CN116337322A

CN116337322A - Array type force sensor calibration method and measuring method and device

Info

Publication number: CN116337322A
Application number: CN202310311103.9A
Authority: CN
Inventors: 何宇星; 黄世华; 许晋诚
Original assignee: Passini Perception Technology Shenzhen Co ltd
Current assignee: Passini Perception Technology Shenzhen Co ltd
Priority date: 2023-03-14
Filing date: 2023-03-14
Publication date: 2023-06-27

Abstract

The embodiment of the application belongs to the technical field of force sensor calibration, and relates to an array force sensor calibration method which is used for calibrating an array force sensor, wherein the array force sensor comprises an integrated force sensing layer and a signal acquisition layer; the signal acquisition layer includes a plurality of independent signal acquisition units, includes: acquiring sensing data and corresponding acting force parameters under different acting forces; the acting force parameters comprise: force position parameters and force vector parameters; the sensing data are a plurality of sensing data which are respectively acquired by a plurality of independent signal acquisition units; and training the initial calibration model by taking the acting force parameters and the sensing data as a training sample set to obtain a trained calibration model. The embodiment of the application also provides a measuring method/device of the array type force sensor. According to the technical scheme, the complexity and difficulty of calibrating the array type force sensor can be reduced.

Description

Array type force sensor calibration method and measuring method and device

Technical Field

The application relates to the technical field of force sensor calibration, in particular to an array force sensor calibration method, a measuring method and a measuring device.

Background

The array type force sensor comprises an integrated force sensing layer and a signal acquisition layer; the signal acquisition layer comprises a plurality of independent signal acquisition units. Typically the force sensing layer is integrally provided, such as: interconnected in the same elastic deformation body. In calibration for example for single point external force contacts, the measurement signals output by the individual signal acquisition units of the signal acquisition layer are an array of signals, wherein each measurement point in the array outputs an intermediate measurement in three directions XYZ. In order to correlate this arrayed signal with a single point external force, conventional methods require physical modeling of the sensor to formulaically establish a mathematical relationship of the arrayed signal to the single point external force.

However, in the case of an array force sensor with an integrated sensing layer, the integrated sensing layer is an integrated complex deformation process in the deformation process of stress, so that the mathematical relationship is extremely complex, and the modeling precision is quite limited; in addition, the model with high precision is poor in real-time performance, so that the traditional force sensor calibration method is not practical.

Disclosure of Invention

The embodiment of the application aims to provide an array type force sensor calibration method, a measurement method and a measurement device, so as to reduce the complexity and difficulty of array type force sensor calibration.

In a first aspect, an embodiment of the present application provides a calibration method for an array force sensor, which adopts the following technical scheme:

the array force sensor calibration method is used for calibrating an array force sensor, and the array force sensor comprises an integrated force sensing layer and a signal acquisition layer; the signal acquisition layer comprises a plurality of independent signal acquisition units, and the method comprises the following steps:

acquiring sensing data and corresponding acting force parameters under different acting forces; the force parameters include: force position parameters and force vector parameters; the sensing data are a plurality of sensing data which are respectively acquired by a plurality of independent signal acquisition units;

and training the initial calibration model by taking the acting force parameters and the sensing data as a training sample set to obtain a trained calibration model.

Further, in one embodiment, the force is a single point force; and/or

The acting force is a multipoint acting force; the acting force parameters under the multipoint acting force comprise: a plurality of force position parameters and force vector parameters corresponding to each position; wherein the multipoint force acts simultaneously on the array force sensor; or the multipoint force acts on the array force sensor in time sequence.

Further, in an embodiment, the calibration model is a multi-layer perceptron.

Further, in one embodiment, before the training of the initial calibration model using the force parameter and the sensing data as a training sample set, the method further includes the following steps:

preprocessing the sensing data based on formula (1);

N＝(M-Vmin)/(Vmax-Vmin) (1)

wherein M is sensing data; vmin is the minimum value of the sensing data output by the signal acquisition unit; vmax is the maximum value of the sensing data output by the signal acquisition unit; n is the sensing data result after pretreatment.

Further, in one embodiment, before acquiring the sensing data and the corresponding force parameters under different forces, the method includes the following steps:

acquiring the sensing data and the corresponding acting force parameters under each acting force;

the sensed data and the corresponding force parameters for each force are stored as a set of data.

Further, in one embodiment, the training the initial calibration model using the force parameter and the sensed data as a training sample set includes the steps of:

inputting current sensing data into the calibration model;

Updating parameters of the calibration model by taking the current acting force parameters corresponding to the current sensing data as marks;

repeating the steps until the preset termination condition is met.

Further, in one embodiment, before the step of acquiring the sensing data and the corresponding acting force parameters under different acting forces, the method further includes the following steps:

setting a value of a preset parameter of the acting force; the preset parameters comprise: a motion range parameter, a force vector range parameter, and a contact number parameter; the value of the contact number parameter is N, and N is an integer greater than or equal to 2.

Generating a plurality of control instructions to instruct the actuator to apply a plurality of forces to the arrayed force sensor based on the values of the preset parameters of the forces; the plurality of forces includes a single point force and/or a multiple point force; and acquiring corresponding sensing data output by a signal sensing layer under each acting force and corresponding acting force vector parameters output by a force transducer.

In a second aspect, an embodiment of the present application provides a method for measuring an array force sensor, configured to measure a force based on the array force sensor, where the array force sensor includes an integrated force sensing layer and a signal acquisition layer; the signal acquisition layer comprises a plurality of independent signal acquisition units, and the method comprises the following steps:

Acquiring sensing data;

inputting the current sensing data into the trained calibration model to obtain acting force parameters; the force parameters include: force location parameters and force vector parameters.

