multi-scale feature fusion and a comprehensive understanding of 3D point clouds. Extensive experiments on S3DIS and ScanNet v2 datasets demonstrate that our RegionPVT achieves competitive or superior performance compared with state-of-the-art approaches, attaining mIoUs of 71.0% and 73.9% respectively, with significantly lower memory footprint. Full article

Localization is a crucial skill in mobile robotics because the robot needs to make reasonable navigation decisions to complete its mission. Many approaches exist to implement localization, but artificial intelligence can be an interesting alternative to traditional localization techniques based on model calculations. This work proposes a machine learning approach to solve the localization problem in the RobotAtFactory 4.0 competition. The idea is to obtain the relative pose of an onboard camera with respect to fiducial markers (ArUcos) and then estimate the robot pose with machine learning. The approaches were validated in a simulation. Several algorithms were tested, and the best results were obtained by using Random Forest Regressor, with an error on the millimeter scale. The proposed solution presents results as high as the analytical approach for solving the localization problem in the RobotAtFactory 4.0 scenario, with the advantage of not requiring explicit knowledge of the exact positions of the fiducial markers, as in the analytical approach. Full article

Despite their enormous potential, the use of indoor localization systems (ILS) remains seldom. One reason is the lack of market transparency and stakeholders’ trust in the systems’ performance as a consequence of insufficient use of test and evaluation (T&E) methodologies. The heterogeneous nature of ILS, their influences, and their applications pose various challenges for the design of a methodology that provides meaningful results. Methodologies for building-wide testing exist, but their use is mostly limited to associated indoor localization competitions. In this work, the T&E 4iLoc Framework is proposed—a methodology for T&E of indoor localization systems in semi-controlled environments based on a system-level and black-box approach. In contrast to building-wide testing, T&E in semi-controlled environments, such as test halls, is characterized by lower costs, higher reproducibility, and better comparability of the results. The limitation of low transferability to real-world applications is addressed by an application-driven design approach. The empirical validation of the T&E 4iLoc Framework, based on the examination of a contour-based light detection and ranging (LiDAR) ILS, an ultra wideband ILS, and a camera-based ILS for the application of automated guided vehicles in warehouse operation, demonstrates the benefits of T&E with the T&E 4iLoc Framework. Full article

Point-based networks have been widely used in the semantic segmentation of point clouds owing to the powerful 3D convolution neural network (CNN) baseline. Most of the current methods resort to intermediate regular representations for reorganizing the structure of point clouds for 3D CNN networks, but they may neglect the inherent contextual information. In our work, we focus on capturing discriminative features with the interactive attention mechanism and propose a novel method consisting of the regional simplified dual attention network and global graph convolution network. Firstly, we cluster homogeneous points into superpoints and construct a superpoint graph to effectively reduce the computation complexity and greatly maintain spatial topological relations among superpoints. Secondly, we integrate cross-position attention and cross-channel attention into a single head attention module and design a novel interactive attention gating (IAG)-based multilayer perceptron (MLP) network (IAG–MLP), which is utilized for the expansion of the receptive field and augmentation of discriminative features in local embeddings. Afterwards, the combination of stacked IAG–MLP blocks and the global graph convolution network, called IAGC, is proposed to learn high-dimensional local features in superpoints and progressively update these local embeddings with the recurrent neural network (RNN) network. Our proposed framework is evaluated on three indoor open benchmarks, and the 6-fold cross-validation results of the S3DIS dataset show that the local IAG–MLP network brings about 1% and 6.1% improvement in overall accuracy (OA) and mean class intersection-over-union (mIoU), respectively, compared with the PointNet local network. Furthermore, our IAGC network outperforms other CNN-based approaches in the ScanNet V2 dataset by at least 7.9% in mIoU. The experimental results indicate that the proposed method can better capture contextual information and achieve competitive overall performance in the semantic segmentation task. Full article

Indoor localization is an enabling technology for pervasive and mobile computing applications. Although different technologies have been proposed for indoor localization, Wi-Fi fingerprinting is one of the most used techniques due to the pervasiveness of Wi-Fi technology. Most Wi-Fi fingerprinting localization methods presented in the literature are discriminative methods. We present a generative method for indoor localization based on Wi-Fi fingerprinting. The Received Signal Strength Indicator received from a Wireless Access Point is modeled by a hidden Markov model. Unlike other algorithms, the use of a hidden Markov model allows ours to take advantage of the temporal autocorrelation present in the Wi-Fi signal. The algorithm estimates the user’s location based on the hidden Markov model, which models the signal and the forward algorithm to determine the likelihood of a given time series of Received Signal Strength Indicators. The proposed method was compared with four other well-known Machine Learning algorithms through extensive experimentation with data collected in real scenarios. The proposed method obtained competitive results in most scenarios tested and was the best method in 17 of 60 experiments performed. Full article

In higher-education world heritage sites, the conservation and energy retrofitting of heritage buildings (HBs) is an important vector for their development, competitiveness and welfare. To guarantee their ongoing use, these buildings must be adapted to face current and emerging societal challenges: (i) the conservation of cultural heritage and the maintenance of their original characteristics and identity; (ii) the transformation of heritage sites into tourist centers that energize the local economy, generating revenue and jobs; (iii) the adaptation of the buildings to new uses and functions that demand energy retrofitting strategies to satisfy today’s standards of thermal comfort, indoor environmental quality (IEQ) and energy efficiency; (iv) tackling impacts of climate change, particularly global warming and extreme weather events; and finally, (v) the implementation of strategies to mitigate the impact of a growing number of tourists. The combined implications of these challenges require a comprehensive approach with interrelated measures strongly reliant on the use of technology and innovation. This work aims to discuss how higher-education cultural HBs can be rethought to serve these expectations. Moreover, a multidisciplinary intervention framework is provided to discuss how HBs can respond to the challenges and risks of rehabilitation, energy retrofitting, climate change and increasing tourism. Full article

In the Panama context, energy consumption in the building sector is mostly related to the conditioning of indoor spaces for cooling and lighting. Different nature strategies can be mimic to strongly impact these two aspects in the building sector, such as the ones presented here. A comprehensive analysis regarding literature related to biomimicry-based approaches destined to improve buildings designs is presented here. This analysis is driven by the increasing energy regulations demands to meet future local goals and to propose a framework for applications in Panama. Such biomimicry-based approaches have been further analyzed and evaluated to propose the incorporation of organism-based design for three of the most climate types found in Panama. Consequently, a SWOT analysis helped realized the potential that biomimicry-based approaches might have in improving the odds of in meeting the local and global regulations demands. The need for multidisciplinary collaboration to accomplish biomimicry-based-designed buildings, brings an increment in the competitivity regarding more trained human-assets, widening the standard-construction-sector thinking. Finally, the analysis presented here can serve as the foundation for further technical assessment, via numerical and experimental means. Full article

Energy resilience can be reached with a secure, sustainable, competitive, and affordable system. In order to achieve energy resilience in the urban environment, urban-scale energy models play a key role in supporting the promotion and identification of effective energy-efficient and low-carbon policies pertaining to buildings. In this work, a dynamic urban-scale energy model, based on an energy balance, has been designed to take into account the local climate conditions and morphological urban-scale parameters. The aim is to present an engineering methodology, applied to clusters of buildings, using the available urban databases. This methodology has been calibrated and optimized through an iterative procedure on 102 residential buildings in a district of the city of Turin (Italy). The results of this work show how a place-based dynamic energy balance methodology can also be sufficiently accurate at an urban scale with an average seasonal relative error of 14%. In particular, to achieve this accuracy, the model has been optimized by correcting the typological and geometrical characteristics of the buildings and the typologies of ventilation and heating system; in addition, the indoor temperatures of the buildings—that were initially estimated as constant—have been correlated to the climatic variables. The proposed model can be applied to other cities utilizing the existing databases or, being an engineering model, can be used to assess the impact of climate change or other scenarios. Full article