Dynamic path autonomous planning system and method for offshore mobile unmanned platform

Abstract

The present application provides a dynamic path autonomous planning system and method for a mobile unmanned platform at sea, which relates to the field of marine path planning technology, including: continuous multi-dimensional monitoring of the target sea area for a predetermined time period; analyzing the target monitoring time sequence and generating path planning constraints, generating a target visual map of the target sea area; performing path planning analysis on the target visual map; dynamically monitoring the real-time neighborhood of the target mobile unmanned platform; extracting the first obstacle in the real-time obstacle set, monitoring and obtaining the first operating state of the first obstacle; obtaining the first predicted operating state of the first obstacle in a predetermined time period; adding the first predicted operating state to the target visual map, and dynamically planning and adjusting the initial mobile path. The present application can solve the technical problem in the prior art that the existing marine unmanned platform technology is insufficient in autonomous navigation capabilities in complex marine environments, and achieve the technical effect of improving autonomous operation efficiency and ensuring navigation safety.

Description

Technical Field

[0001] The present application relates to the technical field of marine path planning, and in particular to a dynamic path autonomous planning system and method for a marine mobile unmanned platform. Background Art

[0002] With the increasing demand for offshore operations, the application of unmanned platforms in fields such as ocean exploration, environmental monitoring and water transportation continues to expand, among which path planning, real-time obstacle avoidance and dynamic navigation control are key technologies.

[0003] At present, the existing path planning algorithms are mostly based on fixed models of static environments, and usually determine the path through grid division or geometric planning. However, in complex environments with high real-time requirements, such as tidal flows, wind speed changes, and sea areas with frequent ship movements, it is difficult to respond quickly and cannot adapt to the real-time changes of dynamic factors at sea. For example, after determining the path, the traditional algorithm cannot make real-time adjustments as the environment changes. When obstacles move or environmental conditions change, the effect of path planning is greatly reduced, which may cause the unmanned platform to deviate from the course or collide. In addition, the marine environmental factors are complex and changeable, such as wind, current, waves, and marine biological activities. These environmental factors have a great impact on the navigation of unmanned platforms, and existing algorithms are difficult to fully model and predict these impacts. For example, in windy weather or large waves, the speed and heading of the unmanned platform will be significantly affected, and existing technologies often cannot achieve precise control and dynamic obstacle avoidance.

[0004] To sum up, there are technical problems in the existing technology that exist due to the insufficient autonomous navigation capability of existing offshore unmanned platform technology in complex marine environments, resulting in limitations in dynamic obstacle recognition and real-time path adjustment, which further affects the autonomous operation efficiency and safety of unmanned platforms in complex environments. Summary of the invention

[0005] The purpose of this application is to provide a dynamic path autonomous planning system and method for a mobile unmanned offshore platform, in order to solve the technical problem that the existing offshore unmanned platform technology is insufficient in autonomous navigation capability in complex marine environments, resulting in limitations in dynamic obstacle recognition and real-time path adjustment, further affecting the autonomous operation efficiency and safety of the unmanned platform in complex environments.

[0006] In view of the above problems, the present application provides a dynamic path autonomous planning system and method for an offshore mobile unmanned platform.

[0007] In the first aspect, the present application provides a dynamic path autonomous planning system for a mobile unmanned platform at sea, including: a target monitoring timing acquisition module, the target monitoring timing acquisition module is used to perform continuous multi-dimensional monitoring of the target sea area for a predetermined time period to obtain a target monitoring timing, wherein the target sea area refers to the activity sea area range of the target mobile unmanned platform; a target visible graph generation module, the target visible graph generation module is used to analyze the target monitoring timing and generate path planning constraints, and generate a target visible graph of the target sea area in combination with the path planning constraints; an initial moving path generation module, the initial moving path generation module is used to perform path planning analysis on the target visible graph in combination with the starting point and target point of the target mobile unmanned platform to generate an initial moving path; a real-time obstacle set acquisition module, the real-time obstacle set acquisition module is used to activate the target A sensor device group is deployed on the target mobile unmanned platform, and the real-time neighborhood of the target mobile unmanned platform is dynamically monitored through the sensor device group to obtain a real-time obstacle set; a first operating state obtaining module, the first operating state obtaining module is used to extract the first obstacle in the real-time obstacle set, and monitor and obtain the first operating state of the first obstacle; a first predicted operating state obtaining module, the first predicted operating state obtaining module is used to analyze the first operating state to obtain the first predicted operating state of the first obstacle in a predetermined time period; a dynamic programming adjustment module, the dynamic programming adjustment module is used to add the first predicted operating state of the first obstacle to the target visual map when the first predicted operating state reaches a predetermined threat constraint, and dynamically program the initial moving path according to the target visual map.

