UAV Navigation Research Articles

Application of Computer Vision Technology in UAV

With the improvement of computer ability and the rapid development of computer vision technology, UAV has been widely used in other industries from the initial military field. At present, the most widely used UAV navigation is the global positioning system (GPS) of the United States, which can be easily interfered by human beings, resulting in certain errors. At present, there are four landing methods for manned aircraft, but no matter which one is, its concealment is poor. At the same time, compared with manned aircraft, UAV is generally smaller in size, so it is unable to follow the domestic microwave radar equipment in China. Through literature review, the origin and development of UAV and computational vision technology are comprehensively reviewed. The application of computer vision technology in agriculture, power, transportation, environmental protection, rescue, entertainment and other fields in recent years is summarized. The relevant survey data are analyzed and presented in the form of chart. Through the data, we can clearly find that the combination of computer vision and UAV still has a lot of room for improvement in the future.

DNCS: New UAV navigation with considering the no-fly zone and efficient selection of the charging station

Due to Drones’ limited battery capacity, quick battery-charging machines (QCMs) are required to ensure long-distance flying. A QCM is chosen for a drone using the shortest path algorithm considering the waiting time at a QCM, charging time, End-to-End (E2E) flight time and other similar time and distance-based criteria for the drone services. However, these methods lack reckoning the varying costs of charging by multiple service providers. Furthermore, such algorithms commonly do not consider the avoidance of no-drone cross-over zones. It is supposed that drone navigation and charging stations (DNCS) are to clearly identify drone trajectories, so both the overall traveled distance and drones charging costs are minimized simultaneously (without crossing over the no-drone zones). The results of simulation reveal the proposed algorithm’s remarkable cost-effectiveness and its avoidance from the risk of crossing over no-fly zone compared to the current planning methods accounting for only the shortest-path- based drone trajectory.

Compact, UAV-mounted hyperspectral imaging system with automatic geometric distortion rectification.

A highly compact hyperspectral imager with an automatic geometric rectification function is developed in this study, which can be mounted on a UAV for ultra-wide range hyperspectral imaging. For better application, the system can provide visible light image transmission and hyperspectral imaging in the real-time mode. A specific design is proposed to allow the visible light camera and hyperspectral camera to share the same telescope optical path, making the system have a high integration level with a total mass of 1.9 kilograms. Thanks to the sharing-optical-path design, the field of view (FOV), frame rate, and spatial resolution are modified the same between the visible light camera and hyperspectral camera. As a result, the geometric rectification is easily performed, and repeated rectifications are eliminated to improve the imaging efficiency. A FOV of 40 degrees in the frame direction and 26 degrees in the flight direction are realized with a focal length of 13mm, providing a large spectral range from 400 nm to 1000 nm and an excellent spectral resolution of 2.5 nm. An automatic geometric rectification workflow is presented and verified in experiments, which can improve the geometric rectification of hyperspectral images in the presence of low-quality UAV navigation data through the incorporation of frame images. Experimental results show that the relative accuracy of geometric rectification is less than 2 pixels when applying the algorithm to our system dataset.

Pose Measurement for Unmanned Aerial Vehicle Based on Rigid Skeleton

Pose measurement is a necessary technology for UAV navigation. Accurate pose measurement is the most important guarantee for a UAV stable flight. UAV pose measurement methods mostly use image matching with aircraft models or 2D points corresponding with 3D points. These methods will lead to pose measurement errors due to inaccurate contour and key feature point extraction. In order to solve these problems, a pose measurement method based on the structural characteristics of aircraft rigid skeleton is proposed in this paper. The depth information is introduced to guide and label the 2D feature points to eliminate the feature mismatch and segment the region. The space points obtained from the marked feature points fit the space linear equation of the rigid skeleton, and the UAV attitude is calculated by combining with the geometric model. This method does not need cooperative identification of the aircraft model, and can stably measure the position and attitude of short-range UAV in various environments. The effectiveness and reliability of the proposed method are verified by experiments on a visual simulation platform. The method proposed can prevent aircraft collision and ensure the safety of UAV navigation in autonomous refueling or formation flight.

Multilayer Mapping Kit for Autonomous UAV Navigation

Mapping, as the back-end of perception and the front-end of path planning in the modern UAV navigation system, draws our interest. Considering the requirements of UAV navigation and the features of the current embedded computation platforms, we designed and implemented a novel multilayer mapping framework. In this framework, we divided the map into three layers: awareness, local, and global. The awareness map is constructed in cylindrical coordinate, enabling fast raycasting. The local map is a probability-based volumetric map. The global map adopts dynamic memory management, allocating memory for the active mapping area, and recycling the memory from the inactive mapping area. We implemented this mapping framework in three parallel threads: awareness thread, local-global thread, and visualization thread. Finally, we evaluated the mapping kit in both the simulation environment and the real-world scenario with the vision-based sensors. The framework supports different kinds of map outputs for the global or local path planners. The implementation is open-source for the research community.

