CN108256180A - Unmanned plane model verification method based on multiple dimensioned Gauss feature error fit - Google Patents
Unmanned plane model verification method based on multiple dimensioned Gauss feature error fit Download PDFInfo
- Publication number
- CN108256180A CN108256180A CN201711489341.XA CN201711489341A CN108256180A CN 108256180 A CN108256180 A CN 108256180A CN 201711489341 A CN201711489341 A CN 201711489341A CN 108256180 A CN108256180 A CN 108256180A
- Authority
- CN
- China
- Prior art keywords
- model
- verified
- error
- fitting
- function
- 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
- 238000012795 verification Methods 0.000 title claims abstract description 75
- 238000000034 method Methods 0.000 title claims abstract description 46
- 230000006870 function Effects 0.000 claims abstract description 86
- 238000011156 evaluation Methods 0.000 claims abstract description 27
- 230000000877 morphologic effect Effects 0.000 claims abstract description 23
- 238000013178 mathematical model Methods 0.000 claims description 66
- 238000005070 sampling Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 5
- NJPPVKZQTLUDBO-UHFFFAOYSA-N novaluron Chemical compound C1=C(Cl)C(OC(F)(F)C(OC(F)(F)F)F)=CC=C1NC(=O)NC(=O)C1=C(F)C=CC=C1F NJPPVKZQTLUDBO-UHFFFAOYSA-N 0.000 claims 4
- 208000011580 syndromic disease Diseases 0.000 claims 2
- 238000012512 characterization method Methods 0.000 claims 1
- 238000013461 design Methods 0.000 abstract description 10
- 238000004088 simulation Methods 0.000 abstract description 5
- 238000004422 calculation algorithm Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 238000012827 research and development Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000001808 coupling effect Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005293 physical law Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- 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
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
- G05D1/104—Simultaneous control of position or course in three dimensions specially adapted for aircraft involving a plurality of aircrafts, e.g. formation flying
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
本发明涉及航空航天领域，为结合无人机执行任务需求和控制器设计需求，给出定量模型验证估结论。并借助于计算机虚拟仿真，在仿真平台上对本发明所提方法进行了仿真验证。为此，本发明采用的技术方案是，基于多尺度高斯特征误差拟合的无人机模型验证方法，包括以下四个步骤：①求取待验证数据基础误差，经过处理后的基础误差将在时间域内反映实际飞行以及待验证模型输出的误差信息；②多尺度高斯特征误差拟合，采用多重高斯函数表征误差信息的形态特征；③尺度拟合函数匹配，从形态特征的角度入手，逐一匹配实际飞行数据与待验证模型输出数据；④定义验证评价函数，给出定量验证准则。本发明主要应用于航空航天场合。
The invention relates to the field of aerospace and provides quantitative model verification and evaluation conclusions in order to combine the requirements of unmanned aerial vehicles to perform tasks and the requirements of controller design. And with the help of computer virtual simulation, the method proposed by the present invention is simulated and verified on the simulation platform. For this reason, the technical solution adopted by the present invention is a UAV model verification method based on multi-scale Gaussian characteristic error fitting, which includes the following four steps: ① Find the basic error of the data to be verified, and the processed basic error will be in The time domain reflects the actual flight and the error information output by the model to be verified; ②Multi-scale Gaussian feature error fitting, using multiple Gaussian functions to characterize the morphological characteristics of the error information; ③Scale fitting function matching, starting from the perspective of morphological features, matching one by one The actual flight data and the output data of the model to be verified; ④ Define the verification evaluation function and give the quantitative verification criteria. The present invention is mainly applied to aerospace occasions.
Description
技术领域technical field
本发明涉及航空航天领域，主要涉及无人机数学模型的模型验证问题，具体讲,涉及多个无人机数学模型相比较时，输出数据多尺度形态特征提取及定量分析。The invention relates to the field of aerospace, and mainly relates to the model verification problem of a mathematical model of an unmanned aerial vehicle, specifically, when comparing multiple mathematical models of an unmanned aerial vehicle, multi-scale morphological feature extraction and quantitative analysis of output data.
背景技术Background technique
无人机智能控制的等级越来越高，其任务的复杂性也越来越高，对控制系统的要求也越来越严格。数学模型作为自动控制算法设计的基础，在控制器设计和系统稳定性证明等方面发挥了很大的作用。随着飞行器飞行技术的发展、飞行任务需求的多样化、飞行速度的提升及飞行机动性的提高，飞行器的设计越来越复杂，结构设计趋于多样化。无人机空气动力系统、飞行控制系统、发动机系统、弹性结构系统之间的耦合作用也越来越明显。The level of intelligent control of drones is getting higher and higher, the complexity of its tasks is getting higher and higher, and the requirements for control systems are becoming more and more stringent. As the basis of automatic control algorithm design, mathematical model plays a great role in controller design and system stability proof. With the development of aircraft flight technology, the diversification of flight mission requirements, the improvement of flight speed and flight maneuverability, the design of aircraft is becoming more and more complex, and the structural design tends to be diversified. The coupling effect among UAV aerodynamic system, flight control system, engine system and elastic structure system is becoming more and more obvious.