In a third aspect, an embodiment of the present application provides an array force sensor calibration device, configured to calibrate an array force sensor, where the array force sensor includes an integrated force sensing layer and a signal acquisition layer; the signal acquisition layer includes a plurality of independent signal acquisition units, the device includes:

the first acquisition module is used for acquiring sensing data and corresponding acting force parameters under different acting forces; the force parameters include: force position parameters and force vector parameters; the sensing data are a plurality of sensing data which are respectively acquired by a plurality of independent signal acquisition units;

and the model training module is used for training the initial calibration model by taking the acting force parameters and the sensing data as a training sample set to obtain a trained calibration model.

In a fourth aspect, embodiments of the present application provide an array force sensor measurement device for performing force measurement based on an array force sensor, the array force sensor including an integrated force sensing layer and a signal acquisition layer; the signal acquisition layer includes a plurality of independent signal acquisition units, the device includes:

The second acquisition module is used for acquiring the sensing data;

the parameter measurement module is used for inputting the current sensing data into the trained calibration model to obtain acting force parameters; the force parameters include: force location parameters and force vector parameters.

In a fifth aspect, embodiments of the present application provide a computer device, including a memory and a processor, where the memory stores a computer program, and the processor implements the steps of the method for calibrating and/or the method for measuring an array force sensor described in any of the above when the computer program is executed.

In a sixth aspect, embodiments of the present application provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the array force sensor calibration method and/or the force measurement method of any of the above.

Compared with the prior art, the embodiment of the application has the following main beneficial effects:

according to the embodiment of the application, the array type force sensor with the integrated signal acquisition layer and the plurality of signal acquisition units is calibrated based on the calibration model, and the calibration model is directly built from data, so that a complex physical modeling process is avoided, and the calibration complexity and difficulty are reduced.

In addition, the calibration method of the embodiment of the application is wide in adaptability, the calibration object is not limited to single-point force, the calibration can be performed on the multi-point force caused by complex actions, the resolution of the different actions is very complex for the traditional modeling method, the calibration is performed based on the calibration model, the complexity and difficulty of the calibration are reduced, and therefore the application range of the calibration method of the embodiment of the application is improved.

Drawings

For a clearer description of the solution in the present application, a brief description will be given below of the drawings that are needed in the description of the embodiments of the present application, it being obvious that the drawings in the following description are some embodiments of the present application, and that other drawings may be obtained from these drawings without inventive effort for a person of ordinary skill in the art.

FIG. 1 is a schematic frame construction of one embodiment of an array force sensor calibration system of the present application;

FIG. 2 is a schematic structural view of one embodiment of an arrayed force sensor of the present application;

FIG. 3 is a flow chart of one embodiment of an array force sensor calibration method of the present application;

FIG. 4 is a flow chart of one embodiment of an arrayed force sensor measurement method of the present application;

FIG. 5 is a schematic frame structure of one embodiment of an array force sensor calibration device of the present application;

FIG. 6 is a schematic diagram of a frame structure of one embodiment of an array force sensor measurement device of the present application;

FIG. 7 is a schematic diagram of an embodiment of a computer device of the present application.

Detailed Description

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the applications herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having" and any variations thereof in the description and claims of the present application and in the description of the figures above are intended to cover non-exclusive inclusions. The terms first, second and the like in the description and in the claims or in the above-described figures, are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the present application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.

In order to better understand the technical solutions of the present application, the following description will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings.

As shown in FIG. 1, FIG. 1 is a schematic diagram of a frame structure of one embodiment of an array force sensor calibration system of the present application.

An embodiment of the present application provides a calibration system for calibrating an array force sensor 20, the system including: actuator 11, clamp 13, bracket 14, load cell 15, and controller 16.

It should be noted that the above-mentioned array force sensor 20 may be any array sensor (hereinafter may also be simply referred to as "force sensor") that performs related force measurement based on a similar principle, for example: magnetic-based three-dimensional force sensors, light-based three-dimensional force sensors.

As shown in fig. 2, fig. 2 is a schematic structural diagram of one embodiment of an array force sensor of the present application. The array force sensor 20 includes an integrated force sensing layer 22 and a signal acquisition layer; the signal acquisition layer comprises a plurality of individual signal acquisition units 21.

Specifically, the "integrated force sensing layer" is an integral force sensing layer corresponding to a plurality of signal acquisition units of the signal acquisition layer.

The "plurality of individual signal acquisition units" are arranged in a random or regular matrix manner, and thus a force sensor comprising a plurality of individual signal acquisition units may be referred to simply as an "array force sensor".

It should be noted that, the plurality of independent signal acquisition units 21 refers to a plurality of independent units (for example, each unit is a hall sensor), but are not necessarily physically independent, for example: multiple independent signal acquisition units 21 may be fixed to the same circuit board.

In one embodiment, the sensing data described in embodiments of the present application is sensing data acquired by a signal acquisition layer.

Taking a magnetic-based force sensor as an example, the force sensor mainly uses magnetic field changes generated by the integrated force sensing layer 22 (also called a magnetic layer) during the stress deformation process, which are sensed by a plurality of independent signal acquisition units, as sensing data. The integrated force sensing layer 22 may be a combination of magnetic and elastic materials; and each individual signal acquisition unit 21 of the signal acquisition layer may be a hall sensor. Based on the elastic deformation of the integrated force sensing layer 22 under the action of external force, the distance between the integrated force sensing layer 22 and the plurality of independent signal acquisition units 21 is changed, so that each signal acquisition unit 21 senses the magnetic field in three directions of XYZ.

