This paper systematically introduces and reviews a scientific exploration of reliability called the belief reliability. Beginning with the origin of reliability engineering, the problems of present theories for reliability engineering are summarized as a query, a dilemma, and a puzzle. Then, through philosophical reflection, we introduce the theoretical solutions given by belief reliability theory, including scientific principles, basic equations, reliability science experiments, and mathematical measures. The basic methods and technologies of belief reliability, namely, belief reliability analysis, function-oriented belief reliability design, belief reliability evaluation, and several newly developed methods and technologies are sequentially elaborated and overviewed. Based on the above investigations, we summarize the significance of belief reliability theory and make some prospects about future research, aiming to promote the development of reliability science and engineering.
Autonomous umanned aerial vehicle (UAV) manipulation is necessary for the defense department to execute tactical missions given by commanders in the future unmanned battlefield. A large amount of research has been devoted to improving the autonomous decision-making ability of UAV in an interactive environment, where finding the optimal maneuvering decision-making policy became one of the key issues for enabling the intelligence of UAV. In this paper, we propose a maneuvering decision-making algorithm for autonomous air-delivery based on deep reinforcement learning under the guidance of expert experience. Specifically, we refine the guidance towards area and guidance towards specific point tasks for the air-delivery process based on the traditional air-to-surface fire control methods. Moreover, we construct the UAV maneuvering decision-making model based on Markov decision processes (MDPs). Specifically, we present a reward shaping method for the guidance towards area and guidance towards specific point tasks using potential-based function and expert-guided advice. The proposed algorithm could accelerate the convergence of the maneuvering decision-making policy and increase the stability of the policy in terms of the output during the later stage of training process. The effectiveness of the proposed maneuvering decision-making policy is illustrated by the curves of training parameters and extensive experimental results for testing the trained policy.
Natural events have had a significant impact on overall flight activity, and the aviation industry plays a vital role in helping society cope with the impact of these events. As one of the most impactful weather typhoon seasons appears and continues, airlines operating in threatened areas and passengers having travel plans during this time period will pay close attention to the development of tropical storms. This paper proposes a deep multimodal fusion and multitasking trajectory prediction model that can improve the reliability of typhoon trajectory prediction and reduce the quantity of flight scheduling cancellation. The deep multimodal fusion module is formed by deep fusion of the feature output by multiple submodal fusion modules, and the multitask generation module uses longitude and latitude as two related tasks for simultaneous prediction. With more dependable data accuracy, problems can be analysed rapidly and more efficiently, enabling better decision-making with a proactive versus reactive posture. When multiple modalities coexist, features can be extracted from them simultaneously to supplement each other’s information. An actual case study, the typhoon Lichma that swept China in 2019, has demonstrated that the algorithm can effectively reduce the number of unnecessary flight cancellations compared to existing flight scheduling and assist the new generation of flight scheduling systems under extreme weather.
In the context of big data, many large-scale knowledge graphs have emerged to effectively organize the explosive growth of web data on the Internet. To select suitable knowledge graphs for use from many knowledge graphs, quality assessment is particularly important. As an important thing of quality assessment, completeness assessment generally refers to the ratio of the current data volume to the total data volume. When evaluating the completeness of a knowledge graph, it is often necessary to refine the completeness dimension by setting different completeness metrics to produce more complete and understandable evaluation results for the knowledge graph. However, lack of awareness of requirements is the most problematic quality issue. In the actual evaluation process, the existing completeness metrics need to consider the actual application. Therefore, to accurately recommend suitable knowledge graphs to many users, it is particularly important to develop relevant measurement metrics and formulate measurement schemes for completeness. In this paper, we will first clarify the concept of completeness, establish each metric of completeness, and finally design a measurement proposal for the completeness of knowledge graphs.
