With the rapid development of UAV technology, research topic remote sensing image segmentation has gradually attracted more and attention. Whether can be accurately segmented is a measure goodness algorithm. In recent years, machine learning methods have been applied to large number fields, deep neural network technology also widely used in field segmentation. This paper introduces specific applications various segmentation, briefly analyzes networks this problem according status several...

Abstract Edge computing has become one of the key enablers for ultra-reliable and low-latency communications in industrial Internet Things fifth generation communication systems is also a promising technology future sixth systems. In this work, we consider application edge to smart factories mission critical task offloading through wireless links. such scenarios, although high end-to-end delays from completion tasks happen with low probability, they may incur severe casualties property loss...

Evaporite salts from saline lakes and playas play active roles in the atmospheric cycles climate system, especially context of changing climate. This study investigates chemical, isotopic, hygroscopic characteristics surface salt samples two lakes, i.e., Mang’ai Dalangtan (MA DLT), in Qaidam Basin. Samples both shared similar ionic compositions, with brines rich Cl-, Mg2+, Na+, lakebed being primarily NaCl-based. Disparities composition between MA DLT crust were...

Evaporite salts from saline lakes and playas play active roles in the atmospheric cycles climate system, especially context of changing climate. This study investigates chemical, isotopic, hygroscopic characteristics surface salt samples two lakes, i.e., Mang'ai Dalangtan (MA DLT), Qaidam Basin. Samples both shared similar ionic compositions, with brines rich Cl–, Mg2+, Na+, lakebed being primarily NaCl-based. Disparities composition between MA DLT crust were observed. Isotopic analyses...

Hybrid particle swarm optimization (PSS) is an intelligent method widely used in machine learning. In this paper, we combine evolutionary game theory and apply them to the of farm planting strategies. Firstly, a linear regression model was established by accurately modeling output, sales volume other relevant indicators agricultural products. Then, using powerful global search ability hybrid solve optimal solution under constraints. Subsequently, Monte Carlo introduced incorporate factors...

Traditional Kalman filter algorithms used for predicting drone positions rely solely on the historical positioning information of itself, which makes it challenging to accurately estimate position over longer time periods. In this paper, an improved IMM-KF (Interacting Multiple Models-Kalman Filter) algorithm is proposed. This incorporates a priori about drone's planned flight route and limited states drone, such as position, velocity, acceleration, achieve real-time prediction information....

Light and small UAVs have attracted widespread attention due to their good adaptability, low cost, high temporal resolution. With the continuous development of technology, multi-UAV cooperation has become a research hot spot. The route planning problem can be decomposed into two sub-problems: task allocation planning. In this paper, method based on reinforcement learning is proposed for cooperation. Considering requirements, capabilities UAV, influence environment conflict task, we construct...

Abstract In the process of wireless network video streaming, especially in more complex scenarios (such as transmission 5G-powered drones), analyzing quality experience (QoE) streaming is a very crucial task. Thus attention should be paid to dynamic interaction between QoE indicators including buffer starvation probability and traffic load. This paper proposes scheduling model based on reinforcement learning. By learning correlation user behavior patterns, series resource allocation...

Abstract Improving the quality of experience (QoE) video streaming is a significant task in wireless network scenario. Buffer starvation transmission process will cause playback freeze, and certain number packets must be prefetched before service restarts. Taking into account shortcomings buffer services, this paper proposes deep learning-based probability calculation model reinforcement packet prefetching model. The learning approach extracts correlation between different timing inputs...

Abstract With regard to video streaming services under wireless networks, how improve the quality of experience (QoE) has always been a challenging task. Especially after arrival 5G era, more attention paid analyze in complex network scenarios (such as 5G-powered drone transmission). Insufficient buffer stream transmission process will cause playback freeze [1]. In order cope with this defect, paper proposes starvation evaluation model based on deep learning and scheduling reinforcement...

Aiming at the needs of civil aviation data sharing and a credible system building, this work designs monitoring based on an encrypted consortium blockchain. This realizes encryption basis blockchain, which has advantages blockchain while preventing leakage sensitive data. The generated in aviation, including flight information, ground passenger services, aircraft maintenance hash value multimedia files, can also be packaged chain real time so that it cannot tampered with tracked. At same...

Abstract The photoelectric pod provides angular information and distance for the UAV (Unmanned Aerial Vehicle), uses it to estimate status of moving target. Since measurement is angle sight relative distance, equation contains some nonlinear functions in Cartesian coordinate system, output frequency low. improved unscented Kalman filter combines function prediction correction, introduces prior target acceleration constraint, offline data obtain steady gain, then estimates state online....

Civilian UAVs have developed rapidly in recent years. However, due to the lack of supervision, cumbersome flight plan application process, and other problems, many people use drones at will, accidents occur as a result. We proposed framework for using blockchain regulate drone behavior implemented demo. With this framework, users don't go through complicated approval process apply plan. Besides, because nature blockchain, system based on our is transparent decentralized, which meets...

Hyperspectral images contain large amounts of information and have high spectral resolution. The performance hyperspectral in describing distinguishing target categories has been greatly improved. With the development unmanned aerial vehicles (UAVs), a lightweight, adaptable low-cost way expanded application field images. This paper proposes spatial spectrum attention mechanism based on Deep Deterministic Policy Gradient (DDPG) for classification. is combined with 3DCNN to assign different...

Evaporite salts from saline lakes and playas play active roles in the atmospheric cycles climate system, especially context of changing climate. Similar processes also occurred on Mars, where large water bodies dried up formed then salt evaporites deposits. In this study, various samples (brines, lakebed salts, crust playa surface a series collected at different depths) were two Martian analogue sites (Mang’ai Dalangtan, MA DLT) Qaidam Basin. The measured for their ionic...

Abstract Edge computing has become one of the key enablers for ultra-reliable and low-latency communications in industrial Internet Things fifth generation communication systems, is also a promising technology future sixth systems. In this work, we consider application edge to smart factories mission-critical task offloading through wireless links. such scenarios, although high end-to-end delays from completion tasks happen with low probability, they may incur severe casualties property...

UAV has become a promising development direction in 5G era because of its flexible deployment and economic efficiency. with communication function can serve many scenes, such as traffic congestion, limited base station conditions, emergency rescue so on. However, airborne energy, throughput energy efficiency are the main bottlenecks an air station. Based on consideration various factors channel, user, speed transmission power, this paper constructs reinforcement learning model for...

In the actual operation scenario, we usually want to measure trajectory of UAV performing task. This can be achieved by many different means. For example, obtain motion state estimation obtaining sensor information such as GPS, RTK and IMU. Alternatively, velocity acceleration measured its displacement calculated integration. The device (including hardware algorithm) complete this is called odometry. Considering efficiency cost, hope devise odometry visually. target visual (VO) estimate...

With the rapid development of drones, unmanned vehicles and robotics industries, VLAM has become a hot technology. In particular, birth 5G-powered UAV promoted emergence more industrial applications, making it most core indispensable role in many scenarios. The loop closure detection can decrease accumulative total error during process VSLAM. Former methods always rely on artificially features, which are not robust, hard to deal with changing complex later deep learning-based considered be...

With the increase of carrier frequency, bandwidth and antenna size communication system, system radar are gradually approaching. Integration sensing technology has become a research hotspot in recent years, can be applied to many fields such as UAV low altitude airspace monitoring traffic management. This paper provides comprehensive overview latest technologies integration communication. For different types schemes, classification improvement details given. refines development application...