CN110737267A - Multi-objective optimization method for unmanned ships and intelligent comprehensive management and control system for unmanned ships - Google Patents
Multi-objective optimization method for unmanned ships and intelligent comprehensive management and control system for unmanned ships Download PDFInfo
- Publication number
- CN110737267A CN110737267A CN201910944783.1A CN201910944783A CN110737267A CN 110737267 A CN110737267 A CN 110737267A CN 201910944783 A CN201910944783 A CN 201910944783A CN 110737267 A CN110737267 A CN 110737267A
- Authority
- CN
- China
- Prior art keywords
- navigation
- ship
- data
- unmanned ship
- unmanned
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 46
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000004458 analytical method Methods 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000004891 communication Methods 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 11
- 238000012360 testing method Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 claims description 6
- 238000007405 data analysis Methods 0.000 claims description 4
- 238000010295 mobile communication Methods 0.000 claims description 4
- 230000002068 genetic effect Effects 0.000 claims description 3
- 230000014759 maintenance of location Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims 1
- 238000003786 synthesis reaction Methods 0.000 claims 1
- 238000007726 management method Methods 0.000 abstract description 13
- 230000019771 cognition Effects 0.000 abstract description 2
- 230000006872 improvement Effects 0.000 description 4
- 230000010354 integration Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000035772 mutation Effects 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 206010063385 Intellectualisation Diseases 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/0206—Control of position or course in two dimensions specially adapted to water vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention provides a multi-target optimization method of an unmanned ship, which comprises the steps of S1 obtaining original data, S2 analyzing and processing the original data to obtain unmanned ship control data in the original data, S3 inputting the unmanned ship control data into a navigation scene analysis model and outputting influence values of navigation conditions on ship navigation, wherein the navigation scene analysis model is a long-short term memory network (LSTM) model trained in advance, S4 inputting the influence values of the navigation conditions on the ship navigation into a multi-target optimization model to obtain at least optimization schemes, controlling and optimizing the unmanned ship according to schemes in at least optimization schemes, and replacing a person to carry out scene cognition to obtain an unmanned ship intelligent equipment control scheme for realizing collaborative planning, collaborative management and collaborative control among intelligent equipment.
Description
Technical Field
The invention relates to the technical field of unmanned ship intelligent systems, in particular to an unmanned ship multi-target optimization method and an unmanned ship intelligent comprehensive management and control system.
Background
75% -96% of marine accidents of ships are caused by human factors, the personnel cost of marine operation of the ships is high, the demand gap of high-grade crews is large, the proportion of the life working space of the crews in the ship space is large, the marine transportation efficiency is reduced, the unmanned ship effectively solves the existing problems, is an important component of the marine intelligent traffic strategy in China, and the autonomous navigation technology is which is the key technology of the unmanned ship.
The intelligent energy and efficiency management system for the ship disclosed in 2018, 1 month and 9 days of China patent application CN107563576A evaluates data acquired in real time, adjusts the running condition of the ship for improving energy efficiency on the basis of actual data, and therefore the energy and efficiency management of the ship is more scientific and effective.
In order to consider the cooperative work among the intelligent systems of the unmanned ship, a cooperative multi-objective optimization method and a cooperative multi-objective optimization system for each intelligent device of unmanned ships are urgently needed.
Disclosure of Invention
() problems to be solved
In order to solve the problems in the prior art, the invention provides unmanned ship multi-target optimization methods, which replace the cognition of people on navigation situations, acquire unmanned ship intelligent equipment control schemes for collaborative planning, collaborative management and collaborative control among intelligent equipment, and improve the intellectualization of unmanned ships, and also provides unmanned ship intelligent comprehensive control systems, so that the comprehensive control and comprehensive command of the intelligent equipment of unmanned ships and the mutual collaborative work among the intelligent equipment are realized.
(II) technical scheme
In order to achieve the purpose, the invention adopts the main technical scheme that:
unmanned ship multi-objective optimization method, comprising the following steps:
s1, acquiring original data of the unmanned ship in a preset time period during driving by the unmanned ship; the original data are data monitored by each intelligent monitoring module on the unmanned ship and comprise ship related parameters, ship external environment data, ship self state data and shore-based center remote control data.
S2: and analyzing and processing the original data by the unmanned ship to obtain unmanned ship control data in the original data.
S3, the unmanned ship inputs the unmanned ship control data into a navigation scene analysis model and outputs the influence value of each navigation condition on the ship navigation; the navigation conditions comprise navigation areas, meteorological conditions, the density of navigation ships around the navigation conditions and the self state of the ships, and the navigation scene analysis model is a long-short term memory network (LSTM) model trained in advance.
