CN116399500A - A contact force detection method and magnetic tactile sensor - Google Patents
A contact force detection method and magnetic tactile sensor Download PDFInfo
- Publication number
- CN116399500A CN116399500A CN202310320776.0A CN202310320776A CN116399500A CN 116399500 A CN116399500 A CN 116399500A CN 202310320776 A CN202310320776 A CN 202310320776A CN 116399500 A CN116399500 A CN 116399500A
- Authority
- CN
- China
- Prior art keywords
- contact force
- flux density
- magnetic flux
- dimensional
- magnets
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01L—MEASURING FORCE, STRESS, TORQUE, WORK, MECHANICAL POWER, MECHANICAL EFFICIENCY, OR FLUID PRESSURE
- G01L5/00—Apparatus for, or methods of, measuring force, work, mechanical power, or torque, specially adapted for specific purposes
- G01L5/16—Apparatus for, or methods of, measuring force, work, mechanical power, or torque, specially adapted for specific purposes for measuring several components of force
- G01L5/169—Apparatus for, or methods of, measuring force, work, mechanical power, or torque, specially adapted for specific purposes for measuring several components of force using magnetic means
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Length, Angles, Or The Like Using Electric Or Magnetic Means (AREA)
Abstract
本发明提供一种接触力检测方法及磁触觉传感器，涉及磁触觉传感器领域，所述方法用于磁触觉传感器，磁触觉传感器包括多个磁体，磁触觉传感器除磁体外的部分由可变形材料构成，磁体均置于相同水平面，每个磁体与相邻磁体的磁化方向相反，磁体的磁化方向垂直于水平面，所述方法包括：根据每个磁体在预设三维坐标系中的检测点的三维磁通密度变化向量和接触力位置分类模型，得到接触力的位置参数；根据三维磁通密度变化向量、位置参数和接触力大小回归模型，得到接触力的大小参数；根据位置参数和大小参数得到接触力的检测结果。本发明的接触力检测方法，根据接触力分类模型和接触力大小回归模型，得到接触力的位置和大小，提高了接触力检测的精准度。
The invention provides a contact force detection method and a magnetic tactile sensor, which relate to the field of magnetic tactile sensors. The method is used in the magnetic tactile sensor. The magnetic tactile sensor includes a plurality of magnets, and parts of the magnetic tactile sensor except the magnets are made of deformable materials. , the magnets are all placed on the same horizontal plane, the magnetization direction of each magnet is opposite to that of the adjacent magnet, and the magnetization direction of the magnet is perpendicular to the horizontal plane, the method includes: according to the three-dimensional magnetization of the detection point of each magnet in the preset three-dimensional coordinate system Flux density change vector and contact force position classification model to obtain the position parameters of the contact force; according to the three-dimensional magnetic flux density change vector, position parameters and the regression model of the contact force size, the size parameters of the contact force are obtained; according to the position parameters and size parameters, the contact force Force test results. The contact force detection method of the present invention obtains the position and magnitude of the contact force according to the contact force classification model and the contact force magnitude regression model, thereby improving the accuracy of the contact force detection.
Description
技术领域technical field
本发明涉及磁触觉传感器领域，具体而言，涉及一种接触力检测方法及磁触觉传感器。The invention relates to the field of magnetic tactile sensors, in particular to a contact force detection method and a magnetic tactile sensor.
背景技术Background technique
在磁触觉传感器检测接触力时，是通过检测磁场在受到接触力时产生的变化，来检测接触力的大小和位置。接触力越大对磁场的影响越大，磁场的变化量越大，测量出的接触力大小和位置也就越准确，而由于较小的接触力对磁场的影响较小，磁场的变化量小，对于接触力的大小和位置的检测结果就不理想。When the magnetic tactile sensor detects the contact force, it detects the magnitude and position of the contact force by detecting the change of the magnetic field when it is subjected to the contact force. The greater the contact force, the greater the impact on the magnetic field, and the greater the change in the magnetic field, the more accurate the measured contact force and position will be. However, because the smaller contact force has less influence on the magnetic field, the change in the magnetic field is small , the detection results for the size and position of the contact force are not ideal.
在现有技术中，采用磁触觉传感器检测较小接触力，由于较小的接触力对磁场的影响较小，磁场的变化不大，因此检测精度不足，其中则包括检测接触里的大小精度和位置精度。In the prior art, a magnetic tactile sensor is used to detect a small contact force. Since a small contact force has little influence on the magnetic field, the change of the magnetic field is not large, so the detection accuracy is insufficient, which includes the detection of the size and precision of the contact. positional accuracy.
发明内容Contents of the invention
本发明主要解决的技术问题是，如何提高磁触觉传感器对接触力检测的精度。为解决上述问题，本发明提供一种接触力检测方法，所述接触力检测方法用于磁触觉传感器，所述磁触觉传感器包括多个磁体，所述磁触觉传感器除所述磁体外的部分由可变形材料构成，多个所述磁体均置于相同水平面上，每个所述磁体与相邻的其他所述磁体的磁化方向相反，且所述磁体的磁化方向垂直于所述水平面，所述接触力检测方法包括：The technical problem mainly solved by the invention is how to improve the precision of the contact force detection by the magnetic tactile sensor. In order to solve the above problems, the present invention provides a contact force detection method, the contact force detection method is used in a magnetic tactile sensor, the magnetic tactile sensor includes a plurality of magnets, and the part of the magnetic tactile sensor other than the magnets is composed of Made of deformable material, a plurality of magnets are placed on the same horizontal plane, the magnetization direction of each magnet is opposite to that of other adjacent magnets, and the magnetization direction of the magnets is perpendicular to the horizontal plane, the Contact force detection methods include:
根据每个所述磁体在预设三维坐标系中的检测点的三维磁通密度变化向量和接触力位置分类模型，得到接触力的位置参数；According to the three-dimensional magnetic flux density change vector and the contact force position classification model of each magnet in the preset three-dimensional coordinate system, the position parameter of the contact force is obtained;
其中，所述三维磁通密度变化向量包括所述检测点在所述预设三维坐标系的三个坐标轴的磁通密度变化向量；Wherein, the three-dimensional magnetic flux density change vector includes the magnetic flux density change vector of the detection point on the three coordinate axes of the preset three-dimensional coordinate system;
根据所述三维磁通密度变化向量、所述位置参数和接触力大小回归模型，得到所述接触力的大小参数；Obtaining the magnitude parameter of the contact force according to the three-dimensional magnetic flux density change vector, the position parameter and the regression model of the magnitude of the contact force;
根据所述位置参数和所述大小参数得到所述接触力的检测结果。The detection result of the contact force is obtained according to the position parameter and the size parameter.
可选地，还包括：Optionally, also include:
根据所述磁体建立所述预设三维坐标系，其中，所述三维坐标系包括XY平面，所述XY平面为所述水平面；Establishing the preset three-dimensional coordinate system according to the magnet, wherein the three-dimensional coordinate system includes an XY plane, and the XY plane is the horizontal plane;
所述根据每个所述磁体在预设三维坐标系中的检测点的三维磁通密度变化向量和接触力位置分类模型，得到接触力的位置参数，包括：According to the three-dimensional magnetic flux density change vector and the contact force position classification model of each of the magnets in the preset three-dimensional coordinate system, the position parameters of the contact force are obtained, including:
将所述三维磁通密度变化向量输入所述接触力位置分类模型；inputting the three-dimensional magnetic flux density change vector into the contact force position classification model;
根据所述接触力位置分类模型的输出，得到所述接触力的所述位置参数。The position parameter of the contact force is obtained according to the output of the position classification model of the contact force.
