Optimization of the Form Factors of Advanced Li-S Pouch Cells.

Lithium-sulfur (Li-S) batteries hold immense promise as next-generation energy storage due to their high theoretical energy density (2600 Wh kg⁻¹), low cost, and non-toxic nature. However, practical implementation faces challenges, primarily from Li polysulfide (LiPS) shuttling within the cathode and Li dendrite growth at the anode. Optimized electrodes/electrolytes design effectively confines LiPS to the cathode, boosting cycling performance in coin cells to up to hundreds of cycles. Scaling up to larger pouch cells presents new obstacles, requiring further research for long-term stability. A 1.45 Ah pouch cell, with optimized sulfur loading and electrolyte/sulfur ratio is developed, which delivers an energy density of 151 Wh kg-1 with 70% capacity retention up to 100 cycles. Targeting higher energy density (180 Wh kg-1), the developed 1Ah pouch cell exhibits 68% capacity retention after 50 cycles. Morphological analysis reveals that pouch cell failure is primarily from Li metal powdering and resulting polarization, rather than LiPS shuttling. This occurs for continuous Li ion stripping/plating during cycling, leading to dendrite growth and formation of non-reactive Li powder, especially under high currents. These issues increase ion diffusion resistance and reduce coulombic efficiency over time. Therefore, the study highlights the importance of a protected Li metal anode for achieving high-energy-dense batteries.

eProbe: Sampling of Environmental DNA within Tree Canopies with Drones.

Environmental DNA (eDNA) analysis is a powerful tool for studying biodiversity in forests and tree canopies. However, collecting representative eDNA samples from these high and complex environments remains challenging. Traditional methods, such as surface swabbing or tree rolling, are labor-intensive and require significant effort to achieve adequate coverage. This study proposes a novel approach for unmanned aerial vehicles (UAVs) to collect eDNA within tree canopies by using a surface swabbing technique. The method involves lowering a probe from a hovering UAV into the canopy and collecting eDNA as it descends and ascends through branches and leaves. To achieve this, a custom-designed robotic system was developed featuring a winch and a probe for eDNA collection. The design of the probe was optimized, and a control logic for the winch was developed to reduce the risk of entanglement while ensuring sufficient interaction force to facilitate transfer of eDNA onto the probe. The effectiveness of this method was demonstrated during the XPRIZE Rainforest Semi-Finals as 10 eDNA samples were collected from the rainforest canopy, and a total of 152 molecular operational taxonomic units (MOTUs) were identified using eDNA metabarcoding. We further investigate how the number of probe interactions with vegetation, the penetration depth, and the sampling duration influence the DNA concentration and community composition of the samples.

The effects of drone transportation on routine laboratory, immunohematology, flow cytometry and molecular analyses.

OBJECTIVES: Transportation of medical samples between laboratories or hospital sites is typically performed by motorized ground transport. Due to the increased traffic congestions in urban environments, drone transportation has become an attractive alternative for fast shipping of samples. In accordance with the CLSI guidelines and the ISO 15189 standard, the impact of this transportation type on sample integrity and performance of laboratory tests must be thoroughly validated. METHODS: Blood samples from 36 healthy volunteers and bacterial spiked urine samples were subjected to a 20-40 min drone flight before they were analyzed and compared with their counterparts that stayed on the ground. Effects on stability of 30 routine biochemical and hematological parameters, immunohematology tests and flow cytometry and molecular tests were evaluated. RESULTS: No clinically relevant effects on blood group typing, flow cytometry lymphocyte subset testing and on the stability of the multicopy opacity-associated proteins (Opa) genes in bacterial DNA nor on the number of Abelson murine leukemia viral oncogene homolog 1 (abl) housekeeping genes in human peripheral blood cells were seen. For three of the 30 biochemistry and hematology parameters a statistically significant difference was found: gamma-glutamyl transferase (gamma-GT), mean corpuscular hemoglobin (MCH) and thrombocyte count. A clinically relevant effect however was only seen for potassium and lactate dehydrogenase (LDH). CONCLUSIONS: Multi-rotor drone transportation can be used for medical sample transportation with no effect on the majority of the tested parameters, including flow cytometry and molecular analyses, with the exception of a limited clinical impact on potassium and LDH.

Comparative studies of knapsack, boom, and drone sprayers for weed management in soybean (Glycine max L.).