And S4, inputting the influence values of the unmanned ship on ship navigation according to each navigation condition into the multi-objective optimization model, acquiring at least optimization schemes, and optimizing the control of the unmanned ship according to schemes in at least optimization schemes.
As an improvement of the method of the present invention, the update equation of the previously trained LSTM model includes:
ft＝σ(Wf·[ht-1,xt]+bf)
it＝σ(Wi·[ht-1,xt]+bi)
ot＝σ(Wo·[ht-1,xt]+bo)
ht＝ot*tanh(ct)
wherein f istForget , W indicating time tfWeight matrix of forgetting , ht-1Hidden layer at time t-1, xtRepresenting said original data, bfBias term for forgetting , sigma represents mapping value to sigmoid layer between (0, 1), 1 represents complete retention, 0 represents complete forgetting, itRepresenting the input , W at time tiIs a weight matrix of the input , biAs an offset term of input otRepresenting the output , W at time toWeight matrix as output , boAn offset term that is output ;representing the input of the current node at time t, WcAs a weight matrix of inputs, bcIs an input bias term; c. CtThe output memory information at the t-th time is shown; h istIndicating a hidden state at time t.
modifications of the method of the present invention, before step S1, further comprising:
a1, constructing an initial LSTM model.
And A2, obtaining a training sailing condition sample and a testing sailing condition sample.
A3, inputting the training navigation condition sample into the initial LSTM model.
A4, judging whether the output navigation condition of the initial LSTM model and the test navigation condition sample are , if , determining that the initial LSTM model is the previously trained LSTM model, if not , optimizing the parameters of the initial LSTM model, and repeating the steps A3 and A4.
As improvements of the method, the training sailing condition samples are input into the initial LSTM model, and the method comprises the step of optimizing the hyper-parameters of the initial LSTM model by adopting a Nadam algorithm.
As improvements to the method of the present invention, the objective function of the multi-objective optimization model includes:
wherein F1 represents the safest sailing, F2 represents the most stable sailing, F3 represents the most economical sailing, and F4 represents the most environment-friendly sailing.
The constraint conditions of the multi-objective optimization model comprise:
the constraint condition 1 represents the influence degree of each navigation condition on the navigation of the ship and the sequence of the influence degree, and the constraint condition 2 represents that the control of the intelligent equipment on the ship cannot exceed the threshold value of the intelligent equipment.
As improvements of the method, the method for solving the multi-objective optimization model is to use the non-dominated sorting genetic algorithm NSGA-II with the elite strategy to carry out the solution.
unmanned ship intelligent integrated management and control system includes:
the system comprises an information acquisition module, a navigation situation analysis module, a decision command module and an information acquisition module, wherein the information acquisition module is used for acquiring original data in a preset time period of an unmanned ship in driving, the original data are data monitored by monitoring modules on intelligent devices on the unmanned ship and comprise ship related parameters, ship external environment data, ship self-state data and shore-based center remote control data, the data analysis module is used for analyzing and processing the original data to acquire unmanned ship control data in the original data, the navigation situation analysis module is used for inputting the unmanned ship control data into a pre-trained LSTM model and outputting influence values of navigation conditions on ship navigation, the navigation conditions comprise navigation areas, meteorological conditions, ship density around the unmanned ship density and the ship self-state, the decision command module is used for inputting the influence values of the navigation conditions on the ship navigation into a multi-target optimization model to acquire at least optimization schemes, and the unmanned ship is controlled and optimized according to schemes in the optimization schemes.
Preferably, the information acquisition module acquires ship related parameters, ship external environment data and ship self state data through a local area network, and the information acquisition module acquires shore-based center remote control data through a communication system; the communication system comprises a mobile communication module, a satellite communication module and a data encryption module; the mobile communication module is used for acquiring shore-based center remote control data by the information acquisition module when the distance between the unmanned ship and the shore-based center is smaller than a preset value; the satellite communication module is used for acquiring shore-based central remote control data by the information acquisition module when the distance between the unmanned ship and the shore-based is larger than a preset value; the data encryption module is used for carrying out safe encryption operation in the process of acquiring shore-based center remote control data by the information acquisition module.
(III) advantageous effects
The invention has the beneficial effects that:
1. the collaborative multi-objective optimization method for each intelligent device of the unmanned ship can consider autonomous navigation from the whole layer of the unmanned ship. The method comprises the steps of firstly, from a macroscopic angle, recognizing navigation situations by a substitute person, and then comprehensively commanding each intelligent device on the ship from a microscopic angle, so that more accurate and effective execution instructions are issued to each intelligent device, collaborative planning, collaborative management and collaborative control among the intelligent devices are realized, and a more comprehensive and reliable unmanned ship autonomous navigation overall decision-making scheme is formed.
2. The method organically combines the LSTM model and the NSGA-II algorithm, the unmanned ship navigation situation obtained through the LSTM model is more in line with the understanding of the artificial driving ship on the ship navigation state, and the ship navigation and decision making are more reasonable through the unmanned ship intelligent equipment control scheme obtained through the NSGA-II algorithm.
3. The cooperative multi-target optimization system for the intelligent devices of the unmanned ship organically combines the intelligent devices of the unmanned ship into pieces of whole bodies, can realize comprehensive management and control, comprehensive command and mutual calling and cooperative work among the intelligent devices of the unmanned ship, improves the intellectualization of the unmanned ship to new heights, and further ensures the safety, economy and environmental protection of autonomous navigation of the unmanned ship.
4. Different communication modes are adopted aiming at different intelligent devices and navigation scenes of the unmanned ship, so that communication is carried out between the intelligent devices of the unmanned ship and between the unmanned ship and a shore base in the most stable and economic mode all the time, and the economy and the reliability of communication of the unmanned ship are improved.
Drawings
The invention is described with the aid of the following figures:
FIG. 1 is a flow chart of a multi-objective optimization method for an unmanned ship in accordance with an embodiment of the present invention;
FIG. 2 is a schematic illustration of the pre-processing of other vessel speed data within a predetermined time period in accordance with an embodiment of the present invention;
FIG. 3 is a flow chart of the NSGA-II algorithm in accordance with the present invention;
fig. 4 is a structural diagram of an intelligent integrated management and control system for an unmanned ship according to an embodiment of the present invention.
Detailed Description
For the purpose of better explaining the present invention and to facilitate understanding, the present invention will be described in detail by way of specific embodiments with reference to the accompanying drawings.
The invention provides an unmanned ship multi-objective optimization method, which comprises the following steps as shown in figure 1:
step S1, the unmanned ship obtains original data in a preset time period of the unmanned ship in driving; the original data are data monitored by each intelligent monitoring module on the unmanned ship and comprise ship related parameters, ship external environment data, ship self state data and shore-based center remote control data.
Specifically, the relevant parameters of the ship comprise parameters such as the ship length and the draught, real-time ship external environment data are information acquired by intelligent equipment such as a radar and a depth finder for detecting external environments, and real-time ship self state data are information acquired by intelligent equipment such as a temperature sensor and an oil temperature and oil pressure sensor for detecting the state of the ship.
And step S2, analyzing and processing the original data by the unmanned ship to obtain unmanned ship control data in the original data.
Specifically, the analysis processing of the raw data includes preprocessing operations and data integration. The preprocessing operation comprises the operations of extracting complex data, cleaning dirty data, supplementing lost data, removing redundant data and the like to obtain preprocessed data. The data integration operation is specifically operations such as processing, classifying, merging, calculating, sorting, converting and the like on the preprocessed data based on technologies such as data mining and the like; and extracting information hidden in the data from disordered and difficult-to-understand data to obtain data which directly plays a role in the cooperative control of the unmanned ship intelligent equipment.
In the embodiment of the invention, an unmanned ship acquires data in a preset time period monitored by AIS equipment on the unmanned ship during driving, wherein the data monitored by the AIS equipment comprises other ship speed data, and the noise data with sudden change of speed possibly exists in the speed data and interferes with the overall navigation judgment of the unmanned ship, so that the noise data needs to be processed before the data is formally applied.
In the specific embodiment of the invention , the data within the preset time period monitored by the AIS equipment is processed, classified, merged, calculated, sequenced, converted, etc. to obtain the relative motion information and the relative motion change rule information (unmanned ship control data) such as the relative azimuth, relative navigational speed, relative heading, etc. between the ship and other ships.
Step S3, the unmanned ship inputs the unmanned ship control data into a navigation scene analysis model, and outputs the influence value of each navigation condition on the ship navigation; the navigation conditions comprise navigation areas, meteorological conditions, the density of navigation ships around the ship and the state of the ship, and the navigation scene analysis model is a long-short term memory network (LSTM) model trained in advance.
The invention analyzes the navigation scene mainly through a navigation AREA (AREA), meteorological conditions (WEATHER), self surrounding navigation ship DENSITY (DENSITY) and ship self STATE (STATE) to judge the influence degree of each navigation condition on the ship navigation.
Specifically, the update equations of the pre-trained LSTM model include:
ft＝σ(Wf·[ht-1,xt]+bf)
it＝σ(Wi·[ht-1,xt]+bi)
ot＝σ(Wo·[ht-1,xt]+bo)
ht＝ot*tanh(ct)
wherein f istForget , W indicating time tfWeight matrix of forgetting , ht-1Hidden layer at time t-1, xtRepresenting said original data, bfBias term for forgetting , sigma represents mapping value to sigmoid layer between (0, 1), 1 represents complete retention, 0 represents complete forgetting, itRepresenting the input , W at time tiIs a weight matrix of the input , biAs an offset term of input otRepresenting the output , W at time toWeight matrix as output , boAn offset term that is output ;
representing the input of the current node at time t, WcAs a weight matrix of inputs, bcIs an input bias term; c. CtThe output memory information at the t-th time is shown; h istIndicating a hidden state at time t.
In the embodiment of the present invention, the training of the LSTM model before step S1 includes the following steps:
a1, constructing an initial LSTM model.
And A2, obtaining a training sailing condition sample and a testing sailing condition sample.
A3, inputting the training navigation condition sample into the initial LSTM model.
A4, judging whether the output navigation condition of the initial LSTM model and the test navigation condition sample are , if , determining that the initial LSTM model is the previously trained LSTM model, if not , optimizing the parameters of the initial LSTM model, and repeating the steps A3 and A4.
In step a2, the training navigation condition samples are a plurality of unmanned ship control data with the same time step, and the test navigation condition samples are navigation areas, meteorological conditions, density of navigation ships around the test navigation condition samples and states of the ships corresponding to the unmanned ship control data with each time steps.
Preferably, in step a3, inputting the training navigation condition samples into the initial LSTM model, including optimizing the hyper-parameters of the initial LSTM model, such as the number of layers, the number of neurons in each layer, and the learning rate, by using the Nadam algorithm.
And S4, inputting the influence values of the unmanned ship on ship navigation according to each navigation condition into a multi-target optimization model, acquiring at least optimization schemes, and optimizing the control of the unmanned ship according to schemes in at least optimization schemes.
Specifically, constructing the objective function of the multi-objective optimization model comprises the following steps:
wherein F1 represents the safest sailing, F2 represents the most stable sailing, F3 represents the most economical sailing, and F4 represents the most environment-friendly sailing.
Specifically, the constraint conditions for constructing the multi-objective optimization model comprise:
the constraint condition 1 represents the influence degree of each navigation condition on the navigation of the ship and the sequence of the influence degree, and the constraint condition 2 represents that the control of the intelligent equipment on the ship cannot exceed the threshold value of the intelligent equipment.
Specifically, the method for solving the multi-objective optimization model is to use a non-dominated sorting genetic algorithm NSGA-II with elite strategy to solve, as shown in fig. 3, and comprises the following steps:
b1, inputting the predefined iteration number, the population size, the cross probability, the counter and the variation probability.
B2, coding the unmanned ship intelligent equipment control scheme (n) to generate an initial population; and performing non-dominant sorting on the father population; the rank and congestion distance values are assigned to the individuals based on the multi-objective function and the constraint conditions in step S4.
B3, selecting binary championship match for the population, and crossing and mutation to obtain sub-population (n). in embodiments of the invention, the crossing operation uses SBX (analog binary crossing) and the mutation operation uses polynomial mutation.
B4, judging whether the specified iteration times are reached, if so, outputting an optimal pareto solution set, otherwise, adding 1 to the counter, and continuing to execute the step B5.
And B5, merging the parent population and the child population to obtain a combined population (2 n).
And B6, performing non-dominant sorting on the combined population, and selecting a new parent population through congestion degree calculation and elite strategy. And jumping to the step B3 for circulation.
Specifically, the congestion degree calculation formula is as follows:
wherein n is the scale of the unmanned ship intelligent equipment control scheme,
the j-th function value representing the i +1 point,
the j-th function value at point i-1 is shown.
In conclusion, the collaborative multi-objective optimization method for each intelligent device of the unmanned ship can consider autonomous navigation from the whole layer of the unmanned ship. The method comprises the steps of firstly, from a macroscopic angle, recognizing navigation scenes by a human instead, and then comprehensively commanding each intelligent device on the ship from a microscopic angle, so that more accurate and effective execution instructions are issued to each intelligent device, and the intelligent devices are subjected to collaborative planning, collaborative management and collaborative control to form a more comprehensive and reliable unmanned ship autonomous navigation integral decision scheme. The method organically combines the LSTM model and the NSGA-II algorithm, the unmanned ship navigation situation obtained through the LSTM model is more in line with the understanding of the artificial driving ship on the ship navigation state, and the ship navigation and decision making are more reasonable through the unmanned ship intelligent equipment control scheme obtained through the NSGA-II algorithm.
The invention also provides unmanned ship intelligent comprehensive control systems, which comprise an information acquisition module, a data analysis module, a navigation situation analysis module, a decision command module and an optimization module, wherein the information acquisition module is used for acquiring original data in a preset time period of an unmanned ship in driving, the original data are data monitored by monitoring modules on intelligent equipment on the unmanned ship and comprise ship related parameters, ship external environment data, ship self-state data and shore-based center remote control data, the data analysis module is used for analyzing and processing the original data to acquire unmanned ship control data in the original data, the navigation situation analysis module is used for inputting the unmanned ship control data into a pre-trained LSTM model and outputting influence values of each navigation condition on ship navigation, the navigation condition comprises a navigation area, a meteorological condition, ship navigation density around the unmanned ship and the ship self-state, the decision command module is used for inputting the unmanned ship navigation influence values into a multi-target optimization model according to each navigation condition, acquiring at least optimization schemes, and controlling the unmanned ship according to at least schemes in optimization schemes.
It should be noted that the intelligent devices referred to in the present invention include an intelligent navigation device, an intelligent energy efficiency management device, an intelligent cabin device, an intelligent cargo management device, an intelligent ship body device, and the like.
Compared with the traditional unmanned ship single intelligent equipment and single module research, the system can organically combine the intelligent equipment of the unmanned ship into integers, can realize the comprehensive control, comprehensive command and mutual calling and cooperative work among the intelligent equipment of the unmanned ship, and can intelligently promote the unmanned ship to new heights, and the step is carried out to ensure the safety, economy and environmental protection of the autonomous navigation of the unmanned ship.
The system can integrate data and applications of the unmanned ship, and compared with the traditional data integration, the system improves the integration of intelligent equipment of the unmanned ship to a high-level stage, provides more comprehensive system decision schemes for the unmanned ship, preferably, an information acquisition module acquires relevant ship parameters, ship external environment data and ship self state data through a local area network, and an information acquisition module acquires shore-based center remote control data through a communication system.
, different data are stored in different storage modes for the original data collected by the information collection module, wherein relevant parameters of the ship are stored in a static database, the size of the database is fixed, real-time ship external environment data, real-time ship self state data and shore-based center remote control data are stored in a memory and a physical dynamic database, unmanned ship navigation historical data, redundant real-time data and redundant historical data are stored in a physical dynamic database, the size of the database can be dynamically changed along with the data volume, the static database and the dynamic database are all built in a server, the capacity of the hard disk of the server is enough, data of at least inspection periods can be stored, data of the same need to be backed up and stored in different hard disks of the same server and different servers, and the accuracy, integrity and usability of the data are guaranteed.
It should be understood that the above description of specific embodiments of the invention is only for the purpose of illustrating the technical lines and features of the invention, and is intended to enable those skilled in the art to understand the content of the invention and to implement the invention, but the invention is not limited to the above specific embodiments. It is intended that all such changes and modifications as fall within the scope of the appended claims be embraced therein.
Claims (8)
- The multi-objective optimization method for the unmanned ships is characterized by comprising the following steps:step S1, the unmanned ship obtains original data in a preset time period of the unmanned ship in driving; the original data are data monitored by each intelligent monitoring module on the unmanned ship and comprise ship related parameters, ship external environment data, ship self state data and shore-based center remote control data;step S2: the unmanned ship analyzes and processes the original data to obtain unmanned ship control data in the original data;step S3, the unmanned ship inputs the unmanned ship control data into a navigation scene analysis model, and outputs the influence value of each navigation condition on the ship navigation; the navigation condition comprises a navigation area, meteorological conditions, the density of navigation ships around the navigation condition and the state of the ships, and the navigation scene analysis model is a pre-trained long-short term memory network (LSTM) model;and S4, inputting the influence values of the unmanned ship on the ship navigation according to the navigation conditions into a multi-objective optimization model, acquiring at least optimization schemes, and optimizing the control of the unmanned ship according to schemes in at least optimization schemes.
- 2. The method of claim 1, wherein the update equations of the pre-trained LSTM model comprise:ft＝σ(Wf·[ht-1，xt]+bf)it＝σ(Wi·[ht-1，xt]+bi)ot＝σ(Wo·[ht-1，xt]+bo)
ht＝ot*tanh(ct)wherein f istForget , W indicating time tfWeight matrix of forgetting , ht-1Hidden layer at time t-1, xtRepresenting said original data, bfBias term for forgetting , sigma represents mapping value to sigmoid layer between (0, 1), 1 represents complete retention, 0 represents complete forgetting, itInput , W representing time tiIs input into weight matrix, biAs an offset term of input otOutput , W representing time toWeight matrix as output , boAn offset term that is output ;representing the input of the current node at time t, WcAs a weight matrix of inputs, bcIs an input bias term; c. CtThe output memory information at the t-th time is shown; h istIndicating a hidden state at time t.
- 3. The method according to claim 2, wherein before step S1, further comprising:a1, constructing an initial LSTM model;a2, obtaining a training navigation condition sample and a test navigation condition sample;step A3, inputting the training navigation condition sample into the initial LSTM model;and A4, judging whether the output navigation condition of the initial LSTM model and the test navigation condition sample are , if , determining that the initial LSTM model is a pre-trained LSTM model, if is not , optimizing parameters of the initial LSTM model, and repeating the steps A3 and A4.
- 4. The method of claim 3, wherein said inputting the training voyage condition samples into the initial LSTM model comprises:and optimizing the hyper-parameters of the initial LSTM model by adopting a Nadam algorithm.
- 5. The method of claim 1, wherein the objective function of the multi-objective optimization model comprises:
wherein F1 represents the safest navigation, F2 represents the most stable navigation, F3 represents the most economical navigation, and F4 represents the most environment-friendly navigation;the constraint conditions of the multi-objective optimization model comprise:
the constraint condition 1 represents the influence degree of each navigation condition on the navigation of the ship and the sequence of the influence degree, and the constraint condition 2 represents that the control of the intelligent equipment on the ship cannot exceed the threshold value of the intelligent equipment. - 6. The method of claim 5, wherein the multi-objective optimization model is solved using the non-dominated ranking genetic algorithm with elite strategy, NSGA-II.
- 7, kinds of unmanned ship intelligent synthesis management and control system, its characterized in that includes:the system comprises an information acquisition module, a data acquisition module and a data processing module, wherein the information acquisition module is used for acquiring original data in a preset time period of the unmanned ship in driving, and the original data are data monitored by monitoring modules on intelligent equipment on the unmanned ship and comprise ship related parameters, ship external environment data, ship self state data and shore-based center remote control data;the data analysis module is used for analyzing and processing the original data to obtain unmanned ship control data in the original data;the navigation scene analysis module is used for inputting the unmanned ship control data into a pre-trained LSTM model and outputting the influence value of each navigation condition on the ship navigation; the navigation conditions comprise navigation areas, meteorological conditions, the density of the navigation ships around the ships and the self states of the ships;the decision command module is used for inputting the influence values of the navigation conditions on the navigation of the ship into the multi-objective optimization model to obtain at least optimization schemes and optimizing the control of the unmanned ship according to schemes in at least optimization schemes;the information acquisition module is further used for sending the optimization scheme and the original data to each intelligent device.
