Search

CN-122015860-A - Navigation planning method, device and storage medium based on multi-mode information fusion

CN122015860ACN 122015860 ACN122015860 ACN 122015860ACN-122015860-A

Abstract

The application discloses a navigation planning method, equipment and storage medium based on multi-modal information fusion, which comprise the steps of obtaining multi-modal navigation information in a navigation area, preprocessing the multi-modal navigation information, uniformly mapping the preprocessed multi-modal navigation information into standardized navigation decision input parameters, constructing a navigation space model through preset navigation constraint conditions, inputting the navigation decision input parameters into the navigation space model, generating a plurality of candidate routes through the navigation space model, carrying out multi-dimensional evaluation on each candidate route, obtaining an evaluation result of each candidate route, responding to a candidate route selection instruction based on the evaluation result, and outputting a candidate route corresponding to the candidate route selection instruction. According to the application, the multi-mode information is mapped into the standard input in a unified way, and a plurality of candidate routes are generated and comprehensively evaluated, so that the intelligent navigation planning of the ship is realized safely, economically and efficiently.

Inventors

  • CHEN FEIER
  • ZHU ZHEN
  • LU JUNGUO

Assignees

  • 上海交通大学

Dates

Publication Date
20260512
Application Date
20260210

Claims (10)

  1. 1. The navigation planning method based on the multi-mode information fusion is characterized by comprising the following steps of: acquiring multi-mode navigation information in a navigation area, wherein the multi-mode navigation information at least comprises navigation traffic flow information, meteorological environment information and navigation risk information; preprocessing the multi-mode navigation information, and uniformly mapping the preprocessed multi-mode navigation information into standardized navigation decision input parameters; Constructing a navigation space model through preset navigation constraint conditions, inputting the navigation decision input parameters into the navigation space model, and generating a plurality of candidate routes through the navigation space model; Performing multidimensional evaluation on each candidate route, and obtaining an evaluation result of each candidate route; and responding to a candidate route selection instruction initiated by a user based on the evaluation result, and outputting a candidate route corresponding to the candidate route selection instruction.
  2. 2. The method for voyage planning based on multimodal information fusion of claim 1, wherein the step of uniformly mapping the preprocessed multimodal voyage information into standardized voyage decision input parameters comprises: Acquiring channel traffic flow information, carrying out data fusion on the channel traffic flow information, a ship historical track and real-time dynamic data, and generating a space-time traffic conflict probability field based on the fused data; Acquiring meteorological environment information, carrying out ship wave resistance and energy efficiency processing based on the meteorological environment information, and then quantitatively generating an environment cost field by processing results; Acquiring navigation risk information, assigning risk weights to corresponding navigation space areas according to risk element categories and risk grades of the navigation risk information, and generating navigation risk weight fields based on the risk weights of the navigation space areas; And taking the space-time traffic conflict probability field, the environment cost field and the navigation risk weight field as the standardized navigation decision input parameters.
  3. 3. The method for voyage planning based on multi-modal information fusion of claim 1, wherein the step of constructing the voyage space model by preset voyage constraints comprises: Acquiring navigation constraint parameters, and determining the navigable water area range of the ship according to the navigation constraint parameters, wherein the navigation constraint parameters comprise electronic chart data, the draft of the ship and safety margin requirements; constructing a static navigation constraint condition through the navigation constraint parameter and the navigable water area; creating the voyage space model based on the static voyage constraints.
  4. 4. A method of voyage planning based on multimodal information fusion as claimed in claim 3 wherein said step of creating said voyage space model based on said static voyage constraints comprises: Discretizing an initial navigation space model into a space-time grid model through the static navigation constraint condition, wherein the space-time grid model comprises a geographic position dimension and a time dimension; And setting navigation characteristic state values for each grid cell according to the space-time position of each grid cell in the space-time grid model so as to generate the navigation space model.
  5. 5. The voyage planning method based on multimodal information fusion as claimed in claim 1, wherein the step of inputting the voyage decision input parameters to the voyage space model, generating a plurality of candidate voyages from the voyage space model, comprises: assigning the navigation decision input parameters to corresponding grid cells of the navigation space model, and generating a navigation state potential field based on assignment information of the grid cells; Adopting a course starting point and a course end point as initial path points, and executing an adaptive search algorithm in the navigation state potential field to generate initial candidate courses; acquiring a plurality of decision input parameters, and adjusting the weight coefficient of the route cost function of the initial candidate route based on each decision input parameter; performing path searching for multiple times in the navigation state potential field based on the navigation path cost function after the weight coefficient adjustment to obtain multiple navigation constraint paths; And combining one or more navigation constraint paths to obtain a plurality of candidate routes.
  6. 6. A method of voyage planning based on multimodal information fusion as claimed in claim 5 wherein said step of obtaining a plurality of decision input parameters and adjusting the weighting coefficients of the voyage cost function of said initial candidate voyage based on each of said decision input parameters comprises: Acquiring a random disturbance factor, and adjusting the decision input parameter through the random disturbance factor; and adjusting the weight coefficient of the road cost function based on the adjusted decision input parameter.
  7. 7. The method for planning a voyage based on multimodal information fusion as claimed in claim 1, wherein said step of performing a multidimensional evaluation on each of said candidate routes and deriving an evaluation result of each of said candidate routes comprises: Calculating risk conflict grades of each candidate route through a preset risk conflict grade algorithm, wherein the risk conflict grades comprise accumulated risk values, potential conflict times and conflict grades; Calculating the estimated voyage time of each candidate route according to the ship main engine oil consumption model and the estimated fuel consumption cost of the whole voyage; Acquiring a ship task type, and distributing an evaluation weight for the risk conflict level and the estimated navigation time of each candidate route based on the ship task type; And calculating the comprehensive evaluation score of each candidate route based on the evaluation weight, the risk conflict level and the predicted voyage time, and obtaining an evaluation result serving as each candidate route according to the comprehensive evaluation.
  8. 8. The voyage planning method based on multi-modal information fusion according to any one of claims 1 to 7, further comprising: Acquiring actual navigation data based on the navigation process of the current candidate route, wherein the actual navigation data comprises an actual track of a ship, actual encountered weather sea conditions, actual traffic conflict events, actual fuel consumption of a host and changes of speed; comparing the collected actual navigation data with the predicted data of the candidate route, and determining a predicted deviation according to a comparison result; and carrying out self-adaptive correction on key parameters of a candidate route planning strategy according to the prediction deviation, wherein the key parameters comprise data fusion confidence degree weights for generating the navigation decision input parameters, grid state updating rules for constructing the navigation space model, weight coefficients in the route cost function and evaluation weights for multidimensional evaluation.
  9. 9. A multimodal information fusion based voyage planning device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the multimodal information fusion based voyage planning method according to any of claims 1 to 8.
  10. 10. A storage medium, characterized in that the storage medium is a computer-readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the steps of the multimodal information fusion based voyage planning method according to any of claims 1 to 8.

