Current Issue

2024, Volume 32,  Issue 2

Display Method:
2024, 32(2): 183-183.
Cross-domain Collaborative Application and Technology Development of Maritime Unmanned Systems
QIU Zhiming, MENG Xiangyao, MA Yan, WANG Liang, XIAO Yujie
2024, 32(2): 184-193. doi: 10.11993/j.issn.2096-3920.2024-0053
Cross-domain collaboration of maritime unmanned systems is the future development trend and important way of application of maritime unmanned systems. With the rapid development of various maritime unmanned systems and their application in local wars and conflicts in the world, how to better use maritime unmanned systems to complete cross-domain combat missions has become the focus of research. With maritime unmanned systems in different spatial domains as the research objects, the current maritime unmanned systems and the development status of cross-domain application of maritime unmanned systems in other countries were summarized. The basic principles and methods of cross-domain collaborative application of maritime unmanned systems were analyzed, and the key issues of cross-domain collaborative application of maritime unmanned systems were put forward. On this basis, the key technologies requiring attention in the development of technologies were sorted out. Finally, several enlightenments for future development were put forward, so as to provide a reference for the cross-domain application and technology development of maritime unmanned systems.
Development and Prospects of Networking Technologies for Autonomous Undersea Vehicles
YU Yang, SUN Siqing, ZHANG Lichuan, PAN Guang, WANG Peng
2024, 32(2): 194-207. doi: 10.11993/j.issn.2096-3920.2024-0055
Utilizing key technologies such as underwater acoustic communication, collaborative detection, and control decision-making, autonomous undersea vehicles(AUVs) can achieve networking and form either homogeneous or heterogeneous clusters for collaborative operations. Through such networking, the capabilities of different vehicle platforms can be fully utilized, enabling information sharing, task collaboration, and resource integration among multiple platforms. Therefore, more complex maritime operations can be completed autonomously. This method based on AUV clusters not only improves the efficiency of task execution but also reduces operational costs and enhances detection, monitoring, and response capabilities in the marine domain. The research status of AUV networking in China and abroad was introduced in this paper, and the key technologies and challenges, such as networking communication, cluster perception, and control decision-making were summarized. The application requirements and the development trend of AUV networking in detection, communication and control were predicted, providing a reference for research on basic theories and practical applications of AUVs.
Cross-Domain Unmanned Cluster Combat Supported by Satellite Internet
LI Mo, GUO Dong, ZHOU Donghong, DONG Guantao, LI Dongyu, CHU Hongbing, SHANG Zhigang
2024, 32(2): 208-214. doi: 10.11993/j.issn.2096-3920.2024-0043
Cross-domain unmanned cluster is a disruptive new combat mode in the future, and its combat effectiveness is greatly affected by communication networking capabilities. In recent years, satellite Internet, with its technological advantages of wide coverage, low latency, high bandwidth, and low cost, has developed rapidly in the world, and it is the key to further developing the cross-domain unmanned cluster combat mode. In view of the development trend and practical needs of the future unmanned undersea system, the development and technical advantages of satellite Internet, as well as the concept of cross-domain unmanned cluster combat were introduced. The advantages of cross-domain unmanned cluster combat supported by satellite Internet were deeply analyzed. The air-sea cross-domain unmanned cluster combat network was proposed by combining satellite Internet technology and unmanned clusters in airspace, sea surface, and undersea combat domains. In addition, a comprehensive analysis of the technical architecture was given, providing a reference for future research on cross-domain unmanned cluster combat.
