CN103412488B - A kind of miniature self-service gyroplane high-accuracy control method based on adaptive neural network - Google Patents
A kind of miniature self-service gyroplane high-accuracy control method based on adaptive neural network Download PDFInfo
- Publication number
- CN103412488B CN103412488B CN201310347956.4A CN201310347956A CN103412488B CN 103412488 B CN103412488 B CN 103412488B CN 201310347956 A CN201310347956 A CN 201310347956A CN 103412488 B CN103412488 B CN 103412488B
- Authority
- CN
- China
- Prior art keywords
- neural network
- adaptive
- adaptive neural
- control
- small unmanned
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Feedback Control In General (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
一种基于自适应神经网络的小型无人旋翼机高精度控制方法，涉及小型无人旋翼机反馈控制、无样本训练的自适应神经网络的构建与优化相结合的复合控制器设计。首先，针对小型无人旋翼机动力学模型，通过极点配置方法构建反馈控制系数矩阵来保证系统的初步稳定性；其次，设计具有自主更新权值特性的自适应神经网络，基于误差信息构建自适应网络权值更新矩阵来在线更新神经网络的权值矩阵，实现对扰动的估计和抑制；并设计自适应阈值优化策略，基于时间窗口内的实际位置与期望位置的误差均方差，对自适应神经网络的控制残差上限阈值进行在线更新，降低控制残差上界不精确对神经网络扰动控制量的影响，进而优化自适应神经网络扰动控制量，实现复杂环境下的小型无人旋翼机高精度姿态控制。本发明具有实时性好、动态参数响应快、对多源干扰适应性强等优点，可用于小型无人旋翼机复杂多源干扰环境下的高精度控制。
A high-precision control method for a small unmanned rotorcraft based on an adaptive neural network involves the design of a composite controller that combines the feedback control of a small unmanned rotorcraft with the construction and optimization of an adaptive neural network without sample training. Firstly, for the dynamic model of small unmanned rotorcraft, the feedback control coefficient matrix is constructed by the pole configuration method to ensure the initial stability of the system; secondly, an adaptive neural network with the characteristics of self-updating weights is designed, and an adaptive network is constructed based on error information The weight update matrix is used to update the weight matrix of the neural network online to realize the estimation and suppression of disturbance; and an adaptive threshold optimization strategy is designed, based on the error mean square error between the actual position and the expected position in the time window, the adaptive neural network The upper threshold of the control residual error is updated online to reduce the influence of the inaccurate upper bound of the control residual error on the disturbance control amount of the neural network, and then optimize the disturbance control amount of the adaptive neural network to realize the high-precision attitude of the small unmanned rotorcraft in a complex environment control. The invention has the advantages of good real-time performance, fast dynamic parameter response, strong adaptability to multi-source interference, etc., and can be used for high-precision control of small unmanned rotorcraft under complex multi-source interference environment.
Description
技术领域technical field
本发明涉及一种基于自适应神经网络的小型无人旋翼机高精度控制方法，适用于工作在空中的无人机器人自主控制领域。The invention relates to a high-precision control method for a small unmanned rotorcraft based on an adaptive neural network, which is suitable for the field of autonomous control of unmanned robots working in the air.
背景技术Background technique
小型无人旋翼机具有垂直起降、悬停等特性，通过自身携带的各类传感器可以在危险区域或市区街道等狭窄空间执行观测、信息收集等任务，具有广泛的应用前景。随着应用领域的拓展，小型无人旋翼机的工作环境也复杂多变，抗扰性强、稳定性高的小型无人旋翼机高精度控制成为研究的热点。The small unmanned rotorcraft has the characteristics of vertical take-off and landing, hovering, etc., and can perform tasks such as observation and information collection in narrow spaces such as dangerous areas or urban streets through various sensors carried by itself, and has a wide range of application prospects. With the expansion of application fields, the working environment of small unmanned rotorcraft is also complex and changeable, and the high-precision control of small unmanned rotorcraft with strong anti-interference and high stability has become a research hotspot.