The study titled, "Comparative Evaluation of Knapsack, Boom, and Drone Sprayers for Weed Management in Soybean (Glycine max L.)" was carried out during the Kharif season 2021-22 at an experimental farm affiliated with the Department of Agronomy, Vasantrao Naik Marathwada Krishi Vidyapeeth, Parbhani. The primary objective was to evaluate the comparative efficacy of various sprayers in controlling weeds in soybeans and their work efficiency. The Randomized Block Design (RBD) included ten treatments of pre-emergence (PE) and post-emergence (POE) herbicides applied by knapsack, boom, and drone sprayers. Pendimethalin 30% EC @ 750 g a.i ha-1 was used for pre-emergence herbicide application, and Imazamox 35% EC + Imazethapyr 35% WG @ 70 g a.i ha-1 were used for post-emergence. These treatments were tested on soybean Monocot and Dicot weed count, weed dry weight, weed index, and weed control efficiency. The sprayers were compared for time, water, labor, herbicide, and overall work efficiency. A knapsack sprayer showed the best results for pre- and post-emergence herbicide application, with the lowest weed count, dry weight, control efficiency, and weed index. Boom and drone sprayers followed in effectiveness. Herbicide application was faster with the drone sprayer than with hand weeding, cultural practices, boom sprayer, and knapsack sprayer. Compared to knapsack and boom sprayers, the drone sprayer used less water and labour. Drone sprayers work most efficiently, followed by boom and knapsack sprayers. This study focuses on the prevalence of herbicides and their impact on non-target ecosystems. It aims to develop mitigation strategies by optimizing spraying efficiency and reducing herbicide usage during pre and post emergence. The dissemination of efficient weed management practices that reduce environmental impacts and increase the efficiency of soybean cultivation is consistent with Sustainable Development Goal 15: life on land.

Yellow Dwarf Virus Resistance in Barley: Phenotyping, Remote Imagery, and Virus-Vector Characterization.

Yellow dwarf viruses (YDVs) spread by aphids are some of the most economically important barley (Hordeum vulgare) virus-vector complexes worldwide. Detection and control of these viruses are critical components in the production of barley, wheat, and numerous other grasses of agricultural importance. Genetic control of plant diseases is often preferable to chemical control to reduce the environmental and economic cost of foliar insecticides. Accordingly, the objectives of this work were to (i) screen a barley population for resistance to YDVs under natural infection using phenotypic assessment of disease symptoms, (ii) implement drone imagery to further assess resistance and test its utility as a disease screening tool, (iii) identify the prevailing virus and vector types in the experimental environment, and (iv) perform a genome-wide association study to identify genomic regions associated with measured traits. Significant genetic differences were found in a population of 192 barley inbred lines regarding their YDV symptom severity, and symptoms were moderately to highly correlated with grain yield. The YDV severity measured with aerial imaging was highly correlated with on-the-ground estimates (r = 0.65). Three aphid species vectoring three YDV species were identified with no apparent genotypic influence on their distribution. A quantitative trait locus impacting YDV resistance was detected on chromosome 2H, albeit undetected using aerial imaging. However, quantitative trait loci for canopy cover and mean normalized difference vegetation index were successfully mapped using the drone. This work provides a framework for utilizing drone imagery in future resistance breeding efforts for YDVs in cereals and grasses, as well as in other virus-vector disease complexes.

Using image segmentation models to analyse high-resolution earth observation data: new tools to monitor disease risks in changing environments.

BACKGROUND: In the near future, the incidence of mosquito-borne diseases may expand to new sites due to changes in temperature and rainfall patterns caused by climate change. Therefore, there is a need to use recent technological advances to improve vector surveillance methodologies. Unoccupied Aerial Vehicles (UAVs), often called drones, have been used to collect high-resolution imagery to map detailed information on mosquito habitats and direct control measures to specific areas. Supervised classification approaches have been largely used to automatically detect vector habitats. However, manual data labelling for model training limits their use for rapid responses. Open-source foundation models such as the Meta AI Segment Anything Model (SAM) can facilitate the manual digitalization of high-resolution images. This pre-trained model can assist in extracting features of interest in a diverse range of images. Here, we evaluated the performance of SAM through the Samgeo package, a Python-based wrapper for geospatial data, as it has not been applied to analyse remote sensing images for epidemiological studies. RESULTS: We tested the identification of two land cover classes of interest: water bodies and human settlements, using different UAV acquired imagery across five malaria-endemic areas in Africa, South America, and Southeast Asia. We employed manually placed point prompts and text prompts associated with specific classes of interest to guide the image segmentation and assessed the performance in the different geographic contexts. An average Dice coefficient value of 0.67 was obtained for buildings segmentation and 0.73 for water bodies using point prompts. Regarding the use of text prompts, the highest Dice coefficient value reached 0.72 for buildings and 0.70 for water bodies. Nevertheless, the performance was closely dependent on each object, landscape characteristics and selected words, resulting in varying performance. CONCLUSIONS: Recent models such as SAM can potentially assist manual digitalization of imagery by vector control programs, quickly identifying key features when surveying an area of interest. However, accurate segmentation still requires user-provided manual prompts and corrections to obtain precise segmentation. Further evaluations are necessary, especially for applications in rural areas.