Maritime Cross-Domain Collaboration Networking Communication Technology Based on Broadband Cognitive Stacked Networks
LI Zhongxiao, LUO Rong, WANG Gang, ZHANG Weiqiang
2024, 32(2): 215-227. doi: 10.11993/j.issn.2096-3920.2024-0036
With the increasingly severe situation of enemy confrontation at sea, multi-platform deployment at the ocean is needed. Maritime wireless communication technology relying on single band including high frequency, very/ultra-high frequency, and microwave frequency can no longer meet the communication support needs of multi-link fusion networking, dynamic resource allocation, and systematic anti-interference for maritime cross-domain collaboration operations. A novel multi-link fusion network communication architecture based on broadband cognitive stacked networks was proposed. A comprehensive integrated radio frequency(RF) front-end was adopted for broadband RF resource management, and a data link fusion meta-model was formed by unifying bandwidth, time slot, and time and space. On this basis, a software communication system backend was adopted for multi-link fusion and stacked networking. Through technologies such as broadband spectrum sensing, cognitive spectrum access, and task-driven adaptive networking, dynamic adaptation to maritime multi-domain communication environments such as air, sea, and submarine was achieved, as well as cognitive decision-making in strong enemy confrontation environments. This could enhance the expansion, cross-domain transmission, and anti-interference communication capabilities of maritime communication networks. In-depth research on the key technologies of broadband cognitive stacked networks was conducted, and a feasible technical approach to broadband cognitive stacked networks was given, which could provide a reference for the system architecture design and technical research in this field.
Cooperative Countermeasure Strategy of Sea-Air Cross-Domain Unmanned Platforms for Saturation Attack of Suicide UAVs
LIANG Xiao, CHEN Cong, LIU Dianyong, YU Changdong, LI Wei
2024, 32(2): 228-236. doi: 10.11993/j.issn.2096-3920.2024-0007
In view of saturation attacks of anti-suicide unmanned aerial vehicles(UAVs) in the marine environment, this paper studied the cooperative countermeasure strategy of sea-air cross-domain unmanned platforms under the condition that the number of targets far exceeded ours and proposed a cooperative algorithm combining improved genetic algorithm and coalition formation game. Firstly, according to the attack and motion characteristics of the sea-air cross-domain unmanned platform, the cost function was designed by combining the maximum and minimum strategies. Then, the genetic algorithm was improved according to the task requirements, and the crossover and mutation processes were guided and restricted. A feasible countermeasure scheme was generated by improving the efficiency of the genetic algorithm. Finally, the coalition formation rules were designed, and the coalitions reached Nash stability via changing members between the coalitions. The countermeasure scheme could still be continuously and stably optimized for many operators. The simulation comparison experiments show that the proposed strategy is feasible and superior and can provide a reasonable and efficient countermeasure scheme when the target is subjected to a saturation attack. This can provide a reference for research on large-scale cross-domain unmanned swarm combat.
Underwater Target Search and Tracking Based on Air-Sea Heterogeneous Unmanned Platform
DING Wenjun, CHAI Yajun, YANG Yuxian, LIU Jiamin, MAO Zhaoyong
2024, 32(2): 237-249. doi: 10.11993/j.issn.2096-3920.2024-0037
The marine heterogeneous unmanned systems can effectively enhance the implementation efficiency of complex missions. In this paper, autonomous undersea vehicles(AUVs) and unmanned aerial vehicles(UAVs) were used for searching and tracking unknown underwater targets in offshore waters. First, the underwater target search and tracking mission was described, and the mission was divided into two stages: target search and target tracking, with the objectives of maximizing the total search space of the AUV&UAV system and minimizing the end position error between the AUV and the underwater target, respectively. Then, a cross-domain collaborative search model of the AUV&UAV system was established, and constraints such as detection range and communication distance for AUVs and UAVs in the model were set. Finally, based on the traditional particle swarm optimization algorithm, an adaptive learning factor regulation strategy and an elite preservation strategy were employed for cross-domain collaborative search and tracking path planning, and search and tracking paths were generated. The simulation experiment demonstrates that the heterogeneous AUV&UAV system based on an improved particle swarm optimization algorithm can more efficiently search and track underwater targets.
Multi-UUV Collaborative Bearing-only Target Tracking Based on Boundary Constrained Particle Filter
HAN Bo, XU Hongli, QIU Shaoxiong, ZHANG Wenrui, RU Jingyu
2024, 32(2): 250-259. doi: 10.11993/j.issn.2096-3920.2024-0050
Due to the difficulties of filter initialization and packet loss in underwater acoustic data transmission faced by existing bearing-only target tracking algorithms, a multi-unmanned undersea vehicle(UUV) collaborative bearing-only target tracking algorithm based on boundary constrained particle filter was proposed, so as to meet the demand for multi-UUV collaborative detection of water surface targets in cross-domain collaboration at sea. Firstly, a master-slave collaborative detection model was proposed, which utilized the follower to report the state estimation results to the navigator for data fusion. Secondly, based on the prior information of UUV sensors and targets, a reliable particle generation method for the initial stage and a particle weight optimization method for the indicator function in the update stage were designed. Finally, a distributed fusion algorithm based on gray prediction was proposed to obtain the target prediction results. The simulation experiment compared this algorithm with other common algorithms and verified its effectiveness and feasibility under communication packet loss and noise interference.
Key Technologies and Verification of Cooperative Positioning Take-off and Landing for USVs and UAVs
ZHENG Bing, Dong Chao, LIU Han, XIONG Junfeng, HUANG Chaoxiong
2024, 32(2): 260-266. doi: 10.11993/j.issn.2096-3920.2024-0026
An unmanned surface vessel(USV) single-platform is easily restricted by the platform during surface mission execution, and the construction of a cross-platform system for USVs and unmanned aerial vehicles(UAVs) can realize the complementary advantages of the platform. In this paper, the take-off and landing and energy replenishment technologies of USVs and UAVs were studied to realize the enduring cooperative operation of USVs and UAVs. Firstly, for the problem of UAV taking off and landing on dynamic USVs, the research on cooperative positioning take-off and landing technologies for USVs and UAVs was carried out, and the dynamic update model of speed and position information of USVs and UAVs was studied, realizing high-precision take-off and landing of UAVs at the USV end. Secondly, to address the problem of the short endurance time of UAVs, the research on UAV replenishment at the USV end was carried out, and the take-off and landing mechanism at the UAV end and the charging mechanism at the USV end were investigated to realize the rapid replenishment of UAVs landing on the USV. Finally, the performance of the cross-platform take-off and landing and energy replenishment system for USVs and UAVs was verified at sea. In the test, UAVs could land at the USV end and get energy replenished at the high-speed movement of 3.0 m/s of the USV-UAV system, meeting the demand of cross-platform mission execution by USVs and UAVs at sea.
Defense Strategy for Suicide Unmanned Surface Vessels Based on Sea and Air Unmanned Clusters
SUN Feng
2024, 32(2): 267-274, 319. doi: 10.11993/j.issn.2096-3920.2024-0034
The recent use of unmanned surface vessels(USVs) in the Russia-Ukraine conflict has received much attention. To prevent the attacks of suicide USVs, the idea of using unmanned platforms to counter USVs was followed, and a new concept of preventing the attacks of suicide USVs by using low-cost platforms such as USVs and unmanned aerial vehicles(UAVs) featuring “collaborative enabling, autonomous control, and agile response” was proposed. Multiple lightweight USVs, UAVs, and unmanned platforms with simple interaction and autonomous decision-making capabilities were used. By analyzing the capabilities of typical suicide USVs, key technologies such as sea and air collaborative task assignment and planning and cross-platform collaborative reconnaissance and positioning were studied, so as to build a new unmanned reconnaissance and defense system. This system not only ensured all-weather and quasi-real-time coverage of key defense areas but also achieved agile confirmation and immediate response to invading suspicious targets. Furthermore, it expanded the defense depth of high-value targets and established multiple layers of defense against the attacks of suicide USVs in clusters.