To solve the problem of risk identification and quantitative assessment for human-computer interaction (HCI) in complex avionics systems, an HCI safety analysis framework based on system-theoretical process analysis (STPA) and cognitive reliability and error analysis method (CREAM) is proposed. STPA-CREAM can identify unsafe control actions and find the causal path during the interaction of avionics systems and pilot with the help of formal verification tools automatically. The common performance conditions (CPC) of avionics systems in the aviation environment is established and a quantitative analysis of human failure is carried out. Taking the head-up display (HUD) system interaction process as an example, a case analysis is carried out, the layered safety control structure and formal model of the HUD interaction process are established. For the interactive behavior “Pilots approaching with HUD”, four unsafe control actions and 35 causal scenarios are identified and the impact of common performance conditions at different levels on the pilot decision model are analyzed. The results show that HUD’s HCI level gradually improves as the scores of CPC increase, and the quality of crew member cooperation and time sufficiency of the task is the key to its HCI. Through case analysis, it is shown that STPA-CREAM can quantitatively assess the hazards in HCI and identify the key factors that impact safety.
As the “engine” of equipment continuous operation and repeated operation, equipment maintenance support plays a more prominent role in the confrontation of symmetrical combat systems. As the basis and guide for the planning and implementation of equipment maintenance tasks, the equipment damage measurement is an important guarantee for the effective implementation of maintenance support. Firstly,this article comprehensively analyses the influence factors to damage measurement from the enemy’s attributes, our attributes and the battlefield environment starting from the basic problem of wartime equipment damage measurement. Secondly, this article determines the key factors based on fuzzy comprehensive evaluation (FCE) and performed principal component analysis (PCA) on the key factors. Finally, the principal components representing more than 85% of the data features are taken as the input and the equipment damage quantity is taken as the output. The data are trained and tested by artificial neural network (ANN) and random forest (RF). In a word, FCE-PCA-RF can be used as a reference for the research of equipment damage estimation in wartime.
Based on the wave attack task planning method in static complex environment and the rolling optimization framework, an online task planning method in dynamic complex environment based on rolling optimization is proposed. In the process of online task planning in dynamic complex environment, online task planning is based on event triggering including target information update event, new target addition event, target failure event, weapon failure event, etc., and the methods include defense area reanalysis, parameter space update, and mission re-planning. Simulation is conducted for different events and the result shows that the index value of the attack scenario after re-planning is better than that before re-planning and according to the probability distribution of statistical simulation method, the index value distribution after re-planning is obviously in the region of high index value, and the index value gap before and after re-planning is related to the degree of posture change.
Aerial threat assessment is a crucial link in modern air combat, whose result counts a great deal for commanders to make decisions. With the consideration that the existing threat assessment methods have difficulties in dealing with high dimensional time series target data, a threat assessment method based on self-attention mechanism and gated recurrent unit (SA-GRU) is proposed. Firstly, a threat feature system including air combat situations and capability features is established. Moreover, a data augmentation process based on fractional Fourier transform (FRFT) is applied to extract more valuable information from time series situation features. Furthermore, aiming to capture key characteristics of battlefield evolution, a bidirectional GRU and SA mechanisms are designed for enhanced features. Subsequently, after the concatenation of the processed air combat situation and capability features, the target threat level will be predicted by fully connected neural layers and the softmax classifier. Finally, in order to validate this model, an air combat dataset generated by a combat simulation system is introduced for model training and testing. The comparison experiments show the proposed model has structural rationality and can perform threat assessment faster and more accurately than the other existing models based on deep learning.
To address the shortcomings of single-step decision making in the existing deep reinforcement learning based unmanned aerial vehicle (UAV) real-time path planning problem, a real-time UAV path planning algorithm based on long short-term memory (RPP-LSTM) network is proposed, which combines the memory characteristics of recurrent neural network (RNN) and the deep reinforcement learning algorithm. LSTM networks are used in this algorithm as Q-value networks for the deep Q network (DQN) algorithm, which makes the decision of the Q-value network has some memory. Thanks to LSTM network, the Q-value network can use the previous environmental information and action information which effectively avoids the problem of single-step decision considering only the current environment. Besides, the algorithm proposes a hierarchical reward and punishment function for the specific problem of UAV real-time path planning, so that the UAV can more reasonably perform path planning. Simulation verification shows that compared with the traditional feed-forward neural network (FNN) based UAV autonomous path planning algorithm, the RPP-LSTM proposed in this paper can adapt to more complex environments and has significantly improved robustness and accuracy when performing UAV real-time path planning.