[0008] In the second aspect, the present application also provides a dynamic path autonomous planning method for a mobile unmanned platform at sea, comprising: performing continuous multi-dimensional monitoring of a target sea area for a predetermined time period to obtain a target monitoring time sequence, wherein the target sea area refers to the activity sea area range of the target mobile unmanned platform; analyzing the target monitoring time sequence and generating path planning constraints, and generating a target visual map of the target sea area in combination with the path planning constraints; performing path planning analysis on the target visual map in combination with the starting point and target point of the target mobile unmanned platform to generate an initial moving path; activating a sensor device group deployed on the target mobile unmanned platform, and dynamically monitoring the real-time neighborhood of the target mobile unmanned platform through the sensor device group to obtain a real-time obstacle set; extracting a first obstacle from the real-time obstacle set, and monitoring to obtain a first operating state of the first obstacle; analyzing the first operating state to obtain a first predicted operating state of the first obstacle in a predetermined time period; when the first predicted operating state reaches a predetermined threat constraint, adding the first predicted operating state of the first obstacle to the target visual map, and dynamically planning and adjusting the initial moving path according to the target visual map.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] The target monitoring sequence is obtained by performing continuous multi-dimensional monitoring of the target sea area for a predetermined time period, wherein the target sea area refers to the activity sea area range of the target mobile unmanned platform; the target monitoring sequence is analyzed and path planning constraints are generated, and a target visual map of the target sea area is generated in combination with the path planning constraints; a path planning analysis is performed on the target visual map in combination with the starting point and the target point of the target mobile unmanned platform to generate an initial moving path; the sensor equipment group deployed on the target mobile unmanned platform is activated, and the real-time neighborhood of the target mobile unmanned platform is dynamically monitored through the sensor equipment group to obtain a real-time obstacle set; the real-time obstacle set is extracted. The first obstacle in the obstacle set is detected, and a first operating state of the first obstacle is obtained by monitoring; the first operating state is analyzed to obtain a first predicted operating state of the first obstacle in a predetermined time period; when the first predicted operating state reaches a predetermined threat constraint, the first predicted operating state of the first obstacle is added to the target visual graph, and the initial moving path is dynamically planned and adjusted according to the target visual graph. That is to say, by achieving the technical goals of accurate obstacle avoidance, adaptive path planning, and real-time adjustment of the navigation path of the unmanned platform in a dynamic and complex marine environment, the technical effect of improving autonomous operation efficiency and ensuring navigation safety is achieved.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0013] Figure 1 This is a schematic diagram of the structure of the dynamic path autonomous planning system of the maritime mobile unmanned platform in this application;

[0014] Figure 2 This is a flow chart of the dynamic path autonomous planning method for the offshore mobile unmanned platform of this application.

[0015] Description of reference numerals:

[0016] Target monitoring time sequence obtaining module 11, target visual graph generating module 12, initial moving path generating module 13, real-time obstacle set obtaining module 14, first operating state obtaining module 15, first predicted operating state obtaining module 16, dynamic planning adjustment module 17. DETAILED DESCRIPTION

[0017] This application provides a dynamic path autonomous planning system and method for a mobile unmanned platform at sea, which solves the technical problem that the existing unmanned platform technology at sea is insufficient in autonomous navigation in complex marine environments, resulting in limitations in dynamic obstacle recognition and real-time path adjustment, which further affects the autonomous operation efficiency and safety of the unmanned platform in complex environments. The technical goal of achieving accurate obstacle avoidance, adaptive path planning, and real-time adjustment of the navigation path of the unmanned platform in a dynamic and complex marine environment is achieved, achieving the technical effect of improving autonomous operation efficiency and ensuring navigation safety.

[0018] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0019] For example, please refer to the attached Figure 1 , this application provides a dynamic path autonomous planning system for a mobile unmanned platform at sea, specifically including:

[0020] The target monitoring timing obtaining module 11 is used to perform continuous multi-dimensional monitoring of the target sea area for a predetermined time period to obtain the target monitoring timing, wherein the target sea area refers to the activity sea area range of the target mobile unmanned platform.

[0021] Specifically, the target sea area refers to the range of active waters covered by the target mobile unmanned platform (such as an unmanned sailboat or drone). For example, for an unmanned sailboat active in the East China Sea or the Yellow Sea, its target sea area refers to its main navigation and monitoring area. Continuous monitoring is carried out within the scope of the target sea area for a predetermined time period, that is, data is continuously collected at a pre-set time interval (such as every 30 minutes or 1 hour) to ensure the time continuity of the data. Multidimensional monitoring refers to collecting environmental information from multiple angles, such as water temperature, ocean current, wind speed, wind direction and other parameters. Each monitoring parameter constitutes a dimension to fully understand the situation in the sea area. Through continuous monitoring, the target monitoring time series can be obtained, that is, the value sequence of each monitoring parameter at different time points, such as the record result of the ocean current speed every 10 minutes in a certain period of time, which is used to help analyze the dynamic changes in the sea area. For example, the result of monitoring the water temperature of an unmanned sailboat every hour may be 22 degrees Celsius, 22.5 degrees Celsius, 23 degrees Celsius, etc., and then the temperature change trend of the target sea area can be obtained.

[0022] The target visibility graph generation module 12 is used to analyze the target monitoring time sequence and generate path planning constraints, and generate a target visibility graph of the target sea area in combination with the path planning constraints.

[0023] Specifically, analyzing the target monitoring time series refers to conducting an in-depth analysis of the time series data collected by the target mobile unmanned platform in the target sea area, and finding key environmental or meteorological conditions by observing and calculating the changing trends of the data. For example, the target monitoring time series includes hourly wind speed, wind direction and water flow speed. The fluctuation and change of the data can directly affect the route of the target mobile unmanned platform. Generating path planning constraints is to use the analysis results to design restrictions for the platform's navigation, helping the target mobile unmanned platform to have more adaptability when responding to environmental changes, such as setting avoidance routes or adjusting the speed in areas with wind speeds exceeding 10 meters per second. Generating a target visualization map of the target sea area in combination with path planning constraints means adding all obstacle information or restrictions to be avoided in path planning to the basic two-dimensional model, making the navigation information of the sea area more intuitive and clear. Among them, static obstacles (such as buoys fixed at 120 degrees longitude and 30 degrees latitude) and dynamically changing areas can be marked in the target visualization map, so that the target mobile unmanned platform can avoid risks more intuitively and effectively when formulating routes. The generated target visualization map not only includes the basic information of the target sea area, but also integrates constraints such as obstacle locations and environmental changes analyzed from monitoring data, providing a safe and efficient path planning basis for the target mobile unmanned platform.

