Real Indoor Environment Research Articles

Identifying Indoor Objects Using Neutrosophic Reasoning for Mobility Assisting Visually Impaired People

Indoor object detection is a fundamental activity for the development of applications of mobility-assistive technology for visually impaired people (VIP). The challenge of seeing interior objects in a real indoor environment is a challenging one since there are numerous complicated issues that need to be taken into consideration, such as the complexity of the background, occlusions, and viewpoint shifts. Electronic travel aids that are composed of the necessary sensors may assist VIPs with their navigation. The sensors have the ability to detect any obstacles, regardless of whether they are static or dynamic, and offer information on the context of an interior scene. The characteristics of an interior scene are not very clear and are subject to a great deal of variation. Recent years have seen the emergence of methods for dealing with issues of this kind, some of which include the use of neural networks, probabilistic methods, and fuzzy logic. This study describes a method for detecting indoor objects using a rotational ultrasonic array and neutrosophic logic. A neutrosophic set has been seen as the next evolution of the fuzzy set because of its indeterminate membership value, which is absent from conventional fuzzy sets. The suggested method is constructed to reflect the position of the walls (obstacle distance) and to direct the VIP to move freely (ahead, to the right, or to the left) depending on the degree of truthiness, the degree of indeterminacy, and the degree of falsity for the reflected distance. The results of the experiments show that the suggested indoor object detecting system has good performance, as its accuracy rate (a mean average precision) is 97.2 ± 1%.

Measurement of the Diffusion and Partition Coefficients of 6-Methyl-5-hepten-2-one in Clothing

Deep understanding of the chemical reaction and physical transport characteristics of 6-methyl-5-hepten-2-one (6-MHO) in different phases (e.g., skin, clothing) is important for modeling the indoor squalene/ozone reaction. In this study, we make the first attempt to accurately measure the physical transport parameters, i.e., the diffusion coefficient (Dm) and partition coefficient (K) of 6-MHO in different clothing (jeans, nylon, pure cotton T-shirt). A ventilated–airtight–ventilated testing procedure was designed, and relevant experiments were performed in 30 L small-scale chambers. The key parameters of 6-MHO (Dm and K) are determined by nonlinearly fitting the measured gas-phase concentration data with physical model using a hybrid optimization method. The high fitting accuracy (R2 > 0.93), good replication (relative deviations of determined Dm and K less than 16% for duplicated experiments), and further sensitivity analysis demonstrate the reliability and robustness of the method. In addition, the variation of the key parameters of 6-MHO in the pure cotton T-shirt with temperature (in the range of 20–37 °C) is investigated. Results indicate that Dm increases while K decreases with increasing temperature. This study provides some of the fundamental data for characterizing the squalene/ozone reaction and should be helpful for the subsequent dermal exposure in realistic indoor environments.

Closed-Loop Wireless Power Transfer With Adaptive Waveform and Beamforming: Design, Prototype, and Experiment

A closed-loop far-field wireless power transfer (WPT) system with adaptive waveform and beamforming using limited feedback is designed, prototyped, and experimented. Spatial domain and frequency domain are jointly exploited by utilizing waveform and beamforming at the transmitter in WPT system to adapt to the multipath fading channel and boost the output dc power. A closed-loop architecture based on a codebook design and an over-the-air limited feedback with low complexity is proposed. The codebook consists of multiple codewords where each codeword represents particular waveform and beamforming. The transmitter sweeps through the codebook and the receiver then feeds the optimal codeword index back to the transmitter, so that the waveform and beamforming can be adaptive for maximizing the output dc power without requiring explicit channel estimation and the knowledge of accurate Channel State Information. The proposed closed-loop WPT with adaptive waveform and beamforming using limited feedback is prototyped using a Software Defined Radio equipment and measured in two real indoor environments. It is experimentally shown that the proposed closed-loop WPT with adaptive waveform and beamforming is able to enhance the output dc power by up to 14.7 dB in comparison with conventional 1-tone 1-antenna WPT system.

Current circumstances and prospects on performance evaluation for infection control technologies of airborne viruses in indoorenvironments.