可选地，所述根据所述三维磁通密度变化向量、所述位置参数和接触力大小回归模型，得到所述接触力的大小参数，包括：Optionally, the step of obtaining the size parameter of the contact force according to the three-dimensional magnetic flux density change vector, the position parameter, and the size regression model of the contact force includes:
根据所述检测点在所述预设三维坐标系的三个所述坐标轴的所述磁通密度变化向量和所述位置参数，得到四维接触力向量；Obtaining a four-dimensional contact force vector according to the magnetic flux density change vector and the position parameter of the detection point on the three coordinate axes of the preset three-dimensional coordinate system;
将所述四维接触力向量输入所述接触力大小回归模型，得到所述接触力的所述大小参数。Inputting the four-dimensional contact force vector into the regression model of the magnitude of the contact force to obtain the magnitude parameter of the contact force.
可选地，在所述根据每个所述磁体在预设三维坐标系中的检测点的三维磁通密度变化向量和接触力位置分类模型，得到接触力的位置参数之前，还包括：Optionally, before obtaining the position parameter of the contact force according to the three-dimensional magnetic flux density change vector and the position classification model of the contact force of the detection point of each magnet in the preset three-dimensional coordinate system, it also includes:
获取所述检测点的磁通密度；Obtain the magnetic flux density of the detection point;
当所述磁体受到所述接触力作用时，将所述接触力作用时的所述检测点在每个所述坐标轴的磁通密度与未受到所述接触力作用时的所述检测点在每个所述坐标轴的所述磁通密度做差得到差值，将所述差值作为所述坐标轴的所述磁通密度变化向量。When the magnet is subjected to the contact force, the magnetic flux density of the detection point on each of the coordinate axes when the contact force is applied is the same as that of the detection point when the contact force is not applied. The magnetic flux densities of each of the coordinate axes are differentiated to obtain a difference, and the difference is used as the magnetic flux density change vector of the coordinate axes.
可选地，所述获取所述检测点的磁通密度，包括：Optionally, the obtaining the magnetic flux density of the detection point includes:
获取每个所述磁体在所述检测点的区域磁通密度；Acquiring the magnetic flux density of each magnet at the detection point;
根据每个所述磁体在所述检测点的所述区域磁通密度，得到所述检测点的所述磁通密度。The magnetic flux density of the detection point is obtained according to the magnetic flux density of each magnet in the region of the detection point.
可选地，所述根据每个所述磁体在所述检测点的所述区域磁通密度，得到所述检测点的所述磁通密度，包括：Optionally, the obtaining the magnetic flux density of the detection point according to the magnetic flux density of each magnet in the region of the detection point includes:
根据每个所述磁体在所述检测点的所述区域磁通密度和叠加公式，得到所述检测点的所述磁通密度；Obtaining the magnetic flux density at the detection point according to the magnetic flux density in the area of each of the magnets at the detection point and a superposition formula;
所述叠加公式为：
其中，BC为所述检测点的所述磁通密度，
可选地，还包括：Optionally, also include:
选定初始分类模型；Select the initial classification model;
获取多组当所述磁体受到所述接触力按压时，得到所述检测点在所述预设三维坐标系中所述三维磁通密度变化向量，以及所述接触力的接触力位置参数；Obtain multiple sets of the three-dimensional magnetic flux density change vector of the detection point in the preset three-dimensional coordinate system and the contact force position parameters of the contact force when the magnet is pressed by the contact force;
将每组所述三维磁通密度变化向量作为所述初始分类模型的样本数据，并将所述接触力的位置参数作为所述初始分类模型的样本标签，对所述初始分类模型进行训练；Using each group of the three-dimensional magnetic flux density change vectors as the sample data of the initial classification model, and using the position parameter of the contact force as the sample label of the initial classification model, to train the initial classification model;
将训练完成的所述初始分类模型作为所述接触力位置分类模型。The trained initial classification model is used as the contact force position classification model.
可选地，所述获取多组当所述磁体受到所述接触力按压时，得到所述检测点在所述预设三维坐标系中所述三维磁通密度变化向量，以及所述接触力的接触力位置参数，包括：Optionally, the acquiring multiple sets of the three-dimensional magnetic flux density change vector of the detection point in the preset three-dimensional coordinate system and the contact force’s Contact force location parameters, including:
根据所述接触力以预设次数按压每个所述磁体的对应区域，得到所述检测点在每个所述坐标轴上的初始磁通密度变化；pressing the corresponding area of each of the magnets for a predetermined number of times according to the contact force to obtain the change of the initial magnetic flux density of the detection point on each of the coordinate axes;
根据所述每个所述坐标轴上的所述初始磁通密度变化，通过消除预设初始偏移量，得到所述三维磁通密度变化向量；Obtaining the three-dimensional magnetic flux density change vector by eliminating a preset initial offset according to the initial magnetic flux density change on each of the coordinate axes;
根据每个所述磁体的所述对应区域，得到所述接触力的所述接触力位置参数。According to the corresponding area of each of the magnets, the contact force position parameter of the contact force is obtained.
可选地，还包括：Optionally, also include:
选定初始回归模型；Select the initial regression model;
获取多组当所述磁体受到所述接触力按压时，所述检测点的所述三维磁通密度变化向量和所述接触力的接触力大小参数，其中，一组所述三维磁通密度变化向量对应一个所述接触力位置参数和一个所述接触力大小参数；Obtain multiple groups of the three-dimensional magnetic flux density change vector of the detection point and the contact force magnitude parameters of the contact force when the magnet is pressed by the contact force, wherein one group of the three-dimensional magnetic flux density change The vector corresponds to a position parameter of the contact force and a magnitude parameter of the contact force;
将所述三维磁通密度变化向量和所述接触力位置参数作为所述初始回归模型的输入，将所述接触力大小参数作为所述初始回归模型的输出，对所述初始回归模型进行训练；Using the three-dimensional magnetic flux density change vector and the contact force position parameter as the input of the initial regression model, using the contact force size parameter as the output of the initial regression model, and training the initial regression model;
将训练完成的所述初始回归模型作为所述接触力大小回归模型。The initial regression model that has been trained is used as the regression model of the contact force magnitude.
本发明的接触力检测方法，通过将磁触觉传感器中检测点检测到的磁通密度变化向量输入至接触力位置分类模型，从而根据训练好的接触力分类模型，得到接触力的位置参数；再根据位置参数和磁通密度变化向量，通过训练好的接触力大小回归模型，得到接触力的大小参数，实现了针对任何大小的接触力都能根据两个模型实现相同准确度的接触力检测，进而提高了对于较小接触力检测的精准度。In the contact force detection method of the present invention, the magnetic flux density change vector detected by the detection point in the magnetic tactile sensor is input to the contact force position classification model, thereby obtaining the position parameter of the contact force according to the trained contact force classification model; According to the position parameter and the magnetic flux density change vector, the size parameter of the contact force is obtained through the trained contact force size regression model, and the contact force detection with the same accuracy can be realized according to the two models for any size of the contact force. In turn, the accuracy of detection of small contact force is improved.
