Research and develop navigation algorithms for unmanned surface vessel in dynamic environments
Searched over 200M research papers
Research Analysis by Consensus
Consensus Meter
Navigation Algorithms for Unmanned Surface Vessels in Dynamic Environments
Introduction to USV Navigation Challenges
Unmanned Surface Vehicles (USVs) are increasingly deployed for various ocean missions, necessitating advanced navigation algorithms to handle dynamic and complex maritime environments. These environments are characterized by time-varying ocean currents, tides, winds, and the presence of both static and dynamic obstacles. Effective navigation algorithms must ensure optimal path planning, energy efficiency, and collision avoidance.
Gaussian Process Motion Planning (GPMP2) for USVs
The anisotropic GPMP2 algorithm is a significant advancement in motion planning for USVs. This algorithm integrates anisotropy into the Gaussian Process Motion Planning (GPMP2) framework, allowing USVs to follow ocean currents to reduce energy consumption. Additionally, it incorporates dynamic fast Gaussian Process interpolation to enhance computational speed and trajectory quality. This method has been validated in a ROS environment, demonstrating its ability to generate smooth, collision-free trajectories that minimize the impact of adverse ocean currents.
Fast Marching Method for USV Formations
The Fast Marching (FM) method, particularly the constrained FM method, is effective for USV formation path planning in dynamic environments. This algorithm models the dynamic behavior of moving ships and has been tested in simulated environments, proving its efficiency in complex navigation scenarios. The constrained FM method ensures that multiple USVs can navigate as a formation fleet, improving overall mission effectiveness.
Dynamic Path Planning in Island-Reef Environments
USVs operating in island-reef environments face significant time-varying characteristics. A dynamic path planning algorithm that considers ocean currents, tides, and winds is essential. This approach constructs an environment model combined with a USV motion model to adapt to environmental disturbances. The velocity obstacle method is used for collision risk assessment, enhancing the USV's ability to avoid dynamic obstacles. The algorithm also employs an improved multi-objective particle swarm optimization to speed up solution accuracy and efficiency.
Autonomous Navigation in Coastal Traffic Areas
Advanced autonomous navigation algorithms are crucial for USVs operating in complex coastal traffic areas. These algorithms incorporate enhanced situational awareness to detect various obstacles, including small floating objects. The integration of the Korean intelligent maritime transportation service, e-Navigation, helps overcome the limitations of onboard sensors. Field tests have demonstrated the capability of these algorithms to handle long-distance navigation and various maritime missions without onboard safety personnel.
Multi-Sensor Data Fusion for Reliable Navigation
To address the inherent uncertainties of navigational sensors and environmental influences, multi-sensor data fusion algorithms are developed. The Unscented Kalman Filter (UKF) is used to process raw sensor measurements, improving the accuracy and reliability of navigational data. This approach ensures that USVs can operate effectively in practical, dynamic environments.
Automatic Navigation Systems
A complete automatic navigation system (ANS) for USVs includes a path planning subsystem (PPS) and a collision avoidance subsystem (CAS). The PPS uses the dynamic domain tunable fast marching square (DTFMS) method to build an environment model from real electronic charts, representing both static and dynamic obstacles. The CAS employs finite control set model predictive control (FCS-MPC) to track trajectories and avoid collisions. This system has been validated in realistic sea environments, demonstrating its effectiveness in various encounter scenarios.
Path Following with Adaptive Neural Networks
The finite-time predictor line-of-sight (LOS)-based integral sliding-mode adaptive neural (FPISAN) scheme is designed for high-accuracy path following in the presence of unknown dynamics and external disturbances. This method combines neural networks with integral sliding-mode control to approximate unknown dynamics and ensure path-following accuracy. Simulation experiments have validated the effectiveness of this approach.
Predictive Navigation Using Fast Marching Method
A Kalman filter-based predictive path planning algorithm is proposed to handle dynamic maritime environments. This algorithm predicts the trajectories of moving ships and the USV's position in real-time, assessing collision risks. The weighted fast marching square method is used for path optimization, balancing mission requirements such as minimum travel distance and safety. Simulations have shown that this approach effectively manages complex traffic environments.
Compliance with COLREGS for Safe Navigation
An autonomous motion planning algorithm based on the velocity obstacles (VO) method ensures USVs navigate safely in compliance with the International Regulations for Preventing Collisions at Sea (COLREGS). This method generates cone-shaped obstacles in the velocity space, encoding COLREGS rules naturally. Field experiments have demonstrated the algorithm's ability to handle multiple vessel encounters without explicit inter-vehicle communication.
Global Optimization with A* Algorithm
The A* with velocity variation and global optimization (A*-VVGO) algorithm addresses the limitations of existing global path planning methods by incorporating temporal dimensions and velocity variations. This approach predicts the paths of other vessels using AIS information and optimizes the USV's path for various mission requirements. Simulation results confirm the algorithm's effectiveness in generating smooth and safe paths in complex environments.
Conclusion
The development of advanced navigation algorithms for USVs is crucial for their effective deployment in dynamic maritime environments. These algorithms must consider various environmental factors, ensure energy efficiency, and comply with international navigation regulations. The integration of advanced techniques such as Gaussian processes, fast marching methods, multi-sensor data fusion, and adaptive neural networks significantly enhances the reliability and safety of USV operations.
Sources and full results
Most relevant research papers on this topic
Anisotropic GPMP2: A Fast Continuous-Time Gaussian Processes Based Motion Planner for Unmanned Surface Vehicles in Environments With Ocean Currents
Path planning algorithm for unmanned surface vehicle formations in a practical maritime environment
Dynamic Path Planning Algorithm for Unmanned Surface Vehicle Under Island–Reef Environment
Field demonstration of advanced autonomous navigation technique for a fully unmanned surface vehicle in complex coastal traffic areas
Filtering based multi-sensor data fusion algorithm for a reliable unmanned surface vehicle navigation
An Automatic Navigation System for Unmanned Surface Vehicles in Realistic Sea Environments
Finite-Time PLOS-Based Integral Sliding-Mode Adaptive Neural Path Following for Unmanned Surface Vessels With Unknown Dynamics and Disturbances
Predictive navigation of unmanned surface vehicles in a dynamic maritime environment when using the fast marching method
Safe Maritime Autonomous Navigation With COLREGS, Using Velocity Obstacles
USV path planning method with velocity variation and global optimisation based on AIS service platform
Try another search
What are the future of urban planning?
What is the impact of high-intensity interval training (HIIT) on cardiovascular and metabolic health?
What are the future of IoT in smart city development?
What is the Historical analysis of economic recessions?
What are the future of personalized medicine?
What are the strategies for using technology to bridge the gap between formal and informal education?