Respiratory infectious diseases have potential of aerosol transmission such as COVID-19. The development of new technologies for infection control against airborne viruses are further required. It is necessary for effective development to evaluate properly the effect and role of these technologies in indoor environment. Here, the author provided essential knowledge for infection control of viral aerosols, i.e., basic concept of infection control, features of COVID-19 and Influenza including the entry receptor in body of each virus, behavior of the viral aerosols released from patient bodies, and Wells-Riley model as a traditional quantitative assessment of the infection risk by aerosol transmission. Previous evaluation studies on airborne viruses were categorized into three types of experiments, namely, in vitro, in vivo, and in humans and real indoor environments. Some prospects were described, including standard evaluation methods for air cleaners, the research group to formulate guidelines for evaluating the hygienic effects of chemical substances on microbes in real indoor space, and personal opinions on evaluation concept linked to three types of experiments. This minireview may help to correctly evaluate the hygienic effects of control technologies against airborne viruses in indoor environment and to contribute development of technologies with required performance according to infection risk.

Experimental and kinetic model evaluation of HONO production from surface nitrate photolysis

Solar irradiation plays a key role in promoting nitrous acid (HONO) production from surface enhanced photochemical reactions such as photolysis of nitrate. Nitrate anions are ubiquitous in ambient particles and on various outdoor and indoor surfaces, with reported surface density from 10−7 to 10−3 mol m−2. In this work, we measured HONO and NO2 production from surface nitrate photolysis under eight narrow bandwidth light sources with peak emission wavelengths ranging from 300 to 605 nm. We defined the probability of HONO and NO2 release per surface nitrate absorbing unit photon density of particular wavelength as “reactive cross section” of surface nitrate. Relative to the absorption cross section of gaseous HNO3, the reactive cross section of surface nitrate was found enhanced by up to 970-fold. The reactive cross section determined in our laboratory decreased from (6.3 ± 0.5) × 10−20 cm2 molecule−1 at 300 nm to (1.0 ± 0.3) × 10−24 cm2 molecule−1 at 605 nm, and showed a nonlinear relationship with the surface density of nitrate. Outdoor HONO production was mostly attributed to solar UV irradiation as expected, while underappreciated indoor HONO accumulation was also feasible considering HONO production induced by visible light and large specific surface area indoors. We then constructed a dynamic model to further evaluate the indoor HONO source in typical indoor environments. Photochemical properties of this HONO source lead to diurnal profiles in indoor HONO accumulation, with daily average HONO accumulation ranging from 1.1 to 17.8 ppb under various conditions. The model calculation confirmed that visible light substantially promotes surface nitrate photolysis and indoor HONO accumulation, with an average daily strength that is 0.28–0.34 mg h−1 at surface nitrate of 10−4 mol m−2 under simulated sunlight illumination. Our results call for direct measurements of indoor HONO in a variety of real indoor environments.

Temperature dependence of volatile organic compound emissions from surface coatings: Inter-species difference observed in real indoor environments and mechanistic understanding

Coatings often cover two-thirds of the surfaces in indoor environments and represent important sources of indoor volatile organic compounds (VOCs). Temperature is known to affect VOC emission rates from coatings, yet inter-species difference in the temperature dependence still needs to be understood. Based on time-resolved VOC measurements in an indoor air campaign conducted in residences in Beijing, China, we identified dibasic ester (DBE), a solvent mixture often used in coatings, and found that the concentration ratios of DBE components exhibited strong temperature dependence in an apartment when the indoor temperature declined stepwise over a multiweek period. To interpret the observational results, we developed a simplified mechanistic model relating the temperature dependence of VOC emission rates from coated surfaces to the temperature dependence of the diffusion coefficient of the emitted VOCs in the coating layer and further to a predicable molecular property of the emitted VOCs, molar volumes at 0 K, based on the free-volume theory. This correlation was quantitatively verified using the DBE data as well as using the data of alkanes, another set of VOCs that might be emitted from coatings, observed in two apartments in the same campaign. Given that indoor temperature varies considerably over seasons and across regions, the correlation proposed herein may help better predict indoor VOC emissions from coatings.