[0024] The initial moving path generating module 13 is used to perform path planning analysis on the target visual map in combination with the starting point and the target point of the target mobile unmanned platform to generate an initial moving path.

[0025] Specifically, the target visualization map is analyzed for path planning in combination with the starting point and target point of the target mobile unmanned platform, that is, the starting position and expected arrival position of the target mobile unmanned platform are used to evaluate the path favorable navigation for static obstacles and dynamic obstacles, and the path is selected or optimized on the target visualization map according to the favorable results to obtain the initial movement path. For example, the obstacles, environmental characteristics, water flow and other influencing factors between the starting point and the target point are evaluated to select the optimal driving route. For example, if the target visualization shows that there is a buoy and a reef on a straight path, the path planning analysis will suggest that the target mobile unmanned platform choose an alternative path to bypass the obstacles, thereby generating a safe and effective initial movement path. By generating the initial movement path, not only can obstacles be effectively avoided, but the impact of the marine environment on navigation can also be taken into account, ensuring that the target mobile unmanned platform can reach the target point smoothly and safely.

[0026] The real-time obstacle set obtaining module 14 is used to activate the sensor equipment group deployed on the target mobile unmanned platform, and dynamically monitor the real-time neighborhood of the target mobile unmanned platform through the sensor equipment group to obtain the real-time obstacle set.

[0027] Specifically, the sensor equipment group deployed on the target mobile unmanned platform is activated, and various sensors installed on the target mobile unmanned platform, including radar, sonar, camera, etc., are started. The real-time neighborhood of the target mobile unmanned platform is dynamically monitored through the sensor equipment group, that is, the surrounding environmental information is collected and analyzed to obtain real-time data related to the space around the target mobile unmanned platform. For example, assuming that the sensor of the target mobile unmanned platform can detect obstacles within fifty meters around, at each data update, the sensor will record the type, distance and relative position of the obstacle, obtain a real-time obstacle set, and generate a list of all detected obstacles, including buoys, other ships, marine life, rocks, etc., providing important support for safe navigation.

[0028] The first operating state obtaining module 15 is used to extract the first obstacle in the real-time obstacle set and monitor and obtain the first operating state of the first obstacle.

[0029] Specifically, the first obstacle in the real-time obstacle set is extracted, and an obstacle is randomly selected from the generated obstacle list for detailed analysis. For example, buoys, fishing boats, and rocks. Then, the first operating state of the first obstacle is monitored, and the real-time dynamics of the first obstacle are obtained as the initial navigation state, which provides the necessary basis for the path planning and obstacle avoidance strategy of the target mobile unmanned platform.

[0030] The first predicted operating state obtaining module 16 is used for analyzing the first operating state to obtain a first predicted operating state of the first obstacle in a predetermined time period.

[0031] Specifically, the first operating state is analyzed to obtain the first predicted operating state of the first obstacle in the predetermined time period, that is, the position and movement direction in the future are inferred based on the motion characteristics of the first obstacle currently observed. Among them, the predicted operating state is the position and state of the obstacle expected in the predetermined time period, that is, the next position of the navigation based on the initial position is predicted according to the duration and speed of the predetermined time period. As the first predicted operating state, it can help the target mobile unmanned platform to judge the potential impact of obstacles in advance, so as to effectively adjust the path planning and obstacle avoidance strategy.

[0032] A dynamic programming adjustment module 17 is used to add the first predicted operating state of the first obstacle to the target visibility graph when the first predicted operating state reaches a predetermined threat constraint, and to dynamically adjust the initial moving path according to the target visibility graph.

[0033] Specifically, when the first predicted operating state reaches the predetermined threat constraint, that is, the future position or state of the obstacle will affect the safe navigation of the target mobile unmanned platform. For example, if it is predicted that the fishing boat will enter the range of 20 meters in front of the target mobile unmanned platform in ten seconds, the pre-set threat distance has been reached. At this time, the first predicted operating state of the first obstacle is added to the target visual map, indicating that this predicted position is marked on the visual map to ensure that the target mobile unmanned platform avoids the potential threat range of the obstacle. Then, the initial moving path is dynamically planned and adjusted according to the target visual map, and a safe path is recalculated to bypass possible collision areas. For example, the platform may choose to deviate slightly from the initial path to increase the safety of reaching the destination, so that it can flexibly respond to dynamic environmental changes and ensure that the target mobile unmanned platform can move forward smoothly and safely during navigation.

[0034] The dynamic path autonomous planning system of the offshore mobile unmanned platform can achieve the technical goals of accurate obstacle avoidance, adaptive path planning, and real-time adjustment of the navigation path of the target mobile unmanned platform in a dynamic and complex marine environment, thereby achieving the technical effect of improving autonomous operation efficiency and ensuring navigation safety.

[0035] Furthermore, the present application also includes:

[0036] Read the predetermined monitoring dimensions, and perform continuous monitoring of the target sea area for the predetermined time period based on the predetermined monitoring dimensions to obtain the target monitoring time sequence, wherein the predetermined monitoring dimensions include environmental dimensions, meteorological dimensions, and obstacle dimensions, and the environmental dimensions include at least ocean terrain features, water flow features, and marine biological features, the meteorological dimensions include at least storm features and wave features, and the obstacle dimensions include at least ship features, buoy features, fishing net features, and coral reef features.

[0037] Specifically, reading the predetermined monitoring dimension means that the unmanned sailboat or other unmanned mobile platform will set specific parameters or information to be monitored in advance. The predetermined monitoring dimension refers to the various data categories that need to be paid attention to in the monitoring task, so that the target mobile unmanned platform can collect and analyze them in a targeted manner in the activity area. The predetermined monitoring dimensions mainly include environmental dimension, meteorological dimension and obstacle dimension. The environmental dimension involves the characteristics of marine terrain, water flow characteristics and the distribution of marine organisms, which affect the navigation of the target mobile unmanned platform. For example, during navigation, changes in sea current speed will cause the target mobile unmanned platform to deviate from the course, and dense areas of marine biological communities may be important avoidance areas. The meteorological dimension focuses on characteristics such as storms and waves. Storm characteristics and wave characteristics change quickly and have a greater impact on navigation, especially in areas where storms occur frequently. Real-time monitoring can help the target mobile unmanned platform adjust its route before bad weather. The obstacle dimension refers to obstacles such as ships, buoys, fishing nets, coral reefs, etc. in the sea area. Moving or stationary on the sea surface may affect the navigation of the target mobile unmanned platform. For example, when a fishing net or buoy is detected in the navigation area, the course needs to be adjusted to avoid collision.

[0038] Based on the predetermined monitoring dimensions, the target mobile unmanned platform will continuously monitor the target sea area according to the predetermined time period, which means that data information of the monitoring dimensions will be repeatedly collected during the activity, such as sea water temperature and water flow velocity measurements every 10 minutes, or meteorological information updates once an hour, to ensure that the latest data information is obtained in real time in the changing ocean environment, and the navigation strategy is adjusted according to the changes in the data to ensure the navigation stability and safety of the target mobile unmanned platform.

[0039] By reading the predetermined monitoring dimensions and conducting continuous time period monitoring of the target sea area based on the predetermined monitoring dimensions, the target monitoring time series data can be obtained, and then real-time adjustments can be made to ensure navigation safety and efficiency.

[0040] Furthermore, the present application also includes:

[0041] Extract any feature of any dimension in the predetermined monitoring dimension; match any parameter time series corresponding to the arbitrary feature in the target monitoring time series; perform energy density distribution analysis on the arbitrary parameter time series to obtain characteristic parameters corresponding to the highest energy density, which are recorded as arbitrary feature constraints of the arbitrary feature; generate the path planning constraint according to the mapping relationship between the arbitrary feature and the arbitrary feature constraint.

[0042] Specifically, extract any feature of any dimension in the predetermined monitoring dimension, and select a specific feature from multiple set monitoring categories for data analysis. For example, in the meteorological dimension of monitoring, you can select "storm features" as a feature, and in the obstacle dimension, you can select "ship features" as a feature. Each feature is a data type that the target mobile unmanned platform needs to focus on, which is helpful for subsequent analysis.

[0043] Find relevant data of specific features from the collected data and organize them into parameter time series in chronological order. For example, if "storm characteristics" is selected as the feature, the time series data of storm changes can be extracted from the monitoring time series, such as the wind speed record results every 10 minutes are 5 meters per second, 7 meters per second, 10 meters per second, etc., to form a storm characteristic parameter time series.

[0044] Perform energy density distribution analysis on any parameter time series, calculate and analyze the energy density of the acquired time series data, find out the energy distribution law of the arbitrary feature in different time periods, that is, obtain the parameter mode in a series of monitoring parameters corresponding to the arbitrary feature as a consideration constraint for the current planned path. For example, for storm characteristic parameter time series, calculate the energy density of the storm to analyze the energy level at different time points, and then help the target mobile unmanned platform identify the time period with concentrated wind speed intensity, and find the wind speed with the highest energy density as the characteristic parameter for subsequent use. Use the characteristic parameter with the highest energy density as a limiting condition for the feature to guide the platform's path planning.

[0045] Path planning constraints are generated based on the mapping relationship between any feature and any feature constraint, that is, by analyzing the correlation between the feature and its constraint, constraints are set for path planning. Taking the "storm feature" as an example, if the wind speed constraint value is 10 meters per second, the target mobile unmanned platform will try to avoid areas exceeding the wind speed range of the "storm feature" when planning the path to ensure the safety and stability of navigation.

[0046] By extracting specific features, analyzing the energy distribution of the time series data of feature parameters, and determining the constraints, the specific constraints for path planning are finally generated. The target mobile unmanned platform can make real-time route adjustments based on the path planning constraints, effectively avoiding risks and maintaining the reliability of navigation.

[0047] Furthermore, the present application also includes:

[0048] Construct an initial two-dimensional model of the target sea area; extract a first static constraint from the path planning constraints, wherein the first static constraint refers to a first position of a first static obstacle; mark the first position corresponding to the first static obstacle to the initial two-dimensional model to obtain the target visibility graph.

[0049] Specifically, the geographical and environmental data of the target sea area are used to draw a basic graphic of the sea area on a two-dimensional plane, including information such as the coastline, sea area scope, and general water depth distribution, providing a basic navigation framework for the operation of the target mobile unmanned platform.

[0050] Randomly extract information about obstacles that will not move from the constraints of path planning, called the "first static constraint". This type of obstacle is fixed in position and needs to be bypassed when the target mobile unmanned platform is sailing. The first static obstacle refers to obstacles such as shoals, reefs, and fixed buoys. The specific location can be marked by GPS data. For example, in a certain sea area, the first position of the first static obstacle is 120 degrees longitude and 30 degrees latitude.

[0051] Marking the position corresponding to the first static obstacle to the initial two-dimensional model to obtain the target visible map means further adding the specific location information of the static obstacle to the initial model so that the obstacle is clearly visible on the map. The model drawn in this way is transformed into a target visible map, which contains the geographical location of the static obstacles, making it easier for operators to identify obstacles in the sea when using the target mobile unmanned platform. For example, when the shoal position at latitude and longitude of 120 degrees and 30 degrees is marked on the target visible map, the target mobile unmanned platform can automatically identify and avoid this area when navigating.

[0052] By constructing an initial two-dimensional model, extracting and marking the positions of static obstacles, we can finally obtain a target visualization diagram containing static obstacle information, which makes the navigation planning of the target mobile unmanned platform visualized, facilitates effective risk avoidance, and improves the safety and accuracy of operations.

[0053] Furthermore, the present application also includes:

[0054] A* algorithm principle is used to perform path planning analysis on the starting point, the target point and the target visibility graph to generate a candidate moving path; a first dynamic constraint in the path planning constraint is extracted, and the first dynamic constraint refers to a first obstruction coefficient of a first dynamic obstacle; the candidate moving path is adjusted based on the first obstruction coefficient to obtain the initial moving path.

[0055] Specifically, the A* algorithm is used to find the optimal path from the starting point to the target point. The A* algorithm is a path-finding algorithm that can select the shortest path based on node costs and distances on a two-dimensional or three-dimensional map. The A* algorithm analyzes factors such as obstacles and distances between the starting point and the target point. For example, if the starting point is at longitude 120 degrees and latitude 30 degrees, and the target point is at longitude 121 degrees and latitude 31 degrees, the A* algorithm will calculate several candidate paths based on the obstacles and path costs on the map.

[0056] Constraints related to dynamic environmental factors are extracted from the constraints of path planning as the first dynamic constraints. Dynamic constraints include obstacles moving in the sea, such as floating ships or changing water currents. The first obstruction coefficient of the first dynamic obstacle is a numerical parameter that represents the intensity of the impact of the obstacle on the path. For example, if a ship is traveling at a speed of 10 meters per second on the route of the target mobile unmanned platform, the A* algorithm will use an obstruction coefficient (such as 0.8) to evaluate the degree of obstruction of the obstacle to the path. For example, when the first dynamic obstacle is a storm, the obstruction coefficient of the storm is weighted according to the intensity and duration of the storm as the first obstruction coefficient, where the obstruction coefficient is a positive number or a complex number, which is positive when the moving direction is with the wind and negative when the moving direction is against the wind.

[0057] In the generated candidate paths, obstacle avoidance optimization is performed according to the degree of influence of dynamic obstacles to obtain a safer initial moving path so that the target mobile unmanned platform can avoid high-risk areas. For example, when a storm with an obstruction coefficient of 1.2 occurs, the area where the storm passes is given priority, and the route is adjusted to a route that is more advantageous for the target mobile unmanned platform; when the obstruction coefficient is between 0 and 1, it means that the storm has little obstruction to the target mobile unmanned platform, and no route adjustment is performed.

[0058] The path between the starting point and the end point is planned through the A* algorithm, and the candidate path is generated by combining the static map. Then, by extracting dynamic constraints and optimizing and adjusting according to the obstruction coefficient of obstacles, the initial path suitable for the target mobile unmanned platform is finally obtained, thereby improving the safety and traffic efficiency of the path.

[0059] Furthermore, the present application also includes:

[0060] The target visible graph is segmented to obtain a target block set; the starting block corresponding to the starting point and the target block corresponding to the target point are matched in sequence in the target block set; based on the A* algorithm principle, the shortest path from the starting block to the target block is obtained, and the shortest path is used as the candidate moving path.

[0061] Specifically, the target visible image is segmented into smaller areas to facilitate more accurate positioning and path analysis over a large area. The path is segmented and dynamically adjusted by dividing the target sea area into several equal-sized blocks.

[0062] Next, the positions of the starting point and the target point are matched in the target block in turn, so that the area where the starting point is located is called the starting block, and the area where the target point is located is the target block, so that the starting and ending areas can be clearly located for subsequent path planning. Then, based on the principle of the A* algorithm, that is, a graph search algorithm with the optimal path as the goal, path selection is performed on the nodes between the target blocks, and the shortest path from the starting block to the target block is calculated. The basic principle of the A* algorithm is to minimize the travel distance and time in the moving path. Therefore, the optimal path is obtained by evaluating the cost of reaching the target point for each path node. Finally, the shortest path from the starting block to the target block will be identified and set as the candidate moving path.

[0063] By combining block processing and the A* algorithm, feasible candidate paths can be quickly generated in complex marine environments, improving the autonomous planning capability and efficiency of unmanned platforms.

[0064] Furthermore, the present application also includes:

[0065] A first adjacent block set of the starting block is formed, and a first block in the first adjacent block set is extracted; whether the first block meets the predetermined block constraint is determined; if so, the first block is used as a first path point, and the shortest path is formed based on the first path point.

[0066] Specifically, adjacent blocks around the departure block are identified and selected as first adjacent blocks, and integrated to obtain a first adjacent block set, so as to expand the path search space during the path planning process. For example, if the first adjacent block set of the departure block includes adjacent blocks in four directions of southeast, northwest, and northeast.

[0067] Next, a first block is randomly extracted from the first adjacent block set so as to check the properties and conditions of each adjacent block one by one to ensure that the path meets the set requirements during the advancement process.

[0068] Then, it is determined whether the first block meets the predetermined block constraints, such as whether it avoids obstacles or meets the safety requirements of the marine environment. If it meets the predetermined conditions, the first block is marked as the first path point, indicating that the first block will become a component of the shortest path.

[0069] Finally, the shortest path is gradually generated based on the first path point and other qualified path points, ultimately guiding the unmanned platform to the target point.

[0070] By extracting adjacent blocks in sequence and checking whether they meet the predetermined constraints, the shortest path is gradually generated when the path planning is met, ensuring the path safety and improving the autonomous navigation efficiency of the unmanned platform.

[0071] Furthermore, the present application also includes:

[0072] According to the first initial position, the first running direction, and the first running speed in the first running state, combined with the length of the predetermined time period, a first predicted position of the first obstacle is calculated; and the first predicted position is used as the first predicted running state of the first obstacle.

[0073] Specifically, the moving distance is calculated by multiplying the first running speed by the length of a predetermined time period. Starting from the first initial position, the moving distance is calculated by navigating in the first running direction, and the first predicted position of the first obstacle is calculated to obtain the predicted position to which the first obstacle may move within a period of time.

[0074] Taking the first predicted position as the first predicted operating state of the first obstacle and considering it as the new position of the first obstacle after a predetermined time period helps to update the understanding of the obstacle behavior, provide a basis for subsequent path planning and obstacle avoidance decisions, and be used to evaluate navigation safety and ensure that the target mobile unmanned platform can effectively avoid this dynamic obstacle.

[0075] By analyzing the first operating state, calculating the movement prediction of the obstacle within a predetermined time period, and updating the predicted position as the operating state of the obstacle, important dynamic information can be provided for subsequent path planning, ensuring that the target mobile unmanned platform can navigate safely and avoid potential dangers.

[0076] Furthermore, the present application also includes:

[0077] According to the initial moving path and the predetermined time period, the target predicted position of the target mobile unmanned platform is obtained; the predetermined threat distance is read, and the predetermined threat range of the target predicted position is obtained according to the predetermined threat distance; the predetermined threat range is used as the predetermined threat constraint for the target mobile unmanned platform when it is at the target predicted position.

[0078] Specifically, based on the navigation path selected by the target mobile unmanned platform, the position where the target mobile unmanned platform is expected to arrive within a predetermined time period is calculated as the target predicted position.

[0079] The potential threat area is evaluated based on the set safety distance. The predetermined threat distance can be, for example, twenty meters, which will form a circular area around the predicted target position, indicating that there are possible threat factors within this range, such as other ships or obstacles.

[0080] The predetermined threat range is used as a predetermined threat constraint for the target mobile unmanned platform when it is at the target predicted position, indicating that the threat range must be taken into account during path planning to ensure the safe operation of the target mobile unmanned platform. That is, when the target mobile unmanned platform reaches the predicted position, it must remain outside the predetermined threat range to avoid collision with any potential threats.

[0081] By calculating the predicted position of the target mobile unmanned platform based on the initial moving path and the predetermined time period, and determining the threat range in combination with the read predetermined threat distance, a safety constraint on the target mobile unmanned platform is ultimately formed, ensuring the safety and effectiveness of the target mobile unmanned platform during navigation.

[0082] In summary, the dynamic path autonomous planning system for a mobile unmanned offshore platform provided by this application has the following technical effects:

[0083] The target monitoring sequence is obtained by performing continuous multi-dimensional monitoring of the target sea area for a predetermined time period, wherein the target sea area refers to the activity sea area range of the target mobile unmanned platform; the target monitoring sequence is analyzed and path planning constraints are generated, and a target visual map of the target sea area is generated in combination with the path planning constraints; a path planning analysis is performed on the target visual map in combination with the starting point and the target point of the target mobile unmanned platform to generate an initial moving path; the sensor equipment group deployed on the target mobile unmanned platform is activated, and the real-time neighborhood of the target mobile unmanned platform is dynamically monitored through the sensor equipment group to obtain a real-time obstacle set; the real-time obstacle set is extracted. The first obstacle in the obstacle set is detected, and a first operating state of the first obstacle is obtained by monitoring; the first operating state is analyzed to obtain a first predicted operating state of the first obstacle in a predetermined time period; when the first predicted operating state reaches a predetermined threat constraint, the first predicted operating state of the first obstacle is added to the target visual graph, and the initial moving path is dynamically planned and adjusted according to the target visual graph. That is to say, by achieving the technical goals of accurate obstacle avoidance, adaptive path planning, and real-time adjustment of the navigation path of the unmanned platform in a dynamic and complex marine environment, the technical effect of improving autonomous operation efficiency and ensuring navigation safety is achieved.

[0084] Embodiment 2: Based on the dynamic path autonomous planning system of the unmanned mobile platform at sea in the above embodiment, the present application also provides a dynamic path autonomous planning method for the unmanned mobile platform at sea, please refer to the attached Figure 2 ,include:

[0085] Conducting continuous multi-dimensional monitoring of the target sea area for a predetermined time period to obtain a target monitoring sequence, wherein the target sea area refers to the activity sea area range of the target mobile unmanned platform; analyzing the target monitoring sequence and generating path planning constraints, and generating a target visual map of the target sea area in combination with the path planning constraints; conducting path planning analysis on the target visual map in combination with the starting point and the target point of the target mobile unmanned platform to generate an initial moving path; activating a sensor device group deployed on the target mobile unmanned platform, and dynamically monitoring the real-time neighborhood of the target mobile unmanned platform through the sensor device group to obtain a real-time obstacle set; extracting a first obstacle from the real-time obstacle set, and monitoring to obtain a first operating state of the first obstacle; analyzing the first operating state to obtain a first predicted operating state of the first obstacle in a predetermined time period; when the first predicted operating state reaches a predetermined threat constraint, adding the first predicted operating state of the first obstacle to the target visual map, and dynamically planning and adjusting the initial moving path according to the target visual map.

[0086] Furthermore, the method for autonomously planning a dynamic path for an unmanned mobile offshore platform further includes:

[0087] Read the predetermined monitoring dimensions, and perform continuous monitoring of the target sea area for the predetermined time period based on the predetermined monitoring dimensions to obtain the target monitoring time sequence, wherein the predetermined monitoring dimensions include environmental dimensions, meteorological dimensions, and obstacle dimensions, and the environmental dimensions include at least ocean terrain features, water flow features, and marine biological features, the meteorological dimensions include at least storm features and wave features, and the obstacle dimensions include at least ship features, buoy features, fishing net features, and coral reef features.

[0088] Furthermore, the method for autonomously planning a dynamic path for an unmanned mobile offshore platform further includes:

[0089] Extract any feature of any dimension in the predetermined monitoring dimension; match any parameter time series corresponding to the arbitrary feature in the target monitoring time series; perform energy density distribution analysis on the arbitrary parameter time series to obtain characteristic parameters corresponding to the highest energy density, which are recorded as arbitrary feature constraints of the arbitrary feature; generate the path planning constraint according to the mapping relationship between the arbitrary feature and the arbitrary feature constraint.

[0090] Furthermore, the method for autonomously planning a dynamic path for an unmanned mobile offshore platform further includes:

[0091] Construct an initial two-dimensional model of the target sea area; extract a first static constraint from the path planning constraints, wherein the first static constraint refers to a first position of a first static obstacle; mark the first position corresponding to the first static obstacle to the initial two-dimensional model to obtain the target visibility graph.

[0092] Furthermore, the method for autonomously planning a dynamic path for an unmanned mobile offshore platform further includes:

[0093] A* algorithm principle is used to perform path planning analysis on the starting point, the target point and the target visibility graph to generate a candidate moving path; a first dynamic constraint in the path planning constraint is extracted, and the first dynamic constraint refers to a first obstruction coefficient of a first dynamic obstacle; the candidate moving path is adjusted based on the first obstruction coefficient to obtain the initial moving path.

[0094] Furthermore, the method for autonomously planning a dynamic path for an unmanned mobile offshore platform further includes:

[0095] The target visible graph is segmented to obtain a target block set; the starting block corresponding to the starting point and the target block corresponding to the target point are matched in sequence in the target block set; based on the A* algorithm principle, the shortest path from the starting block to the target block is obtained, and the shortest path is used as the candidate moving path.

[0096] Furthermore, the method for autonomously planning a dynamic path for an unmanned mobile offshore platform further includes:

[0097] A first adjacent block set of the starting block is formed, and a first block in the first adjacent block set is extracted; whether the first block meets the predetermined block constraint is determined; if so, the first block is used as a first path point, and the shortest path is formed based on the first path point.

[0098] Furthermore, the method for autonomously planning a dynamic path for an unmanned mobile offshore platform further includes:

[0099] According to the first initial position, the first running direction, and the first running speed in the first running state, combined with the length of the predetermined time period, a first predicted position of the first obstacle is calculated; and the first predicted position is used as the first predicted running state of the first obstacle.

[0100] Furthermore, the method for autonomously planning a dynamic path for an unmanned mobile offshore platform further includes:

[0101] According to the initial moving path and the predetermined time period, the target predicted position of the target mobile unmanned platform is obtained; the predetermined threat distance is read, and the predetermined threat range of the target predicted position is obtained according to the predetermined threat distance; the predetermined threat range is used as the predetermined threat constraint for the target mobile unmanned platform when it is at the target predicted position.

[0102] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The dynamic path autonomous planning system for an offshore mobile unmanned platform and the specific examples in the aforementioned embodiment 1 are also applicable to the dynamic path autonomous planning method for an offshore mobile unmanned platform in this embodiment. Through the aforementioned detailed description of the dynamic path autonomous planning system for an offshore mobile unmanned platform, those skilled in the art can clearly understand the dynamic path autonomous planning method for an offshore mobile unmanned platform in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0103] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0104] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. Dynamic path autonomous planning system for offshore mobile unmanned platform, characterized by: include: A target monitoring time sequence obtaining module, wherein the target monitoring time sequence obtaining module is used to perform continuous multi-dimensional monitoring of the target sea area for a predetermined time period to obtain a target monitoring time sequence, wherein the target sea area refers to the activity sea area range of the target mobile unmanned platform; A target visibility graph generation module, the target visibility graph generation module is used to analyze the target monitoring time sequence and generate path planning constraints, and generate a target visibility graph of the target sea area in combination with the path planning constraints; An initial movement path generation module, the initial movement path generation module is used to perform path planning analysis on the target visual map in combination with the starting point and the target point of the target mobile unmanned platform to generate an initial movement path; A real-time obstacle set obtaining module, the real-time obstacle set obtaining module is used to activate the sensor device group deployed on the target mobile unmanned platform, and dynamically monitor the real-time neighborhood of the target mobile unmanned platform through the sensor device group to obtain a real-time obstacle set; a first operating state obtaining module, the first operating state obtaining module being used to extract a first obstacle from the real-time obstacle set, and to monitor and obtain a first operating state of the first obstacle; a first predicted operating state obtaining module, the first predicted operating state obtaining module being used to analyze the first operating state to obtain a first predicted operating state of the first obstacle in a predetermined time period; a dynamic programming adjustment module, the dynamic programming adjustment module being configured to add the first predicted operating state of the first obstacle to the target visibility graph when the first predicted operating state reaches a predetermined threat constraint, and to perform dynamic programming adjustment on the initial moving path according to the target visibility graph; The method comprises: reading a predetermined monitoring dimension, and continuously monitoring the target sea area for the predetermined time period based on the predetermined monitoring dimension to obtain the target monitoring time sequence, wherein the predetermined monitoring dimension comprises an environmental dimension, a meteorological dimension and an obstacle dimension, and the environmental dimension comprises at least marine terrain features, water flow features and marine biological features, the meteorological dimension comprises at least storm features and wave features, and the obstacle dimension comprises at least ship features, buoy features, fishing net features and coral reef features; Among them, include: Extracting any feature of any dimension in the predetermined monitoring dimension; Matching any parameter timing corresponding to the any feature in the target monitoring timing; Performing energy density distribution analysis on the arbitrary parameter time series to obtain a characteristic parameter corresponding to the highest energy density, which is recorded as an arbitrary characteristic constraint of the arbitrary characteristic; The path planning constraint is generated according to a mapping relationship between the arbitrary feature and the arbitrary feature constraint.

2. The dynamic path autonomous planning system for an offshore mobile unmanned platform according to claim 1 is characterized in that: include: Constructing an initial two-dimensional model of the target sea area; Extracting a first static constraint from the path planning constraints, where the first static constraint refers to a first position of a first static obstacle; The first position corresponding to the first static obstacle is marked on the initial two-dimensional model to obtain the target visible graph.

3. The dynamic path autonomous planning system for an offshore mobile unmanned platform according to claim 2 is characterized in that: include: Using the A* algorithm principle to perform path planning analysis on the starting point, the target point and the target visual map to generate a candidate moving path; Extracting a first dynamic constraint from the path planning constraint, where the first dynamic constraint refers to a first obstruction coefficient of a first dynamic obstacle; The candidate moving path is adjusted based on the first obstacle coefficient to obtain the initial moving path.

4. The dynamic path autonomous planning system for an offshore mobile unmanned platform according to claim 3 is characterized in that: include: Blocking the target visible graph to obtain a target block set; sequentially matching the departure block corresponding to the departure point and the target block corresponding to the target point in the target block set; Based on the A* algorithm principle, the shortest path from the starting block to the target block is obtained, and the shortest path is used as the candidate moving path.

5. The dynamic path autonomous planning system for an offshore mobile unmanned platform according to claim 4 is characterized in that: include: Build a first adjacent block set of the starting block, and extract a first block from the first adjacent block set; Determining whether the first block meets a predetermined block constraint; If it is consistent, the first block is used as a first path point, and the shortest path is formed based on the first path point.

6. The dynamic path autonomous planning system for an offshore mobile unmanned platform according to claim 1 is characterized in that: include: Calculate a first predicted position of the first obstacle according to the first initial position, the first running direction, and the first running speed in the first running state and the length of the predetermined time period; The first predicted position is used as the first predicted operating state of the first obstacle.

7. The dynamic path autonomous planning system for an offshore mobile unmanned platform according to claim 1 is characterized in that: include: Obtaining a target predicted position of the target mobile unmanned platform according to the initial moving path and the predetermined time period; Reading a predetermined threat distance, and obtaining a predetermined threat range of the target predicted position according to the predetermined threat distance; The predetermined threat range is used as the predetermined threat constraint on the target mobile unmanned platform when it is at the target predicted position.

8. A method for autonomous dynamic path planning of a mobile unmanned platform at sea, characterized in that: The method is performed by the dynamic path autonomous planning system of the offshore mobile unmanned platform according to any one of claims 1 to 7, comprising: Conducting continuous multi-dimensional monitoring of the target sea area for a predetermined time period to obtain a target monitoring time sequence, wherein the target sea area refers to the activity sea area of ​​the target mobile unmanned platform; Analyzing the target monitoring time sequence and generating path planning constraints, and generating a target visibility graph of the target sea area in combination with the path planning constraints; Performing path planning analysis on the target visual map in combination with the starting point and the target point of the target mobile unmanned platform to generate an initial moving path; Activate the sensor device group deployed on the target mobile unmanned platform, and dynamically monitor the real-time neighborhood of the target mobile unmanned platform through the sensor device group to obtain a real-time obstacle set; Extracting a first obstacle from the real-time obstacle set, and monitoring and obtaining a first operating state of the first obstacle; Analyzing the first operating state to obtain a first predicted operating state of the first obstacle in a predetermined time period; When the first predicted operating state reaches a predetermined threat constraint, the first predicted operating state of the first obstacle is added to the target visibility graph, and the initial moving path is dynamically adjusted according to the target visibility graph.

Images