The weapon and equipment operational requirement analysis (WEORA) is a necessary condition to win a future war, among which the acquisition of knowledge about weapons and equipment is a great challenge. The main challenge is that the existing weapons and equipment data fails to carry out structured knowledge representation, and knowledge navigation based on natural language cannot efficiently support the WEORA. To solve above problem, this research proposes a method based on question answering (QA) of weapons and equipment knowledge graph (WEKG) to construct and navigate the knowledge related to weapons and equipment in the WEORA. This method firstly constructs the WEKG, and builds a neutral network-based QA system over the WEKG by means of semantic parsing for knowledge navigation. Finally, the method is evaluated and a chatbot on the QA system is developed for the WEORA. Our proposed method has good performance in the accuracy and efficiency of searching target knowledge, and can well assist the WEORA.
Aviation accidents are currently one of the leading causes of significant injuries and deaths worldwide. This entices researchers to investigate aircraft safety using data analysis approaches based on an advanced machine learning algorithm. To assess aviation safety and identify the causes of incidents, a classification model with light gradient boosting machine (LGBM) based on the aviation safety reporting system (ASRS) has been developed. It is improved by k-fold cross-validation with hybrid sampling model (HSCV), which may boost classification performance and maintain data balance. The results show that employing the LGBM-HSCV model can significantly improve accuracy while alleviating data imbalance. Vertical comparison with other cross-validation (CV) methods and lateral comparison with different fold times comprise the comparative approach. Aside from the comparison, two further CV approaches based on the improved method in this study are discussed: one with a different sampling and folding order, and the other with more CV. According to the assessment indices with different methods, the LGBM-HSCV model proposed here is effective at detecting incident causes. The improved model for imbalanced data categorization proposed may serve as a point of reference for similar data processing, and the model’s accurate identification of civil aviation incident causes can assist to improve civil aviation safety.
How to mine valuable information from massive multi-source heterogeneous data and identify the intention of aerial targets is a major research focus at present. Aiming at the long-term dependence of air target intention recognition, this paper deeply explores the potential attribute features from the spatiotemporal sequence data of the target. First, we build an intelligent dynamic intention recognition framework, including a series of specific processes such as data source, data preprocessing, target space-time, convolutional neural networks-bidirectional gated recurrent unit-atteneion (CBA) model and intention recognition. Then, we analyze and reason the designed CBA model in detail. Finally, through comparison and analysis with other recognition model experiments, our proposed method can effectively improve the accuracy of air target intention recognition, and is of significance to the commanders’ operational command and situation prediction.
The rapid development of unmanned aerial vehicle (UAV) swarm, a new type of aerial threat target, has brought great pressure to the air defense early warning system. At present, most of the track correlation algorithms only use part of the target location, speed, and other information for correlation. In this paper, the artificial neural network method is used to establish the corresponding intelligent track correlation model and method according to the characteristics of swarm targets. Precisely, a route correlation method based on convolutional neural networks (CNN) and long short-term memory (LSTM) Neural network is designed. In this model, the CNN is used to extract the formation characteristics of UAV swarm and the spatial position characteristics of single UAV track in the formation, while the LSTM is used to extract the time characteristics of UAV swarm. Experimental results show that compared with the traditional algorithms, the algorithm based on CNN-LSTM neural network can make full use of multiple feature information of the target, and has better robustness and accuracy for swarm targets.
The subversive nature of information war lies not only in the information itself, but also in the circulation and application of information. It has always been a challenge to quantitatively analyze the function and effect of information flow through command, control, communications, computer, kill, intelligence, surveillance, reconnaissance (C4KISR) system. In this work, we propose a framework of force of information influence and the methods for calculating the force of information influence between C4KISR nodes of sensing, intelligence processing, decision making and fire attack. Specifically, the basic concept of force of information influence between nodes in C4KISR system is formally proposed and its mathematical definition is provided. Then, based on the information entropy theory, the model of force of information influence between C4KISR system nodes is constructed. Finally, the simulation experiments have been performed under an air defense and attack scenario. The experimental results show that, with the proposed force of information influence framework, we can effectively evaluate the contribution of information circulation through different C4KISR system nodes to the corresponding tasks. Our framework of force of information influence can also serve as an effective tool for the design and dynamic reconfiguration of C4KISR system architecture.
Based on the characteristics of high-end products, crowd-sourcing user stories can be seen as an effective means of gathering requirements, involving a large user base and generating a substantial amount of unstructured feedback. The key challenge lies in transforming abstract user needs into specific ones, requiring integration and analysis. Therefore, we propose a topic mining-based approach to categorize, summarize, and rank product requirements from user stories. Specifically, after determining the number of story categories based on pyLDAvis, we initially classify “I want to” phrases within user stories. Subsequently, classic topic models are applied to each category to generate their names, defining each post-classification user story category as a requirement. Furthermore, a weighted ranking function is devised to calculate the importance of each requirement. Finally, we validate the effectiveness and feasibility of the proposed method using 2 966 crowd-sourced user stories related to smart home systems.
With the development of information technology, a large number of product quality data in the entire manufacturing process is accumulated, but it is not explored and used effectively. The traditional product quality prediction models have many disadvantages, such as high complexity and low accuracy. To overcome the above problems, we propose an optimized data equalization method to pre-process dataset and design a simple but effective product quality prediction model: radial basis function model optimized by the firefly algorithm with Levy flight mechanism (RBFFALM). First, the new data equalization method is introduced to pre-process the dataset, which reduces the dimension of the data, removes redundant features, and improves the data distribution. Then the RBFFALFM is used to predict product quality. Comprehensive experiments conducted on real-world product quality datasets validate that the new model RBFFALFM combining with the new data pre-processing method outperforms other previous methods on predicting pro-duct quality.
Nowadays manufacturers are facing fierce challenge. Apart from the products, providing customers with multiple maintenance options in the service contract becomes more popular, since it can help to improve customer satisfaction, and ultimately promote sales and maximize profit for the manufacturer. By considering the combinations of corrective maintenance and preventive maintenance, totally three types of maintenance service contracts are designed. Moreover, attractive incentive and penalty mechanisms are adopted in the contracts. On this basis, Nash non-cooperative game is applied to analyze the revenue for both the manufacturer and customers, and so as to optimize the pricing mechanism of maintenance service contract and achieve a win-win situation. Numerical experiments are conducted. The results show that by taking into account the incentive and penalty mechanisms, the revenue can be improved for both the customers and manufacturer. Moreover, with the increase of repair rate and improvement factor in the preventive maintenance, the revenue will increase gradually for both the parties.
Dominant technology formation is the key for the high-tech industry to “cross the chasm” and gain an established foothold in the market (and hence disrupt the regime). Therefore, a stimulus-response model is proposed to investigate the dominant technology by exploring its formation process and mechanism. Specifically, based on complex adaptive system theory and the basic stimulus-response model, we use a combination of agent-based modeling and system dynamics modeling to capture the interactions between dominant technology and the socio-technical landscape. The results indicate the following: (i) The dynamic interaction is “stimulus-reaction-selection”, which promotes the dominant technology ’s formation. (ii) The dominant technology’s formation can be described as a dynamic process in which the adaptation intensity of technology standards increases continuously until it becomes the leading technology under the dual action of internal and external mechanisms. (iii) The dominant technology’s formation in the high-tech industry is influenced by learning ability, the number of adopting users and adaptability. Therein, a “critical scale” of learning ability exists to promote the formation of leading technology: a large number of adopting users can promote the dominant technology’s formation by influencing the adaptive response of technology standards to the socio-technical landscape and the choice of technology standards by the socio-technical landscape. There is a minimum threshold and a maximum threshold for the role of adaptability in the dominant technology ’s formation. (iv) The socio-technical landscape can promote the leading technology’s shaping in the high-tech industry, and different elements have different effects. This study promotes research on the formation mechanism of dominant technology in the high-tech industry, presents new perspectives and methods for researchers, and provides essential enlightenment for managers to formulate technology strategies.
At present, although knowledge graphs have been widely used in various fields such as recommendation systems, question and answer systems, and intelligent search, there are always quality problems such as knowledge omissions and errors. Quality assessment and control, as an important means to ensure the quality of knowledge, can make the applications based on knowledge graphs more complete and more accurate by reasonably assessing the knowledge graphs and fixing and improving the quality problems at the same time. Therefore, as an indispensable part of the knowledge graph construction process, the results of quality assessment and control determine the usefulness of the knowledge graph. Among them, the assessment and enhancement of completeness, as an important part of the assessment and control phase, determine whether the knowledge graph can fully reflect objective phenomena and reveal potential connections among entities. In this paper, we review specific techniques of completeness assessment and classify completeness assessment techniques in terms of closed world assumptions, open world assumptions, and partial completeness assumptions. The purpose of this paper is to further promote the development of knowledge graph quality control and to lay the foundation for subsequent research on the completeness assessment of knowledge graphs by reviewing and classifying completeness assessment techniques.
The rapid growth of mobile applications, the popularity of the Android system and its openness have attracted many hackers and even criminals, who are creating lots of Android malware. However, the current methods of Android malware detection need a lot of time in the feature engineering phase. Furthermore, these models have the defects of low detection rate, high complexity, and poor practicability, etc. We analyze the Android malware samples, and the distribution of malware and benign software in application programming interface (API) calls, permissions, and other attributes. We classify the software’s threat levels based on the correlation of features. Then, we propose deep neural networks and convolutional neural networks with ensemble learning (DCEL), a new classifier fusion model for Android malware detection. First, DCEL preprocesses the malware data to remove redundant data, and converts the one-dimensional data into a two-dimensional gray image. Then, the ensemble learning approach is used to combine the deep neural network with the convolutional neural network, and the final classification results are obtained by voting on the prediction of each single classifier. Experiments based on the Drebin and Malgenome datasets show that compared with current state-of-art models, the proposed DCEL has a higher detection rate, higher recall rate, and lower computational cost.
Aiming at the triangular fuzzy (TF) multi-attribute decision making (MADM) problem with a preference for the distribution density of attribute (DDA), a decision making method with TF number two-dimensional density (TFTD) operator is proposed based on the density operator theory for the decision maker (DM). Firstly, a simple TF vector clustering method is proposed, which considers the feature of TF number and the geometric distance of vectors. Secondly, the least deviation sum of squares method is used in the program model to obtain the density weight vector. Then, two TFTD operators are defined, and the MADM method based on the TFTD operator is proposed. Finally, a numerical example is given to illustrate the superiority of this method, which can not only solve the TF MADM problem with a preference for the DDA but also help the DM make an overall comparison.
With the new development trend of multi-resource coordinated Earth observation and the new goal of Earth observation application of “short response time, high observation accuracy, and wide coverage”, space-aeronautics cooperative complex task planning problem has become an urgent problem to be solved. The focus of this problem is to use multiple resources to perform collaborative observations on complex tasks. By analyzing the process from task assignment to receiving task observation results, we propose a multi-layer interactive task planning framework which is composed of a preprocessing method for complex tasks, a task allocation layer, a task planning layer, and a task coordination layer. According to the characteristics of the framework, a hybrid genetic parallel tabu (HGPT) algorithm is proposed on this basis. The algorithm uses genetic annealing algorithm (GAA), parallel tabu (PT) algorithm, and heuristic rules to achieve task allocation, task planning, and task coordination. At the same time, coding improvements, operator design, annealing operations, and parallel calculations are added to the algorithm. In order to verify the effectiveness of the algorithm, simulation experiments under complex task scenarios of different scales are carried out. Experimental results show that this method can effectively solve the problems of observing complex tasks. Meanwhile, the optimization effect and convergence speed of the HGPT is better than that of the related algorithms.
This paper studies a special defense game using unmanned aerial vehicle (UAV) swarm against a fast intruder. The fast intruder applies an offensive strategy based on the artificial potential field method and Apollonius circle to scout a certain destination. As defenders, the UAVs are arranged into three layers: the forward layer, the midfield layer and the back layer. The co-defense mechanism, including the role derivation method of UAV swarm and a guidance law based on the co-defense front point, is introduced for UAV swarm to co-detect the intruder. Besides, five formations are designed for comparative analysis when ten UAVs are applied. Through Monte Carlo experiments and ablation experiment, the effectiveness of the proposed co-defense method has been verified.
Due to people’s increasing dependence on social networks, it is essential to develop a consensus model considering not only their own factors but also the interaction between people. Both external trust relationship among experts and the internal reliability of experts are important factors in decision-making. This paper focuses on improving the scientificity and effectiveness of decision-making and presents a consensus model combining trust relationship among experts and expert reliability in social network group decision-making (SN-GDM). A concept named matching degree is proposed to measure expert reliability. Meanwhile, linguistic information is applied to manage the imprecise and vague information. Matching degree is expressed by a 2-tuple linguistic model, and experts’ preferences are measured by a probabilistic linguistic term set (PLTS). Subsequently, a hybrid weight is explored to weigh experts ’ importance in a group. Then a consensus measure is introduced and a feedback mechanism is developed to produce some personalized recommendations with higher group consensus. Finally, a comparative example is provided to prove the scientificity and effectiveness of the proposed consensus model.
In the field of model-based system assessment, mathematical models are used to interpret the system behaviors. However, the industrial systems in this intelligent era will be more manageable. Various management operations will be dynamically set, and the system will be no longer static as it is initially designed. Thus, the static model generated by the traditional model-based safety assessment (MBSA) approach cannot be used to accurately assess the dependability. There mainly exists three problems. Complex: huge and complex behaviors make the modeling to be trivial manual; Dynamic: though there are thousands of states and transitions, the previous model must be resubmitted to assess whenever new management arrives; Unreusable: as for different systems, the model must be resubmitted by reconsidering both the management and the system itself at the same time though the management is the same. Motivated by solving the above problems, this research studies a formal management specifying approach with the advantages of agility modeling, dynamic modeling, and specification design that can be re-suable. Finally, three typical managements are specified in a series-parallel system as a demonstration to show the potential.
Some attributes are uncertain for evaluation work because of incomplete or limited information and knowledge. It leads to uncertainty in evaluation results. To that end, an evaluation method, uncertainty entropy-based exploratory evaluation (UEEE), is proposed to guide the evaluation activities, which can iteratively and gradually reduce uncertainty in evaluation results. Uncertainty entropy (UE) is proposed to measure the extent of uncertainty. First, the belief degree distributions are assumed to characterize the uncertainty in attributes. Then the belief degree distribution of the evaluation result can be calculated by using uncertainty theory. The obtained result is then checked based on UE to see if it could meet the requirements of decision-making. If its uncertainty level is high, more information needs to be introduced to reduce uncertainty. An algorithm based on the UE is proposed to find which attribute can mostly affect the uncertainty in results. Thus, efforts can be invested in key attribute(s), and the evaluation results can be updated accordingly. This update should be repeated until the evaluation result meets the requirements. Finally, as a case study, the effectiveness of ballistic missiles with uncertain attributes is evaluated by UEEE. The evaluation results show that the target is believed to be destroyed.