- 8. The system of claim 7, wherein the information acquisition module acquires the ship-related parameters, the ship external environment data and the ship self-state data through a local area network, and the information acquisition module acquires the shore-based central remote control data through a communication system;the communication system comprises a mobile communication module, a satellite communication module and a data encryption module; the mobile communication module is used for acquiring shore-based center remote control data by the information acquisition module when the distance between the unmanned ship and the shore-based center is smaller than a preset value; the satellite communication module is used for acquiring shore-based center remote control data by the information acquisition module when the distance between the unmanned ship and the shore-based center is larger than a preset value; the data encryption module is used for carrying out safe encryption operation in the process of acquiring the shore-based center remote control data by the information acquisition module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910944783.1A CN110737267A (en) | 2019-09-30 | 2019-09-30 | Multi-objective optimization method for unmanned ships and intelligent comprehensive management and control system for unmanned ships |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910944783.1A CN110737267A (en) | 2019-09-30 | 2019-09-30 | Multi-objective optimization method for unmanned ships and intelligent comprehensive management and control system for unmanned ships |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110737267A true CN110737267A (en) | 2020-01-31 |
Family
ID=69268486
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910944783.1A Pending CN110737267A (en) | 2019-09-30 | 2019-09-30 | Multi-objective optimization method for unmanned ships and intelligent comprehensive management and control system for unmanned ships |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110737267A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111610789A (en) * | 2020-07-01 | 2020-09-01 | 上海船舶研究设计院（中国船舶工业集团公司第六0四研究院） | Ship comprehensive management and control system and intelligent ship |
| CN112052984A (en) * | 2020-08-07 | 2020-12-08 | 智慧航海（青岛）科技有限公司 | Intelligent ship collision avoidance method and system |
| CN113325896A (en) * | 2021-05-31 | 2021-08-31 | 浙江工业大学 | Multi-target temperature optimization control method of intelligent retail machine |
| CN114185328A (en) * | 2021-12-07 | 2022-03-15 | 天翼物联科技有限公司 | Unmanned ship control method, device, equipment and medium based on sea condition self-adaption |
| CN114462444A (en) * | 2021-12-21 | 2022-05-10 | 上海中船船舶设计技术国家工程研究中心有限公司 | Navigation state identification method and system for polar region inspection ship |
| CN116395105A (en) * | 2023-06-06 | 2023-07-07 | 海云联科技(苏州)有限公司 | Automatic lifting compensation method and system for unmanned ship |
| CN116911560A (en) * | 2023-07-27 | 2023-10-20 | 中国舰船研究设计中心 | Ship task system decision planning method based on multi-objective optimization |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708406A (en) * | 2012-05-10 | 2012-10-03 | 湖北省电力公司 | Scheduling graph optimizing method based on multi-target genetic algorithm |
| CN102831479A (en) * | 2012-07-29 | 2012-12-19 | 江苏大学 | NSGA-II (non-domination sequencing genetic algorithm) multi-objective optimization searching method |
| CN107563576A (en) * | 2017-10-14 | 2018-01-09 | 连云港杰瑞深软科技有限公司 | A kind of ship intelligence energy efficiency management system |
| CN108319293A (en) * | 2018-01-17 | 2018-07-24 | 哈尔滨工程大学 | A kind of UUV Realtime collision free planing methods based on LSTM networks |
| CN109034359A (en) * | 2018-07-03 | 2018-12-18 | 上海电机学院 | The genetic algorithm of non-dominated ranking with elitism strategy changes method |
| CN109634289A (en) * | 2019-01-24 | 2019-04-16 | 大连海事大学 | A kind of unmanned boat autonomous navigation system and its working method |
| CN109931678A (en) * | 2019-03-13 | 2019-06-25 | 中国计量大学 | Air-conditioning fault diagnosis method based on deep learning LSTM |
-
2019
- 2019-09-30 CN CN201910944783.1A patent/CN110737267A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708406A (en) * | 2012-05-10 | 2012-10-03 | 湖北省电力公司 | Scheduling graph optimizing method based on multi-target genetic algorithm |
| CN102831479A (en) * | 2012-07-29 | 2012-12-19 | 江苏大学 | NSGA-II (non-domination sequencing genetic algorithm) multi-objective optimization searching method |
| CN107563576A (en) * | 2017-10-14 | 2018-01-09 | 连云港杰瑞深软科技有限公司 | A kind of ship intelligence energy efficiency management system |
| CN108319293A (en) * | 2018-01-17 | 2018-07-24 | 哈尔滨工程大学 | A kind of UUV Realtime collision free planing methods based on LSTM networks |
| CN109034359A (en) * | 2018-07-03 | 2018-12-18 | 上海电机学院 | The genetic algorithm of non-dominated ranking with elitism strategy changes method |
| CN109634289A (en) * | 2019-01-24 | 2019-04-16 | 大连海事大学 | A kind of unmanned boat autonomous navigation system and its working method |
| CN109931678A (en) * | 2019-03-13 | 2019-06-25 | 中国计量大学 | Air-conditioning fault diagnosis method based on deep learning LSTM |
Non-Patent Citations (6)
| Title |
|---|
| 冯士刚等: "带精英策略的快速非支配排序遗传算法在多目标无功优化中的应用", 《电工技术学报》 * |
| 崔杨等: "出力受限风电场群有功分配多目标优化策略", 《可再生能源》 * |
| 张洪龙等: "基于立体视觉的水下三维测量系统研究", 《集成技术》 * |
| 张福民等: "基于改进NSGA-Ⅱ算法的微网交互式多目标优化", 《电力系统保护与控制》 * |
| 张雪莲: "基于扰动观测的无人水下航行器复合航迹跟踪控制研究", 《中国优秀硕士论文电子期刊网 工程科技Ⅱ辑 信息科技》 * |
| 王鲁等: "基于遗传算法的多目标优化技术", 《科技广场》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111610789A (en) * | 2020-07-01 | 2020-09-01 | 上海船舶研究设计院（中国船舶工业集团公司第六0四研究院） | Ship comprehensive management and control system and intelligent ship |
| CN112052984A (en) * | 2020-08-07 | 2020-12-08 | 智慧航海（青岛）科技有限公司 | Intelligent ship collision avoidance method and system |
| CN113325896A (en) * | 2021-05-31 | 2021-08-31 | 浙江工业大学 | Multi-target temperature optimization control method of intelligent retail machine |
| CN113325896B (en) * | 2021-05-31 | 2022-03-01 | 浙江工业大学 | Multi-target temperature optimization control method of intelligent retail machine |
| CN114185328A (en) * | 2021-12-07 | 2022-03-15 | 天翼物联科技有限公司 | Unmanned ship control method, device, equipment and medium based on sea condition self-adaption |
| CN114462444A (en) * | 2021-12-21 | 2022-05-10 | 上海中船船舶设计技术国家工程研究中心有限公司 | Navigation state identification method and system for polar region inspection ship |
| CN116395105A (en) * | 2023-06-06 | 2023-07-07 | 海云联科技(苏州)有限公司 | Automatic lifting compensation method and system for unmanned ship |
| CN116395105B (en) * | 2023-06-06 | 2023-10-20 | 海云联科技(苏州)有限公司 | Automatic lifting compensation method and system for unmanned ship |
| CN116911560A (en) * | 2023-07-27 | 2023-10-20 | 中国舰船研究设计中心 | Ship task system decision planning method based on multi-objective optimization |
| CN116911560B (en) * | 2023-07-27 | 2024-03-12 | 中国舰船研究设计中心 | Ship task system decision planning method based on multi-objective optimization |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110737267A (en) | Multi-objective optimization method for unmanned ships and intelligent comprehensive management and control system for unmanned ships | |
| CN109214107B (en) | Ship navigation behavior online prediction method | |
| Baruque et al. | A forecasting solution to the oil spill problem based on a hybrid intelligent system | |
| CN113221449B (en) | Ship track real-time prediction method and system based on optimal strategy learning | |
| CN111339229B (en) | Ship autonomous navigation aid decision-making system | |
| CN110555551B (en) | Air quality big data management method and system for smart city | |
| CN108960421A (en) | The unmanned surface vehicle speed of a ship or plane online forecasting method based on BP neural network of improvement | |
| Xiao et al. | Bidirectional data-driven trajectory prediction for intelligent maritime traffic | |
| Sreelakshmi et al. | Neural networks for short term wind speed prediction | |
| CN111538349B (en) | Long-range AUV autonomous decision-making method oriented to multiple tasks | |
| CN117391244A (en) | Ship track prediction method integrating multi-source information | |
| CN117743795A (en) | Multi-ship track prediction method and system based on dynamic space-time refinement network | |
| Newaz et al. | Long-term autonomy for AUVs operating under uncertainties in dynamic marine environments | |
| Kim et al. | Massive scale deep learning for detecting extreme climate events | |
| Hadi et al. | Ship navigation and fuel profiling based on noon report using neural network generative modeling | |
| CN117743487A (en) | RF and LSTM combined ship track prediction model | |
| Cai et al. | Diversity, quality, and quantity of real ship data on the black-box and gray-box prediction models of ship fuel consumption | |
| Kramer et al. | Wind power ramp event prediction with support vector machines | |
| Anne et al. | Prediction of ship track anomaly based on AIS data using long short-term memory (LSTM) and DBScan | |
| Zhang et al. | Prediction of dynamic navigation information of unmanned surface vehicles based on Bayesian optimization LSTM | |
| Jie | Modeling Seafarers’ Navigational Decision-Making for Autonomous Ships’ Safety | |
| Zhang et al. | Time-delay wavelet network predictor based on sensitivity analysis with application to predictive ship course control | |
| Al-Thani et al. | Traffic Transformer: Transformer-based framework for temporal traffic accident prediction | |
| CN115855069B (en) | GEO spacecraft maneuver detection and position prediction method based on SCE | |
| Li et al. | Multi-step Ship Roll Motion Prediction Based on Bi-LSTM and Input Optimization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200131 |