Quantification of species composition in grass-clover swards using RGB and multispectral UAV imagery and machine learning.

Introduction: Growing grass-legume mixtures for forage production improves both yield productivity and nutritional quality, while also benefiting the environment by promoting species biodiversity and enhancing soil fertility (through nitrogen fixation). Consequently, assessing legume proportions in grass-legume mixed swards is essential for breeding and cultivation. This study introduces an approach for automated classification and mapping of species in mixed grass-clover swards using object-based image analysis (OBIA). Methods: The OBIA procedure was established for both RGB and ten band multispectral (MS) images capturedby an unmanned aerial vehicle (UAV). The workflow integrated structural (canopy heights) and spectral variables (bands, vegetation indices) along with a machine learning algorithm (Random Forest) to perform image segmentation and classification. Spatial k-fold cross-validation was employed to assess accuracy. Results and discussion: Results demonstrated good performance, achieving an overall accuracy of approximately 70%, for both RGB and MS-based imagery, with grass and clover classes yielding similar F1 scores, exceeding 0.7 values. The effectiveness of the OBIA procedure and classification was examined by analyzing correlations between predicted clover fractions and dry matter yield (DMY) proportions. This quantification revealed a positive and strong relationship, with R2 values exceeding 0.8 for RGB and MS-based classification outcomes. This indicates the potential of estimating (relative) clover coverage, which could assist breeders but also farmers in a precision agriculture context.

Associations between mental health symptoms, trauma, quality of life and coping in adults living in Ukraine: A cross-sectional study a year after the 2022 Russian invasion.

We aimed to assess the mental health of adults living in Ukraine one year after onset of the Russo-Ukrainian war, along with quality of life and coping strategies. Quota sampling was used to collect online survey data from 2364 adults aged 18-79 years living in Ukraine from April 5, 2023 to May 15, 2023. Among adults living in Ukraine, 14.4 % had probable post-traumatic stress disorder (PTSD), another 8.9 % had complex PTSD (CPTSD), 44.2 % had probable depressive disorder, 23.1 % had anxiety disorder and 38.6 % showed significant loneliness. In adjusted models, the number of trauma events experienced during the war showed a dose-response association with PTSD/CPTSD and was associated with depressive disorder and anxiety disorder. Quality of life domains, particularly physical quality of life, were negatively associated with PTSD/CPTSD, depressive disorder, anxiety disorder, and number of trauma events. Maladaptive coping was positively associated with depressive disorder, anxiety disorder, PTSD/CPTSD and loneliness. All quality of life domains were positively associated with using adaptive coping strategies. Mental health disorders are highly prevalent in adults living in Ukraine one year into the war. Policy and services can promote adaptive coping strategies to improve mental health and quality of life for increased resilience during war.

UAV telephotography elucidates floristic variability and beta diversity of island cliffs under grazing interventions.

Cliffs contain one of the least known plant communities, which has been overlooked in biodiversity assessments due to the inherent inaccessibility. Our study adopted the unmanned aerial vehicle (UAV) with the telephoto camera to remotely clarify floristic variability across unreachable cliffs. Studied cliffs comprised 17 coastal and 13 inland cliffs in Gageodo of South Korea, among which 9 and 5 cliffs were grazed by the introduced cliff-dwelling goats. The UAV telephotography showed 154 and 166 plant species from coastal and inland cliffs, respectively. Inland cliffs contained more vascular plant species (P < 0.001), increased proportions of fern and woody species (P < 0.05), and decreased proportion of herbaceous species (P < 0.001) than coastal cliffs. It was also found that coastal and inland cliffs differed in the species composition (P < 0.001) rather than taxonomic beta diversity (P = 0.29). Furthermore, grazed coastal cliffs featured the elevated proportions of alien and annual herb species than ungrazed coastal cliffs (P < 0.05). This suggests that coastal cliffs might not be totally immune to grazing if the introduced herbivores are able to access cliff microhabitats; therefore, such anthropogenic introduction of cliff-dwelling herbivores should be excluded to conserve the native cliff plant communities.

Tracking of invader drone using hybrid unscented Kalman-Continuous Ant Colony Filter (HUK-CACF).