DCE-Qnet: deep network quantification of dynamic contrast enhanced (DCE) MRI.

INTRODUCTION: Quantification of dynamic contrast-enhanced (DCE)-MRI has the potential to provide valuable clinical information, but robust pharmacokinetic modeling remains a challenge for clinical adoption. METHODS: A 7-layer neural network called DCE-Qnet was trained on simulated DCE-MRI signals derived from the Extended Tofts model with the Parker arterial input function. Network training incorporated B1 inhomogeneities to estimate perfusion (Ktrans, vp, ve), tissue T1 relaxation, proton density and bolus arrival time (BAT). The accuracy was tested in a digital phantom in comparison to a conventional nonlinear least-squares fitting (NLSQ). In vivo testing was conducted in ten healthy subjects. Regions of interest in the cervix and uterine myometrium were used to calculate the inter-subject variability. The clinical utility was demonstrated on a cervical cancer patient. Test-retest experiments were used to assess reproducibility of the parameter maps in the tumor. RESULTS: The DCE-Qnet reconstruction outperformed NLSQ in the phantom. The coefficient of variation (CV) in the healthy cervix varied between 5 and 51% depending on the parameter. Parameter values in the tumor agreed with previous studies despite differences in methodology. The CV in the tumor varied between 1 and 47%. CONCLUSION: The proposed approach provides comprehensive DCE-MRI quantification from a single acquisition. DCE-Qnet eliminates the need for separate T1 scan or BAT processing, leading to a reduction of 10 min per scan and more accurate quantification.

Drones reduce the time to defibrillation in a highly visited non-urban area: A randomized simulation-based trial.

INTRODUCTION: Out-of-hospital cardiac arrest (OHCA) has a high global incidence and mortality rate, with early defibrillation significantly improving survival. Our aim was to assess the feasibility of autonomous drone delivery of automated external defibrillators (AED) in a non-urban area with physical barriers and compare the time to defibrillate (TTD) with bystander retrieval from a public access defibrillator (PAD) point and helicopter emergency medical services (HEMS) physician performed defibrillation. METHODS: This randomized simulation-based trial with a cross-over design included bystanders performing AED retrievals either delivered by automated drone flight or on foot from a PAD point, and simulated HEMS interventions. The primary outcome was the time to defibrillation, with secondary outcomes comparing workload, perceived physical effort, and ease of use. RESULTS: Thirty-six simulations were performed. Drone-delivered AED intervention had a significantly shorter TTD [2.2 (95 % CI 2.0-2.3) min] compared to PAD retrieval [12.4 (95 % CI 10.4-14.4) min] and HEMS [18.2 (95 % CI 17.1-19.2) min]. The self-reported physical effort on a visual analogue scale for drone-delivered AED was significantly lower versus PAD [2.5 (1 - 22) mm vs. 81 (65-99) mm, p = 0.02]. The overall mean workload measured by NASA-TLX was also significantly lower for drone delivery compared to PAD [4.3 (1.2-11.7) vs. 11.9 (5.5-14.5), p = 0.018]. CONCLUSION: The use of drones for automated AED delivery in a non-urban area with physical barriers is feasible and leads to a shorter time to defibrillation. Drone-delivered AEDs also involve a lower workload and perceived physical effort than AED retrieval on foot.

Analyzing unmanned aerial vehicle (drone) attacks; a disaster medicine perspective.