建立无人机数学模型的过程中，采用的建模方法主要分为实验建模方法和机理建模方法。实验建模方法是采用实际飞行实验的方式取实验飞行数据进行数学建模。机理建模的方法是采用机理推导的方式获得无人机的作用规律。在建立数学模型的过程中不可避免的会有简化和近似的过程，不同的建模方法所得到的模型精度不同，复杂度也不尽相同。控制器设计人员需要在设计控制器结构和证明闭环系统稳定性等控制性能时用到数学模型。精度过差的数学模型将会使得理论上优秀的控制算法，在实际应用中效果不理想。精度高的数学模型往往结构更为复杂，非线性特性更强，而结构过于复杂的数学模型，在控制器设计时会使得性能良好控制算法难以应用，且难以证明闭环系统的控制性能。因此，对于无人机数学模型的评价需要综合考虑精度和复杂性等特性。为了给控制器设计人员提供更为合适的数学模型，需要能够客观评价无人机数学模型优劣的模型验证方法。In the process of establishing the UAV mathematical model, the modeling methods used are mainly divided into experimental modeling methods and mechanism modeling methods. The experimental modeling method is to use the actual flight experiment to take the experimental flight data for mathematical modeling. The method of mechanism modeling is to use the method of mechanism derivation to obtain the action law of UAV. In the process of establishing a mathematical model, there will inevitably be a process of simplification and approximation, and the accuracy and complexity of the models obtained by different modeling methods are different. Controller designers need mathematical models when designing controller structures and proving control properties such as closed-loop system stability. A mathematical model with poor precision will make a theoretically excellent control algorithm less effective in practical applications. A mathematical model with high precision tends to have a more complex structure and stronger nonlinear characteristics, while a mathematical model with an overly complex structure will make it difficult to apply a good performance control algorithm when designing a controller, and it is difficult to prove the control performance of a closed-loop system. Therefore, the evaluation of the UAV mathematical model needs to comprehensively consider the characteristics of accuracy and complexity. In order to provide controller designers with a more suitable mathematical model, a model verification method that can objectively evaluate the pros and cons of the UAV mathematical model is needed.
现有的模型验证方法大多从比较实际飞行数据和数学模型输出的差别入手，重点考察误差的大小，误差越小的数学模型，表现越好。然而在模型建立的过程中，不可避免的会出现模型简化，如忽略特定物理规律、在获得解析形式数学模型过程中的函数近似等。无人机数学模型输出与实际飞行数据很难完全吻合，单一的要求误差最小得到的最优数学模型可能会是数学表达非常复杂的数学模型。过于复杂的数学模型将会给控制器设计和地面仿真工作带来难题和不必要的资源消耗。Most of the existing model verification methods start by comparing the difference between the actual flight data and the output of the mathematical model, and focus on the size of the error. The smaller the error, the better the performance of the mathematical model. However, in the process of model building, model simplification inevitably occurs, such as ignoring specific physical laws, function approximation in the process of obtaining an analytical formal mathematical model, and so on. The output of the UAV mathematical model is difficult to completely match the actual flight data, and the optimal mathematical model obtained by a single requirement of minimum error may be a mathematical model with very complex mathematical expressions. An overly complex mathematical model will bring difficulties and unnecessary resource consumption to controller design and ground simulation.
为了解决这一问题，本发明重点关注如何评价复杂度适当的无人机数学模型。当过于复杂的数学模型无法使用时，模型使用者需要一种能够合理的评价数学模型复杂度是否适当的模型验证方法。现有的模型验证方法大多是在时间域内对无人机数学模型的输出进行评价，因为无人机的输出是时间的函数，该输出会受到参考指令的影响。参考指令是地面人员或高级智能体依据无人机飞行任务给出的飞行轨迹信号。这种飞行轨迹信号带有很强的任务特点，如某个特定时间点时的阶跃信号，如某个时间段内的轨迹变更信号等。带有任务特点的参考信号会导致同样带有任务特点的无人机数学模型输出信号，这就使得无人机数学模型的输出在形态上具有一定的特征。单纯从时间域内的误差上考察无人机数学模型的优劣，很难体现这种形态特征。本发明从无人机数学模型输出数据的形态特征角度进行分析，采用多尺度模型验证思路，给予模型使用者多尺度、多自由度评价数学模型的方法。利用多重高斯特征误差拟合的方法，对无人机输出的形态特征在多尺度上进行提取以及定量分析，给出适合的模型验证结论。In order to solve this problem, the present invention focuses on how to evaluate the mathematical model of the unmanned aerial vehicle with appropriate complexity. When an overly complex mathematical model cannot be used, the model user needs a model verification method that can reasonably evaluate whether the complexity of the mathematical model is appropriate. Most of the existing model verification methods evaluate the output of the UAV mathematical model in the time domain, because the output of the UAV is a function of time, and the output will be affected by the reference command. The reference instruction is the flight trajectory signal given by the ground personnel or the advanced intelligent body according to the UAV flight mission. This kind of flight trajectory signal has strong mission characteristics, such as a step signal at a specific time point, such as a trajectory change signal within a certain period of time. The reference signal with mission characteristics will lead to the output signal of UAV mathematical model with mission characteristics, which makes the output of UAV mathematical model have certain characteristics in form. It is difficult to reflect this morphological feature by simply examining the pros and cons of the mathematical model of the UAV from the error in the time domain. The present invention analyzes the morphological characteristics of the output data of the mathematical model of the UAV, adopts the idea of multi-scale model verification, and provides a method for model users to evaluate the mathematical model with multiple scales and multiple degrees of freedom. Using the method of multiple Gaussian feature error fitting, the morphological features output by the UAV are extracted and quantitatively analyzed on multiple scales, and a suitable model verification conclusion is given.
通过对现有技术的检索，并未发现类似发明。特别是针对无人机数学模型的模型验证方法，缺乏关注数据形态特征的方法。该项技术可以为无人机数学模型验证提供新的途径，更好为复杂无人机数学模型进行模型验证工作，为高智能无人机控制算法设计者提供服务。By searching the prior art, no similar inventions have been found. Especially for the model verification method of the mathematical model of the UAV, there is a lack of methods that focus on the morphological characteristics of the data. This technology can provide a new way for UAV mathematical model verification, better model verification for complex UAV mathematical models, and provide services for highly intelligent UAV control algorithm designers.
发明内容Contents of the invention
为克服现有技术的不足，本发明旨在提出适用于无人机数据形态特征评估的模型验证方法，结合无人机执行任务需求和控制器设计需求，给出定量模型验证估结论。并借助于计算机虚拟仿真，在仿真平台上对本发明所提方法进行了仿真验证。为此，本发明采用的技术方案是，基于多尺度高斯特征误差拟合的无人机模型验证方法，包括以下四个步骤：①求取待验证数据基础误差，经过处理后的基础误差将在时间域内反映实际飞行以及待验证模型输出的误差信息；②多尺度高斯特征误差拟合，采用多重高斯函数表征误差信息的形态特征；③尺度拟合函数匹配，从形态特征的角度入手，逐一匹配实际飞行数据与待验证模型输出数据；④定义验证评价函数，给出定量验证准则。In order to overcome the deficiencies of the prior art, the present invention aims to propose a model verification method suitable for evaluating the morphological characteristics of UAV data, combining the UAV mission execution requirements and controller design requirements, and giving quantitative model verification and evaluation conclusions. And with the help of computer virtual simulation, the method proposed by the present invention is simulated and verified on the simulation platform. For this reason, the technical solution adopted by the present invention is, the UAV model verification method based on multi-scale Gaussian characteristic error fitting, including the following four steps: ① Find the basic error of the data to be verified, and the processed basic error will be in The time domain reflects the actual flight and the error information output by the model to be verified; ②Multi-scale Gaussian feature error fitting, using multiple Gaussian functions to characterize the morphological characteristics of the error information; ③Scale fitting function matching, starting from the perspective of morphological features, matching one by one The actual flight data and the output data of the model to be verified; ④ Define the verification evaluation function and give the quantitative verification criteria.
具体地：specifically:
步骤①，求取待验证数据基础误差Step ①, to obtain the basic error of the data to be verified
标准无人机飞行数据为r(t)，t为时间。飞行数据采样点数为Tr，待验证无人机数学模型数据为yi(t)，模型输出数据采样点数为，i为待验证无人机数学模型的个数，标准无人机飞行数据基础误差为待验证无人机数学模型待验证基础误差为： The standard drone flight data is r(t), where t is time. The number of flight data sampling points is T r , the data of the mathematical model of the unmanned aerial vehicle to be verified is y i (t), and the number of sampling points of the model output data is , i is the number of UAV mathematical models to be verified, and the basic error of standard UAV flight data is The basic error of the unmanned aerial vehicle mathematical model to be verified is:
步骤②，多尺度高斯特征误差拟合Step ②, multi-scale Gaussian feature error fitting
采用下式多重高斯函数对误差量er进行拟合Use the following multi-Gaussian function to fit the error amount e r
其中为拟合尺度信息，取值范围为1～Tr。为拟合尺度下的第j项拟合项系数，不同的尺度定义将获得不同的拟合结果，不同的拟合结果将对应不同的模型验证结论，为高斯函数，表达形式为：in To fit the scale information, the value range is 1~T r . fit scale The fitting coefficient of the jth item under , different scale definitions will obtain different fitting results, and different fitting results will correspond to different model verification conclusions. is a Gaussian function, expressed as:
其中为每项拟合项高斯函数的中心值，为每项拟合项高斯函数的宽度。采用下式多重高斯函数对误差量进行拟合in is the central value of each fitted Gaussian function, is the width of the Gaussian function for each fitted term. The following multi-Gaussian function is used to adjust the error amount fit
其中为拟合尺度信息，取值范围为 为拟合尺度下的第j项拟合项系数，不同的尺度定义将获得不同的拟合结果，不同的拟合结果将对应不同的模型验证结论，为高斯函数，表达形式为：in To fit the scale information, the value range is fit scale The fitting coefficient of the jth item under , different scale definitions will obtain different fitting results, and different fitting results will correspond to different model verification conclusions. is a Gaussian function, expressed as:
其中为每项拟合项高斯函数的中心值，为每项拟合项高斯函数的宽度。in is the central value of each fitted Gaussian function, is the width of the Gaussian function for each fitted term.