ІНФОРМАЦІЙНА ТЕХНОЛОГІЯ НАВІГАЦІЇ ТА УПРАВЛІННЯ ПОЛЬОТОМ БПЛА НА БАЗІ СПЕКТРАЛЬНИХ ПОРТРЕТІВ МІСЦЕВОСТІ

The article proposes information technology for navigation and flight control of UAVs based on spectral portraits of the area. The urgency of the work is caused by the appearance in wide access of robotic unmanned aerial vehicles, which creates fundamentally new challenges for the failure of navigation equipment. This makes it possible for a terrorist group to organize even an effective cyber attack on the navigation and control system of several UAVs. In addition, existing solutions for UAV flight control by a radio operator can be relatively easily blocked by electronic warfare (EW) means by arranging electromagnetic interference for the control channel. A possible solution to protect terrorist groups from EW is to use the proposed information navigation technology based on spatial-spectral portraits of the area, when the UAV focuses on pre-loaded libraries of 95 landmarks in the area. In the case of using a small number of reference points or landmarks, they can be falsified or destroyed, but when using a large number of objects, taking into account not only their geometry but also spectral portrait, their falsification in the event of failure of navigation equipment and use of EW by terrorist groups. complicated. It has been experimentally established that the influence of the state of illumination on the spectral parameters of objects should be taken into account when using the most common and cheapest spectral sensors in the optical range. Based on these considerations, the aim of the research is to develop a methodological approach to the correction of lighting effects to determine in the optical range of spectral portraits of objects. The research was carried out in the laboratory and directly on the experimental field hospital, as a result of which the dependences of the intensity of color components on the value of LightValue were established on the example of a gray pattern (saturation 69%). It has been experimentally confirmed that when the lighting changes, the calibration of spectral data based on service data from the camera can be performed for different camera models. For the practical needs of spatial-spectral navigation, the correction when the lighting changes is expedient to carry out using experimentally obtained dependences for a specific model of sensor equipment. When organizing the navigation system of the UAV on the basis of spatial-spectral portraits, it is advisable to choose as objects of orientation those that have the most stable spectral indicators.

Memory-Based Deep Reinforcement Learning for Obstacle Avoidance in UAV With Limited Environment Knowledge

This paper presents our method for enabling a UAV quadrotor, equipped with a monocular camera, to autonomously avoid collisions with obstacles in unstructured and unknown indoor environments. When compared to obstacle avoidance in ground vehicular robots, UAV navigation brings in additional challenges because the UAV motion is no more constrained to a well-defined indoor ground or street environment. Unlike ground vehicular robots, a UAV has to navigate across more types of obstacles - for e.g., objects like decorative items, furnishings, ceiling fans, sign-boards, tree branches, etc., are also potential obstacles for a UAV. Thus, methods of obstacle avoidance developed for ground robots are clearly inadequate for UAV navigation. Current control methods using monocular images for UAV obstacle avoidance are heavily dependent on environment information. These controllers do not fully retain and utilize the extensively available information about the ambient environment for decision making. We propose a deep reinforcement learning based method for UAV obstacle avoidance (OA) which is capable of doing exactly the same. The crucial idea in our method is the concept of partial observability and how UAVs can retain relevant information about the environment structure to make better future navigation decisions. Our OA technique uses recurrent neural networks with temporal attention and provides better results compared to prior works in terms of distance covered without collisions. In addition, our technique has a high inference rate and reduces power wastage as it minimizes oscillatory motion of UAV.

Heuristic Path Planning Method for Multistatic UAV-Borne SAR Imaging System

Multistatic unmanned aerial vehicle-borne synthetic aperture radar (MuUAV-SAR) plays an important role in the applications of environmental monitoring and disaster warning because its distributed platforms can provide high-resolution imagery by fusing the multiple measurements. However, the flight paths of the multiple platforms are limited for such an unmanned system since the flight safety and the path length are basic conditions for guaranteeing the effective observation. This paper first studies the observation signal model of MuUAV-SAR imaging system, and then analyzes the factors that determine the imaging resolution, while these factors are all determined by the flight path of UAVs. Secondly, MuUAV-SAR imaging path planning problem is established as a constrained multi-objective optimization problem (CMOP), which considers the navigation and imaging performance of UAV in the process of completing path planning task in detail. For this CMOP, a heuristic search method is proposed to solve it, which can ensure that each step achieves local optimum, and it can also list all feasible solutions to meet the application requirements for selection. Finally, experimental results verify the effectiveness and practicability of the proposed heuristic path planning method.