INTRODUCTION: Unmanned aerial vehicles (UAVs), more commonly known as drones, have rapidly become more diverse in capabilities and utilization through technology advancements and affordability. While drones have had significant positive impact on healthcare and consumer delivery particularly in remote and austere environments, Violent Non-State Actors (VNSAs) have increasingly used drones as weapons in planning and executing terrorist attacks resulting in significant morbidity and mortality. We aim to analyze drone-related attacks globally against civilians and critical infrastructure for more effective hospital and prehospital care preparedness. METHODS: We retrospectively reviewed the Global Terrorism Database (GTD) from 1970 to 2020 to analyze the worldwide prevalence of drone-related attacks against civilians and critical infrastructure. Cases were excluded if they had insufficient information regarding a drone involvement, and if attacks were conducted by the government entities. The trends in the number of attacks per month, as well as the number of fatalities and injuries, were examined using time series and trend analysis. RESULTS: The database search yielded 253 drone-related incidents, 173 of which met inclusion criteria. These incidents resulted in 92 fatalities and 215 injuries with civilian targets most commonly attacked by drones (76 events, 43.9%), followed by military (46 events, 26.5-%). The Middle East region was most affected (168 events, 97% of attacks) and the Islamic state of Iraq was the most common perpetrator (106 events, 61.2%). Almost all attacks were by explosive devices attached to drones (172 events, 99.4%). Time series with linear trend analyses suggested an upward trends of drone attacks by VNSAs, resulting in a greater number of injuries and fatalities, that became more frequent over the years. CONCLUSIONS: Overtime, there were upward trends of drone attacks, with higher lethality and morbidity. There were more injuries compared to fatalities. Most common region affected was the Middle East, and most common type of weapon employed by drone technology was explosive weapon. Investment in medical personnel training, security, and research is crucial for an effective mass-casualty incident response after the drone attacks.

Securing Your Airspace: Detection of Drones Trespassing Protected Areas.

Unmanned Aerial Vehicle (UAV) deployment has risen rapidly in recent years. They are now used in a wide range of applications, from critical safety-of-life scenarios like nuclear power plant surveillance to entertainment and hobby applications. While the popularity of drones has grown lately, the associated intentional and unintentional security threats require adequate consideration. Thus, there is an urgent need for real-time accurate detection and classification of drones. This article provides an overview of drone detection approaches, highlighting their benefits and limitations. We analyze detection techniques that employ radars, acoustic and optical sensors, and emitted radio frequency (RF) signals. We compare their performance, accuracy, and cost under different operating conditions. We conclude that multi-sensor detection systems offer more compelling results, but further research is required.

A Novel Approach to Detect Drones Using Deep Convolutional Neural Network Architecture.

Over the past decades, drones have become more attainable by the public due to their widespread availability at affordable prices. Nevertheless, this situation sparks serious concerns in both the cyber and physical security domains, as drones can be employed for malicious activities with public safety threats. However, detecting drones instantly and efficiently is a very difficult task due to their tiny size and swift flights. This paper presents a novel drone detection method using deep convolutional learning and deep transfer learning. The proposed algorithm employs a new feature extraction network, which is added to the modified YOU ONLY LOOK ONCE version2 (YOLOv2) network. The feature extraction model uses bypass connections to learn features from the training sets and solves the "vanishing gradient" problem caused by the increasing depth of the network. The structure of YOLOv2 is modified by replacing the rectified linear unit (relu) with a leaky-relu activation function and adding an extra convolutional layer with a stride of 2 to improve the small object detection accuracy. Using leaky-relu solves the "dying relu" problem. The additional convolution layer with a stride of 2 reduces the spatial dimensions of the feature maps and helps the network to focus on larger contextual information while still preserving the ability to detect small objects. The model is trained with a custom dataset that contains various types of drones, airplanes, birds, and helicopters under various weather conditions. The proposed model demonstrates a notable performance, achieving an accuracy of 77% on the test images with only 5 million learnable parameters in contrast to the Darknet53 + YOLOv3 model, which exhibits a 54% accuracy on the same test set despite employing 62 million learnable parameters.

SiaN-VO: Siamese Network for Visual Odometry.

Despite the significant advancements in drone sensory device reliability, data integrity from these devices remains critical in securing successful flight plans. A notable issue is the vulnerability of GNSS to jamming attacks or signal loss from satellites, potentially leading to incomplete drone flight plans. To address this, we introduce SiaN-VO, a Siamese neural network designed for visual odometry prediction in such challenging scenarios. Our preliminary studies have shown promising results, particularly for flights under static conditions (constant speed and altitude); while these findings are encouraging, they do not fully represent the complexities of real-world flight conditions. Therefore, in this paper, we have furthered our research to enhance SiaN-VO, improving data integration from multiple sensors and enabling more accurate displacement predictions in dynamic flight conditions, thereby marking a significant step forward in drone navigation technology.

