CN116893614A - Control method of amphibious unmanned ship based on multi-sensor fusion - Google Patents
Control method of amphibious unmanned ship based on multi-sensor fusion Download PDFInfo
- Publication number
- CN116893614A CN116893614A CN202310660994.9A CN202310660994A CN116893614A CN 116893614 A CN116893614 A CN 116893614A CN 202310660994 A CN202310660994 A CN 202310660994A CN 116893614 A CN116893614 A CN 116893614A
- Authority
- CN
- China
- Prior art keywords
- unmanned ship
- control
- model
- land
- water
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 230000004927 fusion Effects 0.000 title claims abstract description 31
- 238000001514 detection method Methods 0.000 claims abstract description 60
- 238000012549 training Methods 0.000 claims abstract description 27
- 238000004458 analytical method Methods 0.000 claims abstract description 16
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 96
- 239000002245 particle Substances 0.000 claims description 28
- 238000005457 optimization Methods 0.000 claims description 26
- 238000011156 evaluation Methods 0.000 claims description 12
- 238000013507 mapping Methods 0.000 claims description 11
- 230000003044 adaptive effect Effects 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 10
- 230000001105 regulatory effect Effects 0.000 claims description 9
- 238000012512 characterization method Methods 0.000 claims description 8
- 230000001276 controlling effect Effects 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 7
- 238000004132 cross linking Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 abstract description 2
- 230000007123 defense Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 230000005059 dormancy Effects 0.000 description 2
- 230000010365 information processing Effects 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The application discloses a control method of an amphibious unmanned ship based on multi-sensor fusion, which relates to the technical field of intelligent control and comprises the following steps: classifying an amphibious unmanned ship operation database according to a distributed detection sensor network, and respectively training and constructing an unmanned ship control self-adaptive model and a land unmanned ship control self-adaptive model according to the classified unmanned ship sample data set; performing model matching based on the unmanned ship operation mode and the unmanned ship control self-adaptive model, analyzing the acquired unmanned ship operation detection data stream based on a matching model result, and outputting unmanned ship operation control parameters; and obtaining unmanned ship operation deviation information according to the unmanned ship operation control parameters, so as to optimally regulate and control the unmanned ship operation control parameters. The intelligent control of unmanned ship operation is achieved, the analysis accuracy of unmanned ship operation control parameters is improved, the real-time regulation and control of the control parameters is achieved, and further the technical effect of unmanned ship task efficiency is improved.
Description
Technical Field
The application relates to the technical field of intelligent control, in particular to a control method of an amphibious unmanned ship based on multi-sensor fusion.
Background
The amphibious unmanned ship is provided with an advanced navigation, target positioning and fire control system consisting of various radar equipment and a computer information processing terminal, and can be controlled by a remote control system to realize intelligent system combat. The intelligent landing device has the characteristics of high-speed navigation and excellent stealth performance, can carry out hidden dormancy, intelligent cruising, rapid assault and beach landing according to the operational requirements, and can realize the tasks of water delivery, frontier defense patrol, near shore guard, island reef airport protection and the like of special fighters, and can be called as an harpoon in the field of sea defense. Therefore, it is important to conduct intelligent and informationized precise control. However, the unmanned ship operation control in the prior art has the technical problems of complex environment and multiple constraints, low control parameter accuracy and poor control timeliness.
Disclosure of Invention
Based on the above, it is necessary to provide a control method of an amphibious unmanned ship based on multi-sensor fusion, which can improve the analysis accuracy of the unmanned ship operation control parameters, optimally regulate and control the control parameters in time, realize the real-time regulation and control of the control parameters, and further improve the unmanned ship task efficiency.
A method of controlling an amphibious unmanned watercraft based on multisensor fusion, the method comprising: acquiring an amphibious unmanned ship operation database, wherein the amphibious unmanned ship operation database comprises water unmanned ship operation data information and land unmanned ship operation data information; classifying the amphibious unmanned ship operation database according to a distributed detection sensor network to obtain an overwater unmanned ship sample data set and a land unmanned ship sample data set; respectively training and constructing a water unmanned ship control self-adaptive model and a land unmanned ship control self-adaptive model based on the water unmanned ship sample data set and the land unmanned ship sample data set; information sensing is carried out on the whole operation process of the target unmanned ship based on the distributed detection sensor network, and an unmanned ship operation detection data stream is acquired; identifying and determining an unmanned ship operation mode according to the unmanned ship operation detection data stream; performing model matching on the basis of the unmanned ship operation mode, the water unmanned ship control self-adaptive model and the land unmanned ship control self-adaptive model, analyzing the unmanned ship operation detection data stream on the basis of a matching model result, and outputting unmanned ship operation control parameters; and acquiring unmanned ship operation deviation information according to the unmanned ship operation control parameters, and optimally regulating and controlling the unmanned ship operation control parameters based on the unmanned ship operation deviation information.
