A dynamic path planning method based on a gated recurrent unit-recurrent neural network model is proposed for the problem of mobile robot in an unknown space. deep with sensor input used to generate new control strategy output physical movement and thus achieve collision avoidance behavior. Inputs tags are derived from sample sets generated by improved artificial potential field ant colony optimization algorithm. In order make algorithm converge quickly, pheromone trail state transition...

In a complex underwater environment, finding viable, collision-free path for an autonomous vehicle (AUV) is challenging task. The purpose of this paper to establish safe, real-time, and robust method collision avoidance that improves the autonomy AUVs. We propose based on active sonar, which utilizes deep reinforcement learning algorithm learn processed sonar information navigate AUV in uncertain environment. compare performance double Q-network algorithms with genetic learning....

Target state estimation is a key technology for unmanned underwater vehicles (UUVs) to achieve target tracking, collision avoiding, formation control, and other tasks. Compared with measurement methods, has lower reliability due the uncertainty of sonar detection. In this case, performance depends heavily on motion model. However, dynamics UUVs are very complex nonlinear. Although many methods nonlinear systems have been proposed, UUV in detection remain challenges problems. This article...

Autonomous collision avoidance is a critical technology in intelligent control, which of great significance for autonomous navigation and operation Unmanned Underwater Vehicles (UUVs). To enhance the autonomy UUV improve its adaptability to unstable forward-looking sonar observation uncertain environments, we propose novel Transformer-based Dual-Channel Self-attention (TDCS) architecture avoidance. TDCS network composed two encoders decoder, integrates dynamic/static obstacle recognition,...

The challenges facing unmanned underwater vehicle (UUV) target state estimation are observation uncertainty, the unpredictability of motion model, and complex relative moving to observer. Aiming at above problems, a convolutional neural network particle filter (CNNPF) is proposed applied UUV using forward-looking sonar. First, we design prediction based on (CNN) describe nonlinear measurement non-Markov process models. Then, dataset established for build space extract features from...

Predicting the popularity of online content is an important task for recommendation, social influence prediction and so on. Recent deep learning models generally utilize graph neural networks to model complex relationship between information cascade future popularity, have shown better results compared with traditional methods. However, existing adopt simple pooling strategies, e.g., summation or average, which prone generate inefficient representation lead unsatisfactory results. Meanwhile,...

In this paper, we present an online obstacle avoidance planning method for unmanned underwater vehicle (UUV) based on clockwork recurrent neural network (CW‐RNN) and long short‐term memory (LSTM), respectively. essence, UUV is a spatiotemporal sequence problem with the data of sensors as input control instruction to motion controller output. And networks (RNNs) have proven give state‐of‐the‐art performance many labeling prediction tasks. order train networks, dataset generated offline...

Abstract Unmanned Underwater Vehicles (UUVs) are essential equipment for Marine development, widely used in scientific research, resource survey, and security. The autonomous navigation planning ability unknown environments is a critical indicator UUV intelligence. This paper focuses on the particularity of motion complexity underwater environment proposes an adaptive Dynamic Window Approach (DWA) obstacle avoidance planning. DWA introduces novel heading angle evaluation dynamic strategies...

Path planning is one of the important autonomy abilities forautonomous underwater vehicle (AUV), whose main purpose toplan an optimized and safety path autonomously during long-rangenavigation in unknown environment.

Abstract Anomaly-based Intrusion Detection System (ADS) is one of the technologies widely used in network topology. Although many supervised and unsupervised learning methods field machine have been to improve efficiency ADS, achieving good performance still a challenging problem for existing intrusion detection algorithms. Firstly, there are few public datasets available evaluation. Secondly, single classifier may not perform well detecting each type attack. Third, some schemes focus on...

Accurate and stable estimation of the position trajectory noncooperative targets is crucial for safe navigation operation sonar-equipped underwater unmanned vehicles (UUVs). However, uncertainty associated with sonar observations unpredictability target movements often undermine stability traditional Bayesian methods. This paper presents an innovative approach state utilizing 3D Convolutional Kolmogorov–Arnold Networks (3DCKANs). By establishing a non-Markovian model that characterizes UUV...

In the context of multi-autonomous underwater vehicle (multi-AUV) operations, target assignment is addressed as a multi-objective allocation (MOA) problem. The selection strategy for multi-AUV dependent on current non-cooperative environment. This paper establishes situation advantage evaluation system to assess and quantify Based this framework, model using bi-matrix game theory developed, where strategies are considered part strategic framework within game. payoff matrix constructed based...