It should be noted that, the distribution density of the plurality of independent signal acquisition units may be set according to the measurement requirement, for example: the density of the measurement points (generally, the higher the density, the higher the measurement accuracy) and the like are determined according to the area that the sensing layer needs to cover.

The clamp 13 is fixed to a bracket 14.

It should be noted that the clamp 13 may be directly fixed to the bracket 14; or by some fixation means 12 to the bracket (as shown in fig. 1).

It will be appreciated that the clamp may be secured to the bracket at any desired location as desired.

And the clamp 13 is used for fixing the force sensor 20 to be calibrated.

In an alternative embodiment, the clamp 13 includes a base and an assembly location at the base.

The mounting locations cooperate with the array force sensor to secure the array force sensor to the mounting locations.

Specifically, the fixture and the assembly position thereof may be arbitrarily set according to the shape of the force sensor to be actually calibrated, for example: the array force sensor is rectangular in whole, and then the assembly position is rectangular.

In an alternative embodiment, the arrayed force sensor 20 includes an arrayed force sensor body and a carrier disposed outside the arrayed force sensor body.

The fitting of the assembly bit and the array type force sensor can be realized through the following structure: the assembly is adapted to a carrier disposed outside the body of the array force sensor.

According to the embodiment of the application, the array type force sensor is fixed on the assembling position of the clamp through the carrier, so that the carrier can play a role in protecting the array type force sensor on one hand, and damage to the array type force sensor in the clamping process of the clamp is prevented; on the other hand, the carrier can be designed arbitrarily according to the requirement, so as to be convenient for adapting to the clamp, for example: the fixture can be directly matched with the fixture with the same specification through the adjustment carrier; in addition, the carrier can be better designed according to the installation or adaptation requirement between the assembly position and the carrier, and the like, so that the assembly process can be simplified.

In an alternative embodiment, the clamp 13 is detachably fixedly connected to the fixing means 12.

According to the embodiment of the application, the clamp 13 is detachably and fixedly connected with the fixing device 12, when the array force sensor or the component to be tested comprising the array force sensor with different specifications is tested, the clamp can be replaced, the whole calibration device is not required to be replaced, and the calibration cost for calibrating the array force sensor with different specifications is reduced.

The actuator 11 is fixed to the bracket 14.

Specifically, the actuator may be fixed at any desired position of the support through the base, and the specific position of the actuator 11 on the support 14 may be designed by combining the possible movement travel range of the actuator 11 and the specific position of the force sensor 20 to be calibrated fixed on the support 14.

The actuator 11 includes an actuator body 111 and an end force application actuator 112.

The load cell 15 and the end force application actuator 112 are fixed to the end of the actuator body 111.

Specifically, the load cell 15 may be a unidirectional pressure sensor, a three-dimensional force sensor, or the like, as needed, and the present application is not limited thereto.

The actuator body 111 is configured to move the end effector 112 and/or apply force to the array force sensor.

Specifically, the actuator body 111 may be an XYZ platform or similar actuator constructed by a linear motor or a linear module; pneumatic, hydraulic, etc. or actuators constructed in a pneumatic, hydraulic, etc. transmission manner; or a robot, etc. The manipulator may be various types of manipulators, such as: serial manipulators such as a four-axis manipulator, a five-axis manipulator or a six-axis manipulator; or a parallel manipulator. Taking a serial manipulator as an example, the serial manipulator can be formed by sequentially connecting a plurality of driving joints and connecting rods in series.

In an alternative embodiment, as shown in FIG. 1, the actuator body 111 is a robotic arm. Compared with an XYZ platform and other similar executing mechanisms, the manipulator is adopted as an actuator main body, and the motion track of the manipulator can be controlled and changed through track planning, so that the manipulator can be more flexibly adapted to the calibration requirements of different types of sensors, and the universality of calibrating force sensors with different specifications can be further improved. For ease of understanding, the present embodiment is further described in detail by taking the actuator body 111 as the manipulator 111, and specifically, the end force application actuator 112 may be fixed directly or through an intermediate member (such as a load cell 113) to a flange at an output end of an end driving joint of the manipulator 111.

In an alternative embodiment, the end effector 112 may be a force bar 112; further, in an alternative embodiment, the load cell 15 is fixed to the end of the manipulator (e.g., a flange fixed to the end of the manipulator); one end of the force applying rod 112 is fixed at the center of one side of the force sensor 15 corresponding to the array force sensor, the free end of the force applying rod 112 moves under the drive of the manipulator 111 and applies force to the array force sensor 20, and the force sensor 15 is used for measuring the magnitude of the force applied by the force applying rod to the array force sensor 20.

In another alternative embodiment, the end effector 112 is an actuator having a drive structure (e.g., a cylinder) that moves the end effector and the drive structure drives the actuator to apply force to the array force sensor.

The controller 16 may be directly or indirectly communicatively coupled to the array force sensor 20, the actuator 11, and the load cell 15, respectively, by wired or wireless means.

The controller can combine pre-stored information and parameters according to a pre-fixed program; the manually input information and parameters, and force feedback signals collected and sent by the external force sensor and the array force sensor generate corresponding data, instructions and the like. The controller is defined by the description of the force sensor calibration method in the following embodiments.

It should be noted that, the controller 16 may be a unified module, and may respectively implement related functions such as calibration of the array force sensor and control of the actuator by calling different fixed programs and parameters; the system can also be divided into a plurality of independent modules which are respectively positioned on the attached calibration acquisition system and the actuator, and the system is not limited in the application. For ease of understanding, the controllers in the various cases described above will be referred to collectively as controllers 16 in the present embodiments.

As shown in fig. 1, the four-axis manipulator 111 includes a driving and controlling unit (not shown), the controller is communicatively connected to the driving and controlling unit of the four-axis manipulator 111, and the controller 16 sends the generated motion command to the driving and controlling unit, so as to instruct the manipulator and the end force application executor to complete the corresponding target actions.

It should be noted that the wireless connection may include, but is not limited to, 3G/4G/5G connection, wiFi connection, bluetooth connection, wiMAX connection, zigbee connection, UWB (ultra wideband) connection, and other now known or later developed wireless connection.

Specifically, the calibration method and the like of the force sensor provided by the embodiment of the invention can be applied to a computer terminal (Personal Computer, PC); industrial control computer terminals (Industrial Personal Computer, IPC); a mobile terminal; a server; the system comprises a terminal and a server, and is realized through interaction between the terminal and the server; a programmable logic controller (Programmable Logic Controller, PLC); field programmable gate arrays (Field-Programmable Gate Array, FPGA); a Digital signal processor (Digital SignalProcesser, DSP) or a micro control unit (Microcontroller unit, MCU) or the like.

Specifically, the method can be applied to the computer device shown in fig. 7, and the computer device can be a terminal or a server. The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDNs (Content Delivery Network, content delivery networks), basic cloud computing services such as big data and artificial intelligent platforms. The terminal may be, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart stereo, a smart watch, etc. The terminal and the server may be directly or indirectly connected through wired or wireless communication, which is not limited herein.

The present embodiment also provides a calibration method for an array force sensor, which is generally executed by the controller 16, and accordingly, the calibration device for an array force sensor described in the following embodiment is generally disposed in the controller 16.

As shown in fig. 3, fig. 3 is a flowchart of one embodiment of an array force sensor calibration method of the present application, for calibrating an array force sensor, where the array force sensor 20 includes an integrated force sensing layer 22 and a signal acquisition layer; the signal acquisition layer comprises a plurality of independent signal acquisition units 21, and the calibration method specifically comprises the following steps:

Step 210, acquiring sensing data and corresponding acting force parameters under different acting forces; the acting force parameters comprise: force position parameters and force vector parameters; the sensing data are a plurality of sensing data respectively collected by a plurality of independent signal collecting units.

In one embodiment, the controller may sequentially obtain a set of data from the memory or the server for each effort, each set of data including the sensed data and the corresponding effort parameter, according to a preset address.

It should be noted that, the forces acting on a certain point or multiple points generally cause the integrated sensing layer to undergo complex deformation, so each force may correspond to a change of a plurality of sensing data acquired by a plurality of signal acquisition units. In one embodiment, the plurality of sensing data may be stored in a predetermined order, so that different signal acquisition units correspond to one or a group of a series of sensing data, and only need to be sequentially extracted from the inside when in actual use; alternatively, in another embodiment, for better differentiation, a preset code may be preset for each signal acquisition unit, so the sensing data under each acting force further includes a preset code of the signal acquisition unit corresponding to the sensing data.

For example, the sensed data for each force and the corresponding set of force parameters may be recorded as follows: (309.53, -9.37, -89.36,0,0,1,0,24948, -24538,0,24940, -24530 … …), wherein the first three digits "309.53, -9.37, -89.36" represent three-dimensional force position coordinate information for a point; "0, 1" represents vector information of the applied force; "0,24948, -24538,0,24940, -24530 … …" may represent a plurality of sensed data corresponding to the force parameters described above, wherein every three data represent three-dimensional sensed data acquired by a first signal acquisition unit in a sequential arrangement, while "0,24948, -24538" represent three-dimensional magnetic field data acquired by a second signal acquisition unit in a sequential arrangement, and so on.

In one embodiment, prior to step 210, the following steps may be further included:

step 280 obtains sensing data and corresponding force parameters for each force;

in one embodiment, the controller may acquire sensing data acquired by the signal acquisition unit of the array type force sensor from the memory or the server according to a preset address; in one embodiment, the controller may obtain force vector parameters collected by the load cell from a memory or server at a preset address; in addition, in one embodiment, the controller may obtain force location information from the memory or the server at a predetermined address, where the force location information may be generated randomly or according to a certain rule by the controller of the calibration system within a predetermined range of motion, as will be described in further detail below.

Specifically, the sensing data is represented by different forms of data according to different working principles of the force sensor, for example, a certain magnetic signal acquisition unit is taken as an example, and the sensing data can be magnetic field change data in xyz three directions.

Step 290 stores the sensed data for each applied force and the corresponding applied force parameters as a set of data.

In one embodiment, the controller saves each set of data to memory or sends to the server at a preset address.

By the method, the sensing data and the corresponding acting force parameters under each acting force can be respectively stored as one group of data, so that a training sample set formed by multiple groups of data under multiple acting forces is obtained.

In one embodiment, each of the forces described above may be a single point of force.

In another embodiment, each force may be a multipoint force, and the force parameters under the multipoint force include: a plurality of force location parameters and force vector parameters corresponding to each location.

Further, in one embodiment, the multiple forces may be generated simultaneously, such as: acting force generated by simultaneous pressing of multiple fingers; a set of forces occurring in time sequence may also be used, such as: for the force generated by hand sliding and sweeping, the multipoint force does not act on the force sensor at the same time, but acts on the force sensor according to a certain time sequence.

The calibration method of the embodiment of the application has wide adaptability, and the object of the force related to the calibration is not limited to single-point force, but can also be multi-point force, for example: in the application scene of the game handle, the calibration model can be used for distinguishing the force application actions of different hands such as single-finger pressing, two-finger pressing, hand sweeping, light pressing, heavy pressing and the like, the distinguishing of acting forces generated by the different actions is very complex for the traditional modeling method, and the applicability of the force sensor to various conditions can be improved by calibrating the force sensor by adopting the calibration model.

It should be noted that the force parameters generally include a force position parameter and a force vector parameter; in addition, other acting force parameters can be added according to the requirement, and the application is also within the scope of protection.

Step 220, training the initial calibration model by taking the sensing data and the corresponding acting force parameters as a training sample set, and obtaining a trained calibration model.

In one embodiment, the controller may acquire a set of sensing data and corresponding force parameters under the current force in a preset order, and train the model; repeating the steps until the preset termination condition is met.

It should be noted that the calibration model described in the embodiments of the present application may include any network composed of neurons capable of implementing the above functions, for example: feed-Forward Networks, RNNs, LSTM, transformer, GNN, GAN, AE, MLP, convolutional Neural Networks (CNNs), common CNN models may include, but are not limited to: leNet, alexNet, ZFNet, VGG, googLeNet Residual Net, denseNet, R-CNN, SPP-NET, fast-RCNN, FCN, mask-RCNN, YOLO, SSD, GCN, and other now known or later developed network model structures.

In a preferred embodiment, a multi-layer perceptron (MLP) based network model is more suitable for this calibration scenario, where MLP is less likely to be overfitted, training loss is smoother, training/evaluation loss is free of rebound, final model loss is smaller, and inference speed is faster than a very widely applied CNN model.

Specifically, the calibration model training method can adopt various existing or future developed model training methods such as supervised learning, semi-supervised learning and the like for training.

According to the embodiment of the application, the array type force sensor with the integrated signal acquisition layer and the plurality of independent signal acquisition units is calibrated based on the calibration model, sensing data and corresponding acting force parameters under a plurality of acting forces are directly used as a training sample set, the calibration model is trained from the data, the complex physical modeling process is avoided, and the calibration complexity and difficulty are reduced.

In an alternative embodiment, step 220, before training the initial calibration model using the force parameters and the sensed data as the training sample set, further comprises:

step 290 pre-processes the sensed data based on equation (1);

N＝(M-Vmin)/(Vmax-Vmin) (1)

In one embodiment, the sensing data collected by a signal collection unit under a certain acting force is M; the controller selects the maximum value and the minimum value of the sensing data based on the sensing data acquired by the signal acquisition unit under all the pre-stored acting forces.

According to the embodiment of the application, the value of the sensing data input into the calibration model can be adjusted to be within the preset range through the operation, so that the distribution of the data input into the calibration model is certain, the subsequent learning of the calibration model is facilitated, and otherwise, if the distribution of the input calibration model is changed and uncertain frequently, the learning of the calibration model is difficult.

In one embodiment, taking supervised learning as an example, step 220 may specifically include the steps of:

step 221, inputting the current sensing data into a calibration model;

step 222, updating the parameters of the calibration model by taking the current acting force parameters corresponding to the current sensing data as the marks;

step 223 repeats the above steps until a preset termination condition is met.

In one embodiment, the controller inputs current sensed data in a set of current data acquired in sequence into a calibration model with initial parameters, outputs current acting force parameters, compares the outputted acting force parameters of the model with acting force parameters corresponding to the set of sensed data (the acting force parameters are used as marks of the sensed data at the moment) which are stored in advance based on a loss function, and iteratively updates parameters of the calibration model according to the difference until a termination condition (such as that the difference is smaller than a certain threshold value) is met, thereby completing training of the calibration model.

According to the embodiment of the application, the sensing data are used as input data of the calibration model, the acting force parameters corresponding to the sensing data are used as marks of the calibration model to train the model, and training of the model can be completed through a simple training method.

In an alternative embodiment, before the step 210 obtains the sensing data and the corresponding force parameters under different forces, the following steps may be further included:

step 230 sets a value of a preset parameter of the applied force; the preset parameters comprise: a range of motion parameter, a force vector range parameter, and a number of contacts parameter.

In one embodiment, the controller obtains preset parameters of the acting force from the memory or the server according to preset addresses; and obtaining the value of the preset parameter which is received (for example, obtaining the value of all or part of the preset parameter based on the signal input by the human hand) or generated (for example, for the feedback signal collected by some measuring tools or sensors, automatically generating the value of all or part of the preset parameter based on a preset program) from a memory or a server according to the preset address, and giving the value to the corresponding preset parameter so as to complete the setting of the value of the preset parameter.

The movement range parameter may refer to a movement range parameter of an end-effector (hereinafter, may be simply referred to as an "end of effector").

It should be noted that the above-mentioned motion range parameter may be obtained entirely or partially based on actual measurement by an image sensor or a laser sensor (for example, tangential coordinates of two diagonal points of the rectangular force sensor assembly are obtained by actual measurement); in addition, the calculation may be performed based on the actuator.

In one embodiment, the range of motion parameter may be a defined area based on a plurality of limit points (e.g., rectangular area defined by two diagonal limit points) or a combination of limit points and certain preset parameters (e.g., circular area defined by a center limit point and radius parameters), and the controller generates control instructions to instruct the actuator tip to move within the defined area.

It should be noted that, the coordinates of the limit points may be two-dimensional or multidimensional, which are all within the scope of the present application.

In an alternative embodiment, the motion range parameter may be represented by three-dimensional coordinates of the limit point. For example, as shown in fig. 2, two diagonal points of the array force sensor 20 are limit points t and t ', and three-dimensional coordinates of the two limit points t and t ' are respectively set to xyz and coordinates x ' y ' z ', wherein xy/x ' y ' is tangential coordinate (tangential may refer to a direction along a surface of the array force sensor); z/z' is the normal coordinate (normal may refer to the direction perpendicular to the surface of the array force sensor). Exemplary, such as coordinates (375, 11-162) and coordinates (395, 37-164), wherein the range of planar rectangular areas corresponding to the arrayed force sensor 20, i.e., the range of motion of the actuator tip, can be obtained based on tangential coordinates (375, 11), (395, 37) of the two corner points; the range of motion of the actuator tip in the normal direction, i.e., the range of depth of motion of the tip force actuator when applying force to the array force sensor, can be derived based on the normal coordinates, e.g., between-162 and-164 based on the example above.

The force vector range parameter may refer to a range of directions and magnitudes of forces applied to the array force sensor by the end of the actuator, and specifically may be set accordingly according to a practical application range and/or a nominal range of the array force sensor, where the force may be one-dimensional or multidimensional, for example: 0, 10 may represent a force O in the tangential direction and 10 newtons in the normal direction.

The number of contacts, i.e., the number of times the actuator tip applies force to the arrayed force sensor. Specifically, the number of contacts N may be arbitrarily set according to the requirements of the calibration accuracy, the movement range parameter, the force vector range parameter, and the like.

It should be noted that, the preset parameters of the acting force may include other preset parameters as required besides the above-mentioned motion range parameters, acting force vector range parameters and contact number parameters, which all belong to the scope of protection of the present application.

Step 240, generating a plurality of control instructions to instruct the actuator to apply a plurality of forces to the array force sensor based on the values of the preset parameters of the forces; the plurality of acting forces comprise single-point acting forces and/or multi-point acting forces, and corresponding sensing data output by the plurality of signal acquisition units and corresponding acting force vector parameters output by the load cell under each acting force are obtained.

In one embodiment, the controller may select, randomly or according to a preset rule, a certain force position data and a force vector parameter within a range of values of the preset parameter set in the above embodiment, and generate a corresponding control instruction based on a pre-stored fixed program, parameter, and the like in combination with force feedback data of the external load cell 15 on the end of the actuator (the feedback data is finally the same as or close to the selected force vector parameter through the control instruction); and acquiring corresponding sensing data under the acting force through the array type force sensor.

In another embodiment, the controller may select, randomly or according to a preset rule, certain force position data within the range of values of the preset parameters set in the above embodiment, and generate a corresponding control instruction based on a pre-stored fixed program, parameters, and the like; and acquiring force feedback data of the tail end of the actuator output by the external force sensor and sensing data acquired by the array force sensor, thereby acquiring the acting force parameter of each acting force and the corresponding sensing data.

It should be noted that the plurality of control instructions may be set as required, for example: each time the control command is generated, i.e. each time a different force (different position data and/or different force vector parameters) is applied to the array force sensor; in addition, in some cases, multiple identical control commands may be generated separately for each location, multiple sensing data under the identical control commands may be acquired to further and more precisely mark the sensor, and so on, which falls within the scope of the present application.

According to the embodiment of the application, the preset parameters are set, assignment is carried out on the preset parameters based on the array type force sensors with different specifications, and a plurality of control instructions related to the actuator are automatically generated based on the values of the preset parameters so as to instruct the actuator to apply a plurality of acting forces to the array type force sensors; acquiring corresponding sensing data output by the array type force sensor under each acting force and corresponding acting force vector parameters output by the force sensor; based on the method, the mapping relation between acting force and sensing data is constructed, so that different calibration systems do not need to be built for sensors with different specifications, the universality of the calibration of the force sensors with different specifications is improved, and the calibration cost of the array force sensors with different specifications is reduced.

In addition, the embodiment of the application assigns values for the preset parameters on the basis of the array force sensors with different specifications in advance, and then automatically generates a plurality of control instructions related to the actuator on the basis of the values of the preset parameters so as to instruct the actuator to apply a plurality of acting forces to the array force sensors; acquiring corresponding sensing data output by the array type force sensor under each acting force and corresponding acting force vector parameters output by the force sensor; based on the method, the mapping relation between the acting force and the sensing data is constructed, so that the automatic calibration of the force sensor can be realized.

In one embodiment, taking the manipulator as an example, the step of setting the value of the range of motion parameter in step 230 may further include a method step, which is generally performed by the controller 16:

step S250, a first movement instruction is sent to an actuator to indicate that the tail end of the actuator moves to the limit point of the area where the array type force sensor is located; tangential coordinates of the end of the actuator at the limit point are obtained.

In one embodiment, the controller may derive or generate a first motion command based on teaching or robot following, etc., to instruct the end of the actuator to move to a limit point (e.g., two limit points at diagonal positions) located at the boundary of the force sensor to be calibrated as described in the above embodiments; and then, based on a kinematic equation of the manipulator, a conversion relation between the end force application actuator and a manipulator end coordinate system and the like, calculating to obtain tangential coordinates of the actuator end at the limit point.

Step S260, based on force measurement feedback of the force sensor and the value of the force vector range parameter, sending a second movement instruction to the actuator so as to instruct the end of the actuator to move along the normal direction corresponding to the limit point; and acquiring the normal maximum coordinates and the normal minimum coordinates of the tail end of the actuator when the maximum value and the minimum value of the corresponding acting force vector range parameters are obtained.

The minimum coordinate along the normal direction may be the normal coordinate corresponding to the end of the actuator when the value or critical value of the force feedback output by the force sensor is the minimum value (for example, critical value 0) in the set force range value; the maximum coordinate may be a normal coordinate corresponding to the end of the actuator when the force feedback value is at or near the maximum of the preset force range values.

In one embodiment, the controller can drive the end of the actuator to move along the normal direction through the second movement instruction, and when the force measurement value output by the force measurement sensor positioned at the end of the actuator is positioned at the O critical value, the coordinate of the end of the actuator in the normal direction is the minimum coordinate along the normal direction; when the output value of the force sensor is at or near the maximum value in the preset acting force range value, the coordinate of the tail end of the actuator in the normal direction is the maximum coordinate along the normal direction.

Step S270 takes tangential coordinates and normal maximum coordinates and minimum coordinates as values of the movement range parameter. Such as: the limit point coordinates (375, 11, -162) and limit point coordinates (395, 37, -164) described in the above examples.

According to the embodiment of the application, the controller is combined with force measurement feedback of the force sensor, the motion of the actuator is controlled, and based on the motion feedback data of the actuator, the three-dimensional coordinate of the tail end of the actuator can be calculated, so that values of the motion range parameters can be obtained for array force sensors with different specifications or fixed at any positions, and the universality of the force sensor calibration method is further improved.

Based on the force sensor obtained by the force sensor calibration method described in the above embodiments, the present embodiment also provides a force measurement method, which is generally performed by the controller 16, and accordingly, the calibration device of the force sensor described in the following embodiments is generally disposed in the controller 16.

As shown in fig. 4, fig. 4 is a flow chart of one embodiment of an array force sensor measurement method of the present application. Force measurement is carried out based on an array force sensor, wherein the array force sensor comprises a signal acquisition layer and a signal sensing layer; the signal acquisition layer comprises a plurality of independent signal acquisition units, and can comprise the following steps:

step 310 obtains sensing data.

Step 320, inputting the sensing data into the trained calibration model to obtain an acting force parameter of the acting force; the acting force parameters comprise: force location parameters and force vector parameters.

In one embodiment, the force may be a single point force.

In another embodiment, the force is a multi-point force; the acting force parameters under the multipoint acting force comprise: a plurality of force location parameters and force vector parameters corresponding to each location.

Further, in one embodiment, the multiple forces may be generated simultaneously, such as: acting force generated by simultaneous pressing of multiple fingers; a set of forces occurring in time sequence may also be used, such as: the multipoint force does not act on the force sensor at the same time, but acts on the force sensor according to a certain time sequence aiming at the acting force generated by hand sliding and sweeping.

In an alternative embodiment, before the step 320 of inputting the sensing data into the trained calibration model, the following steps may be further included:

step 330 pre-processes the sensed data based on equation (1);

N＝(M-Vmin)/(Vmax-Vmin) (1)

For descriptions of calibration model training, observation data, and the like, reference may be made to the above embodiments, and the description thereof will not be repeated.

The force measurement method provided by the embodiment of the application is wide in adaptability, the measurement object is not limited to single-point force, the measurement can be performed on multi-point force formed by complex actions, and the applicability of the force sensor on measurement of various complex acting forces can be improved.

Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored in a computer-readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. The storage medium may be a nonvolatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a random access Memory (Random Access Memory, RAM).

It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.

With further reference to fig. 5, as an implementation of the method shown in fig. 3, the present application provides an embodiment of an array type force sensor calibration device, where the embodiment of the device corresponds to the embodiment of the method shown in fig. 3, and the device may be specifically applied to various electronic devices.

As shown in fig. 5, the array type force sensor calibration device 400 of the present embodiment includes: a first acquisition module 410, a model training module 420. Wherein:

A first acquiring module 410, configured to acquire sensing data and corresponding force parameters under different forces; the acting force parameters comprise: force position parameters and force vector parameters; the sensing data are a plurality of sensing data respectively collected by a plurality of independent signal collecting units.

The model training module 420 is configured to train the initial calibration model with the acting force parameter and the sensing data as a training sample set, and obtain a trained calibration model.

In an alternative embodiment, the force is a single point force; and/or

The acting force is a multipoint acting force; the force parameters under the multipoint force include: a plurality of force position parameters and force vector parameters corresponding to each position; wherein, the multipoint acting force acts on the array type force sensor at the same time; or the multi-point force acts on the array force sensor in time sequence.

In an alternative embodiment, the calibration model is a multi-layer perceptron.

In an alternative embodiment, the arrayed force sensor calibration apparatus 400 further comprises:

the data processing module is used for preprocessing the sensing data based on the formula (1);

N＝(M-Vmin)/(Vmax-Vmin) (1)

the parameter acquisition module is used for acquiring sensing data and corresponding acting force parameters under each acting force;

and the data storage module is used for storing the sensing data under each acting force and the corresponding acting force parameters as a group of data.

In an alternative embodiment, model training module 420 includes:

the data input sub-module is used for inputting the current sensing data into the calibration model;

the parameter updating sub-module is used for updating parameters of the calibration model by taking the current acting force parameter corresponding to the current sensing data as a mark;

and the step repeating submodule is used for repeating the steps until a preset termination condition is met.

the parameter setting module is used for setting the value of a preset parameter of the acting force; the preset parameters comprise: a motion range parameter, a force vector range parameter, and a contact number parameter; the value of the contact number parameter is N, and N is an integer greater than or equal to 2.

The instruction generation module is used for generating a plurality of control instructions based on the value of the preset parameter of the acting force so as to instruct the executor to apply a plurality of acting forces to the array force sensor; the plurality of forces includes a single point force and/or a multiple point force; corresponding sensing data output by a signal sensing layer under each acting force and corresponding acting force vector parameters output by a force transducer are obtained.

With further reference to fig. 6, as an implementation of the method shown in fig. 4, the present application provides an embodiment of an array type force sensor calibration device, where the embodiment of the device corresponds to the embodiment of the method shown in fig. 4, and the device may be specifically applied to various electronic devices.

As shown in fig. 6, the array type force sensor measuring apparatus 500 of the present embodiment includes: a second acquisition module 510, a data input module 520. Wherein:

a second acquisition module 510, configured to acquire the sensing data.

The parameter measurement module 520 is configured to input the current sensing data into the trained calibration model to obtain an acting force parameter; the acting force parameters comprise: force location parameters and force vector parameters.

In an alternative embodiment, the force is a single point force; and/or

N＝(M-Vmin)/(Vmax-Vmin) (1)

In order to solve the technical problems, the embodiment of the application also provides computer equipment. Referring specifically to fig. 7, fig. 7 is a basic structural block diagram of a computer device according to the present embodiment.

The computer device 7 comprises a memory 71, a processor 72, a network interface 73 communicatively connected to each other via a system bus. It should be noted that only computer device 7 having components 71-73 is shown in the figures, but it should be understood that not all of the illustrated components are required to be implemented and that more or fewer components may be implemented instead. It will be appreciated by those skilled in the art that the computer device herein is a device capable of automatically performing numerical calculations and/or information processing in accordance with predetermined or stored instructions, the hardware of which includes, but is not limited to, microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASICs), programmable gate arrays (fields-Programmable Gate Array, FPGAs), digital processors (Digital Signal Processor, DSPs), embedded devices, etc.

The computer equipment can be a desktop computer, a notebook computer, a palm computer, a cloud server and other computing equipment. The computer equipment can perform man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch pad or voice control equipment and the like.

The memory 71 includes at least one type of readable storage medium including flash memory, hard disk, multimedia card, card memory (e.g., SD or DX memory, etc.), random Access Memory (RAM), static Random Access Memory (SRAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), programmable Read Only Memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the storage 71 may be an internal storage unit of the computer device 7, such as a hard disk or a memory of the computer device 7. In other embodiments, the memory 71 may also be an external storage device of the computer device 7, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the computer device 7. Of course, the memory 71 may also comprise both an internal memory unit of the computer device 7 and an external memory device. In this embodiment, the memory 71 is generally used to store an operating system and various application software installed on the computer device 7, such as program codes of an array force sensor calibration method. Further, the memory 71 may be used to temporarily store various types of data that have been output or are to be output.

The processor 72 may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 72 is typically used to control the overall operation of the computer device 7. In this embodiment, the processor 72 is configured to execute the program code stored in the memory 71 or process data, such as the program code for executing the calibration method of the array force sensor.

The network interface 73 may comprise a wireless network interface or a wired network interface, which network interface 73 is typically used for establishing a communication connection between the computer device 7 and other electronic devices.

The present application also provides another embodiment, namely, a computer readable storage medium, where a calibration method program of a force sensor is stored, where the calibration method program of a force sensor is executable by at least one processor, so that the at least one processor performs the steps of the calibration method of a force sensor as described above.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk), comprising several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method described in the embodiments of the present application.

It is apparent that the embodiments described above are only some embodiments of the present application, but not all embodiments, the preferred embodiments of the present application are given in the drawings, but not limiting the patent scope of the present application. This application may be embodied in many different forms, but rather, embodiments are provided in order to provide a more thorough understanding of the present disclosure. Although the present application has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described in the foregoing, or equivalents may be substituted for elements thereof. All equivalent structures made by the specification and the drawings of the application are directly or indirectly applied to other related technical fields, and are also within the protection scope of the application.

Claims

1. The array type force sensor calibration method is characterized by being used for calibrating an array type force sensor, wherein the array type force sensor comprises an integrated force sensing layer and a signal acquisition layer; the signal acquisition layer comprises a plurality of independent signal acquisition units, and the method comprises the following steps:

2. The method of calibrating an array force sensor of claim 1, wherein the force is a single point force; and/or

3. The method of calibrating an array force sensor according to claim 1 or 2, wherein the calibration model is a multi-layer perceptron.

4. The method of calibrating an array force sensor according to claim 1 or 2, wherein before training the initial calibration model with the force parameter and the sensed data as a training sample set, further comprising the steps of:

Preprocessing the sensing data based on formula (1);

N ＝ (M-Vmin)/(Vmax-Vmin) (1)

5. The method for calibrating an array force sensor according to claim 1 or 2, wherein before acquiring the sensing data and the corresponding force parameters under different forces, the method comprises the following steps:

6. The method of calibrating an array force sensor according to claim 1 or 2, wherein the training of the initial calibration model with the force parameter and the sensed data as a training sample set comprises the steps of:

inputting current sensing data into the calibration model;

repeating the steps until the preset termination condition is met.

7. The method for calibrating an array force sensor according to claim 1 or 2, further comprising the following steps before acquiring the sensing data and the corresponding force parameters under different forces:

8. The measuring method of the array type force sensor is characterized by comprising an integrated force sensing layer and a signal acquisition layer, wherein the measuring method is used for measuring force based on the array type force sensor; the signal acquisition layer comprises a plurality of independent signal acquisition units, and the method comprises the following steps:

Acquiring sensing data;

9. The array type force sensor calibration device is characterized by being used for calibrating an array type force sensor, and the array type force sensor comprises an integrated force sensing layer and a signal acquisition layer; the signal acquisition layer includes a plurality of independent signal acquisition units, the device includes:

10. An array force sensor measuring device is characterized by being used for measuring force based on an array force sensor, and the array force sensor comprises an integrated force sensing layer and a signal acquisition layer; the signal acquisition layer includes a plurality of independent signal acquisition units, the device includes:

The second acquisition module is used for acquiring the sensing data;