步骤③，尺度拟合函数匹配Step ③, scale fitting function matching
对于每个待验证无人机数学模型，经过基础误差的求取以及多尺度高斯特征误差拟合两个步骤，将得到两组数据用以模型验证，尺度下，标准无人机飞行数据拟合参数组以及尺度下，待验证无人机数学模型数据拟合参数组每个拟合参数组代表了多个高斯函数的叠加。为了进行拟合函数匹配，尺度信息取高斯函数代表了待验证数据的形态特征，对两个拟合参数组中的多个高斯函数依据拟合系数、中心值和宽度信息进行匹配，将两组数据中最相近的高斯函数进行一一对应，设代价函数为For each unmanned aerial vehicle mathematical model to be verified, after two steps of calculating the basic error and fitting the multi-scale Gaussian feature error, two sets of data will be obtained for model verification. Next, the standard UAV flight data fitting parameter set and scale Next, the data fitting parameter set of the unmanned aerial vehicle mathematical model to be verified Each fitting parameter set represents the superposition of multiple Gaussian functions. In order to perform fitting function matching, scale information is taken The Gaussian function represents the morphological characteristics of the data to be verified. Multiple Gaussian functions in the two fitting parameter groups are matched according to the fitting coefficient, central value and width information, and the most similar Gaussian functions in the two sets of data are compared one by one. Correspondingly, let the cost function be
以代价函数φi取值最小为目标，进行最小化匹配运算，建立两个拟合参数组中的各个元素Aj与Bj的一一对应关系，得到一一对应处理后的拟合参数组 Taking the minimum value of the cost function φ i as the goal, the minimum matching operation is performed, and the one-to-one correspondence between each element A j and B j in the two fitting parameter groups is established, and the fitting parameter group after one-to-one correspondence processing is obtained
步骤④，验证评价函数Step ④, verify the evaluation function
针对匹配后的拟合参数组进行验证评价函数ψi的计算，计算公式如下：For the fitted parameter set after matching Carry out the calculation of the verification evaluation function ψ i , the calculation formula is as follows:
该函数的取值作为本发明所提模型验证方法的定量验证准则。评价函数取值越低，则待验证模型与标准飞行数据越接近，评价尺度可以依据模型使用者根据需求选取，评价尺度取值越高，验证的形态精度越高。The value of this function is used as the quantitative verification criterion of the model verification method proposed in the present invention. The lower the value of the evaluation function, the closer the model to be verified is to the standard flight data. The evaluation scale can be selected according to the needs of the model user. The higher the value of the evaluation scale, the higher the morphological accuracy of the verification.
本发明的特点及有益效果是：Features and beneficial effects of the present invention are:
本发明可以针对多个验证模型的比较工作开展定量模型验证，针对每个待验证无人机数学模型给出一个定量的模型验证结果。该方法不受到模型结构以及建模方法的限制，可以对阶数高、强非线性、强耦合特征数学模型进行验证。且该验证方法可以随模型使用者的具体应用环境进行尺度调整，尺度取值可以从1取至全采样点数。可以以精度评价为优先，选择小尺度进行高精度模型验证，选出精度最优模型。也可以以适用性评价为优先，选择较大尺度进行形态特征模型验证，选出形态特征适中，模型结构适合的最优模型。The present invention can carry out quantitative model verification for the comparison work of multiple verification models, and provides a quantitative model verification result for each unmanned aerial vehicle mathematical model to be verified. This method is not limited by the model structure and modeling method, and can verify mathematical models with high-order, strong nonlinear, and strong coupling features. Moreover, the verification method can be scaled according to the specific application environment of the model user, and the scale value can be taken from 1 to the full number of sampling points. Accuracy evaluation can be prioritized, small scales can be selected for high-precision model verification, and the model with the best accuracy can be selected. It is also possible to give priority to applicability evaluation, select a larger scale for morphological feature model verification, and select the optimal model with moderate morphological features and suitable model structure.
社会效益及经济效益：此项发明对无人机数学模型的建模工作以及高智能无人机控制算法的提高具有十分重要的推动意义。本发明可以提供有效的无人机模型验证方法，可以为建模工作者提供评价依据，为控制器设计工作者筛选适合控制器设计的复杂度适当无人机数学模型。特别是对于模型结构复杂、控制性能要求高的无人机系统开发具有有效的推动作用。为高智能化、高性能的控制方法在新型无人机系统上的使用工作服务，缩短研发周期，节省大量实际飞行试验消耗，降低无人机开发成本，推动无人机智能化实现进程。Social and economic benefits: This invention has very important significance for the modeling work of the UAV mathematical model and the improvement of the control algorithm of the high-intelligence UAV. The present invention can provide an effective verification method for the UAV model, can provide evaluation basis for modeling workers, and select the appropriate complexity UAV mathematical model suitable for controller design for controller design workers. Especially for the development of UAV systems with complex model structures and high control performance requirements, it has an effective role in promoting. Serving the use of highly intelligent and high-performance control methods on new UAV systems, shortening the research and development cycle, saving a lot of actual flight test consumption, reducing UAV development costs, and promoting the realization of UAV intelligence.
