Search

CN-121995926-A - Multi-attitude unmanned ship autonomous navigation control method based on sail-electric hybrid propulsion

CN121995926ACN 121995926 ACN121995926 ACN 121995926ACN-121995926-A

Abstract

The invention discloses an autonomous navigation control method of a multi-navigation unmanned ship based on sail-electric hybrid propulsion, which comprises the following steps of S1, collecting original physical data, wind energy data and real-time sea state information of ship body equipment in real time, analyzing and preprocessing to obtain a multi-source standardized data packet, S2, constructing an environment situation map based on the multi-source standardized data packet and a joint probability data association algorithm, establishing a wind field model by combining weather forecast and real-time measurement data to generate a continuous risk situation distribution map, S3, receiving and calling an instruction corresponding to a preset task type, and planning a global optimal path and dividing a navigation section through multi-objective optimization by taking the environment risk situation distribution and the real-time state as input to generate an optimal wind sail attack angle instruction. According to the invention, through the constructed three-dimensional wind field model and the generated wind sail attack angle instruction, the whole wind can be dynamically identified and the favorable wind conditions in the sailing process can be fully utilized, and the load of an electric power system is reduced.

Inventors

  • LIANG DONG
  • LI XULONG
  • GAO ZHINING

Assignees

  • 北京海舶无人船科技有限公司

Dates

Publication Date
20260508
Application Date
20260409

Claims (9)

  1. 1. The autonomous navigation control method of the multi-navigation unmanned ship based on sail-electric hybrid propulsion is characterized by comprising the following steps of: S1, acquiring original physical data, wind energy data and real-time sea state information of hull equipment in real time, and performing analysis and preprocessing to obtain a multi-source standardized data packet; S2, constructing an environment situation map based on a multi-source standardized data packet and a joint probability data association algorithm, and constructing a wind field model by combining weather forecast and real-time measurement data to generate a continuous risk situation distribution map and an aviation state suggestion map; s3, receiving and calling a command corresponding to a preset task type, taking environmental risk situation distribution and real-time state as input, planning a global optimal path through multi-objective optimization, dividing a navigation segment, and generating an optimal sail attack angle command; S4, determining the current position of the unmanned ship based on the thrust control module and the wind sail angle instruction, and calculating and generating a control instruction set for left and right propulsion motors of the unmanned ship; S5, translating and encoding the control instruction set into a bottom driving command corresponding to the specific hardware interface through a built-in device registry and a protocol adaptation mechanism.
  2. 2. The autonomous navigation control method of the multi-navigation unmanned ship based on sail-electric hybrid propulsion according to claim 1, wherein the analyzing and preprocessing are used for endowing the original physical data, wind energy data and real-time sea state information with uniform time stamps for time synchronization, and after the time synchronization is completed, the payload and the data type identifier comprising the uniform time stamps, the data source equipment ID and the data serial number are packaged according to a predefined data structure to obtain the multi-source standardized data packet.
  3. 3. The autonomous navigation control method of the multi-attitude unmanned ship based on sail-electric hybrid propulsion according to claim 1, wherein the specific content of S2 is as follows: 1) Building and fusing an environment map; 2) Wind field modeling and energy assessment; 3) Constructing an avionic assessment model; 4) Comprehensive risk assessment; wherein the model for estimating the attitude is capable of calculating an attitude adaptability coefficient The calculation formula is as follows: Wherein, the Is in a state of navigation The corresponding inherent cost of the device is that, The power consumption required for the state switch, As the weight coefficient of the light-emitting diode, Is a preset task demand factor.
  4. 4. The autonomous navigation control method of the multi-attitude unmanned ship based on sail-electric hybrid propulsion according to claim 3, wherein the environment map construction and fusion comprises a data layer and a semantic layer; In the data layer, a joint probability data association algorithm and an extended Kalman filtering algorithm are adopted to uniformly convert the target centers of marine navigation radars and laser radar point clouds, the point trace reported by millimeter wave radars and the central coordinate of a visual detection frame into a body coordinate system taking an unmanned ship as the center; Calculating a correlation probability matrix between all the measurements and all the prediction states by a joint probability data correlation algorithm, and selecting an optimal measurement-track pairing; and in the semantic layer, carrying out association matching on each unmanned ship track output by the data layer and AIS broadcast information of nearby ships, endowing the unique identification code, the ship shape, the size and the sailing state of the ship MMSI with corresponding unmanned ship targets, and integrating and constructing a layered environment situation map.
  5. 5. The autonomous navigation control method of a multi-attitude unmanned ship based on hybrid propulsion of sail-electric power according to claim 3, wherein the comprehensive risk assessment requires discretizing a navigation area into grid cells, expanding all high-risk grids outwards by a certain radius by using a morphological expansion algorithm to form danger areas, calculating repulsive potential from each danger area and attractive potential from a target point, introducing a navigation state adaptability coefficient, taking a navigation mode as a decision variable into a risk state value, calculating the risk state value The calculation formula of (2) is as follows: Wherein, the For the weight coefficient of the attraction potential, Is a grid position point , Is the attractive potential of the target point, In order to set the coefficient to be the preset value, Is the first The parameters of the range of influence of the repulsive potential field, Is the first The center coordinates of the inflated high risk grid cells, M is the total number of all high risk grid cells, The corresponding coefficient of the state of the flight, Is the aero state adaptability coefficient.
  6. 6. The autonomous navigation control method of the multi-attitude unmanned ship based on sail-electric hybrid propulsion according to claim 1, wherein the specific step of S3 is as follows: S3.1, a planning algorithm takes a risk situation distribution diagram and key states in a multi-source standardized data packet as input, and takes a preset task type as a constraint boundary to construct a real-time target database, and a risk situation value is searched Minimizing the weighted sum of the accumulated risk cost, the sailing distance and the expected time of the grid through which the path passes, and determining the optimal path; s3.2, dividing an optimal path into continuous navigation segments according to a certain distance or a key turning point, and inquiring the wind speed and the wind direction of the region where the navigation segments are located; S3.3, according to the heading of the unmanned ship in the current voyage And predicting wind direction Calculating the relative wind direction angle And determining a sail angle instruction of the leg based on a pre-programmed sail angle-thrust coefficient mapping table of experimental data.
  7. 7. The autonomous navigation control method of the multi-attitude unmanned ship based on sail-electric hybrid propulsion according to claim 1, wherein the thrust control module is constructed by adopting a path tracking algorithm and is used for calculating the steering angle of the unmanned ship by combining a real-time heading h Steering angle The calculation formula of (2) is as follows: Wherein, the The real-time position of the unmanned ship is With the target point If b is positive, it indicates that the unmanned ship needs to turn right, and if b is negative, it indicates that the unmanned ship needs to turn left.
  8. 8. The autonomous navigation control method of the multi-attitude unmanned ship based on sail-electric hybrid propulsion according to claim 1, wherein the thrust control module can determine a speed command s according to the current attitude, fuse a steering command with the speed command s, generate independent control amounts of left and right propulsion motors, and pulse width of the left motor And the pulse width of the right motor The calculation formula of (2) is as follows: Wherein, the To adjust the coefficients.
  9. 9. The autonomous navigation control system of the multi-navigation unmanned ship based on sail-electric hybrid propulsion is characterized by comprising a perception layer, an interaction layer, a platform abstraction layer, a cognition layer, a decision layer and an execution layer; The sensing layer is used for collecting original physical data, wind energy data and real-time sea state information of the ship body equipment in real time, and analyzing and preprocessing the original physical data, the wind energy data and the real-time sea state information; The interaction layer is used for receiving various instructions from a command center, a remote base station or a handheld terminal, and can receive a multi-ship cooperative instruction sent by a cooperative component and a third party through a formation interaction interface; the decision layer receives various control instructions and environment risk information, plans a global optimal path through multi-objective optimization, divides the navigation segments, and generates an optimal sail control instruction according to real-time wind energy and hull states of each navigation segment; The cognitive layer can construct an environment situation map through algorithms such as joint probability data association and extended Kalman filtering, and can generate a global risk situation distribution map by combining wind field modeling data; The platform abstraction layer can automatically identify and connect all shipborne hardware and equipment when the unmanned ship is started, and can translate and encode a control instruction issued by the decision layer into a driving command which is suitable for a bottom layer concrete hardware interface; And the execution layer is internally provided with a thrust control module which is used for receiving and executing the concrete control instruction issued by the platform abstraction layer.

Description

Multi-attitude unmanned ship autonomous navigation control method based on sail-electric hybrid propulsion Technical Field The invention relates to the technical field of unmanned ship data processing, in particular to an autonomous navigation control method of a multi-navigation unmanned ship based on sail-electric hybrid propulsion. Background Along with the deep fusion of shipping intellectualization and greening transformation, the unmanned surface vessel (Unmanned Surface Vehicle, USV) has shown remarkable technical value and economic benefit in a plurality of high-end application scenes such as marine mapping, security patrol, environmental monitoring, scientific research and exploration, emergency rescue and the like by virtue of the advantages of strong autonomous operation capability, flexible deployment, adaptability to complex water area environment and the like. By integrating multisource sensors, intelligent decision and high-precision control technologies, unmanned ships are gradually pushing traditional water operation modes to develop towards unmanned, synergistic and sustainable directions, and become key nodes for constructing intelligent ocean systems. However, while unmanned ship technology is rapidly evolving, its energy autonomy and environmental adaptation remain core bottlenecks that restrict its long-term, remote, reliable operation. At present, the main stream unmanned ship adopts a pure electric propulsion scheme, is limited by the current battery energy density, has the outstanding problems of short endurance mileage and limited task cycle, and particularly when working in open sea areas far from a supply point, the working radius and the task elasticity are further limited due to difficult energy supply. In addition, the traditional energy management strategies are mostly based on static models or preset rules, are difficult to adapt to the multi-change environmental disturbance such as offshore wind, wave, current and the like, cannot effectively utilize the offshore rich renewable wind energy resources, and therefore the overall energy efficiency is low, and the energy utilization rate still has a large improvement space. In the aspect of navigation control, the existing unmanned ship autonomous system is mostly dependent on an electric propulsion unit to track and stabilize the gesture, when facing complex sea conditions, environmental disturbance can introduce model uncertainty, external interference and state estimation errors, so that the robustness of a traditional control algorithm is reduced, the track tracking precision is reduced, and even instability risks are caused. Especially in the switching process of multiple sails (such as water surface, semi-submerged and full-submerged), the dynamic characteristics of the system are severely changed, the existing control architecture often lacks modeling and comprehensive optimizing capability for a hybrid propulsion (sail+electric power) system, and the collaborative optimization of energy distribution, risk avoidance and motion control under different sails is difficult to realize, so that the overall intelligent level and task completion reliability of the unmanned ship in dynamic sea conditions are limited. Therefore, there is a need for an autonomous navigation control method of a multi-navigation unmanned ship based on sail-electric hybrid propulsion, so as to break through the limitation of the prior art in terms of energy sustainability, environmental adaptability and control robustness, and promote the development of the next-generation long-endurance, high-autonomy and high-toughness intelligent unmanned ship technology. The method solves the problem of coordination between sustainable energy, environmental adaptation and motion control systematically. Disclosure of Invention Therefore, the invention provides an autonomous navigation control method of a multi-navigation unmanned ship based on sail-electric hybrid propulsion, which aims to solve the problems in the prior art. In order to achieve the above object, the present invention provides the following technical solutions: In a first aspect, a method for controlling autonomous sailing of a multi-propulsion unmanned ship based on hybrid sail-electric propulsion, comprising the steps of: S1, acquiring original physical data, wind energy data and real-time sea state information of hull equipment in real time, and performing analysis and preprocessing to obtain a multi-source standardized data packet; S2, constructing an environment situation map based on a multi-source standardized data packet and a joint probability data association algorithm, and constructing a wind field model by combining weather forecast and real-time measurement data to generate a continuous risk situation distribution map and an aviation state suggestion map; s3, receiving and calling a command corresponding to a preset task type, taking environmental risk situation distribution and real-ti