Developing a Flying Explorer for Autonomous Digital Modelling in Wild Unknowns.

Digital modelling stands as a pivotal step in the realm of Digital Twinning. The future trend of Digital Twinning involves automated exploration and environmental modelling in complex scenes. In our study, we propose an innovative solution for robot odometry, path planning, and exploration in unknown outdoor environments, with a focus on Digital modelling. The approach uses a minimum cost formulation with pseudo-randomly generated objectives, integrating multi-path planning and evaluation, with emphasis on full coverage of unknown maps based on feasible boundaries of interest. The approach allows for dynamic changes to expected targets and behaviours. The evaluation is conducted on a robotic platform with a lightweight 3D LiDAR sensor model. The robustness of different types of odometry is compared, and the impact of parameters on motion planning is explored. The consistency and efficiency of exploring completely unknown areas are assessed in both indoor and outdoor scenarios. The experiment shows that the method proposed in this article can complete autonomous exploration and environmental modelling tasks in complex indoor and outdoor scenes. Finally, the study concludes by summarizing the reasons for exploration failures and outlining future focuses in this domain.

Classification of Unmanned Aerial Vehicles Based on Acoustic Signals Obtained in External Environmental Conditions.

Detection of unmanned aerial vehicles (UAVs) and their classification on the basis of acoustic signals recorded in the presence of UAVs is a very important source of information. Such information can be the basis of certain decisions. It can support the autonomy of drones and their decision-making system, enabling them to cooperate in a swarm. The aim of this study was to classify acoustic signals recorded in the presence of 17 drones while they hovered individually at a height of 8 m above the recording equipment. The signals were obtained for the drones one at a time in external environmental conditions. Mel-frequency cepstral coefficients (MFCCs) were evaluated from the recorded signals. A discriminant analysis was performed based on 12 MFCCs. The grouping factor was the drone model. The result of the classification is a score of 98.8%. This means that on the basis of acoustic signals recorded in the presence of a drone, it is possible not only to detect the object but also to classify its model.

Energy-Saving Multi-Agent Deep Reinforcement Learning Algorithm for Drone Routing Problem.

With the rapid advancement of drone technology, the efficient distribution of drones has garnered significant attention. Central to this discourse is the energy consumption of drones, a critical metric for assessing energy-efficient distribution strategies. Accordingly, this study delves into the energy consumption factors affecting drone distribution. A primary challenge in drone distribution lies in devising optimal, energy-efficient routes for drones. However, traditional routing algorithms, predominantly heuristic-based, exhibit certain limitations. These algorithms often rely on heuristic rules and expert knowledge, which can constrain their ability to escape local optima. Motivated by these shortcomings, we propose a novel multi-agent deep reinforcement learning algorithm that integrates a drone energy consumption model, namely EMADRL. The EMADRL algorithm first formulates the drone routing problem within a multi-agent reinforcement learning framework. It subsequently designs a strategy network model comprising multiple agent networks, tailored to address the node adjacency and masking complexities typical of multi-depot vehicle routing problem. Training utilizes strategy gradient algorithms and attention mechanisms. Furthermore, local and sampling search strategies are introduced to enhance solution quality. Extensive experimentation demonstrates that EMADRL consistently achieves high-quality solutions swiftly. A comparative analysis against contemporary algorithms reveals EMADRL's superior energy efficiency, with average energy savings of 5.96% and maximum savings reaching 12.45%. Thus, this approach offers a promising new avenue for optimizing energy consumption in last-mile distribution scenarios.

Flight Attitude Estimation with Radar for Remote Sensing Applications.

Unmanned aerial vehicles (UAVs) and radar technology have benefitted from breakthroughs in recent decades. Both technologies have found applications independently of each other, but together, they also unlock new possibilities, especially for remote sensing applications. One of the key factors for a remote sensing system is the estimation of the flight attitude. Despite the advancements, accurate attitude estimation remains a significant challenge, particularly due to the limitations of a conventional Inertial Measurement Unit (IMU). Because these sensors may suffer from issues such as drifting, additional effort is required to obtain a stable attitude. Against that background, this study introduces a novel methodology for making an attitude estimation using radar data. Herein, we present a drone measurement system and detail its calculation process. We also demonstrate our results using three flight scenarios and outline the limitations of the approach. The results show that the roll and pitch angles can be calculated using the radar data, and we conclude that the findings of this research will help to improve the flight attitude estimation of remote sensing flights with a radar sensor.