Description

Navigation planning method, device and storage medium based on multi-mode information fusion Technical Field The application relates to the technical field of navigation planning, in particular to a navigation planning method, device and storage medium based on multi-mode information fusion. Background The existing ship navigation planning method mainly relies on static geographic information provided by an electronic chart system, a captain or an empirical route of a shipping company, or limited environmental data (such as a single weather forecast source) for path design. With the development of shipping intelligence, some research attempts have been made to introduce dynamic information in a single dimension, such as real-time traffic flow or refined weather, to aid in decision-making. However, the prior art scheme has the inherent defects that firstly, the multisource heterogeneous navigation information (such as traffic flow, weather and risk areas) has great difference in data format, space-time resolution and updating frequency, and an effective unified characterization and fusion mechanism is lacking, so that a planning system is difficult to form integral and consistent cognition on a navigation environment. Secondly, route planning and subsequent navigation safety and economical evaluation processes are often mutually cut, multi-dimensional evaluation indexes cannot be fully considered during planning, a planning scheme is difficult to trace back and optimize during evaluation, and a decision process lacks systematicness and closed loop. In addition, the existing method has insufficient adaptability to the offshore dynamic and uncertain environment, the generated route scheme is single and stiff, alternatives under various scenes cannot be provided, and robust adjustment is difficult to carry out according to real-time changes. Therefore, one outstanding technical problem faced by the prior art is how to design an intelligent navigation planning method for ships, which can effectively overcome the isomerism of multi-source information, realize the unified fusion and standardized expression of multi-mode navigation information, systematically generate a plurality of candidate routes meeting complex constraint on the basis, and further output the optimal or satisfactory navigation route through a comprehensive multi-dimensional evaluation mechanism so as to comprehensively improve the safety, economy, timeliness and environmental adaptability of planning results. The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present application and is not intended to represent an admission that the foregoing is prior art. Disclosure of Invention The application mainly aims to provide a navigation planning method, equipment and a storage medium based on multi-mode information fusion, which aim at solving the technical problems of how to fuse multi-source heterogeneous navigation information and realize systematic and robust intelligent navigation planning. In order to achieve the above purpose, the present application provides a navigation planning method based on multi-mode information fusion, comprising: acquiring multi-mode navigation information in a navigation area, wherein the multi-mode navigation information at least comprises navigation traffic flow information, meteorological environment information and navigation risk information; preprocessing the multi-mode navigation information, and uniformly mapping the preprocessed multi-mode navigation information into standardized navigation decision input parameters; Constructing a navigation space model through preset navigation constraint conditions, inputting the navigation decision input parameters into the navigation space model, and generating a plurality of candidate routes through the navigation space model; Performing multidimensional evaluation on each candidate route, and obtaining an evaluation result of each candidate route; and responding to a candidate route selection instruction initiated by a user based on the evaluation result, and outputting a candidate route corresponding to the candidate route selection instruction. In order to achieve the above object, the application also provides a navigation planning device based on multi-modal information fusion, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the computer program is configured to realize the steps of the navigation planning method based on multi-modal information fusion. To achieve the above object, the present application further proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the steps of the navigation planning method based on multimodal information fusion as described above. One or more technical schemes