Development Overview and Inspiration of Collaborative Application of Heterogeneous Unmanned Clusters at Sea
SUN Haiwen, WANG Xiye, WANG Zhaochen, MENG Xiangyao, CHEN Ting, JIN Zirong
2024, 32(2): 275-281. doi: 10.11993/j.issn.2096-3920.2024-0032
Unmanned system clusters, with their significant characteristics such as high coordination, complete autonomy, and collective intelligence, can directly affect and change the style of maritime operations. Multiple maritime powers around the world are competing to conduct research on the collaborative application of cross-domain heterogeneous unmanned equipment. The current development status of maritime heterogeneous unmanned clusters in the world and typical operational cases were analyzed, and it was found that maritime heterogeneous unmanned clusters had the characteristics of self-organized network communication of cross-domain multi-heterogeneous nodes, multi-domain and multi-source information fusion situation awareness, and multi-domain and large-scale heterogeneous cluster formation control. Finally, it was proposed that maritime heterogeneous unmanned clusters should fuse manned and unmanned sensory information, refined intelligent decision-making, enhanced operation control simulation training, and construction of decentralized cluster systems.
Review of Visual Control Technology for Undersea Vehicles
GAO Jian, HE Yaozhen, CHEN Yimin, ZHANG Yuanxu, YANG Xubo, LI Yufeng, ZHANG Zhenchi
2024, 32(2): 282-294. doi: 10.11993/j.issn.2096-3920.2023-0061
Visual control is a control method that utilizes visual information for environmental and self-state awareness. In this paper, this technology was applied to control undersea vehicles, and relevant research progress, challenges, and trends in different application scenarios were analyzed. The current development and task scenarios of visual control technology for undersea vehicles were first introduced, mainly focusing on underwater image enhancement, target recognition, and pose estimation technologies. The current development of visual control technology for undersea vehicles was then summarized and analyzed based on three task scenarios: underwater visual dynamic positioning and target tracking, undersea vehicle docking, and underwater operational tasks such as target grasping. Finally, the challenges and development trends of visual control technology for undersea vehicles were outlined.
Digital Development Method for Torpedo Weapons and Equipment Based on MBSE
WEN Zhiwen, WANG Zhong, XIE Bin, CAI Weijun, ZHANG Meiru
2024, 32(2): 295-303. doi: 10.11993/j.issn.2096-3920.2023-0031
With the development and application of torpedo weapons and equipment in future wars, new requirements for digitization, intelligence, and agility have been put forward. The practical problems faced by torpedoes in the traditional text-based development mode, such as the separation of the overall system and subsystems, as well as that of the design process and production process, lead to untimely error detection, insufficient early verification, incomplete information transmission, inconsistent model data, incomplete verification scope, and unclear process traceability. In order to solve these problems, shorten the torpedo development cycle, and improve development efficiency and quality, advanced system engineering methods and new technologies were both utilized. Based on the practical application research on model-based system engineering(MBSE) in China and abroad, this paper proposed a digital development method for torpedo weapons and equipment based on MBSE. The method integrated the MBSE idea with torpedo technology characteristics and research and development processes, constructed cross-stage, cross-discipline, and cross-level MBSE technology system for the entire lifecycle of torpedoes, proposed a digital application framework and technical roadmap for MBSE torpedo, and refined the main key technologies that needed to be addressed in the digital development process of MBSE torpedoes. The method proposed in this paper provides a reference for the digital development practice and comprehensive digital transformation of torpedo weapons and equipment.