作为复杂的多输入多输出控制系统，小型无人旋翼机具有非线性、强耦合、控制难度高等特性。并且小型无人旋翼机在飞行过程中存在多类干扰，如风扰、大气湍流、地面干扰、系统电磁干扰等，因此，小型无人旋翼机在扰动下的高精度控制是飞控系统的关键技术之一。As a complex multiple-input multiple-output control system, small unmanned rotorcraft has the characteristics of nonlinearity, strong coupling, and high difficulty in control. And there are many types of disturbances in the flight process of small unmanned rotorcraft, such as wind disturbance, atmospheric turbulence, ground interference, system electromagnetic interference, etc. Therefore, the high-precision control of small unmanned rotorcraft under disturbance is the key to the flight control system One of the techniques.
为提高性能，智能PID控制方法、鲁棒控制、智能控制方法等各类控制方法被用于小型无人旋翼机的飞行控制。智能PID控制器结构简单，但抗干扰能力差，小型无人旋翼机的控制性能很容易受到外界干扰影响而降低。鲁棒控制可以较好地消除小型无人旋翼机在飞行过程中存在的模型参数不精确和外界干扰问题，但鲁棒控制具有实时性较差、动态参数响应慢的特性。通过大量的样本训练，神经网络可以实现非线性自适应控制，克服小型无人旋翼机所具有的模型不确定性，以及存在多源干扰等问题，实现高精度的姿态控制，但传统的神经网络需要大量的样本数据进行训练，具有实时性差的缺点。In order to improve the performance, various control methods such as intelligent PID control method, robust control method and intelligent control method are used in the flight control of small unmanned rotorcraft. The structure of intelligent PID controller is simple, but its anti-interference ability is poor, and the control performance of small unmanned rotorcraft is easily affected by external interference and degraded. Robust control can better eliminate the inaccurate model parameters and external interference problems existing in the flight process of small unmanned rotorcraft, but robust control has the characteristics of poor real-time performance and slow response of dynamic parameters. Through a large number of sample training, the neural network can realize nonlinear adaptive control, overcome the model uncertainty of small unmanned rotorcraft, and the existence of multi-source interference, etc., to achieve high-precision attitude control, but the traditional neural network A large amount of sample data is required for training, which has the disadvantage of poor real-time performance.
发明内容Contents of the invention
本发明的技术解决问题是：针对小型无人旋翼机在执行任务时控制性能容易受到外界干扰影响的问题，提出一种基于自适应神经网络和极点配置方法相结合的复合控制方法，对小型无人旋翼机在飞行中所受的多源干扰进行估计并抑制，实现大包络范围的高精度控制。The technical solution of the present invention is to solve the problem that the control performance of small unmanned rotorcraft is easily affected by external interference when performing tasks, and proposes a composite control method based on the combination of adaptive neural network and pole configuration method, which is suitable for small unmanned rotorcraft. The multi-source interference suffered by the manned rotorcraft in flight is estimated and suppressed to achieve high-precision control with a large envelope.
本发明的技术解决方案为：针对小型无人旋翼机动力学模型，通过极点配置方法构建反馈控制系数矩阵来保证系统的初步稳定性；设计具有自主更新权值特性的自适应神经网络，基于误差信息构建自适应网络权值更新矩阵来在线更新神经网络的权值矩阵，实现对扰动的估计和抑制；并设计自适应阈值优化策略，基于时间窗口内的实际位置与期望位置的误差均方差，对自适应神经网络的控制残差上限阈值进行在线更新，实现复杂环境下的小型无人旋翼机高精度姿态控制。其实现步骤如下：The technical solution of the present invention is: aiming at the dynamic model of small unmanned rotorcraft, the feedback control coefficient matrix is constructed by the pole configuration method to ensure the preliminary stability of the system; Build an adaptive network weight update matrix to update the weight matrix of the neural network online to realize the estimation and suppression of disturbances; and design an adaptive threshold optimization strategy, based on the error mean square error between the actual position and the expected position in the time window, for The control residual upper threshold of the adaptive neural network is updated online to realize high-precision attitude control of small unmanned rotorcraft in complex environments. Its implementation steps are as follows:
(1)针对小型无人旋翼机动力学模型，通过极点配置方法构建反馈控制系数矩阵来保证系统的初步稳定性；(1) For the dynamic model of small unmanned rotorcraft, the feedback control coefficient matrix is constructed by the pole configuration method to ensure the initial stability of the system;
(2)对飞行中存在的多源干扰，设计具有自主更新权值特性的自适应神经网络，基于误差信息构建自适应网络权值更新矩阵来在线更新神经网络的权值矩阵，实现对小型无人旋翼机在飞行中所受多源干扰进行在线估计，自适应神经网络权值更新矩阵和扰动估计量表达式如下：(2) For the multi-source interference existing in the flight, design an adaptive neural network with the characteristics of self-updating weights, build an adaptive network weight update matrix based on error information to update the weight matrix of the neural network online, and realize small wireless The online estimation of the multi-source interference suffered by the human rotorcraft in flight, the adaptive neural network weight update matrix and the disturbance estimator expressions are as follows:
其中，为自适应神经网络的权值矩阵，为自适应神经网络的扰动估计量；Γi、P为对称正定矩阵，自适应神经网络的输入e＝x-xd为期望状态变量xd和实际状态变量x间的误差，B为小型无人旋翼机控制状态转移矩阵，αw为自适应神经网络的控制残差上限阈值，i*为相应矩阵的第i个行向量，*i为相应矩阵的第i个列向量，s(e)为自适应神经网络隐含层的节点函数，定义为高斯函数，其相应第j个隐含层的节点函数表达式如下：in, is the weight matrix of the adaptive neural network, is the disturbance estimator of the adaptive neural network; Γ i and P are symmetric positive definite matrices, the input of the adaptive neural network e=xx d is the error between the desired state variable x d and the actual state variable x, B is the small unmanned rotor machine control state transition matrix, α w is the control residual upper threshold of the adaptive neural network, i* is the i-th row vector of the corresponding matrix, *i is the i-th column vector of the corresponding matrix, s(e) is the self- The node function adapted to the hidden layer of the neural network is defined as a Gaussian function, and the corresponding node function expression of the jth hidden layer is as follows:
其中，μj,分别为自适应神经网络隐含层高斯函数的中心值和宽度，l 为自适应神经网络隐含层的隐含节点数；Among them, μ j , are the central value and width of the Gaussian function in the hidden layer of the adaptive neural network, respectively, and l is the number of hidden nodes in the hidden layer of the adaptive neural network;
(3)设计自适应阈值优化策略，基于时间窗口内的实际位置与期望位置的误差均方差，对自适应神经网络的控制残差上限阈值进行在线更新，实现复杂环境下的小型无人旋翼机高精度姿态控制。(3) Design an adaptive threshold optimization strategy, based on the error mean square error between the actual position and the expected position within the time window, update the upper threshold of the control residual error of the adaptive neural network online, and realize the small unmanned rotorcraft in a complex environment High precision attitude control.
本发明的基于自适应神经网络的小型无人旋翼机高精度控制方法，其中所述步骤(3)构建自适应阈值优化策略定义如下The high-precision control method of small unmanned rotorcraft based on adaptive neural network of the present invention, wherein said step (3) constructs an adaptive threshold optimization strategy and is defined as follows
式中αw为当前时间窗口采样周期内自适应神经网络的控制残差上限阈值；αw-1为上一时间窗口采样周期内自适应神经网络的控制残差上限阈值；αw-2为上二时间窗口采样周期内自适应神经网络的控制残差上限阈值；χk为上一时间窗口采样周期内小型无人旋翼机实际位置xxm与期望位置xxd的均方差；χk-t为上二时间窗口采样周期内小型无人旋翼机实际位置xxm与期望位置xxd的均方差；eei为时间窗口采样周期内小型无人旋翼机实际位置xxm与期望位置xxd的差值；t为时间窗口采样周期内采样点数；k1为控制参数，η1,η2分别为时间窗口采样周期内小型无人旋翼机实际位置xxm与期望位置xxd的最大绝对误差值和平均误差值。where α w is the control residual upper threshold of the adaptive neural network in the current time window sampling period; α w-1 is the control residual upper threshold of the adaptive neural network in the previous time window sampling period; α w-2 is The control residual upper threshold of the adaptive neural network in the sampling period of the last two time windows; χ k is the mean square error between the actual position xx m and the expected position xx d of the small unmanned rotorcraft in the sampling period of the previous time window; The mean square error of the actual position xx m and the expected position xx d of the small unmanned rotorcraft within the sampling period of the second time window; ee i is the difference between the actual position xx m and the expected position xx d of the small unmanned rotorcraft within the sampling period of the time window; t is the number of sampling points in the time window sampling period; k 1 is the control parameter, η 1 and η 2 are the maximum absolute error value and average error between the actual position xx m and the expected position xx d of the small unmanned rotorcraft in the time window sampling period value.
本发明与现有技术相比的优点在于：The advantage of the present invention compared with prior art is:
(1)本发明在通过极点配置方法构建反馈控制系数矩阵来保证系统的初步稳定性的基础上，通过无样本训练的自适应神经网络，对小型无人旋翼机在飞行过程中所受扰动进行估计和抑制，具有抗干扰性强和便于设计的优点；(1) On the basis of ensuring the initial stability of the system by constructing the feedback control coefficient matrix through the pole configuration method, the present invention conducts a disturbance on the small-sized unmanned rotorcraft during the flight through an adaptive neural network without sample training. Estimation and suppression, with the advantages of strong anti-interference and easy design;
(2)本发明在传统反馈控制保证系统稳定的情况下，进一步利用自适应神经网络估计并抑制小型无人旋翼机在飞行过程中所受的扰动，可以实时根据飞行器的状态信息调整舵量，不仅具有结构简单和控制方便的特性，同时控制方法实时性好、动态参数响应快，能够满足小型无人旋翼机负载环境下高精度控制需求；(2) The present invention further utilizes the self-adaptive neural network to estimate and suppress the disturbance that the small-sized unmanned rotorcraft is subjected to during flight under the condition that the traditional feedback control guarantees system stability, and can adjust the rudder amount according to the state information of the aircraft in real time, It not only has the characteristics of simple structure and convenient control, but also has good real-time control method and fast response to dynamic parameters, which can meet the high-precision control requirements of small unmanned rotorcraft load environments;
(3)本发明仅需要根据小型无人旋翼机实际飞行过程中采集的状态信息，基于解算得到的位置误差，就可以在线更新自适应神经网络的权值，不需要任何样本训练，具有数据获取方便、计算简单的优点。(3) The present invention only needs to update the weights of the self-adaptive neural network online based on the state information collected during the actual flight of the small unmanned rotorcraft, based on the position error obtained from the solution, without any sample training, and has data It has the advantages of convenient acquisition and simple calculation.
附图说明Description of drawings
图1为小型无人旋翼机自主控制流程；Figure 1 is the autonomous control process of a small unmanned rotorcraft;
图2为3.2m/s风扰环境下利用本发明小型无人旋翼机执行四航点巡航飞行效果。Fig. 2 is the effect of using the small unmanned rotorcraft of the present invention to perform four-waypoint cruise flight under the environment of 3.2m/s wind disturbance.
具体实施方式Detailed ways
如图1所示，本发明的具体实现方法如下：As shown in Figure 1, the specific implementation method of the present invention is as follows:
(1)基于极点配置的反馈控制(1) Feedback control based on pole configuration
基于线性化方法，小型无人旋翼机动力学方程表示为Based on the linearization method, the dynamic equation of the small unmanned rotorcraft is expressed as
其中，状态变量x∈Rn表示小型无人旋翼机系统相应的速度、角度和角速度信息。控制变量u∈Rm分别代表小型无人旋翼机侧向周期变距、纵向周期变距、总距控制信号和航向控制信号；A∈Rn×n和B∈Rn×m分别为状态变量和控制变量的状态转移矩阵和控制转移矩阵；d∈Rm表示由风扰、大气湍流、地面干扰、系统电磁干扰、传感器测量误差，以及由小型无人旋翼机系统参数不确定性和无模态动力学特性等因素带来的有界复合干扰。Among them, the state variable x∈R n represents the corresponding speed, angle and angular velocity information of the small unmanned rotorcraft system. The control variables u∈R m represent the lateral periodic pitch, longitudinal periodic pitch, collective pitch control signal and heading control signal of the small unmanned rotorcraft; A∈R n×n and B∈R n×m are the state variables and the state transition matrix and control transition matrix of control variables; d∈R m represents the wind disturbance, atmospheric turbulence, ground disturbance, system electromagnetic Bounded composite disturbances brought about by factors such as dynamic dynamics.
小型无人旋翼机的控制器输入由两部分构成，一部分为是状态反馈输入 Kx(t)，另一部分为自适应神经网络扰动估计量即为The controller input of the small unmanned rotorcraft consists of two parts, one part is the state feedback input Kx(t), and the other part is the adaptive neural network disturbance estimator that is
其中，反馈系数K根据极点配置理论获得，用以保证系统的初步稳定性；Among them, the feedback coefficient K is obtained according to the pole configuration theory to ensure the initial stability of the system;
(2)构建自适应神经网络(2) Build an adaptive neural network
对飞行中存在的多源干扰，设计具有自主更新权值矩阵的自适应神经网络来提高系统的控制精度，实现对小型无人旋翼机在飞行中所受多源干扰进行在线估计和抑制。For the multi-source interference existing in flight, an adaptive neural network with self-updating weight matrix is designed to improve the control accuracy of the system, and realize online estimation and suppression of multi-source interference for small unmanned rotorcraft in flight.
自适应神经网络由输入层、隐含层和输出层构成；自适应神经网络的输入层的输入为期望状态变量xd和实际状态变量x间的误差，即e＝x-xd；Adaptive neural network is made of input layer, hidden layer and output layer; The input of the input layer of adaptive neural network is the error between desired state variable x d and actual state variable x, i.e. e=xx d ;
隐含层由多个高斯函数构成，定义为s(e)，其相应第j个隐含层的节点函数表达式如下：The hidden layer is composed of multiple Gaussian functions, defined as s(e), and the node function expression of the corresponding jth hidden layer is as follows:
其中，μj,分别为自适应神经网络隐含层高斯函数的中心值和宽度，l 为自适应神经网络隐含层的隐含节点数，自适应径向基神经网络节点的中心值μj、宽度由用户决定。Among them, μ j , are the central value and width of the Gaussian function in the hidden layer of the adaptive neural network, l is the number of hidden nodes in the hidden layer of the adaptive neural network, the central value μ j and the width of the adaptive radial basis neural network node up to the user.
自适应的输出层对多源干扰的估计量Adaptive output layer estimator for multi-source interference
其中，为自适应神经网络的权值矩阵；Γi、P为对称正定矩阵，αw为自适应神经网络的控制残差上限阈值，i*为相应矩阵的第i个行向量，*i为相应矩阵的第i个列向量，其中自适应神经网络的权值矩阵按照如下规则进行自主更新in, is the weight matrix of the adaptive neural network; Γ i and P are symmetric positive definite matrices, α w is the upper threshold of the control residual of the adaptive neural network, i* is the ith row vector of the corresponding matrix, and *i is the corresponding matrix The ith column vector of , where the weight matrix of the adaptive neural network is updated autonomously according to the following rules
其中P为下式方程对称正定解where P is the symmetric positive definite solution of the following equation
(A+BK)TP+P(A+BK)＝-Q(A+BK) T P+P(A+BK)＝-Q
其中对称正定矩阵Q＝I。Wherein the symmetric positive definite matrix Q=I.
(3)设计自适应阈值优化策略(3) Design an adaptive threshold optimization strategy
基于时间窗口内的实际位置与期望位置的误差均方差，对自适应神经网络的控制残差上限阈值进行在线更新，实现复杂环境下的小型无人旋翼机高精度姿态控制。Based on the error mean square error between the actual position and the expected position within the time window, the upper threshold of the control residual error of the adaptive neural network is updated online to realize the high-precision attitude control of the small unmanned rotorcraft in a complex environment.
自适应阈值优化策略定义如下：The adaptive threshold optimization strategy is defined as follows:
式中αw为当前时间窗口采样周期内自适应神经网络的控制残差上限阈值；αw-1为上一时间窗口采样周期内自适应神经网络的控制残差上限阈值；αw-2为上二时间窗口采样周期内自适应神经网络的控制残差上限阈值；χk为上一时间窗口采样周期内小型无人旋翼机实际位置xxm与期望位置xxd的均方差；χk-t为上二时间窗口采样周期内小型无人旋翼机实际位置xxm与期望位置xxd的均方差；eei为时间窗口采样周期内小型无人旋翼机实际位置xxm与期望位置xxd的差值；t为时间窗口采样周期内采样点数；k1为控制参数，η1,η2分别为时间窗口采样周期内小型无人旋翼机实际位置xxm与期望位置xxd的最大绝对误差值和平均误差值。where α w is the control residual upper threshold of the adaptive neural network in the current time window sampling period; α w-1 is the control residual upper threshold of the adaptive neural network in the previous time window sampling period; α w-2 is The control residual upper threshold of the adaptive neural network in the sampling period of the last two time windows; χ k is the mean square error between the actual position xx m and the expected position xx d of the small unmanned rotorcraft in the sampling period of the previous time window; The mean square error of the actual position xx m and the expected position xx d of the small unmanned rotorcraft within the sampling period of the second time window; ee i is the difference between the actual position xx m and the expected position xx d of the small unmanned rotorcraft within the sampling period of the time window; t is the number of sampling points in the time window sampling period; k 1 is the control parameter, η 1 and η 2 are the maximum absolute error value and average error between the actual position xx m and the expected position xx d of the small unmanned rotorcraft in the time window sampling period value.
(5)飞行实例(5) Flight examples
基于小型无人旋翼机进行四航点飞行实验验证。四航点巡线飞行定高 20米，以航点(10，-20，20)为起始点，分别经过航点(40，-20，20)， (40，0，20)和(10，0，20)，最终悬停在起始点。基于相同反馈控制矩阵参数的反馈控制方法和自适应神经网络控制方法的对比结果如图2所示，在3.2m/s的最大风扰环境下，基于自适应神经网络的小型无人旋翼机在执行四航点巡线任务时候的压线精度为1.56m,在悬停阶段的压线精度为0.83 m,都优于传统的反馈控制方法。Experimental verification of four-waypoint flight based on a small unmanned rotorcraft. The four-waypoint patrol flight is set at a height of 20 meters, starting from the waypoint (10, -20, 20), passing through the waypoints (40, -20, 20), (40, 0, 20) and (10, 0, 20), and finally hover at the starting point. The comparison results of the feedback control method and the adaptive neural network control method based on the same feedback control matrix parameters are shown in Figure 2. Under the maximum wind disturbance environment of 3.2m/s, the small unmanned rotorcraft based on the adaptive neural network The accuracy of the line pressing during the four-waypoint line patrol mission is 1.56m, and the line pressing accuracy in the hovering stage is 0.83m, both of which are better than the traditional feedback control method.
本发明基于自适应神经网络的小型无人旋翼机高精度控制方法克服了现有控制方法的不足，可以实现小型无人旋翼机复杂多扰环境下的高精度飞行控制等。The high-precision control method of the small unmanned rotorcraft based on the adaptive neural network of the present invention overcomes the shortcomings of the existing control methods, and can realize high-precision flight control of the small unmanned rotorcraft in a complex and disturbed environment.
本发明说明书中未作详细描述的内容属于本领域专业技术人员公知的现有技术。The contents not described in detail in the description of the present invention belong to the prior art known to those skilled in the art.