Cascade filter system tests for airborne nano‐ and microplastic particle analytics in a laboratory dust chamber

AbstractMeasuring nano‐ and microplastic particle (NMP) loads in air requires suitable analytical filter systems that allow a size‐selective separation of airborne nano‐ and microplastic (NMP) fractions. Previously, various filter materials and setups including nano/micro cascade filtration stacks have been presented (P.‐T. Miclea, 2022). To evaluate these novel filter systems, implementing reliable sampling procedures and instrumentation, as well as sample handling, well‐defined test setups, NMP reference materials, and procedures are of major interest. In this study, a novel triple‐stage silicon (Si) and aluminum oxide (Al2O3) cascade filter system for airborne NMP particles is tested in a laboratory environment. Special focus is put on its ability for size‐selective analytical filtration of airborne NMP particles based on a well‐defined mixture of poly‐ and monodisperse NMP reference materials. For this purpose, a laboratory dust test chamber has been adapted for the dispersion of NMP particles in air as well as the vacuum air extraction through the cascade filter system. A defined NMP test material was fabricated from polydisperse polyethylene (PE) and PE terephthalate, and monodisperse polystyrene, polymethylmethacrylate, spanning a size region from 130 nm to >22 µm. Dispersion and filtration experiments are performed by vacuum extraction under defined temperature, pressure, and humidity conditions on single and triple filtration stages. NMP particle analytics are performed with polydisperse PE particles on a single‐stage filter system with subsequent optical microscopy. Here, the aim is to detect the filtration cutoff for retained particles and transmitted particles referring to initial distribution. A second experiment is run on a triple filtration cascade with mixed poly‐ and monodisperse NMP test material. The different filter stages are equipped with analytical Si and Al2O3 filters with pore sizes of 10 µm, 1 µm, and 100 nm, to test the size selectivity and sealing of the complete cascade setup. The NMP particle size distribution on the individual analytical filter substrates is analyzed by optical and scanning electron microscopy to assess size selectivity performance of the triple‐stage setup. The obtained results in our laboratory NMP dust‐dispersion test chamber demonstrate requirements and limitation of size‐selective NMP cascade filter systems and the involved analytical procedures. Future applications in real indoor and outdoor environments will be discussed.

A Non-Line-of-Sight Wireless Indoor Localization System Using Custom-Designed Radio Relay and Uniform Circular Array

Indoor localization using WiFi signals, though getting studied for many years, is not adopted extensively in real indoor environments yet. One possible impediment is that many prior systems perform worse in non-line-of-sight (NLoS) indoor scenarios than in line-of-sight (LoS) scenarios or even fail to work. Nevertheless, realizing localization in both LoS and NLoS scenarios is indispensable for a viable system working in indoor environments. Since existing state-of-the-art solutions mainly aim to localize a target in LoS indoor scenarios, we present a system that uses multipath channel parameters to achieve localization with a single AP in indoor NLoS scenarios. The proposed introduces deployable and simple custom-designed radio relays to assist localization, which can receive signals from the target and forward them to the AP in NLoS scenarios. Field trials conducted in a real NLoS scenario show that our system achieves a median 3-D localization accuracy of 86 cm with a uniform circular array (UCA) in most NLoS conditions, i.e., the direct path from a target to AP is not completely blocked. More importantly, even if in a challenging NLoS condition, i.e., the direct path does not exist, our system can still achieve a median 2-D localization accuracy of 85 cm.

Indoor mm-Wave Coverage Enhancement: Reconfigurable Intelligent Surface Deployment Strategy Based on Human Mobility Model

Reconfigurable intelligent surface (RIS) is regarded as a promising technology to assist indoor millimeter wave (mm-Wave) communication due to its ability to manipulate electromagnetic waves directionally. Due to the complexity of indoor environment, different RIS deployment strategies significantly affect the system performance. However, the existing work investigating RIS indoor deployment ignored the influence of human mobility to indoor signal coverage, therefore a practical indoor deployment strategy remains unknown. In this letter, based on the consideration of human blockage to mm-Wave signal and probability distribution of human location, we propose an indoor RIS deployment strategy with human mobility model to minimize the outage probability, which is more applicable to the realistic indoor environment. An algorithm is presented to minimize the outage probability at a subspace scale of the room geometry and calculate the corresponding RIS location for single RIS deployment and multiple RISs cooperative deployment. Numerical results are provided to display how our deployment strategy works, which indicate that deploying RIS correctly can decrease the indoor communication outage probability significantly.

Dynamic reconstruction in simultaneous localization and mapping based on the segmentation of high variability point zones

Dynamic scene reconstruction in real environments is still an ongoing research challenge; moving objects affect the performance of static environment-based simultaneous localization and mapping and impede a correct scene reconstruction. This paper proposes a method for dynamic scene reconstruction using sensor fusion for dynamic simultaneous localization and mapping. It employs two-dimensional LIDAR statistical behaviour to detect and segment high variability point cloud areas containing a dynamic object. The method is computationally low cost, allowing a 6.6 Hz execution rate. It obtains point cloud reconstruction of a static scene by reducing, segmenting, and concatenating successive point clouds of a dynamic environment. The tests were in real indoor environments with a robotic vehicle and a person traversing a scene. The correlation between the static environment point cloud and successive reconstructed point clouds demonstrates that the proposed method reconstructs different environments in the presence of dynamic objects.

A Semi-Automatic Wheelchair with Navigation Based on Virtual-Real 2D Grid Maps and EEG Signals

A semi-automatic wheelchair allows disabled people to possibly control in an indoor environment with obstacles and targets. The paper proposes an EEG-based control system for the wheelchair based on a grid map designed to allow disabled people to reach any preset destination. In particular, the grid map is constructed by dividing it into grid cells that may contain free spaces or obstacles. The map with the grid cells is simulated to find the optimal paths to the target positions using a Deep Q-Networks (DQNs) model with the Parametric Rectified Linear Unit (PReLU) activation function, in which a novel algorithm for finding the optimal path planning by converting wheelchair actions is applied using the output parameters of the DQNs. For the wheelchair movement in one real indoor environment corresponding to the virtual 2D grid map, the initial position of the wheelchair will be determined based on natural landmarks and a graphical user interface designed for on-screen display can support disabled people in selecting the desired destination from a list of predefined locations using Electroencephalogram (EEG) signals by blinking eyes. Therefore, one user can easily and safely control the wheelchair using an EEG system to reach the desired target when the wheelchair position and destination are determined in the indoor environment. As a result, a grid map was developed and experiments for the semi-automatic wheelchair control were performed in real indoor environments to illustrate the effectiveness of the proposed method. In addition, the system is a platform to develop different types of controls depending on the types of user disabilities and different environmental maps built.

Zigbee Wireless Sensor Networks: Performance Study in an Apartment-Based Indoor Environment

Zigbee is a very popular technology for Internet of things (IoT) networks mainly because of its low power consumption and low-cost features. It shares the unlicensed 2.4 GHz Industrial, Scientific, and Medical (ISM) radio band with other wireless networks such as Wi-Fi. Usually, Zigbee and Wi-Fi networks coexist in indoor environments for their respective applications. Hence, the coexistence introduces interference for both types of networks lowering the performance of the networks, but Zigbee suffers more significant performance losses because of its lower transmission power than Wi-Fi. Since the number of IoT devices is increasing at an unprecedented rate due to numerous emerging applications and thus making the indoor environments very populous, the peaceful coexistence between Zigbee and Wi-Fi networks in proximity becomes an important research study. For this purpose, this paper presents a comprehensive performance study of a Zigbee network in the presence of a Wi-Fi interference network in a real-life apartment-based indoor environment where Wi-Fi access points of dense neighbors exist. The experiments were done in a XBee module-based Zigbee network for measuring the received signal strength indicator (RSSI), packet drop rate (PDR), and loopback throughput with and without nearby Wi-Fi traffic introduced on purpose. Various networking parameters such as the operating channels, the distances between Zigbee devices and Wi-Fi devices, the transmit timeout of Zigbee packets, and the transmission power of the Zigbee transmitter have been used in the experiments to study the network performance. Our results show that in the deployment of IoT networks in a smart home, radio interference from neighboring homes would not be an important factor, but serious considerations may need to be taken inside the same home. The experimental observations of this paper can serve as a good reference study for Zigbee network deployments in real indoor environments, particularly when interference sources are present in proximity.

Single-Antenna Device-to-Device Localization in Smart Environments With Backscatter

A long-standing vision of indoor localization is to eliminate infrastructure and deployment costs. Recent innovations make it possible to enable device-to-device (D2D) localization while requiring multiple antennas for the systems. We ask the following question: can we localize the more generally used single-antenna devices (e.g., IoT) using another single-antenna device (e.g., smartphone or smartwatch) in a smart environment where low-cost backscatter tags are widely deployed on walls or smart objects? In this article, we present TagLoc, a lightweight system that enables D2D localization without relying on large antenna arrays. Our observation is that the reflected signals from the ambient smart environment can be exploited to eliminate the requirement of bulky antenna arrays that are unachievable for the simple-designed IoT devices. Specifically, TagLoc creates multiple direction signatures using backscatter arrays in smart environments. Then, the receiver can accurately estimate the direction signatures from the transmitter to the arrays and then localize the target by cooperating all tag arrays. We prototype TagLoc using two single-antenna Intel NUCs with off-the-shelf Intel 5300 WiFi cards and customized backscatter tags. The results show TagLoc can achieve robust performance in a real indoor environment with a median localization error of 0.82 m.

Received Signal Strength Estimation in Indoor Environment Using High Frequency Rytov Approximation

We present a new and improved formulation of Rytov approximation (RA) using the theory of ray tracing in low loss media for solving the direct scattering problem. The resultant model provides an accurate prediction of wave scattering in the presence of electrically large scatterers with high permittivity and small loss tangent. The high validity range and the straightforward linear formulation of the proposed model make it suitable for accurately predicting the received signal strength (RSS) in indoor environments. We provide experimental results from a real indoor environment to show that the proposed method closely matches the experimental RSS values and achieves up to ten times higher validity range than any other existing linear approximate models. Also, since the proposed method provides an accurate solution to the direct scattering problem, it can be an ideal candidate for inverse scattering applications.

THE DEPLOYMENT OF A WI-FI POSITIONING SYSTEM VIA CROWDSOURCING

Abstract. Wi-Fi fingerprint positioning is widely used because of its ready hardware and high accuracy. However, its application is considerably restricted by time-consuming and labor-intensive works of offline collection and irregular fluctuation of signals. To address the above problems, we proposed a novel method to deploy the Wi-Fi fingerprint database based on implicit crowdsourcing and improved the weighted k-nearest neighbor (WKNN) algorithm to eliminate the influence of neighbor mismatching and device heterogeneity. First, ordinary users continuously gather Wi-Fi information instead of collecting one point after another. Meanwhile, video surveillance cameras record users’ trajectories without any intervention and use monocular vision based on plane constraints to obtain users’ location at the moment of each scanning. At the localization phase, the morphology similarity distance instead of the Euclidean distance is used to measure the similarity of signals to solve the problem of device heterogeneity. Outlier detection is also utilized for a secondary selection of neighbor points. Finally, geometric and signal morphology similarity distances are used to determine the combined weight of all neighbors after the dimensionless treatment. Results of the experiments conducted in a real indoor environment show that the proposed strategy improves the efficiency of fingerprint collection and achieves higher positioning accuracy.

Performance of Fingerprinting-Based Indoor Positioning with Measured and Simulated RSSI Reference Maps

Numerous indoor positioning technologies and systems have been proposed to localize people and objects in large buildings. Wi-Fi and Bluetooth positioning systems using fingerprinting have gained popularity, due to the wide availability of existing infrastructure. Unfortunately, the implementation of fingerprinting-based methods requires time-consuming radio surveys to prepare databases (RSSI maps) that serve as a reference for the radio signal. These surveys must be conducted for each individual building. Here, we investigate the possibility of using simulated RSSI maps with fingerprinting-based indoor localization systems. We discuss the suitability of the two popular radio wave propagation models for the preparation of RSSI reference data: ray tracing and multiwall. Based on an analysis of several representative indoor scenarios, we evaluated the performance of RSSI distribution maps obtained from simulations versus maps obtained from measurement campaigns. An experimental positioning system developed by the authors was used in the study. Based on Bluetooth Low Energy beacons and mobile devices (smartphones), the system uses fingerprinting followed by a particle filter algorithm to estimate the user’s current position from RSSI measurements and a reference spatial RSSI distribution database for each Bluetooth beacon in the building. The novelty of our contribution is that we evaluate the performance of the positioning system with RSSI maps prepared both from measurements and using the two most representative indoor propagation methods, in three different environments in terms of structure and size. We compared not only the three RSSI maps, but also how they influence the performance of the fingerprint-based positioning algorithm. Our original findings have important implications for the development of indoor localization systems and may reduce deployment times by replacing reference measurements with computer simulations. Replacing the labor-intensive and time-consuming process of building reference maps with computer modeling may significantly increase their usefulness and ease of adaptation in real indoor environments.

Focus on Impact: Indoor Exploration With Intrinsic Motivation

Exploration of indoor environments has recently experienced a significant interest, also thanks to the introduction of deep neural agents built in a hierarchical fashion and trained with Deep Reinforcement Learning (DRL) on simulated environments. Current state-of-the-art methods employ a dense extrinsic reward that requires the complete a priori knowledge of the layout of the training environment to learn an effective exploration policy. However, such information is expensive to gather in terms of time and resources. In this work, we propose to train the model with a purely intrinsic reward signal to guide exploration, which is based on the impact of the robot’s actions on its internal representation of the environment. So far, impact-based rewards have been employed for simple tasks and in procedurally generated synthetic environments with countable states. Since the number of states observable by the agent in realistic indoor environments is non-countable, we include a neural-based density model and replace the traditional count-based regularization with an estimated pseudo-count of previously visited states. The proposed exploration approach outperforms DRL-based competitors relying on intrinsic rewards and surpasses the agents trained with a dense extrinsic reward computed with the environment layouts. We also show that a robot equipped with the proposed approach seamlessly adapts to point-goal navigation and real-world deployment.

School-age children show poor use of spatial cues in reverberation for speech-in-speech perception

Understanding speech is particularly difficult for children when there is competing speech in the background. When the target and masker talkers are spatially separated, as compared to co-located, the access to corresponding auditory spatial cues can provide release from masking, resulting in an intelligibility gain for speech-in-speech perception. When tested in free-field environments, previous work showed that children demonstrate adult-like spatial release from masking (SRM) by 9–10 years of age. However, in indoor environments where most critical communications occur such as classrooms, reverberation distorts the critical auditory spatial cues that lead to reduced SRM among adults. Little is known about how children process distorted auditory spatial cues for SRM to aid speech-in-speech perception. In this work, we measure SRM in children in virtual reverberant environments that mimic typical learning spaces. We show free-field measurements overestimate SRM maturation in realistic indoor environments. Children show a more protracted development of SRM in reverberation, likely due to immaturities in using distorted auditory spatial cues.

REVE-CE: Remote Embodied Visual Referring Expression in Continuous Environment

Ithas always been a great challenge for the robot to navigate in the visual world following natural language instructions. Recently, several tasks such as the Vision-and-Language Navigation (VLN) and Remote Embodied Visual Referring Expression in Real Indoor Environments (REVERIE) are proposed trying to solve this challenge. And the most significant difference between VLN and REVERIE tasks is that REVERIE uses a higher guidance level instruction. However, the navigation process of REVERIE is implemented in a discrete environment, which is unrealistic in real world scenarios. To make the REVERIE task more consistent with the real physical world, we develop a new task of Remote Embodied Visual Referring Expression in Continuous Environment, namely REVE-CE, in which the agent executes a much longer sequence of low-level actions given language instructions. Furthermore, we propose a multi-branch cross modal attention (MBCMA) framework to solve the proposed REVE-CE task. Extensive experiments are conducted demonstrating that the proposed framework greatly outperforms the state-of-the-art VLN baselines and a new benchmark for the proposed REVE-CE task is built.

CNN-Based Dense Monocular Visual SLAM for Real-Time UAV Exploration in Emergency Conditions

Unmanned Aerial Vehicles (UAVs) for 3D indoor mapping applications are often equipped with bulky and expensive sensors, such as LIDAR (Light Detection and Ranging) or depth cameras. The same task could be also performed by inexpensive RGB cameras installed on light and small platforms that are more agile to move in confined spaces, such as during emergencies. However, this task is still challenging because of the absence of a GNSS (Global Navigation Satellite System) signal that limits the localization (and scaling) of the UAV. The reduced density of points in feature-based monocular SLAM (Simultaneous Localization and Mapping) then limits the completeness of the delivered maps. In this paper, the real-time capabilities of a commercial, inexpensive UAV (DJI Tello) for indoor mapping are investigated. The work aims to assess its suitability for quick mapping in emergency conditions to support First Responders (FR) during rescue operations in collapsed buildings. The proposed solution only uses images in input and integrates SLAM and CNN-based (Convolutional Neural Networks) Single Image Depth Estimation (SIDE) algorithms to densify and scale the data and to deliver a map of the environment suitable for real-time exploration. The implemented algorithms, the training strategy of the network, and the first tests on the main elements of the proposed methodology are reported in detail. The results achieved in real indoor environments are also presented, demonstrating performances that are compatible with FRs’ requirements to explore indoor volumes before entering the building.