Dynamic Path Generation Method for UUV Terrain Tracking Using Forward-Looking Sonar and Altimetry Sonar
CHEN Tao, WAN Shou
2024, 32(2): 304-310. doi: 10.11993/j.issn.2096-3920.2023-0047
Maintaining fixed altitude tracking navigation of seafloor terrain is a common form of motion, which is used by unmanned undersea vehicles(UUVs) on marine survey and underwater target search missions. The core of this motion is how UUVs can detect unknown undulating seafloor terrain in real time and generate tracking paths online and dynamically based on detection information, so as to achieve fixed altitude tracking navigation on the terrain while avoiding collision with the terrain. To solve the above problems, a method for detecting terrain information based on forward-looking sonar and dynamically generating tracking paths based on polynomial fitting was proposed. First, UUVs used forward-looking sonar to conduct real-time detection of seafloor terrain. After affine processing of the obtained terrain detection data, fixed altitude affine data with discrete characteristics could be obtained. Then, the cubic polynomial method based on the least squares criterion was used to fit the affine data, and the navigation path of UUVs for terrain tracking based on the polynomial function description was generated. Finally, a dynamic execution framework including sonar detection, data affine, path generation, and tracking control was designed, so as to realize the real-time terrain tracking navigation mission of UUVs. In this paper, through simulation of tracking on typical seafloor uphill and mountainous terrain, the effectiveness and feasibility of the proposed tracking path generation and dynamic execution framework were demonstrated.
ROV Motion Control Algorithm Based on RBF Neural Network Compensation
ZHANG Shuaijun, LIU Weidong, LI Le, LIU Jingbin, GUO Liwei, XU Jingming
2024, 32(2): 311-319. doi: 10.11993/j.issn.2096-3920.2023-0033
In view of the motion control problem of the operation-type remotely operated vehicles(ROVs) under the uncertainty of model parameters and the disturbance of the external environment, an adaptive double-loop sliding mode control strategy based on radial basis function(RBF) neural network was proposed. Firstly, the integral sliding mode control method with an improved reaching law was adopted for controlling the position of the ROV’s outer loop, and the integral sliding mode control method with an exponential reaching law was adopted for controlling the speed of the ROV’s inner loop. Secondly, in order to further improve the chattering problem of sliding mode control, the hyperbolic tangent function was introduced as the sliding mode switching term. Subsequently, the RBF neural network control technology was used to estimate and compensate for the uncertain parameters and external disturbances of the ROV model. Finally, the stability of the whole closed-loop system was proved by using the Lyapunov stability theory, and the motion control of the operation-type ROV was simulated numerically. The simulation results verify that the controller designed in this paper can achieve precise control of ROV navigation and effectively suppress the influence of uncertain parameters of the model and external disturbances on ROV motion.
ROV Path Planning of Hydropower Plants Based on Improved Hybrid Motion Sparrow Search Algorithm
ZENG Xuewen, HUANG Xiuhua, CHEN Min, ZHOU Da, ZHANG Fulin
2024, 32(2): 320-327. doi: 10.11993/j.issn.2096-3920.2023-0162
Path planning for underwater remotely operated vehicles(ROVs) is a prerequisite for underwater inspection operation of hydropower plants. The reservoir of hydropower plants has complex environments, and the existing path planning algorithms face the problems of long planning time, poor stability of algorithms, easy fall into the local optimum, and unsmooth path generation. In view of these issues, this paper put forward a ROV path planning method for hydropower plants based on the improved hybrid motion sparrow search algorithm. Firstly, the good point set was introduced to improve the sparrow population initialization method, which enhanced the population diversity; secondly, the hybrid motion strategy was proposed to optimize the sparrow population position updating method, increasing the algorithm’s convergence accuracy and stability; then, the multivariate objective function, which contained time cost, path threat, current disturbance, and penalty function, was established by combining with the actual engineering problems and considering the factors of large flow velocity of reservoirs, strong magnetic field, obstacles, and cost; finally, the triple B-spline interpolation was used to obtain the optimal smooth path. The simulation results show that compared with other path planning algorithms, the proposed method performs better in terms of computational accuracy, convergence speed, and stability, and it is suitable for underwater inspection tasks of hydropower plants.