Claims (2)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310347956.4A CN103412488B (en) | 2013-08-12 | 2013-08-12 | A kind of miniature self-service gyroplane high-accuracy control method based on adaptive neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310347956.4A CN103412488B (en) | 2013-08-12 | 2013-08-12 | A kind of miniature self-service gyroplane high-accuracy control method based on adaptive neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103412488A CN103412488A (en) | 2013-11-27 |
| CN103412488B true CN103412488B (en) | 2018-10-30 |
Family
ID=49605513
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310347956.4A Active CN103412488B (en) | 2013-08-12 | 2013-08-12 | A kind of miniature self-service gyroplane high-accuracy control method based on adaptive neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103412488B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104865829B (en) * | 2015-03-26 | 2017-06-23 | 哈尔滨工业大学 | Multi-robot system distributed self-adaption neutral net continuous tracking control method of electro |
| CN104880945B (en) * | 2015-03-31 | 2018-05-29 | 成都市优艾维机器人科技有限公司 | The adaptive inverse control of rotor wing unmanned aerial vehicle based on neutral net |
| CN105785762B (en) * | 2016-03-17 | 2020-12-11 | 北京航空航天大学 | A high-precision control method for two-axis inertial stabilization platform based on adaptive backstepping sliding mode |
| CN105843243B (en) * | 2016-05-06 | 2019-02-22 | 深圳一电航空技术有限公司 | Control method, device and the multi-rotor aerocraft of multi-rotor aerocraft |
| CN106094518B (en) * | 2016-06-22 | 2018-12-21 | 北京航空航天大学 | A kind of structural vibration pole placement control method based on Multidisciplinary systems optimization |
| CN106682733B (en) * | 2016-11-07 | 2018-10-19 | 上海资誉电子科技有限公司 | Unmanned plane motion state analysis method and device |
| CN110832408B (en) * | 2017-07-03 | 2022-03-25 | 深圳市大疆创新科技有限公司 | Neural network based image target tracking by aircraft |
| CN107985569A (en) * | 2017-12-06 | 2018-05-04 | 余姚市荣事特电子有限公司 | A kind of unmanned plane |
| CN108032994A (en) * | 2017-12-06 | 2018-05-15 | 余姚市荣事特电子有限公司 | A kind of unmanned helicopter |
| CN108563124B (en) * | 2018-04-26 | 2021-01-05 | 宋永端 | Self-adaptive control method of rotor unmanned aerial vehicle based on API (application program interface) user-settable track |
| CN109375642B (en) * | 2018-09-20 | 2021-06-22 | 太原理工大学 | Energy-saving control method for unmanned aerial vehicle |
| CN112445238B (en) * | 2020-11-19 | 2021-09-10 | 西南大学 | Unmanned aerial vehicle tracking system based on neural network |
| CN113267998B (en) * | 2021-03-19 | 2024-02-02 | 北京航空航天大学 | A high-precision modeling and control method for atomic gyroscope temperature control system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07277286A (en) * | 1994-04-11 | 1995-10-24 | Mitsubishi Heavy Ind Ltd | Learning flight control device for aircraft |
| EP0901053A1 (en) * | 1997-09-04 | 1999-03-10 | Rijksuniversiteit te Groningen | Method for modelling and/or controlling a production process using a neural network and controller for a production process |
| US6092919A (en) * | 1995-08-01 | 2000-07-25 | Guided Systems Technologies, Inc. | System and method for adaptive control of uncertain nonlinear processes |
| US6185470B1 (en) * | 1997-11-07 | 2001-02-06 | Mcdonnell Douglas Corporation | Neural network predictive control method and system |
| CN102289714A (en) * | 2011-06-02 | 2011-12-21 | 北京航空航天大学 | Method for controlling autonomous take-off and landing of small unmanned rotorcraft based on behavioral model |
| CN102385316A (en) * | 2011-09-16 | 2012-03-21 | 哈尔滨工程大学 | Deepening controlling method of underactuated automatic underwater vehicle based on neural network back stepping method |
| CN102411305A (en) * | 2011-11-11 | 2012-04-11 | 南京航空航天大学 | Design method of comprehensive disturbance rejection control system for single-rotor wing helicopter/turboshaft engine |
-
2013
- 2013-08-12 CN CN201310347956.4A patent/CN103412488B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07277286A (en) * | 1994-04-11 | 1995-10-24 | Mitsubishi Heavy Ind Ltd | Learning flight control device for aircraft |
| US6092919A (en) * | 1995-08-01 | 2000-07-25 | Guided Systems Technologies, Inc. | System and method for adaptive control of uncertain nonlinear processes |