本发明还提供一种磁触觉传感器，所述磁触觉传感器包括多个磁体，所述磁触觉传感器除所述磁体外的部分由可变形材料构成，多个所述磁体均置于相同水平面上，每个所述磁体与相邻的其他所述磁体的磁化方向相反，且所述磁体的磁化方向垂直于所述水平面，所述磁触觉传感器包括：The present invention also provides a magnetic tactile sensor, the magnetic tactile sensor includes a plurality of magnets, the part of the magnetic tactile sensor except the magnets is made of deformable material, and the plurality of magnets are placed on the same horizontal plane, The magnetization direction of each of the magnets is opposite to that of the adjacent other magnets, and the magnetization direction of the magnets is perpendicular to the horizontal plane, and the magnetic tactile sensor includes:
接触力位置检测单元，用于根据每个所述磁体在预设三维坐标系中的检测点的三维磁通密度变化向量和接触力位置分类模型，得到接触力的位置参数；The contact force position detection unit is used to obtain the position parameter of the contact force according to the three-dimensional magnetic flux density change vector and the contact force position classification model of each detection point of the magnet in the preset three-dimensional coordinate system;
其中，所述三维磁通密度变化向量包括所述检测点在所述预设三维坐标系的三个坐标轴的磁通密度变化向量；Wherein, the three-dimensional magnetic flux density change vector includes the magnetic flux density change vector of the detection point on the three coordinate axes of the preset three-dimensional coordinate system;
接触力大小检测单元，用于根据所述三维磁通密度变化向量、所述位置参数和接触力大小回归模型，得到所述接触力的大小参数；A contact force size detection unit, configured to obtain the size parameter of the contact force according to the three-dimensional magnetic flux density change vector, the position parameter and the contact force size regression model;
处理单元，用于根据所述位置参数和所述大小参数得到所述接触力的检测结果。A processing unit, configured to obtain the detection result of the contact force according to the position parameter and the size parameter.
本发明的磁触觉传感器，通过将磁触觉传感器中检测点检测到的磁通密度变化向量输入至接触力位置分类模型，从而根据训练好的接触力分类模型，得到接触力的位置参数；再根据位置参数和磁通密度变化向量，通过训练好的接触力大小回归模型，得到接触力的大小参数，实现了针对任何大小的接触力都能根据两个模型实现相同准确度的接触力检测，进而提高了对于较小接触力检测的精准度。In the magnetic tactile sensor of the present invention, the magnetic flux density change vector detected by the detection point in the magnetic tactile sensor is input to the contact force position classification model, thereby obtaining the position parameter of the contact force according to the trained contact force classification model; and then according to The position parameter and the magnetic flux density change vector, through the trained contact force size regression model, get the size parameter of the contact force, and realize the contact force detection with the same accuracy according to the two models for any size of the contact force, and then Improved accuracy for small contact force detection.
附图说明Description of drawings
图1为本发明实施例中磁触觉传感器结构示意图；Fig. 1 is a schematic structural diagram of a magnetic tactile sensor in an embodiment of the present invention;
图2为本发明实施例中接触力检测方法流程图；Fig. 2 is a flow chart of a contact force detection method in an embodiment of the present invention;
图3为本发明实施例中接触力检测方法流程图；3 is a flowchart of a contact force detection method in an embodiment of the present invention;
图4为本发明实施例中接触力检测方法流程图；4 is a flowchart of a contact force detection method in an embodiment of the present invention;
图5为本发明实施例中接触力检测方法流程图；5 is a flowchart of a contact force detection method in an embodiment of the present invention;
图6为本发明实施例中接触力检测方法流程图；6 is a flowchart of a contact force detection method in an embodiment of the present invention;
图7为本发明实施例中磁触觉传感器的单元示意图；7 is a schematic diagram of a unit of a magnetic tactile sensor in an embodiment of the present invention;
图8为本发明实施例中磁触觉传感器的单元示意图。FIG. 8 is a schematic diagram of a unit of a magnetic touch sensor in an embodiment of the present invention.
具体实施方式Detailed ways
为使本发明实施例的目的、技术方案和优点更加清楚，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整的描述，显然，所描述的实施例是本发明的一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例，都属于本发明保护的范围。In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
本发明的接触力检测方法，所述接触力检测方法用于磁触觉传感器，所述磁触觉传感器包括多个磁体，多个所述磁体均置于相同水平面上，每个所述磁体与相邻的其他所述磁体的磁化方向相反，且所述磁体的磁化方向垂直于所述水平面，所述磁触觉传感器除所述磁体外的部分由可变形材料构成，结合图1所示，以图1的磁触觉传感器为例，磁触觉传感器具有四个磁体，每个磁体分别设置在每个正方形单元内，磁触觉传感器共有4个单元，每个单元为具有正方形横截面的长方体，O1与O2以及O4的磁化方向相反，O1与O3的磁化方向相同，磁触觉传感器的底面作为水平面，当磁触觉传感器置于水平面上时，实现了磁触觉传感器的磁体的对称分布，磁触觉传感器除磁体外的可变形材料可以为软浇铸泡沫；In the contact force detection method of the present invention, the contact force detection method is used in a magnetic tactile sensor, and the magnetic tactile sensor includes a plurality of magnets, and the plurality of magnets are placed on the same horizontal plane, and each of the magnets is connected to the adjacent The magnetization directions of the other magnets are opposite, and the magnetization directions of the magnets are perpendicular to the horizontal plane, and the part of the magnetic tactile sensor except the magnets is made of a deformable material, as shown in FIG. Take the magnetic tactile sensor as an example. The magnetic tactile sensor has four magnets, and each magnet is respectively arranged in each square unit. The magnetic tactile sensor has 4 units in total, and each unit is a cuboid with a square cross section. O 1 and O 2 and O 4 have opposite magnetization directions, O 1 and O 3 have the same magnetization direction, and the bottom surface of the magnetic tactile sensor is used as a horizontal plane. The deformable material of the sensor other than the magnet may be soft cast foam;
结合图2所示，所述接触力检测方法包括：As shown in Figure 2, the contact force detection method includes:
S1：根据每个所述磁体在预设三维坐标系中的检测点的三维磁通密度变化向量和接触力位置分类模型，得到接触力的位置参数；S1: According to the three-dimensional magnetic flux density change vector and the contact force position classification model of each magnet in the preset three-dimensional coordinate system, the position parameter of the contact force is obtained;
其中，所述三维磁通密度变化向量包括所述检测点在所述预设三维坐标系的三个坐标轴的磁通密度变化向量；Wherein, the three-dimensional magnetic flux density change vector includes the magnetic flux density change vector of the detection point on the three coordinate axes of the preset three-dimensional coordinate system;
具体地，三维磁通密度变化向量即为检测点在预设三维坐标系的三个坐标轴的磁通密度变化向量，通过磁编码器等磁通密度检测装置在磁触觉传感器中的检测点进行检测，当接触力作用到可变形材料上，可变形材料发生变形，以此来影响磁通密度，测量在磁触觉传感器受到接触力影响后磁通密度在预设三维坐标系的三个坐标轴的变化向量，并通过接触力位置分类模型根据三个坐标轴的磁通密度的变化向量，得到接触力的位置参数，其中位置参数为接触力作用在磁触觉传感器上的区域，该区域可用编号来表示，以图1为例，编号可根据磁触觉传感器中的磁体分为1，2，3，4来代表每个磁体对应的区域，其中，接触力位置分类模型作为分类器是通过对K近邻分类模型进行训练得到的，用于对接触力的作用位置进行分类。Specifically, the three-dimensional magnetic flux density change vector is the magnetic flux density change vector of the detection point on the three coordinate axes of the preset three-dimensional coordinate system. Detection, when the contact force acts on the deformable material, the deformable material is deformed to affect the magnetic flux density, and the magnetic flux density is measured on the three coordinate axes of the preset three-dimensional coordinate system after the magnetic tactile sensor is affected by the contact force The change vector of the contact force, and the position parameter of the contact force is obtained according to the change vector of the magnetic flux density of the three coordinate axes through the contact force position classification model, where the position parameter is the area where the contact force acts on the magnetic tactile sensor, and the area can be numbered To represent, taking Figure 1 as an example, the numbers can be divided into 1, 2, 3, and 4 according to the magnets in the magnetic tactile sensor to represent the corresponding area of each magnet, wherein the contact force position classification model is used as a classifier through the K The nearest neighbor classification model is trained to classify the location of the contact force.
S2：根据所述磁通密度变化向量、所述位置参数和接触力大小回归模型，得到所述接触力的大小参数；S2: Obtain the magnitude parameter of the contact force according to the magnetic flux density change vector, the position parameter, and a regression model of the magnitude of the contact force;
具体地，当得到上述接触力位置分类模型输出的位置参数后，即得到对应的位置编号和在检测点测得的三个坐标轴的磁通密度的变化向量后，根据上述两个参数，通过接触力大小回归模型来获取具体的接触力的大小，其中，接触力大小回归模型作为一个回归器是通过前馈神经网络进行训练得到的，用于对接触力的大小参数进行回归，得到较为准确的数值。Specifically, after obtaining the position parameters output by the above-mentioned contact force position classification model, that is, after obtaining the corresponding position number and the change vector of the magnetic flux density of the three coordinate axes measured at the detection point, according to the above two parameters, through The contact force size regression model is used to obtain the specific size of the contact force. As a regressor, the contact force size regression model is trained through a feedforward neural network, and is used to regress the size parameters of the contact force to obtain a more accurate value.
S3：根据所述位置参数和所述大小参数得到所述接触力的检测结果；S3: Obtain the detection result of the contact force according to the position parameter and the size parameter;
具体地，当得到接触力的位置参数和大小参数，即可得知接触力的大小以及接触力的作用位置，当得到接触力的大小以及位置后即可完成对接触力的检测。Specifically, when the position parameter and size parameter of the contact force are obtained, the magnitude of the contact force and the position where the contact force acts can be known, and the detection of the contact force can be completed after the magnitude and position of the contact force are obtained.
本发明的接触力检测方法，通过将磁触觉传感器中检测点检测到的磁通密度变化向量输入至接触力位置分类模型，从而根据训练好的接触力分类模型，得到接触力的位置参数；再根据位置参数和磁通密度变化向量，通过训练好的接触力大小回归模型，得到接触力的大小参数，实现了针对任何大小的接触力都能根据两个模型实现相同准确度的接触力检测，进而提高了对于较小接触力检测的精准度。In the contact force detection method of the present invention, the magnetic flux density change vector detected by the detection point in the magnetic tactile sensor is input to the contact force position classification model, thereby obtaining the position parameter of the contact force according to the trained contact force classification model; According to the position parameter and the magnetic flux density change vector, the size parameter of the contact force is obtained through the trained contact force size regression model, and the contact force detection with the same accuracy can be realized according to the two models for any size of the contact force. In turn, the accuracy of detection of small contact force is improved.
结合图3所示，本发明实施例中，还包括：As shown in Figure 3, in the embodiment of the present invention, it also includes:
根据所述磁体建立所述预设三维坐标系，其中，所述三维坐标系包括XY平面，所述XY平面为所述水平面；Establishing the preset three-dimensional coordinate system according to the magnet, wherein the three-dimensional coordinate system includes an XY plane, and the XY plane is the horizontal plane;
所述S1：根据每个所述磁体在预设三维坐标系中的检测点的三维磁通密度变化向量和接触力位置分类模型，得到接触力的位置参数，包括：The S1: According to the three-dimensional magnetic flux density change vector and the contact force position classification model of each magnet in the preset three-dimensional coordinate system, the position parameters of the contact force are obtained, including:
S11：将所述三维磁通密度变化向量输入所述接触力位置分类模型；S11: Input the three-dimensional magnetic flux density change vector into the contact force position classification model;
S12：根据所述接触力位置分类模型的输出，得到所述接触力的所述位置参数。S12: According to the output of the contact force position classification model, obtain the position parameter of the contact force.
在本实施例中，首先需要根据磁触觉传感器中的每个磁体建立一个三维坐标系，将XY平面作为水平面，每个磁体都置于水平面上，由于磁触觉传感器布置多个磁体，而检测点只有一个，因此检测点在每个磁体所在的三维坐标系的位置不同，根据检测点，检测磁通密度的变化向量，获取该磁通密度的变化向量在预设三维坐标系中三轴的分量，即磁通密度变化分量，分别包括磁通密度变化向量在X轴、Y轴和Z轴的分量；将三个分量输入训练好的接触力位置分类模型，对接触力位置进行分类，得到接触力的具体施加位置，以对应区域的编号来表示。In this embodiment, firstly, a three-dimensional coordinate system needs to be established according to each magnet in the magnetic tactile sensor, and the XY plane is used as a horizontal plane, and each magnet is placed on the horizontal plane. Since the magnetic tactile sensor arranges multiple magnets, the detection point There is only one, so the position of the detection point in the three-dimensional coordinate system where each magnet is located is different. According to the detection point, the change vector of the magnetic flux density is detected, and the three-axis component of the change vector of the magnetic flux density in the preset three-dimensional coordinate system is obtained. , that is, the magnetic flux density change component, including the components of the magnetic flux density change vector on the X-axis, Y-axis and Z-axis respectively; input the three components into the trained contact force position classification model, classify the contact force position, and obtain the contact force position The specific location where the force is applied is represented by the number of the corresponding area.
本发明的接触力检测方法，通过将磁触觉传感器检测到的磁通密度变化向量进行计算，得到在三维坐标系中每个轴的分量，由于在Z轴分量的磁通密度变化具有最重要的整体响应，因此可以根据最重要的分量的变化，通过分类器得到接触力的位置，并通过训练好的分类模型，提高了对接触力位置的检测精度。The contact force detection method of the present invention calculates the magnetic flux density change vector detected by the magnetic touch sensor to obtain the component of each axis in the three-dimensional coordinate system, because the magnetic flux density change in the Z axis component has the most important The overall response, so the position of the contact force can be obtained through the classifier according to the change of the most important component, and the detection accuracy of the position of the contact force can be improved through the trained classification model.
结合图4所示，本发明实施例中，所述S2：根据所述磁通密度变化向量、所述位置参数和接触力大小回归模型，得到所述接触力的大小参数，包括：As shown in FIG. 4, in the embodiment of the present invention, said S2: Obtain the size parameter of the contact force according to the magnetic flux density change vector, the position parameter and the contact force size regression model, including:
S21：根据所述检测点在所述预设三维坐标系的三个所述坐标轴的所述磁通密度变化向量和所述位置参数，得到四维接触力向量；S21: Obtain a four-dimensional contact force vector according to the magnetic flux density change vector and the position parameter of the detection point on the three coordinate axes of the preset three-dimensional coordinate system;
S22：将所述四维接触力向量输入所述接触力大小回归模型，得到所述接触力的所述大小参数。S22: Input the four-dimensional contact force vector into the contact force magnitude regression model to obtain the magnitude parameter of the contact force.
在本实施例中，在得到磁通密度变化向量在三轴的分量后，再与接触力位置分类模型输出的接触力位置参数，组合成为一组包括分量以及接触力对应区域编号的四维向量，将该四维向量输入至训练好的接触力大小回归模型中，接触力大小回归模型输出相应的接触力大小参数。In this embodiment, after obtaining the three-axis components of the magnetic flux density change vector, they are combined with the contact force position parameters output by the contact force position classification model to form a set of four-dimensional vectors including the components and the corresponding area numbers of the contact force. The four-dimensional vector is input into the trained contact force magnitude regression model, and the contact force magnitude regression model outputs corresponding contact force magnitude parameters.
本发明的接触力检测方法，通过训练好的前馈神经网络根据磁通密度变化和接触力位置，回归得到接触里的大小值，根据大小参数和位置参数即可完成接触力的检测。In the contact force detection method of the present invention, through the trained feedforward neural network, according to the change of magnetic flux density and the position of the contact force, the size value of the contact area is regressed, and the detection of the contact force can be completed according to the size parameter and the position parameter.
本发明实施例中，在所述根据每个所述磁体在预设三维坐标系中的检测点的三维磁通密度变化向量和接触力位置分类模型，得到接触力的位置参数之前，还包括：In the embodiment of the present invention, before the position parameter of the contact force is obtained according to the three-dimensional magnetic flux density change vector and the position classification model of the contact force of each magnet in the preset three-dimensional coordinate system, it also includes:
获取所述检测点的磁通密度；Obtain the magnetic flux density of the detection point;
当所述磁体受到所述接触力作用时，将所述接触力作用时的所述检测点在每个所述坐标轴的磁通密度与未受到所述接触力作用时的所述检测点在每个所述坐标轴的所述磁通密度做差得到差值，将所述差值作为所述坐标轴的所述磁通密度变化向量。When the magnet is subjected to the contact force, the magnetic flux density of the detection point on each of the coordinate axes when the contact force is applied is the same as that of the detection point when the contact force is not applied. The magnetic flux densities of each of the coordinate axes are differentiated to obtain a difference, and the difference is used as the magnetic flux density change vector of the coordinate axes.
在本实施例中，可以通过磁编码器来获取检测点的磁通密度，由于没有接触力作用时，磁通密度是不会变化的，因此当出现接触力对磁体进行压力时，导致磁通密度改变，通过未受接触力干扰前的磁通密度与改变后的磁通密度的差值，得到磁通密度变化向量。In this embodiment, the magnetic flux density of the detection point can be obtained by the magnetic encoder. Since the magnetic flux density does not change when there is no contact force, when the contact force exerts pressure on the magnet, the magnetic flux Density change, through the difference between the magnetic flux density before being disturbed by the contact force and the changed magnetic flux density, the magnetic flux density change vector is obtained.
本发明的接触力检测方法，通过磁通密度检测设备对磁通密度进行检测，并计算磁通密度的变化量，方便后续根据变化量来计算接触力。In the contact force detection method of the present invention, the magnetic flux density is detected by the magnetic flux density detection device, and the change amount of the magnetic flux density is calculated, so as to facilitate subsequent calculation of the contact force according to the change amount.
本发明实施例中，所述获取所述检测点的磁通密度，包括：In the embodiment of the present invention, the obtaining the magnetic flux density of the detection point includes:
获取每个所述磁体在所述检测点的区域磁通密度；Acquiring the magnetic flux density of each magnet at the detection point;
根据每个所述磁体在所述检测点的所述区域磁通密度，得到所述检测点的所述磁通密度。The magnetic flux density of the detection point is obtained according to the magnetic flux density of each magnet in the region of the detection point.
在本实施例中，由于每个磁体都会收接触力作用产生一个磁通密度的变化，因此若要得到检测点的整体的磁通密度变化向量，首先需要针对每个磁体进行分析，得到每个磁体在检测点产生的区域磁通密度，并根据所有磁体在相同检测点的区域磁通密度得到检测点的磁通密度。In this embodiment, since each magnet will absorb the contact force to produce a change in magnetic flux density, if you want to obtain the overall magnetic flux density change vector of the detection point, you first need to analyze each magnet to obtain each The area magnetic flux density generated by the magnet at the detection point, and the magnetic flux density of the detection point is obtained according to the area magnetic flux density of all magnets at the same detection point.
本发明的接触力检测方法，通过计算每个磁体在检测点的磁通密度，进而得到实际检测点所检测到的磁通密度的值，方便后续判断磁通密度的变化值。The contact force detection method of the present invention calculates the magnetic flux density of each magnet at the detection point, and then obtains the value of the magnetic flux density detected at the actual detection point, which facilitates subsequent judgment of the change value of the magnetic flux density.
本发明实施例中，所述根据每个所述磁体在所述检测点的所述区域磁通密度，得到所述检测点的所述磁通密度，包括：In the embodiment of the present invention, the obtaining the magnetic flux density of the detection point according to the magnetic flux density of each of the magnets in the region of the detection point includes:
根据每个所述磁体在所述检测点的所述区域磁通密度和叠加公式，得到所述检测点的所述磁通密度；Obtaining the magnetic flux density at the detection point according to the magnetic flux density in the area of each of the magnets at the detection point and a superposition formula;
所述叠加公式为：
其中，BC为所述检测点的所述磁通密度，
在本实施例中，由于一个磁触觉传感器内部布置有多个磁体，且磁体布置在同一平面上，以图1的磁触觉传感器为例，磁体对称分布，因此每个磁体在本磁触觉传感器检测点处的磁通密度叠加可以得知本检测点所检测到的磁通密度，包括了四个磁体在该点处的磁通密度的叠加，因此根据叠加公式可计算出检测点的磁通密度，以图1的磁触觉传感器为例，若此磁体为四个，则
本发明的接触力检测方法，通过叠加公式计算每个磁体在检测点的磁通密度，进而得到实际检测点所检测到的磁通密度的值，方便后续判断磁通密度的变化值。The contact force detection method of the present invention calculates the magnetic flux density of each magnet at the detection point through a superposition formula, and then obtains the value of the magnetic flux density detected at the actual detection point, which facilitates subsequent judgment of the change value of the magnetic flux density.
结合图5所示，本发明的实施例中，还包括：As shown in FIG. 5, in the embodiment of the present invention, it also includes:
S41：选定初始分类模型；S41: Select an initial classification model;
S42：获取多组当所述磁体受到所述接触力按压时，在所述预设三维坐标系中的所述检测点的所述磁通密度变化向量，其中，一组所述磁通密度变化向量对应一个接触力位置参数；S42: Obtain multiple sets of the magnetic flux density change vectors of the detection points in the preset three-dimensional coordinate system when the magnet is pressed by the contact force, wherein one set of the magnetic flux density change vectors The vector corresponds to a contact force position parameter;
S43：将每组所述三维磁通密度变化向量作为所述初始分类模型的样本数据，并将所述接触力的位置参数作为所述初始分类模型的样本标签，对所述初始分类模型进行训练；S43: Using each group of the three-dimensional magnetic flux density change vectors as the sample data of the initial classification model, and using the position parameter of the contact force as the sample label of the initial classification model, to train the initial classification model ;
S44：将训练完成的所述初始分类模型作为所述接触力位置分类模型。S44: Use the trained initial classification model as the contact force position classification model.
在本实施例中，在初次使用接触力位置分类模型前，需要对接触力位置分类模型进行训练，使得模型的输出符合要求，方可进行实际的接触力的检测。因此首先需要选定初始分类模型，将每组三维磁通密度变化向量作为初始分类模型的样本数据，将接触力的位置参数作为初始分类模型的样本标签，对初始分类模型进行训练，是初始分类模型输入样本数据后输出样本标签，后将训练好的初始分类模型作为接触力位置分类模型，其中，当初始分类模型的测试结果中显示模型分类精度达到99.5％及以上时，则判断初始分类模型训练完成。In this embodiment, before using the contact force position classification model for the first time, it is necessary to train the contact force position classification model so that the output of the model meets the requirements before the actual contact force detection can be performed. Therefore, the initial classification model needs to be selected first, and each group of three-dimensional magnetic flux density change vectors is used as the sample data of the initial classification model, and the position parameter of the contact force is used as the sample label of the initial classification model to train the initial classification model. After the model inputs the sample data, it outputs the sample label, and then uses the trained initial classification model as the contact force position classification model. When the test results of the initial classification model show that the classification accuracy of the model reaches 99.5% or above, the initial classification model is judged Training is complete.