Homography-based camera pose estimation with known gravity direction for UAV navigation

Homography-based camera pose estimation with known gravity direction for UAV navigation

Robust Data Fusion of UAV Navigation Measurements with Application to the Landing System

To perform precise approach and landing concerning an aircraft in automatic mode, local airfield-based landing systems are used. For joint processing of measurements of the onboard inertial navigation systems (INS), altimeters and local landing systems, the Kalman filter is usually used. The application of the quadratic criterion in the Kalman filter entails the well-known problem of high sensitivity of the estimate to anomalous measurement errors. During the automatic approach phase, abnormal navigation errors can lead to disaster, so the data fusion algorithm must automatically identify and isolate abnormal measurements. This paper presents a recurrent filtering algorithm that is resistant to anomalous errors in measurements and considers its application in the data fusion problem for landing system measurements with onboard sensor measurements—INS and altimeters. The robustness of the estimate is achieved through the combined use of the least modulus method and the Kalman filter. To detect and isolate failures the chi-square criterion is used. It makes possible the customization of the algorithm in accordance with the requirements for false alarm probability and the alarm missing probability. Testing results of the robust filtering algorithm are given both for synthesized data and for real measurements.

Determining the Location of UAVs Automatically Using Aerial or Remotely Sensed High-Resolution Images for Intelligent Navigation of UAVs at the Time of Disconnection with GPS

Currently, UAVs are being used in various industries increasingly. UAV’s navigation under the conditions in which their GPS signals are disturbed is among the main discussions regarding them. In an attempt to remove this problem, several methods have been presented, which yielded good results in restricted conditions. These methods face the restrictions such as lack of moving object in the imaging range and impossibility of using various reference images, in particular satellite images. In this study, the comprehensive system has been developed that is able to use the satellite, aerial, and UAV georeferenced images as the reference image and determine UAV location by combining reference image with the area elevation model instantly. In this regard, a suitable matching algorithm has been used for matching UAV images with satellite and aerial images, which allows for determining UAV location using any relevant reference image type. According to the presence possibility of moving object in the images taken by airplane and UAV, especially in urban areas that led to error in matching process during positioning, an algorithm has been developed in this system for removing such errors. This system has been tested using aerial and satellite reference data from Bushehr port, Iran. It has been successful to locate UAV over the area with a horizontal difference less than 5 m using the coordinates obtained by UAV’s GPS.

Investigation of Alternative Parameters for Immunity-based UAV Navigation in GNSS-denied Environment

This paper is focused on analyzing effects of several significant parameters on the performance of an immunity-inspired methodology for autonomous navigation of unmanned air vehicles when measurements from global navigation satellite systems (GNSS) or similar current sources, including external information of opportunity, are not available. An artificial immune system (AIS) provides corrections to a dead reckoning algorithm for adequate estimates of vehicle position and velocity. Parameter effects are assessed and analyzed through simulation in terms of trajectory tracking errors during autonomous flight.

Research on Key Technologies of Multi-rotor UAV High-precision Positioning and Navigation System

For the continuous high-precision navigation and positioning needs of rotorcraft UAVs based on industry applications, based on low-cost GNSS single-frequency positioning modules and MEMS inertial navigation, the design high-precision GNSS / INS integrated navigation module. At the same time using computer vision algorithms. Aiming at the requirement of high precision, the original data acquisition scheme that fully guarantees the accuracy of MEMS inertial navigation is designed. No one against the rotor for the high-frequency vibration problem of the mobile platform, the effect of IMU white noise parameters on the performance of MEMS inertial navigation and integrated navigation was analysed. The test results show that the IMU white noise parameters can effectively improve the combined navigation performance under high-frequency vibration conditions; the positioning accuracy of the module can reach centimetre level and be continuous and stable.

Erratum to “Opportunistic UAV Navigation With Carrier Phase Measurements From Asynchronous Cellular Signals” [Aug 20 3285-3301

It is noted that the original article noted in the title was incorrectly formatted as a Correspondence item. This work was accepted as a full paper. The author photos and biographies, omitted from the correspondence item, are published here.

UAV Navigation System Autonomous Correction Algorithm Based on Road and River Network Recognition

Abstract—The paper considers an original autonomous correction algorithm for UAV navigation system based on comparison between terrain images obtained by onboard machine vision system and vector topographic map images. Comparison is performed by calculating the homography of vision system images segmented using the convolutional neural network and the vector map images. The presented results of mathematical and flight experiments confirm the algorithm effectiveness for navigation applications.

Cooperative Navigation of UAVs in GNSS-Denied Area With Colored RSSI Measurements

In this article, we investigate how the proposed navigation mode “dead reckoning + relative ranging + heading constraint” can be used to cooperatively localize a UAV swarm when GNSS (Global Navigation Satellite System) is denied. By focusing on the relative ranging performance, the wireless RF modules mounted on the cost-efficient UAVs are used to provide communication, ranging and positioning at the same time; however, its positioning accuracy is usually degraded by the noise in the actual RSSI (Received Signal Strength Indicator) measurements. To solve this problem, we analyze the influence of antenna pattern inhomogeneity and channel variation, respectively. The former mainly determines an antenna radiation function related to the yaw angle and relative position between the two measuring UAVs. The latter uses overlapping Allan variance to analyze and identify the measurement noises from outfield tests, that is, quantization noise, flicker noise, random walk noise and Gaussian white noise, which to some extent bridges the difference between the theoretical model and the practical measurement of RSSI. We also evaluate the sub-swarm size that guarantees robustness of RSS-based localization. In this way, an improved extended Kalman filter is proposed to predict and correct the colored noise by adaptively integrating the current peer-to-peer radio ranging performance and its Allan variance. To prove the effectiveness of this approach, simulation results based on real environment datasets and practical noise modeling are demonstrated.

Design and Development of Multirotor UAV with Automatic Collision Avoidance using Stereo Camera

The operation of drones in cluttered environments like forests and hilly areas is extremely difficult; it is impossible to use drones autonomously without having built-in information to detect and avoid obstacles. The vision based obstacle avoidance algorithm is presented in this paper, with extensions to UAV navigation. The proposed method is incorporated on a stereo vison multi copter using a block matching algorithm. The stereo vision baseline is based on horizontal configuration and computes the depth using a sum of absolute difference algorithm. The image processing node (LabVIEW vi) and the controller node are run on a remote laptop. This vi computes the distance between the multirotor and an obstacle and transmits depth data to an onboard flight controller through the MAVLink protocol. The algorithm efficiency was tested using the software in the loop on Gazebo simulator to analyze the performance of the UAV. The hardware in loop results are also shown in this paper after the successful flight test.

A Service-Based Architecture for Enabling UAV Enhanced Network Services

This article provides an overview of enhanced network services, while emphasizing the role of UAVs as core network equipment with radio and backhaul capabilities. Initially, we elaborate the various deployment options, focusing on UAVs as airborne radio, backhaul and core network equipment, pointing out the benefits and limitations. We then analyze the required enhancements in the SBA to support UAV services including UAV navigation and air traffic management, weather forecasting and UAV connectivity management. The use of airborne UAVs network services is assessed via qualitative means, considering the impact on vehicular applications. Finally, an evaluation has been conducted via a testbed implementation, to explore the performance of UAVs as edge cloud nodes, hosting an ACS function responsible for the control and orchestration of a UAV fleet.

Pipeline Gaussian Particle Filter and Hardware Design for Attitude Estimation with UAV

This paper based on the Gaussian particle filter (GPF) deals with the attitude estimation of UAV. GPF algorithm has better estimation accuracy than the general nonlinear non-Gaussian state estimation and is usually used to improve the system’s real-time performance whose noise is specific such as Gaussian noise during the mini UAV positioning and navigation. The attitude estimation algorithm is implemented on FPGA to verify the effectiveness of the Gaussian particle filter. Simulation results have illustrated that the GPF algorithm is effective and has better real-time performance than that of the particle filter.

Preface

The International Workshop “Navigation and Motion Control” (NMC) is organized by the State Research Centre of the Russian Federation – Concern CSRI Elektropribor, JSC, with participation of ITMO University and St. Petersburg Electrotechnical University. Also, the Workshop would not have been held without support of the Academy of Navigation and Motion Control and its President, Academician of the Russian Academy of Sciences (RAS) Vladimir G. Peshekhonov.The Workshop is aimed at making the participants familiar with new technologies in presentday navigation and motion control systems, in the fields of data processing and systems development. The scope of problems discussed at the Workshop covers the upcoming trends such as inertial sensors based on new physical principles, cyber-physical systems, UAV navigation, and many others. Plenary papers are presented by well-known scientists from the leading institutes of Russia and Europe.List of Logo, International Workshop photograph, program committee, Local organizing committee and The editors are availabel in the pdf.