A control system for an amphibious unmanned watercraft based on multisensor fusion, the system comprising: the unmanned ship operation database obtaining module is used for obtaining an amphibious unmanned ship operation database, and the amphibious unmanned ship operation database comprises water unmanned ship operation data information and land unmanned ship operation data information; the database classification module is used for classifying the amphibious unmanned ship operation database according to the distributed detection sensor network to obtain an overwater unmanned ship sample data set and a land unmanned ship sample data set; the self-adaptive model construction module is used for respectively training and constructing a water unmanned ship control self-adaptive model and a land unmanned ship control self-adaptive model based on the water unmanned ship sample data set and the land unmanned ship sample data set; the detection information sensing module is used for sensing information of the whole operation process of the target unmanned ship based on the distributed detection sensor network, and acquiring an unmanned ship operation detection data stream; the unmanned ship operation mode determining module is used for identifying and determining an unmanned ship operation mode according to the unmanned ship operation detection data stream; the operation control parameter obtaining module is used for carrying out model matching on the basis of the unmanned ship operation mode and the water unmanned ship control self-adaptive model and the land unmanned ship control self-adaptive model, analyzing the unmanned ship operation detection data stream on the basis of a matching model result and outputting unmanned ship operation control parameters; and the control parameter optimization regulation and control module is used for obtaining unmanned ship operation deviation information according to the unmanned ship operation control parameters and carrying out optimization regulation and control on the unmanned ship operation control parameters based on the unmanned ship operation deviation information.
According to the amphibious unmanned ship control method based on multi-sensor fusion, the technical problems that in the prior art, unmanned ship operation control environments are complex and multi-constraint, control parameter accuracy is low, and control timeliness is poor are solved, intelligent control on unmanned ship operation is achieved through constructing an unmanned ship control self-adaptive model, unmanned ship operation control parameter analysis accuracy is improved, control parameters are optimally regulated and controlled in time, control parameter regulation timeliness is achieved, and further the unmanned ship task efficiency is improved.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
FIG. 1 is a flow diagram of a method of controlling an amphibious unmanned watercraft based on multi-sensor fusion in one embodiment;
FIG. 2 is a schematic flow chart of training and constructing an unmanned aerial vehicle control adaptive model in an amphibious unmanned aerial vehicle control method based on multi-sensor fusion in one embodiment;
fig. 3 is a block diagram of a control system for an amphibious unmanned boat based on multi-sensor fusion in one embodiment.
Reference numerals illustrate: the unmanned ship operation system comprises an unmanned ship operation database obtaining module 11, a database classifying module 12, an adaptive model constructing module 13, a detection information sensing module 14, an unmanned ship operation mode determining module 15, an operation control parameter obtaining module 16 and a control parameter optimizing and regulating module 17.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
As shown in fig. 1, the application provides a control method of an amphibious unmanned ship based on multi-sensor fusion, which comprises the following steps:
step S100: acquiring an amphibious unmanned ship operation database, wherein the amphibious unmanned ship operation database comprises water unmanned ship operation data information and land unmanned ship operation data information;
step S200: classifying the amphibious unmanned ship operation database according to a distributed detection sensor network to obtain an overwater unmanned ship sample data set and a land unmanned ship sample data set;
specifically, the amphibious unmanned ship is provided with an advanced navigation, target positioning and fire control system consisting of various radar equipment and a computer information processing terminal, and can be controlled by a remote control system to realize intelligent system combat. The intelligent landing device has the characteristics of high-speed navigation and excellent stealth performance, can carry out hidden dormancy, intelligent cruising, rapid assault and beach landing according to the operational requirements, and can realize the tasks of water delivery, frontier defense patrol, near shore guard, island reef airport protection and the like of special fighters, and can be called as an harpoon in the field of sea defense. Therefore, it is important to conduct intelligent and informationized precise control.
The amphibious unmanned ship operation database is obtained through a big data technology and unmanned ship operation history data, and comprises water unmanned ship operation data information and land unmanned ship operation data information, wherein the operation data information comprises sensor measurement data of the unmanned ship and corresponding operation control parameters. For determining the position of the unmanned ship and sensing the change condition of surrounding environment information thereof in real time, various sensor group components are arranged to form a detection sensor network, including a direction sensor, a position sensor, an image sensor, a position sensor, a temperature sensor, a radar and the like. Classifying the amphibious unmanned ship operation database according to the distributed detection sensor network, namely classifying the unmanned ship operation database according to the sensor type, so as to obtain a classified sample data set of the unmanned ship on water and a classified sample data set of the unmanned ship on land, and improve the construction accuracy of a follow-up model.
Step S300: respectively training and constructing a water unmanned ship control self-adaptive model and a land unmanned ship control self-adaptive model based on the water unmanned ship sample data set and the land unmanned ship sample data set;
in one embodiment, as shown in fig. 2, the training builds the unmanned water craft control adaptive model and the unmanned land craft control adaptive model, and the applying step S300 further includes:
step S310: respectively carrying out equal weight layer training on the water unmanned ship sample data set and the land unmanned ship sample data set to obtain a basic water unmanned ship control branch model set and a basic land unmanned ship control branch model set;
step S320: verifying based on the model output results of the basic unmanned water vehicle control branch model set and the basic land unmanned water vehicle control branch model set to obtain a water model control error rate information set and a land model control error rate information set;
step S330: updating weight distribution of the water unmanned ship sample data set and the land unmanned ship sample data set based on the water model control error rate information set and the land model control error rate information set, and iteratively obtaining a water unmanned ship control branch model set and a basic land unmanned ship control branch model set;
step S340: and fusing all branch models in the water unmanned ship control branch model set and the basic land unmanned ship control branch model set to generate the water unmanned ship control self-adaptive model and the land unmanned ship control self-adaptive model.
In one embodiment, the updating the weight distribution of the sample data set of the unmanned water craft and the sample data set of the unmanned land craft further includes:
step S331: carrying out importance analysis on each sensor node in the distributed detection sensor network to obtain a sensor network importance coefficient;
step S332: carrying out parameter fusion on the water model control error rate information set and the land model control error rate information set according to the importance coefficient of the sensor network to obtain a water model global parameter set and a land model global parameter set;
step S333: and updating weight distribution of the water unmanned ship sample data set and the land unmanned ship sample data set by using the water model global parameter set and the land model global parameter set.
