A dynamic path planning method based on a Gated Recurrent Unit-Recurrent Neural Network model is proposed for the problem of path planning of a mobile robot in an unknown space. A deep neural network with sensor input is used to generate a new control strategy output to the physical model to control the movement of the robot and thus achieve collision avoidance behavior. Inputs and tags are derived from sample sets generated by an improved artificial potential field and an improved ant colony optimization algorithm. In order to make the ant colony algorithm converge quickly, the pheromone trail and state transition probability are improved. The field function of the artificial potential field method is modified. Using the end-to-end network model to learn the mapping between input and output in the sample data, the direction and speed of the mobile robot are obtained. Simulation experiments and realistic simulations show that the network model can plan a reasonable path in an unknown environment. Compared with other traditional path planning algorithms, the proposed method is more robust than traditional path planning algorithms to differences in robot structure.

Target state estimation is a key technology for unmanned underwater vehicles to achieve target tracking, collision avoiding, formation control, and other tasks. Compared with other measurement methods, underwater measurement has lower reliability due to the uncertainty of sonar detection. In this case, the performance of target state estimation depends heavily on the target motion model. However, the dynamics of unmanned underwater vehicles are very complex and nonlinear. Although many state estimation methods for nonlinear systems have been proposed, the complex dynamics of unmanned underwater vehicle and uncertainty in sonar detection remain challenges for underwater target state estimation problems. This paper proposes a gated recurrent units-based particle filter to improve the performance of the target state estimator for unmanned underwater vehicles. A deep neural network framework based on gated recurrent units is used to establish the mapping between the previous measurements and the current target states. This neural network learns the dynamics model of unmanned underwater vehicles and recognizes the measurement noise. The proposed filter samples from previous measurements of the target unmanned underwater vehicle, and the fully trained deep neural network predicts the current states of the sampled particles. The proposed method solves the low accuracy and instability caused by motion modeling errors and system nonlinearity. Simulation results show that the proposed gated recurrent units-based particle filter for unmanned underwater vehicle state estimation has better accuracy and stability than the traditional state estimation methods. All simulations are done with the noise which is Gaussian noise pass through a nonlinear transformation.

Autonomous collision avoidance is a critical technology in intelligent control, which is of great significance for autonomous navigation and operation of Unmanned Underwater Vehicles (UUVs). To enhance the autonomy of UUV collision avoidance and improve its adaptability to unstable forward-looking sonar observation and uncertain environments, we propose a novel Transformer-based Dual-Channel Self-attention (TDCS) architecture for UUV collision avoidance. TDCS is a network architecture composed of two encoders and a decoder, which integrates dynamic/static obstacle recognition, obstacle motion prediction, collision risk assessment, and collision avoidance decision. The two encoders extract the observation features of different sensors in parallel, and this dual-channel structure avoids the mutual interference of input information. Moreover, the two encoders are beneficial for adaptive learning to focus on sensor information with varying forms of representation. The fusion layer integrates the bimodal features extracted from different sensors by the dual-channel encoder. Finally, the fusion features and historical decisions are recombined through the decoder to form a new decision.

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. In essence, UUV online obstacle avoidance planning is a spatiotemporal sequence planning problem with the spatiotemporal data sequence of sensors as input and control instruction to motion controller of UUV as output. And recurrent neural networks (RNNs) have proven to give state‐of‐the‐art performance on many sequence labeling and sequence prediction tasks. In order to train the networks, a UUV obstacle avoidance dataset is generated and an offline training and testing is adopted in this paper. Finally, the proposed two types of RNN based online obstacle avoidance planners are compared in path cost, obstacle avoidance planning success rate, training time, time‐consumption, learning, and generalization, respectively. And the good performance of the proposed methods is demonstrated with a series of simulation experiments in different environments.

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

Unmanned underwater vehicles (UUVs) face significant challenges in achieving safe and efficient autonomous navigation in complex marine environments due to uncertain perception, dynamic obstacles, and nonlinear coupled motion control. This study proposes a hierarchical autonomous navigation framework that integrates improved particle swarm optimization (PSO) for 3D global route planning, and a deep deterministic policy gradient (DDPG) algorithm enhanced by noisy networks and proportional prioritized experience replay (PPER) for local collision avoidance. To address dynamic sideslip and current-induced deviations during execution, a novel 3D adaptive line-of-sight (ALOS) guidance method is developed, which decouples nonlinear motion in horizontal and vertical planes and ensures robust tracking. The global planner incorporates a multi-objective cost function that considers yaw and pitch adjustments, while the improved PSO employs nonlinearly synchronized adaptive weights to enhance convergence and avoid local minima. For local avoidance, the proposed DDPG framework incorporates a memory-enhanced state–action representation, GRU-based temporal processing, and stratified sample replay to enhance learning stability and exploration. Simulation results indicate that the proposed method reduces route length by 5.96% and planning time by 82.9% compared to baseline algorithms in dynamic scenarios, it achieves an up to 11% higher success rate and 10% better efficiency than SAC and standard DDPG. The 3D ALOS controller outperforms existing guidance strategies under time-varying currents, ensuring smoother tracking and reduced actuator effort.