Model-Free Adaptive Path Tracking Control Based on Active Disturbance Rejection Control for AUVs
FU Shaobo, GUAN Xiawei, ZHANG Hao
2024, 32(2): 328-336, 375. doi: 10.11993/j.issn.2096-3920.2023-0120
In view of the task requirements of accurate recovery of autonomous undersea vehicles(AUVs), a model-free adaptive control based on the active disturbance rejection control(ADRC-MFAC) algorithm was proposed from the perspective of model-free control, so as to improve the insufficient path tracking accuracy caused by model uncertainty and vulnerability to environmental interference in AUV motion. According to the characteristics of the second-order model system and line-of-sight guidance, the MFAC was improved by redesigning the control input criterion function, solving the problem that MFAC was only applicable to the self-balancing system. A tracking differentiator was introduced to smooth the desired signal, and a linear extended state observer was designed by considering the influence of unknown compound interference. The estimated disturbance was compensated for in the controller. The stability of the controller was verified, and system robustness was improved. Under the same interference, the proposed control scheme could improve the anti-interference ability and control precision by 42.37% and 45%, compared with the traditional proportional-integral-differential controller. The result shows that ADRC-MFAC can significantly improve the anti-interference performance of AUVs and enhance path tracking accuracy.
Energy-Saving Path Planning for AUVs in Current Environment Based on Improved PSO-Lévy Algorithm
YANG Huizhen, WANG Zijiang, ZHOU Zhuoyu, YANG Jun, LI Jianguo
2024, 32(2): 337-344. doi: 10.11993/j.issn.2096-3920.2023-0062
To obtain energy-saving obstacle avoidance paths of autonomous undersea vehicles(AUVs) in the current environment, a three-dimensional dynamic current environment model based on current field velocity information and underwater topographic obstacles was established. The augmented objective function was established based on navigation energy consumption, AUV maneuvering performance, and obstacle constraints, and an improved particle swarm optimization-Lévy(PSO-Lévy) algorithm based on weight adjustment mechanism and random wandering mechanism was proposed. The parameter adjustment strategy based on the optimal damping ratio and the random step wandering strategy based on the Lévy-flight process were introduced into the PSO algorithm. By executing a long-step wandering operation and regulating the velocity inertial of particles with a certain probability, the PSO algorithm could get a longer step and jump out of the local optimum. The simulation results show that the proposed algorithm can plan the optimal path with low energy consumption according to the current information while effectively avoiding obstacles.
Nonlinear Programming-Based Fault-Tolerant Control for X-Rudder AUVs
2024, 32(2): 345-353. doi: 10.11993/j.issn.2096-3920.2023-0133
To fully utilize the fault-tolerant capability of the X-rudder autonomous undersea vehicles (AUVs), a fault-tolerant motion control algorithm for AUVs oriented towards rudder failures was proposed, and it was deployed on a prototype of an X-rudder AUV. The fault-tolerant motion control algorithm consisted of two parts: dynamics control and control allocation. In dynamics control, the introduction of a multi-loop incremental feedback control algorithm could make the output virtual rudder instruction smooth and gentle. The control allocation algorithm converted the virtual rudder to the control input of the X-rudder actuator by solving a nonlinear programming problem, with the optimization goal of minimizing the allocation error and control output, and the constraints of rudder failure, rudder angle saturation, and other physical limitations were considered. This also enabled the X-rudder AUV to have fault-tolerant motion capabilities. Field trial results show that the rudder instructions generated by the fault-tolerant motion control algorithm proposed in this paper are smooth, and the X-rudder AUV still maintains a certain navigation control capability after the rudder failure. This has a certain guiding significance for the design of a fault-tolerant steering system applied to X-rudder AUVs.
Calculation Method for UKF Target Motion Elements Based on Detection Information of Active and Passive Sonars
ZHANG Hongrui, SU Jun, LI Qian, LI Bin, KOU Xiaoming
2024, 32(2): 354-361. doi: 10.11993/j.issn.2096-3920.2023-0045
The target motion element is important information in anti-submarine warfare, and its calculation results have a great influence on the hitting probability of the target, thus affecting combat decision-making. At present, active sonar is the main source of information in the calculation method for target motion elements in anti-submarine warfare of surface ships. However, active sonar uses a fixed number of sending periods, and there are gaps in the target information during the continuous tracking process. As a result, there are large errors and slow convergence in the calculation results of the target motion elements. In order to obtain the target motion elements more quickly and accurately, the detection information of passive sonar was added to the filtering process. The unscented Kalman filter(UKF) method was used to simulate the information detection methods using only active sonar and both active and passive sonars, and the results were compared. The simulation results show that under the same conditions, the proposed method can significantly improve the convergence accuracy and speed compared with the traditional method. It can improve the calculation accuracy of speed, azimuth, and heading angle by 33.55%, 38.99%, and 35.29% on average, verifying its effectiveness.
Calculation of Favorable Advance Angle Based on Maximum Entropy Distribution under Uncertain Homing Range
DAI Zhen, HAN Qingwei, LI Yang, WU Jiwei, SHEN Xiangzhong
2024, 32(2): 362-367. doi: 10.11993/j.issn.2096-3920.2023-0102
In order to solve the favorable advance angle in the case of an uncertain homing range of torpedoes, the influence of homing range on torpedo firing was first analyzed. It was pointed out that when the actual homing range of torpedoes was less than its binding value, the favorable advance angle obtained from the calculation became smaller, and the probability of torpedo discovery was significantly reduced. Secondly, the weighted average discovery probability of torpedoes was defined. On this basis, a favorable advance angle calculation method was proposed to maximize the weighted average discovery probability. Then, according to the principle of maximum entropy, two maximum entropy distribution models for the actual homing range of torpedoes were given, and the formula for solving the favorable advance angle under this distribution model was obtained. Finally, the favorable advance angle in the case of an uncertain homing range of torpedoes was obtained through simulation.
Finding Probability of Submarine-Launched Acoustic Homing Torpedoes Based on Gaussian Process Regression
REN Bin, XIE Chao
2024, 32(2): 368-375. doi: 10.11993/j.issn.2096-3920.2023-0113
The determination of the finding probability of submarine-launched acoustic homing torpedoes significantly affects tactical formulation. Conventional analytical and statistical algorithms fail to balance the speed and precision of probability assessment. In response to this issue, this paper introduced a model for assessing the finding probability based on Gaussian process regression. Additionally, a method was proposed for generating a training dataset based on the analytical model. Numerical simulations for assessing the finding probability were conducted within a specific battlefield scenario. The outcomes illustrate the superior assessment effect of the proposed method, offering theoretical support for decision-making in relevant battlefield contexts.
Adaptive Neural Network-Based Prescribed Performance Control of AUVs with Input Saturation
XU Wenfeng, LIU Jiapeng, YU Jinpeng, HAN Yaning
2024, 32(2): 376-382. doi: 10.11993/j.issn.2096-3920.2023-0041
In view of system uncertainty and input saturation of autonomous undersea vehicles(AUVs), an improved adaptive neural network-based prescribed performance control strategy was proposed to track the desired trajectory. Firstly, the nonlinear transformation was introduced to ensure that the position error remained within the preset time-varying range, improving control accuracy. Based on backstepping and Lyapunov functions, a virtual control law for the system was designed. Then, the neural network technology was used to process the unknown parameters of the system model, and the real control law of the system was reconstructed, which simplified the traditional backstepping control strategy and effectively reduced the computational complexity. Then, based on the Lyapunov stability theory, all the error signals of the AUV system were confirmed to be bounded. Finally, compared with traditional dynamic surface control methods, the simulation results show that the proposed control strategy has better control performance and can effectively overcome the impact of uncertainty on system performance by considering input saturation, effectively tracking target trajectories.