In one embodiment, the obtaining the importance coefficient of the sensor network in step S331 of the present application further includes:
step S3311: making a weighting rule of the sensor node;
step S3312: the unmanned ship expert group carries out importance analysis on each sensor node in the distributed detection sensor network based on the sensor node weighting rule to obtain a sensor node importance coefficient evaluation set;
step S3313: marking the trust degree of each expert in the unmanned ship expert group to obtain an expert group trust degree set;
step S3314: and carrying out cross-linking fusion on the sensor node importance coefficient evaluation set and the expert group trust degree set to obtain the sensor network importance coefficient.
Specifically, unmanned ship control self-adaptive model construction is respectively carried out based on the unmanned ship sample data set on water and the unmanned ship sample data set on land, and the unmanned ship control self-adaptive model has high detection accuracy and high data processing rate. The construction process is that the sample data sets of the unmanned water craft and the sample data sets of the land craft are respectively subjected to equally-divided weight layer training, namely initial training weights of the sample data sets of the unmanned water craft are identical, sample data weights in each group of model training data are equally divided, a neural network model is trained by using the model training data, a basic unmanned water craft control branch model set and a basic land craft control branch model set are obtained, namely weak control branch models corresponding to various sensor samples in a distributed detection sensor network are included, and model output accuracy is low. And verifying the model output results of the basic unmanned water vehicle control branch model set and the basic land unmanned water vehicle control branch model set, and respectively calculating error rates of the model output detection results through model test sample data so as to obtain corresponding water model control error rate information sets and land model control error rate information sets.
And updating weight distribution of the water unmanned ship sample data set and the land unmanned ship sample data set based on the water model control error rate information set and the land model control error rate information set so as to improve the weight of the samples which are incorrectly classified by the previous round of weak control branch model and reduce the weight of the correctly classified samples. The sample weight updating process is to analyze the importance of each sensor node in the distributed detection sensor network, and firstly, a sensor node weighting rule is formulated, wherein the sensor node weighting rule is the basis for weight distribution of each sensor node type of the detection sensor network, comprises a weighting value range, a weighting index and the like, and can be set according to the actual application situation. And the unmanned ship expert group performs importance analysis on each sensor node in the distributed detection sensor network based on the sensor node weighting rule to obtain a sensor node importance coefficient evaluation set obtained by each expert evaluation.
And marking the trust degree of each expert in the unmanned ship expert group, namely marking the trust degree of each expert, and determining according to the historical evaluation accuracy, wherein the stronger the evaluation specialty is, the higher the corresponding trust degree is, so as to determine the trust degree set of the expert group. And carrying out cross-linking fusion on the sensor node importance coefficient evaluation set and the expert group trust degree set, namely carrying out weighted fusion on the sensor node importance coefficients according to the expert trust degree, calculating to obtain the sensor network importance coefficients, and taking the sensor network importance coefficients as the importance coefficients of each branch model, so that the sensor node weight evaluation accuracy is improved, and the subsequent model training accuracy is further improved. Carrying out parameter fusion on the water model control error rate information set and the land model control error rate information set according to the importance coefficient of the sensor network, wherein the parameter fusion function is particularly preferablyWherein->Identifying branch model error rate,/">Identifying a sensor network importance coefficient, +.>Along with->And (3) calculating a water model global parameter set and a land model global parameter set of the basic branch model after parameter fusion.
Respectively updating weight distribution of the water unmanned ship sample data set and the land unmanned ship sample data set by using the water model global parameter set and the land model global parameter set so as to improve training output accuracy of the weak control branch model, wherein a weight updating function of model training data is as followsWherein, the method comprises the steps of, wherein,updated weights indicating model training data, +.>Weight indicating the ith sample of the nth control branch model, +.>Predicted value for ith sample for nth control branch model, +.>The correct output value of the ith sample, namely, the training data individuals with failed classification are given larger weight, and the training individuals are more concerned in the next iteration operation, so that the sample weight of the previous weak control branch model with wrong classification can be enhanced for the next iteration.
Iteration is continued untilAnd (5) training to obtain the water unmanned ship control self-adaptive model and the land unmanned ship control self-adaptive model when the preset error rate is reached or the specified maximum iteration number is reached. Fusing all branch models in the water unmanned ship control branch model set and the basic land unmanned ship control branch model set, wherein the fusion coefficient of each branch model is a model key parameterThe weight of the weak control branch model with small model error rate and great importance is increased, so that the weak control branch model plays a great role in voting; the weight of the weak control branch model with large model error rate and small importance is reduced, so that the weak control branch model plays a small role in voting, and the weight fusion generates a water unmanned ship control self-adaptive model and a land unmanned ship control self-adaptive model, namely a strong unmanned ship control self-adaptive model formed by the weak control branch models. And the control parameter analysis of the double-channel unmanned ship is carried out by constructing the control self-adaptive model of the unmanned ship on water and the control self-adaptive model of the land unmanned ship, so that the analysis accuracy and the analysis efficiency of the operation control parameter of the unmanned ship are improved.
Step S400: information sensing is carried out on the whole operation process of the target unmanned ship based on the distributed detection sensor network, and an unmanned ship operation detection data stream is acquired;
step S500: identifying and determining an unmanned ship operation mode according to the unmanned ship operation detection data stream;
step S600: and carrying out model matching on the basis of the unmanned ship operation mode, the water unmanned ship control self-adaptive model and the land unmanned ship control self-adaptive model, analyzing the unmanned ship operation detection data stream on the basis of a matching model result, and outputting unmanned ship operation control parameters.
Specifically, information sensing is carried out on the whole operation process of the target unmanned ship based on the distributed detection sensor network, and a corresponding unmanned ship operation detection data stream detected by the sensor network is acquired. And identifying and determining an unmanned ship operation mode according to the unmanned ship operation detection data stream, namely judging whether the unmanned ship is in a water operation mode or a land operation mode. And performing model matching on the unmanned ship operation mode, the water unmanned ship control self-adaptive model and the land unmanned ship control self-adaptive model, and determining the self-adaptive model corresponding to the unmanned ship operation mode. And analyzing the unmanned ship operation detection data stream based on a matching model result, and outputting unmanned ship operation control parameters, wherein the unmanned ship operation control parameters comprise a linear speed control parameter, an angular speed control parameter, a thrust control parameter, a rudder deflection control parameter and the like, so as to comprehensively realize unmanned ship control intellectualization and improve unmanned ship control accuracy.
Step S700: and acquiring unmanned ship operation deviation information according to the unmanned ship operation control parameters, and optimally regulating and controlling the unmanned ship operation control parameters based on the unmanned ship operation deviation information.
In one embodiment, the optimizing and controlling the unmanned ship operation control parameter based on the unmanned ship operation deviation information further includes:
step S710: acquiring an association mapping relation between an unmanned ship running state and unmanned ship control parameters;
step S720: matching is carried out based on the association mapping relation and the unmanned ship operation deviation information, and target operation parameter optimization information is obtained;
step S730: constructing a control parameter optimizing space according to the target operation parameter optimizing information;
step S740: constructing an unmanned ship loss fitness function, performing parameter optimization in the control parameter optimization space based on the unmanned ship loss fitness function, and outputting unmanned ship operation optimization control parameters.
In one embodiment, the outputting the unmanned ship operation optimization control parameter, step S740 of the present application further includes:
step S741: initializing the control parameter optimizing space to obtain a particle swarm constraint parameter;
step S742: iteratively calculating and updating the particle swarm constraint parameters based on the unmanned ship loss fitness function, and obtaining an output result of the unmanned ship loss fitness function when a preset iteration termination condition is reached, wherein the output result comprises optimal result particles;
step S743: and determining the unmanned aerial vehicle operation optimization control parameters based on the optimal result particles.
In one embodiment, the unmanned ship loss fitness function is specifically:+/>wherein->Characterizing energy loss weight,/->Is an empirical function of energy loss->Characterization of efficiency loss, +.>As an empirical function of efficiency loss->Characterization of the ith control parameter,/->And->The sum of weights is 1, < >>The loss constant is characterized.
Specifically, the operation control of the unmanned ship is performed according to the unmanned ship operation control parameters, the operation of the unmanned ship is monitored, the unmanned ship monitoring data information is compared with the operation reference target, and unmanned ship operation deviation information, such as the occurrence of the route position deviation of the unmanned ship, is obtained. And carrying out optimized regulation and control on the unmanned ship operation control parameters based on the unmanned ship operation deviation information, and firstly, acquiring an association mapping relation between an unmanned ship operation state and the unmanned ship control parameters, namely, a mapping relation between operation state types and a plurality of association control parameters, wherein the state of an unmanned ship operation route is related to a linear speed control parameter and an angular speed control parameter by way of example. And matching the incidence mapping relation with the unmanned ship operation deviation information to obtain target operation parameter optimization information, wherein the target operation parameter optimization information comprises an operation control parameter type and an optimization parameter degree which need to be optimized. And determining an optimizable threshold range of the operation control parameters according to the target operation parameter optimization information, and constructing a control parameter optimizing space according to the optimizable threshold range for the operation control parameters so as to be used for optimizing the control parameters.
Constructing an unmanned ship loss fitness function, wherein the unmanned ship loss fitness function is used for evaluating an adaptive value of optimizing control parameters, and screening the control parameters according to the adaptive value, and the fitness function is specifically as follows:+/>wherein->Characterizing energy loss weight,/->Is an empirical function of energy loss->Characterization of efficiency loss, +.>As an empirical function of efficiency loss->Characterization of the ith control parameter,/->And->The sum of weights is 1, < >>And the loss constant is characterized and determined according to the operation of the actual unmanned ship. Parameter optimization is carried out in the control parameter optimizing space based on the unmanned ship loss fitness function, firstly, the control parameter optimizing space is initialized, the control parameter optimizing space comprises random positions and speeds of particles, and particle swarm constraint parameters are obtained. And iteratively calculating and updating the constraint parameters of the particle swarm based on the unmanned ship loss fitness function, further updating the positions and the speeds of the particles in the particle swarm, inputting all the particles into a model for training, evaluating the quality of the particles by calculating the loss fitness function of the particle swarm, and adjusting the positions and the speeds of each particle by using the loss fitness function.
And when the preset iteration termination condition is reached, obtaining an output result of the unmanned ship loss fitness function, wherein the output result comprises optimal result particles. Further, the PSO algorithm stops including two possibilities, one is that the particles are balanced or in an optimal state, the other is that the operation limit is exceeded, no specific analysis is performed on the condition that the operation limit is exceeded, and the optimal result particles are in the optimal state of the particles. And determining unmanned aerial vehicle operation optimization control parameters based on the optimal result particles, wherein the unmanned aerial vehicle operation optimization control parameters are operation control parameters obtained by optimizing according to the current state of the unmanned aerial vehicle. The unmanned ship operation is intelligently controlled by constructing an unmanned ship control self-adaptive model, and then the control parameters are optimally regulated and controlled in time by combining a particle swarm algorithm, so that the real-time regulation and control of the control parameters is realized, and the unmanned ship task efficiency is further improved.
In one embodiment, as shown in fig. 3, there is provided a control system for an amphibious unmanned boat based on multi-sensor fusion, comprising: the unmanned ship operation database obtaining module 11, the database classifying module 12, the self-adaptive model constructing module 13, the detection information sensing module 14, the unmanned ship operation mode determining module 15, the operation control parameter obtaining module 16 and the control parameter optimizing and regulating module 17, wherein:
the unmanned ship operation database obtaining module 11 is used for obtaining an amphibious unmanned ship operation database, wherein the amphibious unmanned ship operation database comprises water unmanned ship operation data information and land unmanned ship operation data information;
a database classification module 12, configured to classify the amphibious unmanned ship operation database according to a distributed detection sensor network, and obtain an above-water unmanned ship sample data set and a land unmanned ship sample data set;
the self-adaptive model construction module 13 is used for respectively training and constructing a water unmanned ship control self-adaptive model and a land unmanned ship control self-adaptive model based on the water unmanned ship sample data set and the land unmanned ship sample data set;
the detection information sensing module 14 is used for sensing information of the whole operation process of the target unmanned ship based on the distributed detection sensor network, and acquiring an unmanned ship operation detection data stream;
the unmanned ship operation mode determining module 15 is used for identifying and determining an unmanned ship operation mode according to the unmanned ship operation detection data stream;
the operation control parameter obtaining module 16 is configured to perform model matching with the unmanned aerial vehicle control adaptive model and the land unmanned aerial vehicle control adaptive model based on the unmanned aerial vehicle operation mode, analyze the unmanned aerial vehicle operation detection data stream based on a matching model result, and output unmanned aerial vehicle operation control parameters;
the control parameter optimization regulation and control module 17 is used for obtaining unmanned ship operation deviation information according to the unmanned ship operation control parameters and carrying out optimization regulation and control on the unmanned ship operation control parameters based on the unmanned ship operation deviation information.
In one embodiment, the system further comprises:
the equally-weighted layer training unit is used for respectively carrying out equally-weighted layer training on the water unmanned ship sample data set and the land unmanned ship sample data set to obtain a basic water unmanned ship control branch model set and a basic land unmanned ship control branch model set;
the model verification unit is used for verifying based on model output results of the basic unmanned water vehicle control branch model set and the basic land unmanned water vehicle control branch model set to obtain an overwater model control error rate information set and a land model control error rate information set;
the model iteration obtaining unit is used for updating weight distribution of the water unmanned ship sample data set and the land unmanned ship sample data set based on the water model control error rate information set and the land model control error rate information set, and iteratively obtaining a water unmanned ship control branch model set and a basic land unmanned ship control branch model set;
and the branch model fusion unit is used for fusing each branch model in the water unmanned ship control branch model set and the basic land unmanned ship control branch model set to generate the water unmanned ship control self-adaptive model and the land unmanned ship control self-adaptive model.
In one embodiment, the system further comprises:
the importance analysis unit is used for carrying out importance analysis on each sensor node in the distributed detection sensor network to obtain an importance coefficient of the sensor network;
the parameter fusion unit is used for carrying out parameter fusion on the water model control error rate information set and the land model control error rate information set according to the importance coefficient of the sensor network to obtain a water model global parameter set and a land model global parameter set;
and the weight distribution updating unit is used for updating the weight distribution of the water unmanned ship sample data set and the land unmanned ship sample data set by utilizing the water model global parameter set and the land model global parameter set.
In one embodiment, the system further comprises:
the weighting rule making unit is used for making a weighting rule of the sensor node;
the importance coefficient obtaining unit is used for carrying out importance analysis on each sensor node in the distributed detection sensor network based on the sensor node weighting rule by the unmanned ship expert group to obtain a sensor node importance coefficient evaluation set;
the trust marking unit is used for marking the trust of each expert in the unmanned ship expert group to obtain an expert group trust set;
and the coefficient cross-linking and fusion unit is used for cross-linking and fusing the sensor node importance coefficient evaluation set and the expert group trust degree set to obtain the sensor network importance coefficient.
In one embodiment, the system further comprises:
the incidence mapping relation obtaining unit is used for obtaining the incidence mapping relation between the unmanned ship running state and the unmanned ship control parameter;
the deviation information mapping matching unit is used for matching the unmanned ship operation deviation information based on the association mapping relation to obtain target operation parameter optimization information;
the optimizing space construction unit is used for constructing a control parameter optimizing space according to the target operation parameter optimizing information;
and the space parameter optimizing unit is used for constructing an unmanned ship loss fitness function, carrying out parameter optimizing in the control parameter optimizing space based on the unmanned ship loss fitness function, and outputting unmanned ship operation optimizing control parameters.
In one embodiment, the system further comprises:
the space initialization unit is used for initializing the control parameter optimizing space to obtain particle swarm constraint parameters;
the optimal result particle obtaining unit is used for carrying out iterative computation and updating on the particle swarm constraint parameters based on the unmanned ship loss fitness function, and obtaining an output result of the unmanned ship loss fitness function when a preset iteration termination condition is reached, wherein the output result comprises optimal result particles;
and the optimal control parameter determining unit is used for determining the unmanned aerial vehicle operation optimal control parameters based on the optimal result particles.
In one embodiment, the system further comprisesThe method comprises the following steps: the fitness function unit specifically comprises:+/>wherein->Characterizing energy loss weight,/->Is an empirical function of energy loss->Characterization of efficiency loss, +.>As an empirical function of efficiency loss->Characterization of the ith control parameter,/->And->The sum of weights is 1, < >>The loss constant is characterized.
The application provides a control method of an amphibious unmanned ship based on multi-sensor fusion, which comprises the following steps: acquiring an amphibious unmanned ship operation database, wherein the amphibious unmanned ship operation database comprises water unmanned ship operation data information and land unmanned ship operation data information; classifying the amphibious unmanned ship operation database according to a distributed detection sensor network to obtain an overwater unmanned ship sample data set and a land unmanned ship sample data set; respectively training and constructing a water unmanned ship control self-adaptive model and a land unmanned ship control self-adaptive model based on the water unmanned ship sample data set and the land unmanned ship sample data set; information sensing is carried out on the whole operation process of the target unmanned ship based on the distributed detection sensor network, and an unmanned ship operation detection data stream is acquired; identifying and determining an unmanned ship operation mode according to the unmanned ship operation detection data stream; performing model matching on the basis of the unmanned ship operation mode, the water unmanned ship control self-adaptive model and the land unmanned ship control self-adaptive model, analyzing the unmanned ship operation detection data stream on the basis of a matching model result, and outputting unmanned ship operation control parameters; and acquiring unmanned ship operation deviation information according to the unmanned ship operation control parameters, and optimally regulating and controlling the unmanned ship operation control parameters based on the unmanned ship operation deviation information. The unmanned aerial vehicle control self-adaptive model is built to intelligently control the unmanned aerial vehicle operation, the analysis accuracy of unmanned aerial vehicle operation control parameters is improved, the control parameters are optimally regulated and controlled in time, the real-time regulation and control of the control parameters is realized, and the technical effect of improving the unmanned aerial vehicle task efficiency is further achieved.
The specification and drawings are merely exemplary illustrations of the present application, and the present application is intended to include such modifications and variations if they fall within the scope of the present application and the equivalent techniques thereof.
Claims (8)
1. A method for controlling an amphibious unmanned ship based on multi-sensor fusion, the method comprising:
acquiring an amphibious unmanned ship operation database, wherein the amphibious unmanned ship operation database comprises water unmanned ship operation data information and land unmanned ship operation data information;
classifying the amphibious unmanned ship operation database according to a distributed detection sensor network to obtain an overwater unmanned ship sample data set and a land unmanned ship sample data set;
respectively training and constructing a water unmanned ship control self-adaptive model and a land unmanned ship control self-adaptive model based on the water unmanned ship sample data set and the land unmanned ship sample data set;
information sensing is carried out on the whole operation process of the target unmanned ship based on the distributed detection sensor network, and an unmanned ship operation detection data stream is acquired;
identifying and determining an unmanned ship operation mode according to the unmanned ship operation detection data stream;
performing model matching on the basis of the unmanned ship operation mode, the water unmanned ship control self-adaptive model and the land unmanned ship control self-adaptive model, analyzing the unmanned ship operation detection data stream on the basis of a matching model result, and outputting unmanned ship operation control parameters;
and acquiring unmanned ship operation deviation information according to the unmanned ship operation control parameters, and optimally regulating and controlling the unmanned ship operation control parameters based on the unmanned ship operation deviation information.
2. The method of claim 1, wherein the training constructs a water unmanned aerial vehicle control adaptive model and a land unmanned aerial vehicle control adaptive model, comprising:
respectively carrying out equal weight layer training on the water unmanned ship sample data set and the land unmanned ship sample data set to obtain a basic water unmanned ship control branch model set and a basic land unmanned ship control branch model set;
verifying based on the model output results of the basic unmanned water vehicle control branch model set and the basic land unmanned water vehicle control branch model set to obtain a water model control error rate information set and a land model control error rate information set;
updating weight distribution of the water unmanned ship sample data set and the land unmanned ship sample data set based on the water model control error rate information set and the land model control error rate information set, and iteratively obtaining a water unmanned ship control branch model set and a basic land unmanned ship control branch model set;
and fusing all branch models in the water unmanned ship control branch model set and the basic land unmanned ship control branch model set to generate the water unmanned ship control self-adaptive model and the land unmanned ship control self-adaptive model.
3. The method of claim 2, wherein the updating the weight distribution of the water craft sample data set and land craft sample data set comprises:
carrying out importance analysis on each sensor node in the distributed detection sensor network to obtain a sensor network importance coefficient;
carrying out parameter fusion on the water model control error rate information set and the land model control error rate information set according to the importance coefficient of the sensor network to obtain a water model global parameter set and a land model global parameter set;
and updating weight distribution of the water unmanned ship sample data set and the land unmanned ship sample data set by using the water model global parameter set and the land model global parameter set.