附图说明：Description of drawings:
附图1无人机模型验证系统结构图。Accompanying drawing 1 is the structural diagram of the UAV model verification system.
附图2基于多尺度高斯特征误差拟合的无人机模型验证方法流程图。Accompanying drawing 2 is the flow chart of the UAV model verification method based on multi-scale Gaussian feature error fitting.
附图3基于多尺度高斯特征误差拟合的无人机模型验证方法软件实现界面图。Accompanying drawing 3 is based on multi-scale Gaussian feature error fitting UAV model verification method software implementation interface diagram.
具体实施方式Detailed ways
本发明研究无人机数学模型的验证问题，重点关注多个无人机数学模型向比较下的定量验证问题。考虑模型结构复杂性，同时考虑非线性特性、耦合特性、高阶特性下的模型验证问题。分析待验证数学模型的形态特征，给出多尺度模型验证准则。The invention studies the verification problem of the mathematical model of the UAV, and focuses on the quantitative verification problem under the comparison of multiple UAV mathematical models. Consider the complexity of the model structure, and at the same time consider the model verification problems under nonlinear characteristics, coupling characteristics, and high-order characteristics. The morphological characteristics of the mathematical model to be verified are analyzed, and the multi-scale model verification criterion is given.
本发明提出的基于多尺度高斯特征误差拟合的无人机模型验证方法，旨在探索无人机高智能算法开发时对复杂数学模型提出的新要求，所建立的模型验证方法有重要的理论意义和实际应用前景。多尺度的自由设定将给控制器开发人员模型性能方面的分析信息，促进先进控制算法在复杂无人机系统中的应用进程，并为数学模型的建立提供验证工作方面的支撑。可以推进无人机智能化发展，缩短新型无人机系统的研发周期，既快速高效又节省开支，具有很好的应用前景与经济价值。The UAV model verification method based on multi-scale Gaussian feature error fitting proposed by the present invention aims to explore new requirements for complex mathematical models in the development of UAV high-intelligence algorithms. The established model verification method has important theoretical Significance and practical application prospects. The free setting of multiple scales will provide controller developers with analytical information on model performance, promote the application of advanced control algorithms in complex UAV systems, and provide verification support for the establishment of mathematical models. It can promote the intelligent development of UAVs and shorten the research and development cycle of new UAV systems. It is fast, efficient and cost-saving, and has good application prospects and economic value.
本发明旨在克服现有技术的不足，以理论方法和虚拟仿真技术相结合为主要研究手段，针对无人机数学模型验证问题，提出适用于无人机数据形态特征评估的模型验证方法，结合无人机执行任务需求和控制器设计需求，给出定量模型验证评估结论。并借助于计算机虚拟仿真，在仿真平台上对本发明所提方法进行了仿真验证。The present invention aims to overcome the deficiencies of the prior art, and uses the combination of theoretical methods and virtual simulation technology as the main research means to address the problem of mathematical model verification of UAVs, and proposes a model verification method suitable for the evaluation of UAV data morphological characteristics. UAV execution mission requirements and controller design requirements, and quantitative model verification evaluation conclusions are given. And with the help of computer virtual simulation, the method proposed by the present invention is simulated and verified on the simulation platform.
基于多尺度风扰分析的无人机机群协调控制系统性能评估方法包括以下四个步骤：①求取待验证数据基础误差，经过处理后的基础误差将在时间域内反映实际飞行以及待验证模型输出的误差信息。②多尺度高斯特征误差拟合，采用多重高斯函数表征误差信息的形态特征。③尺度拟合函数匹配，从形态特征的角度入手，逐一匹配实际飞行数据与待验证模型输出数据。④定义验证评价函数，给出定量验证准则，该值的大小直接对应待验证模型的优劣评价。The performance evaluation method of UAV fleet coordination control system based on multi-scale wind disturbance analysis includes the following four steps: ① Obtain the basic error of the data to be verified, and the processed basic error will reflect the actual flight and the output of the model to be verified in the time domain error information. ② Multi-scale Gaussian feature error fitting, using multiple Gaussian functions to characterize the morphological characteristics of error information. ③ Scale fitting function matching, starting from the perspective of morphological characteristics, matching the actual flight data and the output data of the model to be verified one by one. ④Define the verification evaluation function and give the quantitative verification criterion. The size of this value directly corresponds to the evaluation of the quality of the model to be verified.
结合附图对本发明作进一步详述。The present invention will be described in further detail in conjunction with the accompanying drawings.
参见图1，无人机模型验证结构中，标准飞行数据以及多个待验证无人机数学模型数据，将接受相同的参考指令信号。该参考指令信号为飞行位置和飞行姿态随时间的期望取值，依据无人机的飞行任务、地形特征、环境特征等由地面站或高级控制单元给出。在相同的参考指令，以及相同的控制结构作用下，记录飞行任务时段内的飞行数据，对标准飞行数据进行采样作为待验证系统的基准数据。同时分别对待验证无人机数学模型输出数据进行采样，作为验证模型性能优劣的数据集。Referring to Figure 1, in the UAV model verification structure, the standard flight data and multiple UAV mathematical model data to be verified will receive the same reference command signal. The reference command signal is the expected value of the flight position and flight attitude over time, and is given by the ground station or the advanced control unit according to the flight mission, terrain characteristics, and environmental characteristics of the UAV. Under the same reference command and the same control structure, the flight data during the flight mission period is recorded, and the standard flight data is sampled as the benchmark data of the system to be verified. At the same time, the output data of the mathematical model of the unmanned aerial vehicle to be verified is sampled separately as a data set for verifying the performance of the model.
参见图2，为本算法的具体实施流程图，具体步骤为：Referring to Figure 2, it is a flow chart of the specific implementation of this algorithm, and the specific steps are:
步骤①，求取待验证数据基础误差Step ①, to obtain the basic error of the data to be verified
标准无人机飞行数据为r(t)，t为时间。飞行数据采样点数为Tr，待验证无人机数学模型数据为yi(t)，模型输出数据采样点数为i为待验证无人机数学模型的个数，标准无人机飞行数据基础误差为待验证无人机数学模型待验证基础误差为： The standard drone flight data is r(t), where t is time. The number of flight data sampling points is T r , the data of the mathematical model of the unmanned aerial vehicle to be verified is y i (t), and the number of sampling points of the model output data is i is the number of UAV mathematical models to be verified, and the basic error of standard UAV flight data is The basic error of the unmanned aerial vehicle mathematical model to be verified is:
步骤②，多尺度高斯特征误差拟合Step ②, multi-scale Gaussian feature error fitting
采用下式多重高斯函数对误差量er进行拟合Use the following multi-Gaussian function to fit the error amount e r
其中为拟合尺度信息，取值范围为1～Tr。为拟合尺度下的第j项拟合项系数，不同的尺度定义将获得不同的拟合结果，不同的拟合结果将对应不同的模型验证结论，为高斯函数，表达形式为：in To fit the scale information, the value range is 1~T r . fit scale The fitting coefficient of the jth item under , different scale definitions will obtain different fitting results, and different fitting results will correspond to different model verification conclusions. is a Gaussian function, expressed as:
其中为每项拟合项高斯函数的中心值，为每项拟合项高斯函数的宽度。采用下式多重高斯函数对误差量进行拟合in is the central value of each fitted Gaussian function, is the width of the Gaussian function for each fitted term. The following multi-Gaussian function is used to adjust the error amount to fit
其中为拟合尺度信息，取值范围为 为拟合尺度下的第j项拟合项系数，不同的尺度定义将获得不同的拟合结果，不同的拟合结果将对应不同的模型验证结论，为高斯函数，表达形式为：in To fit the scale information, the value range is fit scale The fitting coefficient of the jth item under , different scale definitions will obtain different fitting results, and different fitting results will correspond to different model verification conclusions. is a Gaussian function, expressed as:
其中为每项拟合项高斯函数的中心值，为每项拟合项高斯函数的宽度。in is the central value of each fitted Gaussian function, is the width of the Gaussian function for each fitted term.
步骤③，尺度拟合函数匹配Step ③, scale fitting function matching
对于每个待验证无人机数学模型，经过基础误差的求取以及多尺度高斯特征误差拟合两个步骤，将得到两组数据用以模型验证，尺度下，标准无人机飞行数据拟合参数组以及尺度下，待验证无人机数学模型数据拟合参数组每个拟合参数组代表了多个高斯函数的叠加。为了进行拟合函数匹配，尺度信息取高斯函数代表了待验证数据的形态特征，对两个拟合参数组中的多个高斯函数依据拟合系数、中心值和宽度信息进行匹配，将两组数据中最相近的高斯函数进行一一对应，设代价函数为For each unmanned aerial vehicle mathematical model to be verified, after two steps of calculating the basic error and fitting the multi-scale Gaussian feature error, two sets of data will be obtained for model verification. Next, the standard UAV flight data fitting parameter set and scale Next, the data fitting parameter set of the unmanned aerial vehicle mathematical model to be verified Each fitting parameter set represents the superposition of multiple Gaussian functions. In order to perform fitting function matching, scale information is taken The Gaussian function represents the morphological characteristics of the data to be verified. Multiple Gaussian functions in the two fitting parameter groups are matched according to the fitting coefficient, central value and width information, and the most similar Gaussian functions in the two sets of data are compared one by one. Correspondingly, let the cost function be
以代价函数φi取值最小为目标，进行最小化匹配运算，建立两个拟合参数组中的各个元素Aj与Bj的一一对应关系，得到一一对应处理后的拟合参数组 Taking the minimum value of the cost function φ i as the goal, the minimum matching operation is performed, and the one-to-one correspondence between each element A j and B j in the two fitting parameter groups is established, and the fitting parameter group after one-to-one correspondence processing is obtained
步骤④，验证评价函数Step ④, verify the evaluation function
针对匹配后的拟合参数组进行验证评价函数ψi的计算，计算公式如下：For the fitted parameter set after matching Carry out the calculation of the verification evaluation function ψ i , the calculation formula is as follows:
该函数的取值作为本发明所提模型验证方法的定量验证准则。评价函数取值越低，则待验证模型与标准飞行数据越接近，评价尺度可以依据模型使用者根据需求选取，评价尺度取值越高，验证的形态精度越高。The value of this function is used as the quantitative verification criterion of the model verification method proposed in the present invention. The lower the value of the evaluation function, the closer the model to be verified is to the standard flight data. The evaluation scale can be selected according to the needs of the model user. The higher the value of the evaluation scale, the higher the morphological accuracy of the verification.
参见图3，基于多尺度高斯特征误差拟合的无人机模型验证方法人机交互主控软件界面采用Matlab引擎技术进行开发。界面中包含六个功能区：无人机数学模型区、验证参数设定区、闭环分析区、验证结果区、实时飞行数据区和输出数据区。无人机数学模型区功能包括待验证数学模型的导入、仿真次数设定、实际飞行数据设置和Simulink程序运行等功能。该功能主要进行数据的采集及准备工作，预留了待验证数学模型导入接口，可接入多个待验证数学模型。待验证数学模型将以数据形式进行验证分析，因此该发明将不会受到数学模型具体解析形式及建模方法的影响，适用范围广。验证参数设定区功能包括验证尺度的选择功能。验证尺度的选择作为验证重要开放参数，可以进行自己设置，尺度的改变将直接影响验证结论。闭环分析区功能包括参考指令希望值设定功能、匹配性和稳定性分析功能。验证结果区功能包括数据匹配测试、拟合结果显示、基础误差显示等功能。该功能区给出了验证过程中数据的图形展示功能，这些分析结果图形的展示将给出详细的模型误差形态信息，为模型使用者提供全面的待验证无人机数学模型信息，并可以依据这些结果适当调整尺度信息。实时飞行区功能包括实时飞行数据导入功能。输出数据区功能包括在时间域内显示数据结果。See Figure 3, the UAV model verification method based on multi-scale Gaussian feature error fitting The human-computer interaction main control software interface is developed using Matlab engine technology. The interface contains six functional areas: UAV mathematical model area, verification parameter setting area, closed-loop analysis area, verification result area, real-time flight data area and output data area. The functions of the UAV mathematical model area include the import of the mathematical model to be verified, the setting of the number of simulations, the setting of the actual flight data, and the operation of the Simulink program. This function is mainly for data collection and preparation, and an interface for importing mathematical models to be verified is reserved, and multiple mathematical models to be verified can be connected. The mathematical model to be verified will be verified and analyzed in the form of data, so the invention will not be affected by the specific analytical form and modeling method of the mathematical model, and has a wide range of applications. The function of the verification parameter setting area includes the selection function of the verification scale. The choice of verification scale is an important open parameter for verification, which can be set by yourself, and the change of the scale will directly affect the verification conclusion. The function of the closed-loop analysis area includes the function of setting the desired value of the reference command, matching and stability analysis. The functions of the verification result area include data matching test, fitting result display, basic error display and other functions. This functional area provides the graphic display function of the data in the verification process. The graphic display of these analysis results will provide detailed model error information, provide model users with comprehensive information on the mathematical model of the unmanned aerial vehicle to be verified, and can be based on These results are properly adjusted for scaling information. The real-time flight zone function includes the real-time flight data import function. The output data area function includes displaying data results in the time domain.
Claims (2)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711489341.XA CN108256180B (en) | 2017-12-30 | 2017-12-30 | Unmanned aerial vehicle model verification method based on multi-scale Gaussian characteristic error fitting |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711489341.XA CN108256180B (en) | 2017-12-30 | 2017-12-30 | Unmanned aerial vehicle model verification method based on multi-scale Gaussian characteristic error fitting |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108256180A true CN108256180A (en) | 2018-07-06 |
| CN108256180B CN108256180B (en) | 2023-04-18 |
Family
ID=62725364
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711489341.XA Active CN108256180B (en) | 2017-12-30 | 2017-12-30 | Unmanned aerial vehicle model verification method based on multi-scale Gaussian characteristic error fitting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108256180B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114446418A (en) * | 2020-11-04 | 2022-05-06 | 浙江荷清柔性电子技术有限公司 | Packet loss processing method, device, computer equipment and storage medium for uterine contraction data |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103353985A (en) * | 2013-03-27 | 2013-10-16 | 西华大学 | Measurement method for image Gaussian Blur |
| CN105004337A (en) * | 2015-08-19 | 2015-10-28 | 青岛科技大学 | Straight line matching based autonomous navigation method for agricultural unmanned aerial vehicle |
| US20160068267A1 (en) * | 2014-09-05 | 2016-03-10 | SZ DJI Technology Co., Ltd | Context-based flight mode selection |
| WO2017030737A1 (en) * | 2015-08-20 | 2017-02-23 | Motionloft, Inc. | Object detection and analysis via unmanned aerial vehicle |
| CN106595603A (en) * | 2016-11-23 | 2017-04-26 | 华南农业大学 | Method for detecting canopy lodging cone caused by unmanned aerial vehicle rotor airflow |
| CN106682361A (en) * | 2017-01-13 | 2017-05-17 | 沈阳航空航天大学 | System and method for simulating flight tracks of unmanned aerial vehicles on basis of GPS (global positioning system) simulation |
| CN106844887A (en) * | 2016-12-29 | 2017-06-13 | 深圳市道通智能航空技术有限公司 | The dynamic modeling method and device of rotor wing unmanned aerial vehicle |
| US20170349267A1 (en) * | 2016-06-01 | 2017-12-07 | Regents Of The University Of Minnesota | Fault-tolerant aircraft flight control using a subset of aerodynamic control surfaces |
-
2017
- 2017-12-30 CN CN201711489341.XA patent/CN108256180B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103353985A (en) * | 2013-03-27 | 2013-10-16 | 西华大学 | Measurement method for image Gaussian Blur |
| US20160068267A1 (en) * | 2014-09-05 | 2016-03-10 | SZ DJI Technology Co., Ltd | Context-based flight mode selection |
| CN105004337A (en) * | 2015-08-19 | 2015-10-28 | 青岛科技大学 | Straight line matching based autonomous navigation method for agricultural unmanned aerial vehicle |
| WO2017030737A1 (en) * | 2015-08-20 | 2017-02-23 | Motionloft, Inc. | Object detection and analysis via unmanned aerial vehicle |
| US20170349267A1 (en) * | 2016-06-01 | 2017-12-07 | Regents Of The University Of Minnesota | Fault-tolerant aircraft flight control using a subset of aerodynamic control surfaces |
| CN106595603A (en) * | 2016-11-23 | 2017-04-26 | 华南农业大学 | Method for detecting canopy lodging cone caused by unmanned aerial vehicle rotor airflow |
| CN106844887A (en) * | 2016-12-29 | 2017-06-13 | 深圳市道通智能航空技术有限公司 | The dynamic modeling method and device of rotor wing unmanned aerial vehicle |
| CN106682361A (en) * | 2017-01-13 | 2017-05-17 | 沈阳航空航天大学 | System and method for simulating flight tracks of unmanned aerial vehicles on basis of GPS (global positioning system) simulation |
Non-Patent Citations (2)
| Title |
|---|
| 屈耀红等: "曲线拟合滤波在无人机导航数据处理中的应用", 《系统工程与电子技术》 * |
| 郭荣幸等: "基于非对称sinc函数拟合的激光测距算法优化", 《红外与激光工程》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114446418A (en) * | 2020-11-04 | 2022-05-06 | 浙江荷清柔性电子技术有限公司 | Packet loss processing method, device, computer equipment and storage medium for uterine contraction data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108256180B (en) | 2023-04-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105115692B (en) | The pneumatic Forecasting Methodology that a kind of CFD numerical simulations are combined with wind tunnel test | |
| CN108170162B (en) | Multi-scale wind disturbance analysis performance evaluation method of UAV fleet coordination control system | |
| CN106529093A (en) | Pneumatic/structure/static aeroelasticity coupling optimizing method for high-aspect-ratio wing | |
| CN114065662A (en) | Method suitable for rapidly predicting airfoil flow field with variable grid topology | |
| CN110223565A (en) | A kind of flight simulation method, device, equipment and storage medium | |
| CN113536458B (en) | Quick geometric modeling method of deformable wing based on class shape function transformation | |
| CN116738891A (en) | LU-SGS improvement method for enhancing simulation stability of aircraft flow field | |
| CN106092496A (en) | For the APDSMC flow field detection method flowed across yardstick | |
| CN106407601B (en) | A data processing method of aerodynamic characteristics based on data mining technology | |
| CN108256180A (en) | Unmanned plane model verification method based on multiple dimensioned Gauss feature error fit | |
| Yu et al. | Comparing potential flow solvers for aerodynamic characteristics estimation of the T-Flex UAV | |
| Periaux et al. | Fast reconstruction of aerodynamic shapes using evolutionary algorithms and virtual nash strategies in a CFD design environment | |
| CN118013891B (en) | A method and system for optimizing aerodynamic performance based on numerical wind tunnel | |
| CN107808040B (en) | Unmanned aerial vehicle model verification method based on variable-scale space neighborhood evaluation | |
| Zhu et al. | Design of an RBF surrogate model for low Reynolds number airfoil based on transfer learning | |
| CN114169228A (en) | A Real-time Wind Field Modeling Method for Helicopter Engineering Simulators | |
| CN109635376B (en) | A UAV-based modeling method and system | |
| CN118917036A (en) | Aircraft multimode fusion approximate characterization method based on interest parameter domain | |
| CN111506959A (en) | Universal flight control system and method for flight simulation training | |
| CN114995191B (en) | An eVTOL aircraft simulation verification method, device, electronic equipment and storage medium | |
| CN107944201B (en) | A fast modeling method for fixed-wing aircraft | |
| CN106021640B (en) | Method and system for co-simulating vehicle collision process with multi-body dynamics and finite element explicit dynamics | |
| Tian et al. | Application of Image Classification Algorithm Based on Deep Learning in Residential Interior Design Style Recognition | |
| CN117291123B (en) | Uncertainty quantification method for numerical simulation of flying wing layout considering the influence of intake and exhaust | |
| Ghazlane | Adjoint-based aerostructural sensitivity analysis for wing design |
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 |