| EP0901053A1 (en) * | 1997-09-04 | 1999-03-10 | Rijksuniversiteit te Groningen | Method for modelling and/or controlling a production process using a neural network and controller for a production process |
| US6185470B1 (en) * | 1997-11-07 | 2001-02-06 | Mcdonnell Douglas Corporation | Neural network predictive control method and system |
| CN102289714A (en) * | 2011-06-02 | 2011-12-21 | 北京航空航天大学 | Method for controlling autonomous take-off and landing of small unmanned rotorcraft based on behavioral model |
| CN102385316A (en) * | 2011-09-16 | 2012-03-21 | 哈尔滨工程大学 | Deepening controlling method of underactuated automatic underwater vehicle based on neural network back stepping method |
| CN102411305A (en) * | 2011-11-11 | 2012-04-11 | 南京航空航天大学 | Design method of comprehensive disturbance rejection control system for single-rotor wing helicopter/turboshaft engine |
Non-Patent Citations (3)
| Title |
|---|
| Neural networks as a tool for the assessment of human pilot behavkmr in wind shear;D. Martens 等;《Aerospace Science and Technology》;19991231;第39-48页 * |
| The errors in simultaneous approximation by feed-forward neural networks;Tingfan Xie 等;《Neurocomputing》;20100131;第903-907页 * |
| 无人旋翼飞行器自适应飞行控制系统设计与仿真;夏青元 等;《机器人》;20130131;第35卷(第1期);第98-106页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103412488A (en) | 2013-11-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103412488B (en) | A kind of miniature self-service gyroplane high-accuracy control method based on adaptive neural network | |
| CN108445898B (en) | Four-rotor unmanned aerial vehicle system motion planning method based on differential flatness characteristic | |
| CN109144084B (en) | An Attitude Tracking Control Method for Vertical Takeoff and Landing Reusable Vehicle Based on Fixed Time Convergence Observer | |
| CN103885450B (en) | Depopulated helicopter attitude nonlinear control method and verification platform | |
| CN110502033B (en) | Fixed-wing unmanned aerial vehicle cluster control method based on reinforcement learning | |
| CN103760906B (en) | Neural network and non-linear continuous depopulated helicopter attitude control method | |
| CN105676641B (en) | The design method of Nonlinear Robust Controller based on contragradience and sliding formwork control | |
| CN111880567B (en) | Formation coordination control method and device for fixed-wing UAV based on deep reinforcement learning | |
| CN102289714B (en) | A Behavioral Model-Based Autonomous Take-off and Landing Control Method for Small Unmanned Rotorcraft | |
| CN102591358A (en) | A dynamic formation control method for multiple UAVs | |
| Wang et al. | Controller for UAV to oppose different kinds of wind in the environment | |
| CN106874617B (en) | An Efficient Evaluation Method of Helicopter Maneuvering Flight Quality | |
| CN107765553A (en) | For the nonlinear control method of rotor wing unmanned aerial vehicle hanging transportation system | |
| CN106647584A (en) | Fault tolerant control method of four-rotor-wing aircraft based on optimal sliding mode | |
| CN106681345A (en) | Crowd-searching-algorithm-based active-disturbance-rejection control method for unmanned plane | |
| CN104932514A (en) | Posture nonlinear self-adaptive control method for small unmanned helicopter | |
| CN105607473A (en) | Self-adaptive control method of rapid attitude error convergence for small unmanned helicopter | |
| CN106527462A (en) | Unmanned aerial vehicle (UAV) control device | |
| Zhu et al. | Landing control of fixed-wing uav based on adrc | |
| CN107703953A (en) | A kind of attitude control method of unmanned plane, device, unmanned plane and storage medium | |
| Yang et al. | A decentralised control strategy for formation flight of unmanned aerial vehicles | |
| CN108762070A (en) | A kind of fractional order control method of drive lacking unmanned plane | |
| CN107991884A (en) | Small-sized fixed-wing unmanned plane advanced stall recovery method based on range information | |
| CN114879490A (en) | Iterative optimization and control method for unmanned aerial vehicle perching maneuver | |
| CN103809433B (en) | The multistage PID robust Controller Design method of aircraft multiloop model bunch compound root locus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |