Object detection is a basic task on computer vision, recently drone-captured scenarios had wide range of applications in the industry. In this paper, firstly we will introduce some characteristics yolov5s structure and defects tiny-size object detection. Secondly, to solve mentioned above focus improvement based VisDrone dataset. Due UAV dataset, target scale changes greatly at different flight altitude, areas captured by drone show high density, which may bring blurring or occlusion images...

Super-resolution (SR) is an effective and reasonable way to improve the spatial resolution of remote sensing images, which serve as important information carriers for Earth observations. Compared natural more complex distributions detailed ground contained within data place higher demands on feature-representation ability model. Moreover, considering deployment these systems mobile hardware, complexity model also urgent issue. To overcome problems, this study proposes multi-size distillation...

Due to different historical conditions, the social values of Chinese and western countries are very different. Starting from definition distinction collectivism individualism, this paper analyzes reasons for emergence two geographical environment, economic basis cultural philosophy middle West, discusses performance in reality, compares collective standard Western individual standard.

Abstract Reusable Launch Vehicles are crucial for reducing space mission costs and improving efficiency through multiple flights. The unpowered return automatic landing phase pose challenges due to aerodynamic uncertainties, the absence of propulsion, strict constraints. This research integrates Gauss pseudospectral method with a three-degree-of-freedom simulation model plan trajectory. includes Earth’s curvature, ground effects, gear deployment. A Linear Active Disturbance Rejection...

The problem of decreasing stability margins in L1 adaptive control systems is discussed and an out-of-loop scheme based on Lyapunov's theorem proposed. This enhances the effectiveness adaptation, which ensures that system has sufficient to achieve desired performance under parametric uncertainty, additional delays, actuator faults. developed demonstrated through a series simulations. Compared with existing scheme, constant adjustment by proposed allows their range be extended factor 4–5,...

A super-resolution (SR) reconstruction of remote sensing images is becoming a highly active area research. With increasing upscaling factors, richer and more abundant details can progressively be obtained. However, in comparison with natural images, the complex spatial distribution data increases difficulty its reconstruction. Furthermore, most SR methods suffer from low feature information utilization equal processing all regions an image. To improve performance this paper proposes deep...

With the increase in number of Global Navigation Satellite System (GNSS) satellites and their operating frequencies, richer observation data are provided for tightly coupled System/Inertial (GNSS/INS). In this paper, we propose an efficient robust combined navigation scheme to address key issues system accuracy, robustness, computational efficiency. The fuses multi-source such as pseudo-range, pseudo-range rate, dual-antenna observations from GNSS horizontal attitude angle vertical gyro (VG)...

This paper describes the design and implementation of a three-axis acceleration control autopilot for an asymmetric tail-controlled, skid-to-turn tactical missile. In earlier flight test, degraded performance was attributed to multiple disturbances uncertainties presence hidden coupling terms, giving rise miss distance greater than 20 m. To address these issues, missile dynamics are decomposed into angular rate as fast slow subsystem using singular perturbation theory analyze...

This paper describes the design and implementation of a novel three-axis attitude control autopilot scheme for tube-launched, air-deployed UAVs. In early flight tests, various factors, such as model uncertainties during launch, aerodynamic uncertainties, geometric parameter changes deployment, significant in booster rocket installation, exceeded capabilities autopilot, causing instability. order to address these issues, numerical simulation full launch process considering deviations was...

A vertical take-off and landing (VTOL) Unmanned Aerial Vehicle (UAV) can hover, take off, land vertically. The platform structure includes both multi-rotors fixed-wing (FW) conventional control surfaces, therefore named as VTOL-FW. aircraft has the advantages of airplanes. It hover need no runway. In first part, this paper describes mathematical model VTOL-FW in perspective forces moments dynamic with six degrees freedom. second transition scheme is described to illustrate logic. third some...

As a data-driven approach, deep learning requires large amount of annotated data for training to obtain sufficiently accurate and generalized model, especially in the field computer vision. However, when compared with generic object recognition datasets, aerial image datasets are more challenging acquire expensive label. Obtaining high-quality understanding is an urgent problem. Existing studies show that synthetic can effectively reduce required. Therefore, this paper, we propose first...

Huang, B.; He, Wu, L., and Lin, Y., 2020. A deep learning approach to detecting ships from high-resolution aerial remote sensing images. In: Liu, X. Zhao, L. (eds.), Today's Modern Coastal Society: Technical Sociological Aspects of Research. Journal Research, Special Issue No. 111, pp. 16–20. Coconut Creek (Florida), ISSN 0749-0208.Ship activity has become progressive frequent with maritime safety's evolution. The accurate fast localization marine is therefore growing increasingly important....

Vertical takeoff and landing (VTOL) is an essential feature of unmanned aerial vehicles (UAVs). On the one hand, VTOL can expand enhance applications UAVs; yet, on other it makes design control systems for UAVs more complicated. The most challenging demand in designing system to achieve satisfactory response sharpness fixed-wing commands ensure that aircraft mode channels are effectively decoupled. In this work, a six-degree-of-freedom (6-DoF) model with forces moments established based...

The method of collecting aerial images or videos by unmanned vehicles (UAVs) for object search has the advantages high flexibility and low cost, been widely used in various fields, such as pipeline inspection, disaster rescue, forest fire prevention. However, case a wide area, scanning efficiency real-time performance UAV are often difficult to satisfy at same time, which may lead missing best time perform task. In this paper, we design wide-area system based on deep learning problem. first...

In order to improve the control of vertical take-off and landing (VTOL) UAV's airspeed speed, this paper formulates a fixed-wing longitudinal system based on total energy control. Then, we carry out digital simulation in which parameters are selected by trial robustness is verified observing response step demand condition aerodynamic deviation. Next, conduct series flight tests, including stability acceleration ability cruise phase speed ascent phase, as well terrain following. The...

Aircraft surface inspection is critical to the safe flight of aircraft. At present, aircraft still mainly conducted by manual visual inspection. In order improve efficiency efficiency, in this paper, we designed an unmanned aerial vehicle (UAV) intelligent system collect images and used deep learning object detection algorithm independently efficiently solved include:(1) Used whole image wholly according pre-planning information. (2) A high-performance for real-time a high-precision accurate...

Due to the complexity of landing environments, precision guidance and high-precision control technology have become key rope-hook recovery shipborne unmanned aerial vehicles (UAVs). The process was divided into three stages a reasonable strategy had been designed for them, respectively. This study separated issues an outer loop inner loop. (attitude loop) controled UAV follow acceleration commands generated by (trajectory tracking loop). longitudinal controller lateral were based on active...

This paper proposes a new type of design method to improve the performance Unmanned Aerial Vehicle (UAV), which combines twin-rotor tail-sitter Vertical Take-off and Landing (VTOL) UAV system with vector motor frames. Firstly, three-dimensional shape body is established by CATIA software, 3D printer also used process parts Furthermore, this analyzes flight control principles tilting characteristics VTOL UAV, small scale MPC5534 controller chip built as main platform, transplant Mavlink...

The main focus of this article is on the jump-takeoff method for autogyros. On basis a high-confidence autogyro model, we design simulation experiment to study and optimize performance. Using simplified version blade element theory, conduct secondary development YASim dynamics library in FlightGear software construct highly accurate auto-rotation rotor model. implementation requires appropriate control parameters collective angle pre-rotation speed. We explore minimum condition speed obtain...

Landing control of unmanned aerial vehicles (UAVs) is challenging because the strong nonlinear dynamics, multivariable, model uncertainties, wind variations, and sensor noise. Motivated by this fact, paper investigates an automatic landing system (ALS) that includes trajectory generation guidance law for first flight test a turbine-based combined cycle technology demonstrator. Specifically, scheme increases original model’s order to generate reasonable monotone-decreasing throttle reference...

Image super-resolution (SR) technology aims to recover high-resolution images from low-resolution originals, and it is of great significance for the high-quality interpretation remote sensing images. However, most present SR-reconstruction approaches suffer network training difficulties challenge increasing computational complexity with numbers layers. This indicates that these are not suitable application scenarios limited computing resources. Furthermore, complex spatial distributions rich...

In wind tunnel tests, the pose of aircraft model in wire-driven parallel suspension system (WDPSS) is determined by driving several wires. Pose measurement very important for study WDPSS. Using machine vision technology, Monocular Vision Measurement System has been constructed to estimate applying a camera calibration, extracting corresponding control points model, and homogeneous transformations. This article describes programs system, principle data processing methods which based on HALCON...