Multi-rotor Unmanned Aerial Vehicle Research Articles

Experimental investigation on turbulence effects on unsteady aerodynamics performances of two horizontally placed small-size UAV rotors

Wind turbulent disturbance is one of the critical challenges to flying unmanned aerial vehicles (UAVs); especially when the size, weight, and power on board such aircraft are limited. Therefore, to shed light on such turbulent wind effects on UAVs' aerodynamic performances, wind tunnel experiments were conducted on the single- and multi-rotor configurations by varying the turbulence intensity up to 15%, while setting the wind velocity at 5 m/s. The turbulent disturbances were generated inside the wind tunnel using two different passive grids. One is with 29% blockage, and the other is 49%. As the single-rotor experimental tests were conducted, the onset turbulence was found to give rise to a significant change in thrust and the power coefficient in the presence of a lower revolution per minute (rpm). However, at higher rpm, only the wind effect is observed to be the dominant contributor to the aerodynamic performance. Additionally, three different rotor spacings are experimentally examined to measure the propeller wake interactions at the same velocity and different onset turbulence intensities. It was experimentally found that a significant wake interaction occurs between 0.5 D (diameter) and 1.25 D below the propellers, depending on the rotor spacing and the flow conditions. The present work opens up an experimental approach to examine the unsteady aerodynamic performance of multi-rotor UAVs in the presence of wind disturbances, i.e. at different turbulence intensities.

Experimental Characterization of Composite-Printed Materials for the Production of Multirotor UAV Airframe Parts.

In this paper, the characterization of 3D-printed materials that are considered in the design of multirotor unmanned aerial vehicles (UAVs) for specialized purposes was carried out. The multirotor UAV system is briefly described, primarily from the aspect of system dynamics, considering that the airframe parts connect the UAV components, including the propulsion configuration, into a functional assembly. Three additive manufacturing (AM) technologies were discussed, and a brief overview was provided of selective laser sintering (SLS), fused deposition modeling (FDM), and continuous fiber fabrication (CFF). Using hardware and related software, 12 series of specimens were produced, which were experimentally tested utilizing a quasi-static uniaxial tensile test. The results of the experimental tests are provided graphically with stress-strain diagrams. In this work, the focus is on CFF technology and the testing of materials that will be used in the production of mechanically loaded airframe parts of multirotor UAVs. The experimentally obtained values of the maximum stresses were compared for different technologies. For the considered specimens manufactured using FDM and SLS technology, the values are up to 40 MPa, while for the considered CFF materials and range of investigated specimens, it is shown that it can be at least four times higher. By increasing the proportion of fibers, these differences increase. To be able to provide a wider comparison of CFF technology and investigated materials with aluminum alloys, the following three-point flexural and Charpy impact tests were selected that fit within this framework for experimental characterization.

Design and Analysis of a Vertical Electric Ducted Fan for Unmanned Aerial Vehicles

Several types of multirotor unmanned aerial vehicles (UAVs) are widely used in various industrial activities. UAVs with electric ducted fans (EDFs) have been developed to mitigate the risks associated with rotary blades, which are installed on the exterior structure of conventional UAVs and exposed to the ambient environment. Particularly, vertical EDFs have become a focus in the field of UAV design because they require less space for the flow channel compared with horizontal EDFs. In this study, a vertical EDF UAV model was developed. Structural and flow field analysis was performed in Ansys software to determine the optimal design configuration. The results confirmed that the designed model facilitated smooth air flow with favorable inlet and outlet velocity. The vehicle structure required minor adjustment to reduce the frequency of occurrence of vortices. Overall, this study made two research contributions. First, an EDF UAV model was developed to mitigate the risk of rotary blades being exposed to the ambient environment. Second, a vertical EDF system was proposed to reduce the flow channel space and maintain the outlet velocity, thereby improving the takeoff, landing, and thrust performance of UAVs.

Antenna Directional Pattern Testing System Based on Multirotor Unmanned Aerial Vehicles

Obtaining azimuth data of antennas after installation on actual platforms (such as large aircraft, ships, etc.) is crucial for effectively analyzing the platform's impact on the antennas and facilitating optimization and improvement. Traditional testing methods are unable to accurately and efficiently acquire azimuth data due to significant differences between measurements in a dark room and the actual conditions. With the rapid development of drone technology, unmanned aerial platforms have been widely utilized in various engineering fields. Utilizing the flexibility of drones in three-dimensional space and their ability to be precisely controlled through RTK technology, it is possible to use drones equipped with test antennas to measure the azimuth pattern by receiving spectrum signals transmitted by antennas on real platforms. In order to achieve precise angle control, a gimbal system is designed to control the angle of the test antenna. The effectiveness of the entire testing system is verified through experimental results.

The Control Method of Autonomous Flight Avoidance Barriers of UAVs in Confined Environments

This paper proposes an improved 3D-Vector Field Histogram (3D-VFH) algorithm for autonomous flight and local obstacle avoidance of multi-rotor unmanned aerial vehicles (UAVs) in a confined environment. Firstly, the method employs a target point coordinate system based on polar coordinates to convert the point cloud data, considering that long-range point cloud information has no effect on local obstacle avoidance by UAVs. This enables UAVs to effectively utilize obstacle information for obstacle avoidance and improves the real-time performance of the algorithm. Secondly, a sliding window algorithm is used to estimate the optimal flight path of the UAV and implement obstacle avoidance control, thereby maintaining the attitude stability of the UAV during obstacle avoidance flight. Finally, experimental analysis is conducted, and the results show that the UAV has good attitude stability during obstacle avoidance flight, can autonomously follow the expected trajectory, and can avoid dynamic obstacles, achieving precise obstacle avoidance.

Advances and Modern Applications of Frequency-Domain Aircraft and Rotorcraft System Identification

This paper provides an overview of advances and applications of aircraft and rotorcraft flight dynamics system identification at the U.S. Army Combat Capabilities Development Command Aviation & Missile Center over the past 20 years. An overarching topic is how frequency-domain system identification has evolved to aid modeling work of modern aircraft and rotorcraft configurations. The paper starts with a brief overview of the frequency-response method for system identification. Subsequently, the paper highlights and demonstrates several new advances using flight-test data from a conventional helicopter, fixed-wing aircraft, advanced coaxial compound rotorcraft, and a small multirotor unmanned aerial vehicle. The first advancement is the use of identification results to improve simulation model fidelity. The next advancement is the Joint Input-Output Method for identification of aircraft with highly correlated inputs. Next, the paper discusses the advancement of flexible constraints, which provide the ability to directly determine physical parameters from flight-test data. The paper then discusses a methodology to identify lower-order linearized dynamic inflow models from higher-order free wake aerodynamic models. Finally, the paper discusses model stitching, which is the technique of combining or stitching together individual linear models and trim data for discrete flight conditions to produce a continuous, full flight-envelope real-time stitched simulation model.

Low-level control with actuator dynamics for multirotor UAVs

PurposeA low-level controller is critical to the overall performance of multirotor unmanned aerial vehicles. The purpose of this paper is to propose a nonlinear low-level angular velocity controller for multirotor unmanned aerial vehicles in various operating conditions (e.g. different speed and different mode).Design/methodology/approachTo tackle the above challenge, the authors have designed a nonlinear low-level controller taking the actuator dynamics into account. The authors first build the actuator subsystem by combining the actuator dynamics with the angular velocity dynamics model. Then, a recursive low-level controller is developed by designing a high-gain observer to estimate unmeasurable states. Furthermore, a detailed stability analysis is given with the Lyapunov theory.FindingsSimulation tests and real-world flying experiments are provided to validate the proposed approach. In particular, we illustrate the performance of the proposed controller using violent random command test, attitude mode flight and high-speed flight of up to 18.7 m/s in real world. Compared with the classical method used in PX4 autopilot and the estimation-based incremental nonlinear dynamic inversion method, experimental results show that the proposed method can further reduce the control error.Research limitations/implicationsLow-level control of multirotor UAVs is challenging due to the complex dynamic characteristics of UAVs and the diversity of tasks. Although some progress has been made, the performance of existing methods will deteriorate as operating conditions change due to the disregard for the electromechanical characteristics of the actuator.Originality/valueTo solve the low-level angular velocity control problem in various operating conditions of multirotor UAVs, this paper proposes a nonlinear low-level angular velocity controller which takes the actuator dynamics into account.

Assessment of vineyard vigour and yield spatio-temporal variability based on UAV high resolution multispectral images

Accurate, timely assessment of the vineyard on a field scale is essential for successful grape yield and quality. Remote sensing can be an effective and useful monitoring tool, as data from sensors on board Unmanned Aerial Vehicles (UAV) can measure vegetative and reproductive growth and thus directly or indirectly detect variability. Through the images obtained from UAV, the Vegetation Indices (VIs) can be calculated and compared with various agronomic characteristics of the vineyard. The objective of this study was to evaluate the multispectral response of the vineyard in three specific phenological phases and to analyse the spatial distribution of vegetative vigour. A multirotor UAV equipped with a camera featuring multispectral sensors was used. Four VIs namely Normalised Difference Vegetation Index (NDVI), Normalised Difference Red Edge (NDRE), Green Normalised Difference Vegetation Index (GNDVI), Modified Soil Adjusted Vegetation Index (MSAVI), were calculated using the georeferenced orthomosaic UAV images. Computer vision techniques were used to segment these orthoimages to extract only the vegetation canopy pixels. High level of agronomic variability within the vineyard was identified. Pearson's coefficient showed a significant correlation between NDVI and NDRE indices and yield since early phenological stages (r = 0.80 and 0.72 respectively), GNDVI at grape ripening (r = 0.83). Shoot pruning weight (SPW) shows the highest values of correlation (r = 0.84) with NDVI during the phenological stage of berries pea size. Simple linear regression techniques were evaluated using VIs as predictors of the SPW, and accurate predictive results were obtained for NDVI and NDRE with RMSE values of 0.18 and 0.24, respectively. Geostatistical analysis was applied to model the spatial variability of SPW, and thus vineyard vigour. Assessing spatial variability and appreciating the level of vigour enables improved vineyard management by increasing sustainability and production efficiency.

Water Quality Sampling and Multi-Parameter Monitoring System Based on Multi-Rotor UAV Implementation

Water quality sampling and monitoring are fundamental to water environmental protection. The purpose of this study was to develop a water quality sampling and multi-parameter monitoring system mounted on a multi-rotor unmanned aerial vehicle (UAV). The system consisted of the UAV, water sampling and multi-parameter detection device, and path planning algorithm. The water sampling device was composed of a rotating drum, a direct current (DC) reduction motor, water suction hose, high-pressure isolation pump, sampling bottles, and microcontroller. The multi-parameter detection device consisted of sensors for potential of hydrogen (pH), turbidity, total dissolved solids (TDS), and a microcontroller. The flight path of the UAV was optimized using the proposed layered hybrid improved particle swarm optimization (LHIPSO) and rapidly-exploring random trees (RRT) obstacle avoidance path planning algorithm, in order to improve the sampling efficiency. Simulation experiments were conducted that compared the LHIPSO algorithm with the particle swarm optimization (PSO) algorithm and the dynamic adjustment (DAPSO) algorithm. The simulation results showed that the LHIPSO algorithm had improved global optimization capability and stability compared to the other algorithms, validating the effectiveness of the proposed algorithm. Field experiments were conducted at an aquaculture fish farm, and the device achieved real-time monitoring of three water quality parameters (pH, TDS, turbidity) at depths of 1 m and 2 m. A rapid analysis of three parameters (ammonia nitrogen, nitrite, dissolved oxygen) was performed in the laboratory on the collected water samples, and validated the feasibility of this study.

Intelligent Position Controller for Unmanned Aerial Vehicles (UAV) Based on Supervised Deep Learning

In recent years, multi-rotor UAVs have become valuable tools in several productive fields, from entertainment to agriculture and security. However, during their flight trajectory, they sometimes do not accurately perform a specific set of tasks, and the implementation of flight controllers in these vehicles is required to achieve a successful performance. Therefore, this research describes the design of a flight position controller based on Deep Neural Networks and subsequent implementation for a multi-rotor UAV. Five promising Neural Network architectures are developed based on a thorough literature review, incorporating LSTM, 1-D convolutional, pooling, and fully-connected layers. A dataset is then constructed using the performance data of a PID flight controller, encompassing diverse trajectories with transient and steady-state information such as position, speed, acceleration, and motor output signals. The tuning of hyperparameters for each type of architecture is performed by applying the Hyperband algorithm. The best model obtained (LSTMCNN) consists of a combination of LSTM and CNN layers in one dimension. This architecture is compared with the PID flight controller in different scenarios employing evaluation metrics such as rise time, overshoot, steady-state error, and control effort. The findings reveal that our best models demonstrate the successful generalization of flight control tasks. While our best model is able to work with a wider operational range than the PID controller and offers step responses in the Y and X axis with 97% and 98% similarity, respectively, within the PID’s operational range. This outcome opens up possibilities for efficient online training of flight controllers based on Neural Networks, enabling the development of adaptable controllers tailored to specific application domains.

Concept of a Modular Multirotor Heavy Lift Unmanned Aerial Vehicle Platform

This paper presents a novel concept of a modular multirotor aerial robotic platform that can be used for specific profiles of heavy payload missions. A comprehensive mathematical model of the multirotor unmanned aerial vehicle (UAV) is presented, which is divided into the dynamic model and the control allocation scheme that describes the configuration of the aircraft. The components of the propulsion and energy module are selected, and the characterization of the propulsion units is carried out, as well as a preliminary analysis of the module parameters with regard to the considered payloads from 5 to 50 kg. The main goal is to design a modular aircraft that consists of easy-to-assemble modules and that enables the assembly of different propulsion and energy module configurations. Based on the analysis of system parameters, a modular aerial system is designed and the process of manufacturing using rapid prototyping technologies is presented. Based on the parameters obtained from the aircraft assembly CAD model and the implemented mathematical model, simulations are conducted for the two considered missions in the field of smart agriculture. For the purpose of conducting preliminary experiments, a quadrotor aircraft is built and tested in the case of attitude and remote control.

Influence of Operationally Consumed Propellers on Multirotor UAVs Airworthiness: Finite Element and Experimental Approach

Unmanned aerial vehicles (UAVs) are widely used in many applications due to the current technological trend. In these applications, understanding the concept of the operating frequency range is vital to ensure UAV safety. Although most current studies are focused on health monitoring and fault diagnostic methodologies, none of them has included a comprehensive knowledge of the behavior of the natural frequency behavior. This research examines the effectiveness of computer-aided design (CAD) and computer-aided engineering (CAE) in advancing finite element modal analysis. Theoretical results were further evaluated on an experimentally operated drone to understand the interference concept of natural and operational frequencies. Extensive, high-quality, and comprehensive soft-labeled data was acquired in the time-domain from the selected hovering quad-copter equipped with an accelerometer sensor connected to a data acquisition system and a PC platform. Healthy propellers and operationally consumed propellers vibration signal datasets were recorded. The collected datasets were transformed to frequency-domain using Fast Fourier Transform (FFT) spectrum analysis. In summary, the finite element modal analysis provided a thorough behavior understanding of the faulty operating drone status by comparing the natural frequencies to those acquired experimentally based on a detailed discussion. The relationship between natural and operational frequencies is then investigated and presented.

Wind Pressure Orthogonal Decomposition Anemometer: A Wind Measurement Device for Multi-Rotor UAVs

Wind speed and direction are critical meteorological elements. Multi-rotor unmanned aerial vehicles UAVs are widely used as a premium payload platform in meteorological monitoring. The meteorological UAV is able to improve the spatial and temporal resolution of the elements collected. However, during wind measurement missions, the installed anemometers are susceptible to interference caused by rotor turbulence. This paper puts forward a wind pressure orthogonal decomposition (WPOD) strategy to overcome this limitation in three ways: the location of the sensors, a new wind measurement method, and supporting equipment. A weak turbulence zone (WTZ) is found around the airframe, where the turbulence strength decays rapidly and is more suitable for installing wind measurement sensors. For the sensors to match the spatial structure of this area, a WPOD wind measurement method is proposed. An anemometer based on this principle was mounted on a quadrotor UAV to build a wind measurement system. Compared with a standard anemometer, this system has satisfactory performance. Analysis of the resulting data indicates that the error of the system is ±0.3 m/s and ±2° under hovering conditions and ±0.7 m/s and ±5° under moving conditions. In summary, WPOD points to a new orientation for wind measurement under a small spatial–temporal scale.

Identifying Tree Species in a Warm-Temperate Deciduous Forest by Combining Multi-Rotor and Fixed-Wing Unmanned Aerial Vehicles

Fixed-wing unmanned aerial vehicles (UAVs) and multi-rotor UAVs are widely utilized in large-area (>1 km2) environmental monitoring and small-area (<1 km2) fine vegetation surveys, respectively, having different characteristics in terms of flight cost, operational efficiency, and landing and take-off methods. However, large-area fine mapping in complex forest environments is still a challenge in UAV remote sensing. Here, we developed a method that combines a multi-rotor UAV and a fixed-wing UAV to solve this challenge at a low cost. Firstly, we acquired small-scale, multi-season ultra-high-resolution red-green-blue (RGB) images and large-area RGB images by a multi-rotor UAV and a fixed-wing UAV, respectively. Secondly, we combined the reference data of visual interpretation with the multi-rotor UAV images to construct a semantic segmentation model and used the model to expand the reference data. Finally, we classified fixed-wing UAV images using the large-area reference data combined with the semantic segmentation model and discuss the effects of different sizes. Our results show that combining multi-rotor and fixed-wing UAV imagery provides an accurate prediction of tree species. The model for fixed-wing images had an average F1 of 92.93%, with 92.00% for Quercus wutaishanica and 93.86% for Juglans mandshurica. The accuracy of the semantic segmentation model that uses a larger size shows a slight improvement, and the model has a greater impact on the accuracy of Quercus liaotungensis. The new method exploits the complementary characteristics of multi-rotor and fixed-wing UAVs to achieve fine mapping of large areas in complex environments. These results also highlight the potential of exploiting this synergy between multi-rotor UAVs and fixed-wing UAVs.

Design of an Over-Actuated Hexacopter Tilt-Rotor for Landing and Coupling in Power Transmission Lines

It is known that new power transmission line inspection techniques have been developed over the last few years with great potential to improve and, in some cases, even replace traditional inspection procedures such as using helicopters and cars. A series of Unmanned Aerial Vehicles (UAVs) such as fixed-wing or rotary-wing UAVs, and vehicles that climb on the power transmission line, promise to revolutionize the inspection market. In this light, at least 39 new research studies and/or products have been conducted and/or introduced to the market, respectively. However, in an incipient way, some works point to the fusion of some technologies: the development of multi-rotor UAVs and the ability to connect and move over the power transmission line. In line with this, the current work was proposed, with significant unprecedented advances (such as an over-actuated control capacity with tilt rotors, the capability of a displacement in the angle, and the maintenance of active motors on the power transmission line), and the design, modeling, and control of an over-actuated UAV able to move over the conductor cable without the need for a new locomotion system is presented. The aircraft allows for a greater response and the indispensable ability to approximate landing in a power transmission line arbitrary position rather than the catenary lowest point (due to its ability to forward/backward move using the tilting rotors). Its design is detailed, its subsystems are described, and its normal and coupled flight mode dynamics are modeled. The results show good stability and reliable maneuvers for the coupling-to-power-transmission-line flight mode, without any overshoots, and the ability to follow the entire catenary through different Real Control Action (RCA) sets.

Active fault-tolerant control of quadrotor UAVs with nonlinear observer-based sliding mode control validated through hardware in the loop experiments

Multirotor unmanned aerial vehicles (UAV) are highly prone to motor faults, which can arise from defective motors or damaged propellers. Motor faults severely change the multirotor UAV’s dynamics and therefore endanger flight safety and reliability since the controller loses its efficiency. To cope with such crucial problems, a fault-tolerant controller is proposed in this paper for full control of a quadrotor UAV with motor faults. The proposed fault-tolerant approach consists of the nonlinear observer technique and the Sliding Mode Control (SMC). The designed novel nonlinear observer predicts the effects of motor faults on quadrotor dynamics and it is augmented with an SMC to create a fault-tolerant controller. In addition, the nonlinear observer enhances the robustness of the SMC against the uncertainties and disturbances acting on the quadrotor during flight. Any actuator fault will be treated as a disturbance detected by the nonlinear observer and will be attenuated directly by the proposed SMC. The disturbance attenuation capability achieved by the nonlinear observer decreases the amount of control action expected from the SMC, which results in advanced robustness without sacrificing the nominal control performance. The performance of the proposed nonlinear observer SMC (NOSMC) is demonstrated through simulations and testbed experiments. The results show that the proposed fault-tolerant controller effectively recovers the full control of the quadrotor with motor faults up to 40% while tracking the predefined trajectory and rotation angles as desired.

A hybrid thrusting system for increasing the endurance time of multirotor unmanned aerial vehicles

One of the most significant disadvantages of electric multirotor unmanned aerial vehicles is their short flight time compared to fuel-powered unmanned aerial vehicles. This is mainly due to the low energy density of electric batteries. Fuel has much more energy density when compared to batteries. Electric-powered motors in multirotor unmanned aerial vehicles cannot be replaced with fuel-based engines because the stability and control of multirotor unmanned aerial vehicles rely on the high response rates of electric motors. One of the possible solutions to overcome this problem of short endurance times is by using hybrid thrusting systems that combine the advantages of both fuel and electrical propulsion systems, where high maneuverability and long endurance flight time could be achieved. In this work, hybrid thrusting and power systems for multirotor unmanned aerial vehicles are studied. Targeted hybrid thrusting systems consist of combustion engines, electric motors, and their power sources. Then a hybrid thrusting system-based quadrotor unmanned aerial vehicle model is developed. The article presents the altitude and attitude control systems of the developed hybrid thrusting system-based unmanned aerial vehicle. The presented hybrid quadcopter model comprises four electric motors and one fuel engine. The fuel engine used in this work is a 4.07 cc internal combustion engine targeting 2–3 kg unmanned aerial vehicles with up to 5 kg maximum takeoff weight. The developed hybrid quadrotor unmanned aerial vehicle achieved a 139% improvement in flight time when compared with traditional electric-based quadrotor unmanned aerial vehicles. The article also reports on other flight time-related issues such as the optimal fuel mass to battery size ratio to maximize the endurance time of the quadrotor unmanned aerial vehicles.

Application of aeromagnetic survey to mineral exploration of Jinping, Yunnan, China by using multirotor UAV

Jingping, Yunnan, China, is an unfriendly area for ground survey due to the high altitude, rough terrain and dense vegetation. To detect the concealed magnetite ore bodies in this area, the aeromagnetic survey in a 7.8 km2 region was carried out. The field experiments were conducted to investigate the source and level of the noise. Then, the proton magnetometer was placed 3 m under the multirotor unmanned aerial vehicle (UAV) to decrease the static interference. The inverse distance weighting interpolation algorithm with a nonlinear filter was proposed to suppress the noise and dynamic interference caused by the strong wind. The results show that these methods can alleviate the interference of the UAV rotor and strong wind effectively. The patterns of the data from our aeromagnetic survey agree well with the horizontal distribution of the magnetic strata deduced from the geological background. Furthermore, the concealed mafic magmatic rocks and the titanium magnetite inferred from the aeromagnetic survey are confirmed by 4 drill logs in the study area, which supports the validity of the UAV aeromagnetic survey.

Classification of Rice Growth Stage on UAV Image Based on Convolutional Neural Network Method

Currently, the majority of the agricultural sector in Indonesia is carried out by small communities. Half of the Indonesian people (approximately 10 million people) work in the agricultural sector and utilize agricultural land. Some of the tools used by farmers are still using traditional tools, but some are already using modern farming tools. In general, agricultural tools are divided into 3 categories, namely agricultural tools used before the seeds are planted, agricultural tools used when caring for seedlings that are growing and developing, and agricultural tools used when harvesting. One of the technologies used in agriculture is the use of drones or Unmanned Aerial Vehicles (UAV) in the process of sowing fertilizers and seeds and spraying pesticides. The current use of UAVs supports agriculture with manual operation and based on GPS waypoint positioning. In the process, the visual aspects that can be obtained from the UAV have not been considered, so the treatment carried out on agricultural land is the same. The problem of similarity in treatment can lead to similar treatment on heterogeneous agricultural land. Agricultural land should be treated according to the conditions of the land. Because the condition of the land will affect the growth of the planted vegetation. Another problem found in agricultural land is the different rice growth in each paddy field. Rice growth can be seen by farmers through visual aspects but farmers cannot directly see the visual condition of rice growth as a whole because of the large area of land. Utilization of UAV by taking high-resolution aerial imagery can provide visuals of the overall condition of rice from various angles of image capture. The general objective of this research is to classify rice growth on high resolution UAV images based on the Convolutional Neural Network (CNN). The data used in this study were acquired using a multirotor UAV in the same rice field area. The data consists of 500 images consisting of 5 groups. Group 1-2 is the vegetative phase, group 3 is the generative phase and group 4-5 is the ripening phase. CNN is used to conduct training with variations of epochs are 100, 250 and 500. The best accuracy results are obtained in the training epoch 500 with 96% of Accuration

Structural fault diagnosis of UAV based on convolutional neural network and data processing technology

ABSTRACT This study presents a novel method for damage detection and identification in unmanned aerial vehicles (UAVs) using vibration data gathering and processing technologies based on deep learning. To conduct the study, a quad-rotor UAV was manufactured, and a vibration data acquisition system was developed to collect vibration data along three axes under normal and three damage scenarios. Empirical mode decomposition (EMD) was employed to reduce high-frequency noise in the signals, and the root mean square error (RMSE) feature was utilised to select the Y-axis acceleration data, which exhibits significant changes across different damage cases. Finally, a convolutional neural network was used to identify the damage based on the vibration data. Experimental results demonstrate that the proposed method achieved 97.5% accuracy using selected and noise-reduced Y-axis acceleration data, thereby indicating its usefulness in diagnosing damage types in multi-rotor UAVs.