Splendid Unmanned Aerial Vehicle (UAV) applications upshot its enormous use in densely inhabited areas, which is a matter of concern. In such areas, a proper tracking system is required to track an unauthorized/invader drone to ensure safety. With the flexibility of reaching inaccessible places, an Unmanned Aerial Vehicle Mounted Adaptable Radar Antenna Array (UAVMARAA) could be used. In this regard, a Hybrid Unscented Kalman-Continuous Ant Colony Filter (HUK-CACF) is proposed to estimate the position of the invader drone efficiently. Simulation results demonstrate the efficiency and robustness of the proposed filter for tracking system compared to the existing filters in terms of success rate. Further, for various Adaptable Radar Antenna Array (ARAA) patterns such as Uniform Linear Array (ULA), Uniform Rectangular Array (URA), and Uniform Circular Array (UCA), analysis is done for pertaining actual tracking effect for various parameters such as bearing, Doppler shift, ranging, and Radar Cross Section (RCS) by considering wobbling and mutual coupling (MC) effect. The result shows that the proposed filter outperforms in all the scenarios. Among the various ARAA, URA performs better than the other configurations.

The Influence of Temporal Disturbances in EKF Calculations on the Achieved Parameters of Flight Control and Stabilization of UAVs.

This article investigates the causes of occasional flight instability observed in Unmanned Aerial Vehicles (UAVs). The issue manifests as unexpected oscillations that can lead to emergency landings. The analysis focuses on delays in the Extended Kalman Filter (EKF) algorithm used to estimate the drone's attitude, position, and velocity. These delays disrupt the flight stabilization process. The research identifies two potential causes for the delays. First cause is magnetic field distrurbances created by UAV motors and external magnetic fields (e.g., power lines) that can interfere with magnetometer readings, leading to extended EKF calculations. Second cause is EKF fusion step implementation of the PX4-ECL library combining magnetometer data with other sensor measurements, which can become computionally expensive, especially when dealing with inconsistent magnetic field readings. This can significantly increase EKF processing time. The authors propose a solution of moving the magnetic field estimation calculations to a separate, lower-priority thread. This would prevent them from blocking the main EKF loop and causing delays. The implemented monitoring techniques allow for continuous observation of the real-time operating system's behavior. Since addressing the identified issues, no significant problems have been encountered during flights. However, ongoing monitoring is crucial due to the infrequent and unpredictable nature of the disturbances.

Aerial imagery reveals abnormal stingrays, Taeniura lymma (Myliobatiformes: Dasyatidae), in the central Red Sea.

While morphological abnormalities have been widely reported in batomorphs, ontogenetic deformities of the posterior pectoral fin are rare. In this paper, we present two individuals of the bluespotted ribbontail ray, Taeniura lymma (Forsskål, 1775), with symmetrically deformed posterior pectoral fins. Both individuals were observed through aerial imagery on a coastal sandflat in the central Red Sea (22.30° N, 39.09° E). The similarity of this symmetrical deformity in both individuals indicates it likely has a genetic base. However, lacking access to the specimens, the ultimate cause of the abnormality remains uncertain. The incomplete disk closure did not seem to affect survival, as both individuals had reached a disk width of 22 cm, well above the typical birth size of the species. Our observations constitute both the first report of a morphological abnormality in T. lymma and the first record of a batomorph with a symmetrically deformed posterior pectoral fin.

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 10 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 5-51% depending on the parameter. Parameter values in the tumor agreed with previous studies despite differences in methodology. The CV in the tumor varied between 1-47%. 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 minutes per scan and more accurate quantification.

Impact of drone-specific dispatch instructions on the safety and efficacy of drone-delivered emergency medical treatments: A randomized simulation pilot study.

Introduction: Medical drones have potential for improving the response times to out-of-hospital emergencies. However, widespread adoption is hindered by unanswered questions surrounding medical dispatch and bystander safety. This study evaluated the impact of novel drone-specific dispatch instructions (DSDI) on bystanders' ability to interact effectively with a medical drone and provide prompt, safe, and high-quality treatment in a simulated emergency scenario. We hypothesized DSDI would improve bystanders' performance and facilitate safer bystander-drone interactions. Methods: Twenty-four volunteers were randomized to receive either DSDI and standard Medical Priority Dispatch (MPD) instructions or MPD alone in a simulated out-of-hospital cardiac arrest (OHCA) or pediatric anaphylaxis.,3 Participants in the DSDI group received detailed instructions on locating and interacting with the drone and its enclosed medical kit. The simulations were video recorded. Participants completed a semi-structured interview and survey. Results: The addition of DSDI did not lead to statistically significant changes to the overall time to provide care in either the anaphylaxis or OHCA simulations. However, DSDI did have an impact on bystander safety. In the MPD only group, 50% (6/12) of participants ignored the audio and visual safety cues from the drone instead of waiting for it to be declared safe compared to no DSDI participants ignoring these safety cues. Conclusions: All participants successfully provided patient care. However, this study indicates that DSDI may be useful to ensure bystander safety and should be incorporated in the continued development of emergency medical drones.

Characterizing heterogeneous forest structure in ponderosa pine forests via UAS-derived structure from motion.

Increasingly, dry conifer forest restoration has focused on reestablishing horizontal and vertical complexity and ecological functions associated with frequent, low-intensity fires that characterize these systems. However, most forest inventory approaches lack the resolution, extent, or spatial explicitness for describing tree-level spatial aggregation and openings that were characteristic of historical forests. Uncrewed aerial system (UAS) structure from motion (SfM) remote sensing has potential for creating spatially explicit forest inventory data. This study evaluates the accuracy of SfM-estimated tree, clump, and stand structural attributes across 11 ponderosa pine-dominated stands treated with four different silvicultural prescriptions. Specifically, UAS-estimated tree height and diameter-at-breast-height (DBH) and stand-level canopy cover, density, and metrics of individual trees, tree clumps, and canopy openings were compared to forest survey data. Overall, tree detection success was high in all stands (F-scores of 0.64 to 0.89), with average F-scores > 0.81 for all size classes except understory trees (< 5.0 m tall). We observed average height and DBH errors of 0.34 m and - 0.04 cm, respectively. The UAS stand density was overestimated by 53 trees ha-1 (27.9%) on average, with most errors associated with understory trees. Focusing on trees > 5.0 m tall, reduced error to an underestimation of 10 trees ha-1 (5.7%). Mean absolute errors of bole basal area, bole quadratic mean diameter, and canopy cover were 11.4%, 16.6%, and 13.8%, respectively. While no differences were found between stem-mapped and UAS-derived metrics of individual trees, clumps of trees, canopy openings, and inter-clump tree characteristics, the UAS method overestimated crown area in two of the five comparisons. Results indicate that in ponderosa pine forests, UAS can reliably describe large- and small-grained forest structures to effectively inform spatially explicit management objectives.

The effect of bee drone brood on the motility and viability of stallion spermatozoa-an in vitro study.

Bee drone brood is a beehive by-product with high hormonal activity used in natural medicine to treat male infertility. The aim of the study was to assess the effect of drone brood on stallion spermatozoa during a short-term incubation for its potential use in the equine semen extenders. Three different forms of fixed drone brood (frozen (FR), freeze-dried (FD), and dried extract (DE)) were used. Solutions of drone brood were compared in terms of testosterone, protein, total phenolic content, and antioxidant activity. The stallion semen was diluted with prepared drone brood solutions. The computer-assisted semen analysis (CASA) method was employed to evaluate the movement characteristics of the diluted ejaculate. To determine spermatozoa viability, the mitochondrial toxicity test (MTT) and Alamar Blue test were performed. In terms of testosterone content and antioxidant activity, a close likeness between FR and FD was found whereas DE's composition differed notably. FR had a positive effect mainly on progressive motility, but also on sperm distance and speed parameters after 2 and 3 h of incubation. On the contrary, FD and DE acted negatively, depending on increasing dose and time. For the first time, a positive dose-dependent effect of fixed drone brood on spermatozoa survival in vitro was demonstrated.

Harbour porpoises respond to recreational boats by speeding up and moving away from the boat path.

Recreational boats are common in many coastal waters, yet their effects on cetaceans and other sensitive marine species remain poorly understood. To address this knowledge gap, we used drone video footage recorded from a recreational boat to quantify how harbour porpoises (Phocoena phocoena) responded to the boat approaching at different speeds (10 or 20 knots). Furthermore, we used a hydrophone to record boat noise levels at full bandwidth (0.1-150 kHz) and at the 1/3 octave 16 kHz frequency band for both experimental speeds. The experiments were carried out in shallow waters near Funen, Denmark (55.51° N, 10.79° E) between July and September 2022. Porpoises were more likely to move further away from the path of the boat when approached at 10 knots, but not when approached at 20 knots. In contrast, they swam faster when approached at 20 knots, but not when approached at 10 knots. The recorded received sound level did not depend on how fast the boat approached, suggesting that differences in porpoise responses were related to the speed of the approaching boat rather than to sound intensity. In addition, porpoises generally reacted within close proximity (<200 m) to the approaching boat and quickly (<50 s) resumed their natural behaviour once the boat had passed, indicating that the direct impact of small vessels on porpoise behaviour was most likely small. Nevertheless, repeated exposure to noise from small vessels may influence porpoises' activity or energy budget, and cause them to relocate from disturbed areas. The approach used in this study increases our understanding of recreational boats' impact on harbour porpoises and can be used to inform efficient mitigation measures to help focus conservation efforts.

EscaYard: Precision viticulture multimodal dataset of vineyards affected by Esca disease consisting of geotagged smartphone images, phytosanitary status, UAV 3D point clouds and Orthomosaics.

The "EscaYard" dataset comprises multimodal data collected from vineyards to support agricultural research, specifically focusing on vine health and productivity. Data collection involved two primary methods: (1) unmanned aerial vehicle (UAV) for capturing multispectral images and 3D point clouds, and (2) smartphones for detailed ground-level photography. The UAV used was DJI Matrice 210 V2 RTK, equipped with a Micasense Altum sensor, flying at 30 m above ground level to ensure detailed coverage. Ground-level data were collected using smartphones (iPhone X and Xiaomi Poco X3 Pro), which provided high-resolution images of individual plants. These images were geotagged, enabling location mapping, and included data on the phytosanitary status and number of grape clusters per plant. Additionally, the dataset contains RTK GNSS data, offering high-precision location information for each vine, enhancing the dataset's value for spatial analysis. Moreover, the dataset is structured to support various research applications, including agronomy, remote sensing, and machine learning. It is particularly suited for studying disease detection, yield estimation, and vineyard management strategies. The high-resolution and multispectral nature of the data allows for a detailed analysis of vineyard conditions. Potential reuse of the dataset spans multiple disciplines, enabling studies on environmental monitoring, geographic information systems (GIS), and precision agriculture. Its comprehensive nature makes it a valuable resource for developing and testing algorithms for disease classification, yield prediction, and plant phenotyping. For instance, the images of bunches and grape leaves can be used to train object detection algorithms for accurate disease detection and consequent precise spraying. Moreover, yield prediction algorithms can be trained by extracting the phenotypic traits of the grape bunches. The "EscaYard" dataset provides a foundation for advancing research in sustainable farming practices, optimising crop health, and improving productivity through precise agricultural technologies.

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.

Investigating the Impact of Drone Transport on the Stability of Monoclonal Antibodies for Inter-Hospital Transportation.

In the field of healthcare logistics, the reliance on conventional transport methods such as cars for the delivery of monoclonal antibodies (mAbs) is susceptible to challenges posed by traffic and infrastructure, leading to increased and unpredictable transport times. Recognizing the potential role of drones in mitigating these challenges, we aimed to investigate the impact of medical drone transport on the stability of mAbs. Compromised stability could lead to aggregation and immunogenicity, thereby jeopardizing the efficacy and safety of mAbs. We studied the transportation of vials as well as ready-to-administer infusion bags with blinatumomab, tocilizumab, and daratumumab. The methodology involved the measurement of both temperature and mechanical shock during drone transport. Moreover, the analytical techniques High Performance Size-Exclusion Chromatography (HP-SEC), Dynamic Light Scattering (DLS), Light Obscuration (LO), Micro-Flow Imaging (MFI), and Nanoparticle Tracking Analysis (NTA) were employed to comprehensively assess the presence of aggregates and particle formation. The key findings revealed no significant differences between car and drone transport, indicating that the stability of mAbs in both vials and infusion bags was adequately maintained during drone transport. This suggests that medical drones are a viable and reliable means for the inter-hospital transport of mAbs, paving the way for more efficient and predictable logistics in healthcare delivery.

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.

An Improved TransMVSNet Algorithm for Three-Dimensional Reconstruction in the Unmanned Aerial Vehicle Remote Sensing Domain.

It is important to achieve the 3D reconstruction of UAV remote sensing images in deep learning-based multi-view stereo (MVS) vision. The lack of obvious texture features and detailed edges in UAV remote sensing images leads to inaccurate feature point matching or depth estimation. To address this problem, this study improves the TransMVSNet algorithm in the field of 3D reconstruction by optimizing its feature extraction network and costumed body depth prediction network. The improvement is mainly achieved by extracting features with the Asymptotic Pyramidal Network (AFPN) and assigning weights to different levels of features through the ASFF module to increase the importance of key levels and also using the UNet structured network combined with an attention mechanism to predict the depth information, which also extracts the key area information. It aims to improve the performance and accuracy of the TransMVSNet algorithm's 3D reconstruction of UAV remote sensing images. In this work, we have performed comparative experiments and quantitative evaluation with other algorithms on the DTU dataset as well as on a large UAV remote sensing image dataset. After a large number of experimental studies, it is shown that our improved TransMVSNet algorithm has better performance and robustness, providing a valuable reference for research and application in the field of 3D reconstruction of UAV remote sensing images.

UAV-Borne Mapping Algorithms for Low-Altitude and High-Speed Drone Applications.

This article presents an analysis of current state-of-the-art sensors and how these sensors work with several mapping algorithms for UAV (Unmanned Aerial Vehicle) applications, focusing on low-altitude and high-speed scenarios. A new experimental construct is created using highly realistic environments made possible by integrating the AirSim simulator with Google 3D maps models using the Cesium Tiles plugin. Experiments are conducted in this high-realism simulated environment to evaluate the performance of three distinct mapping algorithms: (1) Direct Sparse Odometry (DSO), (2) Stereo DSO (SDSO), and (3) DSO Lite (DSOL). Experimental results evaluate algorithms based on their measured geometric accuracy and computational speed. The results provide valuable insights into the strengths and limitations of each algorithm. Findings quantify compromises in UAV algorithm selection, allowing researchers to find the mapping solution best suited to their application, which often requires a compromise between computational performance and the density and accuracy of geometric map estimates. Results indicate that for UAVs with restrictive computing resources, DSOL is the best option. For systems with payload capacity and modest compute resources, SDSO is the best option. If only one camera is available, DSO is the option to choose for applications that require dense mapping results.

MinION Sequencing of Fungi in Sub-Saharan African Air and a Novel LAMP Assay for Rapid Detection of the Tropical Phytopathogenic Genus Lasiodiplodia.

To date, there have been no DNA-based metabarcoding studies into airborne fungi in tropical Sub-Saharan Africa. In this initial study, 10 air samples were collected onto Vaseline-coated acrylic rods mounted on drones flown at heights of 15-50 meters above ground for 10-15 min at three sites in Ghana. Purified DNA was extracted from air samples, the internal transcribed spacer (ITS) region was amplified using fungal-specific primers, and MinION third-generation amplicon sequencing was undertaken with downstream bioinformatics analyses utilizing GAIA cloud-based software (at genus taxonomic level). Principal coordinate analyses based on Bray-Curtis beta diversity dissimilarity values found no clear evidence for the structuring of fungal air communities, nor were there significant differences in alpha diversity, based on geographic location (east vs. central Ghana), underlying vegetation type (cocoa vs. non-cocoa), or height above ground level (15-23 m vs. 25-50 m), and despite the short flight times (10-15 min), ~90 operational taxonomic units (OTUs) were identified in each sample. In Ghanaian air, fungal assemblages were skewed at the phylum taxonomic level towards the ascomycetes (53.7%) as opposed to basidiomycetes (24.6%); at the class level, the Dothideomycetes were predominant (29.8%) followed by the Agaricomycetes (21.8%). The most common fungal genus in Ghanaian air was cosmopolitan and globally ubiquitous Cladosporium (9.9% of reads). Interestingly, many fungal genera containing economically important phytopathogens of tropical crops were also identified in Ghanaian air, including Corynespora, Fusarium, and Lasiodiplodia. Consequently, a novel loop-mediated isothermal amplification (LAMP) assay, based on translation elongation factor-1α sequences, was developed and tested for rapid, sensitive, and specific detection of the fungal phytopathogenic genus Lasiodiplodia. Potential applications for improved tropical disease management are considered.

E-DQN-Based Path Planning Method for Drones in Airsim Simulator under Unknown Environment.

To improve the rapidity of path planning for drones in unknown environments, a new bio-inspired path planning method using E-DQN (event-based deep Q-network), referring to introducing event stream to reinforcement learning network, is proposed. Firstly, event data are collected through an airsim simulator for environmental perception, and an auto-encoder is presented to extract data features and generate event weights. Then, event weights are input into DQN (deep Q-network) to choose the action of the next step. Finally, simulation and verification experiments are conducted in a virtual obstacle environment built with an unreal engine and airsim. The experiment results show that the proposed algorithm is adaptable for drones to find the goal in unknown environments and can improve the rapidity of path planning compared with that of commonly used methods.