Aimed at a multiple traveling salesman problem (MTSP) with multiple depots and closed paths, this paper proposes a k-means clustering donkey and a smuggler algorithm (K-DSA). The algorithm first uses the k-means clustering method to divide all cities into several categories based on the center of various samples; the large-scale MTSP is divided into multiple separate traveling salesman problems (TSPs), and the TSP is solved through the DSA. The proposed algorithm adopts a solution strategy of clustering first and then carrying out, which can not only greatly reduce the search space of the algorithm but also make the search space more fully explored so that the optimal solution of the problem can be more quickly obtained. The experimental results from solving several test cases in the TSPLIB database show that compared with other related intelligent algorithms, the K-DSA has good solving performance and computational efficiency in MTSPs of different scales, especially with large-scale MTSP and when the convergence speed is faster; thus, the advantages of this algorithm are more obvious compared to other algorithms.
As one of the most important part of weapon system of systems (WSoS), quantitative evaluation of reconnaissance satellite system (RSS) is indispensable during its construction and application. Aiming at the problem of nonlinear effectiveness evaluation under small sample conditions, we propose an evaluation method based on support vector regression (SVR) to effectively address the defects of traditional methods. Considering the performance of SVR is influenced by the penalty factor, kernel type, and other parameters deeply, the improved grey wolf optimizer (IGWO) is employed for parameter optimization. In the proposed IGWO algorithm, the opposition-based learning strategy is adopted to increase the probability of avoiding the local optima, the mutation operator is used to escape from premature convergence and differential convergence factors are applied to increase the rate of convergence. Numerical experiments of 14 test functions validate the applicability of IGWO algorithm dealing with global optimization. The index system and evaluation method are constructed based on the characteristics of RSS. To validate the proposed IGWO-SVR evaluation method, eight benchmark data sets and combat simulation are employed to estimate the evaluation accuracy, convergence performance and computational complexity. According to the experimental results, the proposed method outperforms several prediction based evaluation methods, verifies the superiority and effectiveness in RSS operational effectiveness evaluation.
Aiming at the suppression of enemy air defense (SEAD) task under the complex and complicated combat scenario, the spatiotemporal cooperative path planning methods are studied in this paper. The major research contents include optimal path points generation, path smoothing and cooperative rendezvous. In the path points generation part, the path points availability testing algorithm and the path segments availability testing algorithm are designed, on this foundation, the swarm intelligence-based path point generation algorithm is utilized to generate the optimal path. In the path smoothing part, taking terminal attack angle constraint and maneuverability constraint into consideration, the Dubins curve is introduced to smooth the path segments. In cooperative rendezvous part, we take estimated time of arrival requirement constraint and flight speed range constraint into consideration, the speed control strategy and flight path control strategy are introduced, further, the decoupling scheme of the circling maneuver and detouring maneuver is designed, in this case, the maneuver ways, maneuver point, maneuver times, maneuver path and flight speed are determined. Finally, the simulation experiments are conducted and the acquired results reveal that the time-space cooperation of multiple unmanned aeriel vehicles (UAVs) is effectively realized, in this way, the combat situation suppression against the enemy can be realized in SEAD scenarios.
The deep deterministic policy gradient (DDPG) algorithm is an off-policy method that combines two mainstream reinforcement learning methods based on value iteration and policy iteration. Using the DDPG algorithm, agents can explore and summarize the environment to achieve autonomous decisions in the continuous state space and action space. In this paper, a cooperative defense with DDPG via swarms of unmanned aerial vehicle (UAV) is developed and validated, which has shown promising practical value in the effect of defending. We solve the sparse rewards problem of reinforcement learning pair in a long-term task by building the reward function of UAV swarms and optimizing the learning process of artificial neural network based on the DDPG algorithm to reduce the vibration in the learning process. The experimental results show that the DDPG algorithm can guide the UAVs swarm to perform the defense task efficiently, meeting the requirements of a UAV swarm for non-centralization, autonomy, and promoting the intelligent development of UAVs swarm as well as the decision-making process.