本发明的接触力检测方法，通过初始分类模型进而得到接触力位置分类模型，进而检测施加在磁触觉传感器上的任何接触力的位置，通过分类模型提高了对接触力位置检测的精度。The contact force detection method of the present invention obtains the contact force position classification model through the initial classification model, and then detects the position of any contact force applied to the magnetic tactile sensor, and improves the detection accuracy of the contact force position through the classification model.
本发明实施例中，所述获取多组当所述磁体受到所述接触力按压时，得到所述检测点在所述预设三维坐标系中所述三维磁通密度变化向量，以及所述接触力的接触力位置参数，包括：In the embodiment of the present invention, the acquiring multiple sets of the three-dimensional magnetic flux density change vector of the detection point in the preset three-dimensional coordinate system when the magnet is pressed by the contact force, and the contact force Contact force location parameters for forces, including:
根据所述接触力以预设次数按压每个所述磁体的对应区域，得到所述检测点在每个所述坐标轴上的初始磁通密度变化；pressing the corresponding area of each of the magnets for a predetermined number of times according to the contact force to obtain the change of the initial magnetic flux density of the detection point on each of the coordinate axes;
根据所述每个所述坐标轴上的所述初始磁通密度变化，通过消除预设初始偏移量，得到所述三维磁通密度变化向量；Obtaining the three-dimensional magnetic flux density change vector by eliminating a preset initial offset according to the initial magnetic flux density change on each of the coordinate axes;
根据每个所述磁体的所述对应区域，得到所述接触力的所述接触力位置参数。According to the corresponding area of each of the magnets, the contact force position parameter of the contact force is obtained.
在本实施例中，通过控制接触力以预设次数对磁触觉传感器进行按压来获取磁通密度变化量，并计算磁通密度变化量在三个坐标轴的分量，并针对每个轴分量的变化值进行标准化处理，即减去固定预设偏移量得到最终的分量，将根据该分量作为初始分类模型的输入。In this embodiment, by controlling the contact force to press the magnetic tactile sensor for a preset number of times to obtain the change in magnetic flux density, and calculate the components of the change in magnetic flux density on the three coordinate axes, and for each axis component The change value is standardized, that is, the final component is obtained by subtracting the fixed preset offset, which will be used as the input of the initial classification model.
在本发明的优选实施例中，在数据处理方面，使用低通无限冲激响应滤波器(IIR)处理收集的数据，通带和阻带频率分别设置为3Hz和10Hz，通带纹波和阻带衰减分别设置为1dB和60dB；采用Chebyshev I型方法过滤后的数据被局部峰值和最小值切割，形成每个按压周期的加在和卸载数据集，将收集到的数据集根据饱和度数据的均值和标准差进行标准化，删除掉一些实验过程中因为意外导致数据突然发生很大变化的周期数据并用相应的位置类别标记为训练数据，得到的数据集分为训练、测试和验证数据集，其比例为70：15：15，为了识别关于不同接触位置的模式，使用闵可夫斯基距离度量训练K近邻分类模型其中，近邻数设置为1。In a preferred embodiment of the present invention, in terms of data processing, a low-pass infinite impulse response filter (IIR) is used to process the collected data, the passband and stopband frequencies are set to 3Hz and 10Hz respectively, and the passband ripple and stopband The band attenuation was set to 1dB and 60dB respectively; the data filtered by the Chebyshev type I method were cut by local peaks and minimums to form the addition and unloading data sets of each compression cycle, and the collected data sets were collected according to the saturation data Standardize the mean and standard deviation, delete some periodic data that caused a sudden change in the data due to accidents during the experiment, and mark it as training data with the corresponding position category. The obtained data set is divided into training, testing and verification data sets, and With a ratio of 70:15:15, to identify patterns with respect to different contact locations, a K-nearest neighbor classification model was trained using the Minkowski distance metric, where the number of neighbors was set to 1.
在本发明优选实施例中，在数据收集方面，结合图1的磁触觉传感器为例，磁触觉传感器通过三轴压迫平台进行了测试，X和Y平台的位置由两个测微手动线性平台控制，Z轴则由伺服电机与运动控制器共同控制。力传感器安装在Z平台上，在其下方，具有圆形尖端的探头连接到力传感器，其中，在一些优选实施例中，力传感器可以选用ATI-Nano17六轴力传感器进行测量。原型被放置在实验室千斤顶上并固定在一个用于定位四个磁体对应的测试位置的底座上。在检测到压头和磁触觉传感器之间的水平接触位置后，进行按压测试，每个位置进行20次循环，加载速度设置为0.2mm/s，采样率为100z。In the preferred embodiment of the present invention, in terms of data collection, taking the magnetic tactile sensor in Figure 1 as an example, the magnetic tactile sensor is tested through a three-axis compression platform, and the positions of the X and Y platforms are controlled by two micrometric manual linear platforms , Z-axis is jointly controlled by servo motor and motion controller. The force sensor is installed on the Z platform, and below it, a probe with a round tip is connected to the force sensor, wherein, in some preferred embodiments, the force sensor can be selected from the ATI-Nano17 six-axis force sensor for measurement. The prototype is placed on laboratory jacks and secured to a base that positions the four magnets for corresponding test positions. After detecting the horizontal contact position between the indenter and the magnetic tactile sensor, a press test was performed with 20 cycles per position, the loading speed was set at 0.2 mm/s, and the sampling rate was 100 z.
本发明的的接触力检测方法，对得到的三个坐标轴的磁通密度变化分量进行标准化处理，得到比较规律变化的磁通数据以增大模型训练的成功率，以提升训练效果，同时减轻由PCB集成引起的传感器上的不同刚度分布。The contact force detection method of the present invention standardizes the obtained magnetic flux density variation components of the three coordinate axes to obtain relatively regularly changing magnetic flux data to increase the success rate of model training, to improve the training effect, and at the same time reduce the Different stiffness distributions on the sensor caused by PCB integration.
结合图6所示，本发明实施例中，还包括：As shown in FIG. 6, in the embodiment of the present invention, it also includes:
S51：选定初始回归模型；S51: Select an initial regression model;
S52：获取多组当所述磁体受到所述接触力按压时，所述检测点的所述三维磁通密度变化向量和所述接触力的接触力大小参数，其中，一组所述三维磁通密度变化向量对应一个所述接触力位置参数和一个所述接触力大小参数；S52: Obtain multiple sets of the three-dimensional magnetic flux density change vector of the detection point and the contact force magnitude parameters of the contact force when the magnet is pressed by the contact force, wherein one set of the three-dimensional magnetic flux The density change vector corresponds to a position parameter of the contact force and a size parameter of the contact force;
S53：将所述三维磁通密度变化向量和所述接触力位置参数作为所述初始回归模型的输入，将所述接触力大小参数作为所述初始回归模型的输出，对所述初始回归模型进行训练；S53: Using the three-dimensional magnetic flux density change vector and the contact force position parameter as the input of the initial regression model, using the contact force size parameter as the output of the initial regression model, and performing an operation on the initial regression model train;
S54：将训练完成的所述初始回归模型作为所述接触力大小回归模型。S54: Use the trained initial regression model as the contact force magnitude regression model.
在本实施例中，与上述训练初始分类模型的步骤类似，将上述训练初始分类模型的数据集中的磁通密度变化量在三轴的分量以及数据集中接触力位置参数输入到初始回归模型中，进行训练，得到与三轴的分量对应的接触力的大小，模型输出接触力大小参数后，测量初始分类模型的回归相关性系数，当初始分类模型的回归相关性系数达到0.99及以上时，则证明该模型训练完成，偏差大于或等于预设结果阈值，则代表模型的输出没有达到训练标准，需要再次根据数据集对模型进行训练，直至偏差小于预设结果阈值，即可完成训练。In this embodiment, similar to the above-mentioned steps of training the initial classification model, the three-axis component of the magnetic flux density variation in the data set of the above-mentioned training initial classification model and the contact force position parameter in the data set are input into the initial regression model, Carry out training to obtain the size of the contact force corresponding to the three-axis component. After the model outputs the size parameter of the contact force, measure the regression correlation coefficient of the initial classification model. When the regression correlation coefficient of the initial classification model reaches 0.99 or above, then It proves that the model training is completed, and the deviation is greater than or equal to the preset result threshold, which means that the output of the model does not meet the training standard, and the model needs to be trained again according to the data set until the deviation is less than the preset result threshold, and the training can be completed.
本发明的的接触力检测方法，通过训练前馈神经网络根据接触力的位置以及磁通密度的变化，得到对应的接触力的大小，实现了即便在磁通密度变化不大时，接触力依旧可以被准确的测量，提高接触力检测精度。The contact force detection method of the present invention obtains the corresponding contact force size by training the feedforward neural network according to the position of the contact force and the change of the magnetic flux density, and realizes that the contact force remains the same even when the magnetic flux density changes little. It can be accurately measured and the precision of contact force detection can be improved.
在训练完成后，为了评估磁触觉传感器的静态性能，以图1的磁触觉传感器为例，再次对四个磁体对应的区域进行20个循环的按压测试，收集标准化后的磁通密度变化数据，并将其输入至训练好的接触力位置分类模型(K近邻分类模型)和接触力大小回归模型(前馈神经网络)中，得到用于进一步评估的接触力位置和力测量值，在评估中得到了磁触觉传感器满量程输出、迟滞误差、重复性误差和静态误差，根据上述误差可以得到磁触觉传感器对接触力的定位精度上升的结果，有效提高了对接触力的检测精度。After the training is completed, in order to evaluate the static performance of the magnetic tactile sensor, take the magnetic tactile sensor in Figure 1 as an example, perform 20 cycles of pressing tests on the areas corresponding to the four magnets again, and collect the standardized magnetic flux density change data. And input it into the trained contact force position classification model (K nearest neighbor classification model) and contact force size regression model (feedforward neural network), to obtain the contact force position and force measurement value for further evaluation, in the evaluation The full-scale output, hysteresis error, repeatability error and static error of the magnetic tactile sensor are obtained. According to the above errors, the positioning accuracy of the magnetic tactile sensor to the contact force can be obtained, which effectively improves the detection accuracy of the contact force.
结合图7所示，本发明还提供一种磁触觉传感器100，所述磁触觉传感器100包括多个磁体，所述磁触觉传感器除所述磁体外的部分由可变形材料构成，多个所述磁体均置于相同水平面上，每个所述磁体与相邻的其他所述磁体的磁化方向相反，且所述磁体的磁化方向垂直于所述水平面，所述磁触觉传感器100包括：As shown in FIG. 7 , the present invention also provides a magnetic tactile sensor 100, the magnetic tactile sensor 100 includes a plurality of magnets, the part of the magnetic tactile sensor except the magnets is made of deformable material, and the plurality of the magnetic tactile sensors The magnets are all placed on the same horizontal plane, and the magnetization direction of each magnet is opposite to that of other adjacent magnets, and the magnetization direction of the magnets is perpendicular to the horizontal plane. The magnetic tactile sensor 100 includes:
接触力位置检测单元130，用于根据每个所述磁体在预设三维坐标系中的检测点的三维磁通密度变化向量和接触力位置分类模型，得到接触力的位置参数；其中，所述三维磁通密度变化向量包括所述检测点在所述预设三维坐标系的三个坐标轴的磁通密度变化向量；The contact force
接触力大小检测单元140，用于根据所述三维磁通密度变化向量、所述位置参数和接触力大小回归模型，得到所述接触力的大小参数；The contact
处理单元150，用于根据所述位置参数和所述大小参数得到所述接触力的检测结果。The
结合图8所示，所述磁触觉传感器100还包括模型训练单元110和磁通密度检测单元120；As shown in FIG. 8, the magnetic tactile sensor 100 also includes a
接触力位置检测单元130还用于根据所述磁体建立所述预设三维坐标系，其中，所述三维坐标系包括XY平面，所述XY平面为所述水平面，将所述三维磁通密度变化向量输入所述接触力位置分类模型；根据所述接触力位置分类模型的输出，得到所述接触力的所述位置参数。The contact force
接触力大小检测单元140还用于根据所述检测点在所述预设三维坐标系的三个所述坐标轴的所述磁通密度变化向量和所述位置参数，得到四维接触力向量；将所述四维接触力向量输入所述接触力大小回归模型，得到所述接触力的所述大小参数。The contact
磁通密度检测单元120用于获取所述检测点的磁通密度；当所述磁体受到所述接触力作用时，将所述接触力作用时的所述检测点在每个所述坐标轴的磁通密度与未受到所述接触力作用时的所述检测点在每个所述坐标轴的所述磁通密度做差得到差值，将所述差值作为所述坐标轴的所述磁通密度变化向量；所述获取所述检测点的磁通密度，包括：获取每个所述磁体在所述检测点的区域磁通密度；根据每个所述磁体在所述检测点的所述区域磁通密度，得到所述检测点的所述磁通密度；所述根据每个所述磁体在所述检测点的所述区域磁通密度，得到所述检测点的所述磁通密度，包括：根据每个所述磁体在所述检测点的所述区域磁通密度和叠加公式，得到所述检测点的所述磁通密度；所述叠加公式为：
模型训练单元110用于选定初始分类模型；获取多组当所述磁体受到所述接触力按压时，得到所述检测点在所述预设三维坐标系中所述三维磁通密度变化向量，以及所述接触力的接触力位置参数；将每组所述三维磁通密度变化向量作为所述初始分类模型的样本数据，并将所述接触力的位置参数作为所述初始分类模型的样本标签，对所述初始分类模型进行训练；将训练完成的所述初始分类模型作为所述接触力位置分类模型。The
所述获取多组当所述磁体受到所述接触力按压时，得到所述检测点在所述预设三维坐标系中所述三维磁通密度变化向量，以及所述接触力的接触力位置参数，包括：根据所述接触力以预设次数按压每个所述磁体的对应区域，得到所述检测点在每个所述坐标轴上的初始磁通密度变化；根据所述每个所述坐标轴上的所述初始磁通密度变化，通过消除预设初始偏移量，得到所述三维磁通密度变化向量；根据每个所述磁体的所述对应区域，得到所述接触力的所述接触力位置参数；The acquiring multiple sets of the three-dimensional magnetic flux density change vector of the detection point in the preset three-dimensional coordinate system and the contact force position parameters of the contact force are obtained when the magnet is pressed by the contact force , comprising: pressing the corresponding area of each of the magnets for a preset number of times according to the contact force to obtain the change of the initial magnetic flux density of the detection point on each of the coordinate axes; according to each of the coordinates The change of the initial magnetic flux density on the axis, by eliminating the preset initial offset, the three-dimensional magnetic flux density change vector is obtained; according to the corresponding area of each of the magnets, the contact force is obtained contact force position parameter;
选定初始回归模型；获取多组当所述磁体受到所述接触力按压时，所述检测点的所述三维磁通密度变化向量和所述接触力的接触力大小参数，其中，一组所述三维磁通密度变化向量对应一个所述接触力位置参数和一个所述接触力大小参数；将所述三维磁通密度变化向量和所述接触力位置参数作为所述初始回归模型的输入，将所述接触力大小参数作为所述初始回归模型的输出，对所述初始回归模型进行训练；将训练完成的所述初始回归模型作为所述接触力大小回归模型。Selecting an initial regression model; obtaining a plurality of groups when the magnet is pressed by the contact force, the three-dimensional magnetic flux density change vector of the detection point and the contact force size parameters of the contact force, wherein a set of The three-dimensional magnetic flux density change vector corresponds to one of the contact force position parameters and one of the contact force size parameters; the three-dimensional magnetic flux density change vector and the contact force position parameter are used as the input of the initial regression model, and the The contact force magnitude parameter is used as the output of the initial regression model, and the initial regression model is trained; the trained initial regression model is used as the contact force magnitude regression model.
本发明的磁触觉传感器，通过将磁触觉传感器中检测点检测到的磁通密度变化向量输入至接触力位置分类模型，从而根据训练好的接触力分类模型，得到接触力的位置参数；再根据位置参数和磁通密度变化向量，通过训练好的接触力大小回归模型，得到接触力的大小参数，实现了针对任何大小的接触力都能根据两个模型实现相同准确度的接触力检测，进而提高了对于较小接触力检测的精准度。In the magnetic tactile sensor of the present invention, the magnetic flux density change vector detected by the detection point in the magnetic tactile sensor is input to the contact force position classification model, thereby obtaining the position parameter of the contact force according to the trained contact force classification model; and then according to The position parameter and the magnetic flux density change vector, through the trained contact force size regression model, get the size parameter of the contact force, and realize the contact force detection with the same accuracy according to the two models for any size of the contact force, and then Improved accuracy for small contact force detection.
需要说明的是，在本文中，诸如“第一”和“第二”等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。It should be noted that in this article, relative terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply these No such actual relationship or order exists between entities or operations. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements of or also include elements inherent in such a process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising said element.
以上所述仅是本发明的具体实施方式，使本领域技术人员能够理解或实现本发明。对这些实施例的多种修改对本领域的技术人员来说将是显而易见的，本文中所定义的一般原理可以在不脱离本发明的精神或范围的情况下，在其它实施例中实现。因此，本发明将不会被限制于本文所示的这些实施例，而是要符合与本文所申请的原理和新颖特点相一致的最宽的范围。The above descriptions are only specific embodiments of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Accordingly, the present invention will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310320776.0A CN116399500A (en) | 2023-03-29 | 2023-03-29 | A contact force detection method and magnetic tactile sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310320776.0A CN116399500A (en) | 2023-03-29 | 2023-03-29 | A contact force detection method and magnetic tactile sensor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116399500A true CN116399500A (en) | 2023-07-07 |
Family
ID=87011708
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310320776.0A Pending CN116399500A (en) | 2023-03-29 | 2023-03-29 | A contact force detection method and magnetic tactile sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116399500A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014085310A (en) * | 2012-10-26 | 2014-05-12 | Canon Inc | Magnetic type force sensor and robot device |
| CN111993446A (en) * | 2020-07-03 | 2020-11-27 | 北京大学 | Magnetic field-based flexible tactile sensor |
| WO2021005755A1 (en) * | 2019-07-10 | 2021-01-14 | 三菱電機株式会社 | Sensing device and sensing device system |
| WO2021061240A2 (en) * | 2019-06-21 | 2021-04-01 | Carnegie Mellon University | Systems and methods for sensing deformation of a magnetic material and fabrication methods thereof |
| CN114661169A (en) * | 2019-05-16 | 2022-06-24 | 腾讯科技（深圳）有限公司 | Vision-based tactile measurement method, device, device and storage medium |
| US20220276038A1 (en) * | 2021-03-01 | 2022-09-01 | City University Of Hong Kong | Electromechanical sensor and a method of sensing an object or a tactile input using the sensor |
-
2023
- 2023-03-29 CN CN202310320776.0A patent/CN116399500A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014085310A (en) * | 2012-10-26 | 2014-05-12 | Canon Inc | Magnetic type force sensor and robot device |
| CN114661169A (en) * | 2019-05-16 | 2022-06-24 | 腾讯科技（深圳）有限公司 | Vision-based tactile measurement method, device, device and storage medium |
| WO2021061240A2 (en) * | 2019-06-21 | 2021-04-01 | Carnegie Mellon University | Systems and methods for sensing deformation of a magnetic material and fabrication methods thereof |
| CN114144688A (en) * | 2019-06-21 | 2022-03-04 | 卡内基·梅隆大学 | System and method for sensing deformation of magnetic material and method of making the same |
| WO2021005755A1 (en) * | 2019-07-10 | 2021-01-14 | 三菱電機株式会社 | Sensing device and sensing device system |
| CN111993446A (en) * | 2020-07-03 | 2020-11-27 | 北京大学 | Magnetic field-based flexible tactile sensor |
| US20220276038A1 (en) * | 2021-03-01 | 2022-09-01 | City University Of Hong Kong | Electromechanical sensor and a method of sensing an object or a tactile input using the sensor |
Non-Patent Citations (1)
| Title |
|---|
| 杜承金: "《基于FBG的微创手术机器人力与形状感知系统研究》", 中国优秀硕士学位论文全文数据库信息科技辑, 15 January 2021 (2021-01-15), pages 1 - 144 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104769411B (en) | The rigid assay method of dynamic tensile and measure device of the outside plate of Vehicle component | |
| CN104657526B (en) | The Forecasting Methodology of the processing quality of toolroom machine | |
| TW200415340A (en) | Method and apparatus for measuring object thickness | |
| CN111999314B (en) | Device and method for automatically checking deformation of flexible PCB in processing process | |
| CN110998253A (en) | Apparatus and method for performing shock excitation techniques | |
| US20220349695A1 (en) | Systems and Methods for Sensing Deformation of a Magnetic Material and Fabrication Methods Thereof | |
| CN109902326A (en) | A kind of finite element simulation experiment effect assessment method | |
| CN109307582A (en) | Wind field detection device and detection method for additive manufacturing equipment | |
| CN107991177B (en) | A method for testing biaxial yielding and hardening behavior of sheet metal | |
| CN116399500A (en) | A contact force detection method and magnetic tactile sensor | |
| CN113532341B (en) | Method and device for determining a measurement strategy for measuring a measurement object and program | |
| CN117232452B (en) | Comparison instrument for detecting size of part and application method thereof | |
| CN114881938A (en) | Grain size detection method and system based on wavelet analysis and neural network | |
| CN104697478A (en) | Ruled surface adaptively measuring method based on constant contact force | |
| CN101592692A (en) | Measure the appraisal procedure of board | |
| CN117450883B (en) | A vertical probe rapid measurement method | |
| CN102262456A (en) | Information input device and method | |
| CN107421424A (en) | ω type spring symmetry slide block type Quick detection tools and the method for inspection | |
| CN114084764B (en) | Elevator transportation quality detection method and detection system | |
| CN112880966B (en) | Fluorescent oil film thickness obtaining method based on CFD numerical simulation calculation | |
| CN213398040U (en) | Tactile measuring device | |
| CN119321848B (en) | Pressure sensor core testing system with high-frequency response structure | |
| CN110738180B (en) | Method for evaluating signal accuracy and system precision in detection process | |
| JP2019144213A (en) | Tactile detection device | |
| CN222014741U (en) | Vibration detection device |
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 |