Adaptive Sensing Data Augmentation for Drones Using Attention-Based GAN.

Drones have become essential tools across various industries due to their ability to provide real-time data and perform automated tasks. However, integrating multiple sensors on a single drone poses challenges such as payload limitations and data management issues. This paper proposes a comprehensive system that leverages advanced deep learning techniques, specifically an attention-based generative adversarial network (GAN), to address data scarcity in drone-collected time-series sensor data. By adjusting sensing frequency based on operational conditions while maintaining data resolution, our system ensures consistent and high-quality data collection. The spatiotemporal The attention mechanism within the GAN enhances the generation of synthetic data, filling gaps caused by reduced sensing frequency with realistic data. This approach improves the efficiency and performance of various applications, such as precision agriculture, environmental monitoring, and surveillance. The experimental results demonstrated the effectiveness of our methodology in extending the operational range and duration of drones and providing reliable augmented data utilizing a variety of evaluation metrics. Furthermore, the superior performance of the proposed system was verified by comparing it with various comparative GAN models.

Performance Evaluation of an Object Detection Model Using Drone Imagery in Urban Areas for Semi-Automatic Artificial Intelligence Dataset Construction.

Modern image processing technologies, such as deep learning techniques, are increasingly used to detect changes in various image media (e.g., CCTV and satellite) and understand their social and scientific significance. Drone-based traffic monitoring involves the detection and classification of moving objects within a city using deep learning-based models, which requires extensive training data. Therefore, the creation of training data consumes a significant portion of the resources required to develop these models, which is a major obstacle in artificial intelligence (AI)-based urban environment management. In this study, a performance evaluation method for semi-moving object detection is proposed using an existing AI-based object detection model, which is used to construct AI training datasets. The tasks to refine the results of AI-model-based object detection are analyzed, and an efficient evaluation method is proposed for the semi-automatic construction of AI training data. Different FBeta scores are tested as metrics for performance evaluation, and it is found that the F2 score could improve the completeness of the dataset with 26.5% less effort compared to the F0.5 score and 7.1% less effort compared to the F1 score. Resource requirements for future AI model development can be reduced, enabling the efficient creation of AI training data.

Development and Testing of a Rocket-Based Sensor for Atmospheric Sensing Using an Unmanned Aerial System.

Measurements of the vertical structure of the lower atmosphere are important to the understanding of air quality. Unmanned Aerial Systems (UASs, drones) can provide low cost, repeatable measurements of the temperature, pressure, and relative humidity. A set of inexpensive sensors controlled with an Arduino microprocessor board were tested on a UAS against a meteorology grade sensor. Two modes of operation for sampling were tested: a forward moving sampler and a vertical ascent sampler. A small particle sensor (Sensiron SPS30) was integrated and was capable of retrieving vertical aerosol distributions during an inversion event. The thermocouple-based temperature probe and the relative humidity measurement on the Bosch BME280 sensor correlated well with the meteorological sensor. The temperature and relative humidity sensors were then deployed on a rocket sounding platform. The rocket sounding system performed well up to a height of 400 m. The inexpensive sensors were found to perform adequately for low-cost development and uses in education and research.

YOLO-IHD: Improved Real-Time Human Detection System for Indoor Drones.

In the field of unmanned systems, the combination of artificial intelligence with self-operating functionalities is becoming increasingly important. This study introduces a new method for autonomously detecting humans in indoor environments using unmanned aerial vehicles, utilizing the advanced techniques of a deep learning framework commonly known as "You Only Look Once" (YOLO). The key contribution of this research is the development of a new model (YOLO-IHD), specifically designed for human detection in indoor using drones. This model is created using a unique dataset gathered from aerial vehicle footage in various indoor environments. It significantly improves the accuracy of detecting people in these complex environments. The model achieves a notable advancement in autonomous monitoring and search-and-rescue operations, highlighting its importance for tasks that require precise human detection. The improved performance of the new model is due to its optimized convolutional layers and an attention mechanism that process complex visual data from indoor environments. This results in more dependable operation in critical situations like disaster response and indoor rescue missions. Moreover, when combined with an accelerating processing library, the model shows enhanced real-time detection capabilities and operates effectively in a real-world environment with a custom designed indoor drone. This research lays the groundwork for future enhancements designed to significantly increase the model's accuracy and the reliability of indoor human detection in real-time drone applications.