With the development of artificial intelligence and integrated sensor technologies, unmanned aerial vehicles (UAVs) are more applied in air combats. A bottleneck that constrains capability UAVs against manned is autonomous maneuver decision, which a very challenging problem short-range combat undergoing highly dynamic uncertain maneuvers enemies. In this paper, an decision model proposed for UAV based on reinforcement learning, mainly includes aircraft motion model, one-to-one evaluation...

In order to improve the autonomous ability of unmanned aerial vehicles (UAV) implement air combat mission, many artificial intelligence-based maneuver decision-making studies have been carried out, but these are often aimed at individual in 1v1 scenarios which rarely happen actual combat. Based on research decision, this paper builds a multi-UAV cooperative decision model based multi-agent reinforcement learning. Firstly, bidirectional recurrent neural network (BRNN) is used achieve...

With the rapid development of electronic science and technology, research on wearable devices is constantly updated. Still, current are not yet comprehensive in recognizing analyzing movement patterns special sports. Based this, this article improves for table tennis, a single sport, realizes recognition evaluation tennis players' motor skills based pattern method hierarchical system. First, player's motion information acquisition device was designed collected data were processed. Second,...

To address the shortcomings of single-step decision making in existing deep reinforcement learning based unmanned aerial vehicle (UAV) real-time path planning problem, a UAV algorithm on long short-term memory (RPP-LSTM) network is proposed, which combines characteristics recurrent neural (RNN) and algorithm. LSTM networks are used this as Q-value for Q (DQN) algorithm, makes has some memory. Thanks to network, can use previous environmental information action effectively avoids problem...

With the burgeoning impact of artificial intelligence on traditional UAV industry, pursuit autonomous flight has emerged as a focal point contemporary research. Addressing imperative for advancing critical technologies in flight, this paper delves into realm state recognition and trajectory prediction. Presenting an innovative approach focused improving precision unmanned aerial vehicle (UAV) path forecasting via identification states, study demonstrates its efficacy through implementation...

To solve the real-time complex mission-planning problem for Multiple heterogeneous Unmanned Aerial Vehicles (UAVs) in dynamic environments, this paper addresses a new approach by effectively adapting Consensus-Based Bundle Algorithms (CBBA) under constraints of task timing, limited UAV resources, diverse types tasks, addition and requirements. We introduce generation mechanism, which satisfied timing constraints. The tasks that require cooperation multiple UAVs are simplified into sub-tasks...

UAV trajectory prediction is the core technology for autonomous flight and a prerequisite control navigation. In this paper, path model established by collecting data of actual UAV. Firstly, information collection preprocessing are carried out; Secondly, state recognition based on PCA-SVM to identify five states; Finally, established, neural network recognition. The experimental results show that: 1) accuracy more than 90%. 2) average error traditional 0.422m, maximum circling 0.84m. 3)...

With the impact of artificial intelligence on traditional UAV industry, autonomous flight has become a popular research field at present. Based demand for key technologies flight, this paper studies state recognition. This is based multi-sensor acquisition on-board information, and uses collected information data fusion to complete identification. Firstly, preprocessing are carried out; secondly, trajectory features extracted multidimensional fusion; finally, recognition model PCA-DAGSVM...

The Grey Wolf Optimizer (GWO) algorithm is recognized for its simplicity and ease of implementation, has become a preferred method solving global optimization problems due to adaptability search capabilities. Despite these advantages, existing Unmanned Aerial Vehicle (UAV) path planning algorithms are often hindered by slow convergence rates, susceptibility local optima, limited robustness. To surpass limitations, we enhance the application GWO in UAV improving trajectory evaluation...

In modern Beyond-Visual-Range (BVR) aerial combat, unmanned loyal wingmen are pivotal, yet their autonomous capabilities limited. Our study introduces an advanced control algorithm based on hierarchical reinforcement learning to enhance these for critical missions like target search, positioning, and relay guidance. Structured a dual-layer model, the algorithm's lower layer manages basic aircraft maneuvers optimal flight, while upper processes battlefield dynamics, issuing precise...

With the rapid changes in battlefield situation, requirement of time for UAV groups to deal with complex tasks is getting higher, which puts forward higher requirements dynamic allocation group. However, most existing methods focus on task pre-allocation, and research technology during execution not sufficient. Aiming at high real-time multi-UAV collaborative problem, this paper introduces market auction mechanism design a discrete particle swarm algorithm based quality clustering by hybrid...

In this paper, a LabVIEW-based online monitoring and safety evaluation system for UAVs is designed to address the deficiencies in UAV flight state parameter evaluation. The consists of lower unit recording an upper on ground. collects detects data connects through wireless digital transmission module via serial port. receives carries out situation UAV. adopts multi-sensors collect navigation information real time achieve detection, while LabVIEW design prediction system, enabling during...

In order to achieve the fastest fire-fighting purpose, warehouse autonomous mobile robots need make an overall optimal planning based on principle of shortest time for their traveling path. A<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="M1"><mml:mi>∗</mml:mi></mml:math> algorithm is considered as a very ideal path algorithm, but not necessarily robots. Furthermore, conventional id="M2"><mml:mi>∗</mml:mi></mml:math> affected by search neighborhood restriction and theoretical...

In order to enhance the capability of tracking targets autonomously unmanned aerial vehicle (UAV), partially observable Markov decision process (POMDP) model for UAV path planning is established based on POMDP framework. The elements are analyzed and described. state transfer law in can be described by method interactive multiple (IMM) due diversity target motion law, which used switch accommodate maneuvers, hence improving accuracy. simulation results show that achieve efficient route,...

With the development of artificial intelligence technology and demand air combat, autonomous maneuver decision UAV has become a popular research direction. Scholars at home abroad have researched on combat in depth based various technologies achieved some results, among which decisions reinforcement learning are more efficient. But present, simulation targets used to test effectiveness method relatively fixed, cannot reflect complexity enemy's strategy under real conditions. This paper...

In order to improve the performance of UAV's autonomous maneuvering decision-making, this paper proposes a decision-making method based on situational continuity. The algorithm in designs situation evaluation function with strong guidance, then trains Long Short-Term Memory (LSTM) under framework Deep Q Network (DQN) for air combat decision-making. Considering continuity between adjacent situations, takes multiple consecutive situations as one input neural network. To reflect difference...

In this article, both the fixed-time distributed consensus tracking and average problems for double-integrator-type multiagent systems with bounded input disturbances are studied. First, a new practical robust sliding-mode control method based on time-based generator is proposed. Second, two observers designed to estimate state disagreement between leader followers under undirected directed communication, respectively. Third, observer measure value of multiple reference signals...

Field experiments were carried out at Research Farm, ICAR Sikkim Centre, Tadong during two consecutive Rabi seasons of 2012 and 2013 to determine the effect different microbial inoculants on selected soil biological properties, growth, yield, quality common buckwheat, then identify best inoculant for application local buckwheat production in hilly ecosystem North-East India. The results indicated that seed applied effectively increased plant chlorophyll content (SPAD), yield attributing...

In this brief, the fixed-time practical distributed average tracking (DAT) problem for multiple nonlinear signals is studied, where are with bounded- and Lipschitz-type derivatives respectively. Two DAT algorithms proposed agents local interactions to track of within a fixed convergence time by using time-base generator techniques. Different from existing results, in brief applicable problems both derivatives, which keeps more reality. Also, able provide an explicit estimation upper-bounded...

Anomaly detection plays a critical role in intelligent video surveillance. However, real-world data obtained always contains large numbers of normal data, along with unlabeled data. A promising solution one-class classification and semi-supervised learning may not be satisfactory as they fail to make good use only available. In this paper, we introduced new framework, called Positive Unlabeled learning-based event Detection (PU-AD), exploit the weakly-supervised information. To best our...