4. The method of claim 3, wherein the obtaining the sensor network importance coefficients comprises:
making a weighting rule of the sensor node;
the unmanned ship expert group carries out importance analysis on each sensor node in the distributed detection sensor network based on the sensor node weighting rule to obtain a sensor node importance coefficient evaluation set;
marking the trust degree of each expert in the unmanned ship expert group to obtain an expert group trust degree set;
and carrying out cross-linking fusion on the sensor node importance coefficient evaluation set and the expert group trust degree set to obtain the sensor network importance coefficient.
5. The method of claim 1, wherein the optimally adjusting the unmanned ship operation control parameters based on the unmanned ship operation deviation information comprises:
acquiring an association mapping relation between an unmanned ship running state and unmanned ship control parameters;
matching is carried out based on the association mapping relation and the unmanned ship operation deviation information, and target operation parameter optimization information is obtained;
constructing a control parameter optimizing space according to the target operation parameter optimizing information;
constructing an unmanned ship loss fitness function, performing parameter optimization in the control parameter optimization space based on the unmanned ship loss fitness function, and outputting unmanned ship operation optimization control parameters.
6. The method of claim 5, wherein outputting unmanned ship operation optimization control parameters comprises:
initializing the control parameter optimizing space to obtain a particle swarm constraint parameter;
iteratively calculating and updating the particle swarm constraint parameters based on the unmanned ship loss fitness function, and obtaining an output result of the unmanned ship loss fitness function when a preset iteration termination condition is reached, wherein the output result comprises optimal result particles;
and determining the unmanned aerial vehicle operation optimization control parameters based on the optimal result particles.
7. The method of claim 5, wherein the unmanned ship loss fitness function is specifically:+/>wherein->Characterizing energy loss weight,/->Is an empirical function of energy loss->Characterization of efficiency loss, +.>As an empirical function of efficiency loss->Characterization of the ith control parameter,/->And->The sum of weights is 1, < >>The loss constant is characterized.
8. A control system for an amphibious unmanned watercraft based on multisensor fusion, the system comprising:
the unmanned ship operation database obtaining module is used for obtaining an amphibious unmanned ship operation database, and the amphibious unmanned ship operation database comprises water unmanned ship operation data information and land unmanned ship operation data information;
the database classification module is used for classifying the amphibious unmanned ship operation database according to the distributed detection sensor network to obtain an overwater unmanned ship sample data set and a land unmanned ship sample data set;
the self-adaptive model construction module is used for respectively training and constructing a water unmanned ship control self-adaptive model and a land unmanned ship control self-adaptive model based on the water unmanned ship sample data set and the land unmanned ship sample data set;
the detection information sensing module is used for sensing information of the whole operation process of the target unmanned ship based on the distributed detection sensor network, and acquiring an unmanned ship operation detection data stream;
the unmanned ship operation mode determining module is used for identifying and determining an unmanned ship operation mode according to the unmanned ship operation detection data stream;
the operation control parameter obtaining module is used for carrying out model matching on the basis of the unmanned ship operation mode and the water unmanned ship control self-adaptive model and the land unmanned ship control self-adaptive model, analyzing the unmanned ship operation detection data stream on the basis of a matching model result and outputting unmanned ship operation control parameters;
and the control parameter optimization regulation and control module is used for obtaining unmanned ship operation deviation information according to the unmanned ship operation control parameters and carrying out optimization regulation and control on the unmanned ship operation control parameters based on the unmanned ship operation deviation information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310660994.9A CN116893614B (en) | 2023-06-06 | 2023-06-06 | Control method of amphibious unmanned ship based on multi-sensor fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310660994.9A CN116893614B (en) | 2023-06-06 | 2023-06-06 | Control method of amphibious unmanned ship based on multi-sensor fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116893614A true CN116893614A (en) | 2023-10-17 |
| CN116893614B CN116893614B (en) | 2023-12-15 |
Family
ID=88312728
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310660994.9A Active CN116893614B (en) | 2023-06-06 | 2023-06-06 | Control method of amphibious unmanned ship based on multi-sensor fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116893614B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH05143108A (en) * | 1991-11-20 | 1993-06-11 | Hitachi Ltd | Adaptive control method for process |
| JP2008102720A (en) * | 2006-10-18 | 2008-05-01 | Omron Corp | Controller, and apparatus, method and program for adjusting control parameter |
| CN104331083A (en) * | 2014-11-21 | 2015-02-04 | 大连大学 | Method for optimizing wide-angle attitude control parameters of spacecraft |
| US20150309486A1 (en) * | 2010-03-29 | 2015-10-29 | Siemens Aktiengesellschaft | Engineering Tool and Method for Parameterizing a Model-Based Predictive Controller |
| CN107255920A (en) * | 2017-06-21 | 2017-10-17 | 武汉理工大学 | PID control method and apparatus and system based on network optimization algorithm |
| CN108681487A (en) * | 2018-05-21 | 2018-10-19 | 千寻位置网络有限公司 | The distributed system and tuning method of sensing algorithm arameter optimization |
| CN112099344A (en) * | 2020-07-31 | 2020-12-18 | 紫光云(南京)数字技术有限公司 | Hysteresis compensation method based on adaptive projection algorithm |
| CN112329338A (en) * | 2020-10-26 | 2021-02-05 | ***通信集团内蒙古有限公司 | Cold source system control method and device based on fish swarm optimization BP neural network |
| CN112947477A (en) * | 2021-03-23 | 2021-06-11 | 哈尔滨工程大学 | Unmanned ship course active disturbance rejection control system and control method |
| CN113467225A (en) * | 2021-06-24 | 2021-10-01 | 王珠 | Intelligent PID controller online optimization method and system |
| CN113534788A (en) * | 2021-05-13 | 2021-10-22 | 大连海事大学 | Stability control method of electric propulsion ship under four sea condition environments |
| WO2021248769A1 (en) * | 2020-06-11 | 2021-12-16 | 深圳技术大学 | Method, apparatus and system for monitoring operating state of electromechanical device |
| CN114362595A (en) * | 2022-03-18 | 2022-04-15 | 杭州四叶智能设备有限公司 | Particle swarm self-adaption-based multi-motor control method and device and electronic equipment |
-
2023
- 2023-06-06 CN CN202310660994.9A patent/CN116893614B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH05143108A (en) * | 1991-11-20 | 1993-06-11 | Hitachi Ltd | Adaptive control method for process |
| JP2008102720A (en) * | 2006-10-18 | 2008-05-01 | Omron Corp | Controller, and apparatus, method and program for adjusting control parameter |
| US20150309486A1 (en) * | 2010-03-29 | 2015-10-29 | Siemens Aktiengesellschaft | Engineering Tool and Method for Parameterizing a Model-Based Predictive Controller |
| CN104331083A (en) * | 2014-11-21 | 2015-02-04 | 大连大学 | Method for optimizing wide-angle attitude control parameters of spacecraft |
| CN107255920A (en) * | 2017-06-21 | 2017-10-17 | 武汉理工大学 | PID control method and apparatus and system based on network optimization algorithm |
| CN108681487A (en) * | 2018-05-21 | 2018-10-19 | 千寻位置网络有限公司 | The distributed system and tuning method of sensing algorithm arameter optimization |
| WO2021248769A1 (en) * | 2020-06-11 | 2021-12-16 | 深圳技术大学 | Method, apparatus and system for monitoring operating state of electromechanical device |
| CN112099344A (en) * | 2020-07-31 | 2020-12-18 | 紫光云(南京)数字技术有限公司 | Hysteresis compensation method based on adaptive projection algorithm |
| CN112329338A (en) * | 2020-10-26 | 2021-02-05 | ***通信集团内蒙古有限公司 | Cold source system control method and device based on fish swarm optimization BP neural network |
| CN112947477A (en) * | 2021-03-23 | 2021-06-11 | 哈尔滨工程大学 | Unmanned ship course active disturbance rejection control system and control method |
| CN113534788A (en) * | 2021-05-13 | 2021-10-22 | 大连海事大学 | Stability control method of electric propulsion ship under four sea condition environments |
| CN113467225A (en) * | 2021-06-24 | 2021-10-01 | 王珠 | Intelligent PID controller online optimization method and system |
| CN114362595A (en) * | 2022-03-18 | 2022-04-15 | 杭州四叶智能设备有限公司 | Particle swarm self-adaption-based multi-motor control method and device and electronic equipment |
Non-Patent Citations (1)
| Title |
|---|
| 褚式新等: "基于PSO的无人艇操纵响应模型参数辨识结果优化", 《武汉理工大学学报(交通科学与工程版)》, vol. 44, no. 5 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116893614B (en) | 2023-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110033051B (en) | Fishing trawler behavior discrimination method based on multi-step clustering | |
| Zhang et al. | A machine learning method for the prediction of ship motion trajectories in real operational conditions | |
| CN111399541B (en) | Unmanned aerial vehicle whole-region reconnaissance path planning method of unsupervised learning type neural network | |
| CN110018453B (en) | Intelligent model identification method based on aircraft track characteristics | |
| CN110851957A (en) | Atmospheric data sensing system resolving method based on deep learning | |
| CN114048889A (en) | Aircraft trajectory prediction method based on long-term and short-term memory network | |
| CN108960421A (en) | The unmanned surface vehicle speed of a ship or plane online forecasting method based on BP neural network of improvement | |
| Zhang et al. | Probabilistic planning and risk evaluation based on ensemble weather forecasting | |
| CN111694913A (en) | Ship AIS (automatic identification System) track clustering method and device based on convolution self-encoder | |
| CN116907282B (en) | Unmanned target aircraft ultra-low altitude flight control method based on artificial intelligence algorithm | |
| CN113283653A (en) | Ship track prediction method based on machine learning and AIS data | |
| CN113589272A (en) | Automatic generation method for target tracking equipment on-duty log | |
| CN110503148B (en) | Point cloud object identification method with scale invariance | |
| CN116954264B (en) | Distributed high subsonic unmanned aerial vehicle cluster control system and method thereof | |
| CN116893614B (en) | Control method of amphibious unmanned ship based on multi-sensor fusion | |
| CN112800075B (en) | Ship manipulation prediction database updating method based on six-degree-of-freedom attitude data of real ship | |
| CN110779526B (en) | Path planning method, device and storage medium | |
| CN117390498A (en) | Flight capability assessment method of fixed wing cluster unmanned aerial vehicle based on Transformer model | |
| Lui et al. | Long Short-Term Memory-Based Neural Networks for Missile Maneuvers Trajectories Prediction⋆ | |
| CN110908404A (en) | AUV intelligent observation motion method based on data driving | |
| Karataş et al. | Trajectory prediction for maritime vessels using ais data | |
| KR102343507B1 (en) | Variable pre-swirl stator and method for regulating angle thereof | |
| CN113030849B (en) | Near field source positioning method based on self-encoder and parallel network | |
| CN112698666B (en) | Aircraft route optimization method based on meteorological grid | |
| Zhang et al. | Multi-dimensional local weighted regression ship motion identification modeling based on particle swarm 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |