Grasping devices, grasping control methods, electronic devices, storage media and products

Abstract

This application relates to a gripping device, gripping control method, electronic device, storage medium, and product. The gripping device collects bioelectrical signals of the object being gripped by the device through an electrical signal detection component. The gripping device adjusts the torque of the gripping device on the object by means of the bioelectrical signals. This can reduce the control delay of the gripper when the object being gripped experiences a sudden situation, thereby improving the response speed to sudden situations of the object being gripped.

Description

Technical Field

[0001] This application relates to the field of gripping control technology, and in particular to a gripping device, gripping control method, electronic device, storage medium and product. Background Technology

[0002] With the continuous development of industrial automation technology, intelligent control of grippers such as robotic arms and suction cups is being used more and more widely in people's lives.

[0003] Currently, to achieve intelligent control of the fixture, a piezoresistive sensor needs to be installed on the fixture to detect the normal pressure when the fixture grasps an object. However, if only a piezoresistive sensor is used to detect the normal pressure of the fixture, the limited spatial resolution of the piezoresistive sensor will result in the inability to detect the sliding of the grasped object in time during the grasping process. This leads to a large delay in the control of the fixture when the grasped object experiences a sudden situation, and it cannot improve the response speed to sudden situations of the grasped object. Summary of the Invention

[0004] This application provides a gripping device that can reduce the control delay of the clamp when the object being gripped experiences a sudden situation, thereby improving the response speed to the sudden situation of the object being gripped and at least partially solving the above-mentioned technical problems.

[0005] To achieve the above objectives, according to a first aspect of this application, a grasping device is provided, the grasping device comprising: an electrical signal detection component, the electrical signal detection component being used to acquire a bioelectrical signal of a grasping object targeted by the grasping device, the bioelectrical signal being used to adjust the torque of the grasping device on the grasping object.

[0006] Optionally, the electrical signal detection component includes: a microelectrode array, the microelectrode array being arranged in a preset shape, and / or the microelectrode array being composed of a plurality of preset electrodes arranged at preset intervals, wherein the spatial resolution corresponding to the preset electrodes is greater than or equal to a preset threshold.

[0007] According to a second aspect of this application, a crawling control method is provided, comprising:

[0008] Based on the electrical signal detection component on the grasping device, the bioelectrical signal of the grasping object targeted by the grasping device is collected;

[0009] Based on the bioelectric signal of the object being grasped, the torque of the grasping device on the object being grasped is adjusted.

[0010] Optionally, adjusting the torque of the grasping device on the grasped object based on the bioelectrical signal of the grasped object includes:

[0011] The bioelectric signal is identified by a preset neural network model, or the bioelectric signal and the environmental information of the grasped object are identified to determine at least one object label of the grasped object;

[0012] Based on the object tag, determine the object control parameters for the crawled object;

[0013] The grasping device is controlled based on the object control parameters to adjust the torque of the grasping device on the grasped object.

[0014] Optionally, determining the object control parameters for the crawled object based on the object tag includes:

[0015] The target scene where the crawled object is located is determined based on the object label, and the object control parameters for the crawled object are determined based on the target scene;

[0016] In the case where the target scene is a biological scene, a first parameter value table corresponding to the biological scene is determined, and the object control parameters of the crawling object are determined from the first parameter value table based on the object tag.

[0017] Specifically, when the target scene is a non-biological scene, a second parameter value table corresponding to the non-biological scene is determined; based on the force information corresponding to the bioelectric signal, the object control parameters of the grasping object are determined from the second parameter value table.

[0018] Optionally, before identifying the bioelectric signal using a preset neural network model, or identifying the bioelectric signal and environmental information of the grasped object to determine at least one object label of the grasped object, the method further includes:

[0019] The bioelectric signal is separated by a signal classification algorithm to obtain a static signal and a dynamic signal. The static signal and the dynamic signal are then used as the processed bioelectric signal and input into the neural network model.

[0020] Optionally, before identifying the bioelectric signal using a preset neural network model, or identifying the bioelectric signal and environmental information of the grasped object to determine at least one object label of the grasped object, the method further includes:

[0021] Electrochemical feature extraction is performed on the bioelectric signal to obtain the extracted signal;

[0022] The extracted signal is subjected to feature recognition to obtain feature recognition results, which are then used as processed bioelectrical signals input into the neural network model; and / or

[0023] The bioelectric signal is subjected to biopotential decoding to obtain decoded information;

[0024] Biological characteristics are extracted from the decoded information to obtain target decoded information, which is then input into the neural network model as a processed bioelectrical signal.

[0025] According to a third aspect of this application, a gripping control device is provided, comprising:

[0026] The acquisition module is used to acquire the bioelectrical signals of the object to be grasped by the grasping device based on the electrical signal detection component on the grasping device;

[0027] An adjustment module is used to adjust the torque of the grasping device on the grasped object based on the bioelectric signal of the grasped object.

[0028] According to a fourth aspect of this application, an electronic device is provided, including one or more processors and a memory, and a grasping device, wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform any of the grasping control methods provided in the embodiments of this application.

[0029] According to a fifth aspect of this application, a computer-readable storage medium is provided, including a computer program that, when run on a controller, causes the controller to perform any of the grasping control methods provided in the embodiments of this application.

[0030] According to a sixth aspect of this application, a computer program product is provided, including a computer program or instructions, which, when executed by a processor, implement any of the grasping control methods provided in the embodiments of this application.

[0031] In the grasping control method of this application embodiment, an electrical signal detection component is provided in the grasping device. The electrical signal detection component is used to collect the bioelectrical signal of the grasping object targeted by the grasping device. The bioelectrical signal is used to adjust the torque of the grasping device on the grasping object, thereby reducing the control delay of the clamp when the grasped object has a sudden situation, so as to improve the response speed to the sudden situation of the grasped object.

[0032] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] To gain a more complete understanding of this application and its beneficial effects, the following description will be provided in conjunction with the accompanying drawings, wherein the same reference numerals in the following description denote the same parts.

[0035] Figure 1 This is a flowchart illustrating a capture control method provided in an exemplary embodiment of this disclosure;

[0036] Figure 2 This is a schematic diagram of the grasping control system provided in an exemplary embodiment of this disclosure;

[0037] Figure 3 This is a schematic flowchart of a method for collecting and transmitting bioelectrical signals provided in an exemplary embodiment of this disclosure;

[0038] Figure 4 This is another flowchart illustrating the grabbing control method provided in an exemplary embodiment of this disclosure;

[0039] Figure 5 This is a schematic diagram of the gripping control device provided in an exemplary embodiment of this disclosure;

[0040] Figure 6 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0041] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0042] This application provides a grasping device, a grasping control method, an electronic device, a storage medium, and a product. The grasping control device can be integrated into electronic devices such as terminal devices and / or cloud servers. For example, the terminal device can be a vehicle control unit, an in-vehicle data processing system, an in-vehicle communication system, a drone controller (handle, etc.), a smartphone, tablet computer, laptop computer, desktop computer, smart speaker, smartwatch, or other devices installed in a vehicle.

[0043] Furthermore, in the embodiments of this application, "multiple" refers to two or more. The terms "first" and "second," etc., in the embodiments of this application are used for distinguishing descriptions and should not be construed as implying relative importance.

[0044] The following sections provide detailed descriptions of each example. It should be noted that the order in which the embodiments are described is not intended to limit the preferred order of the embodiments.

[0045] One embodiment of this application provides a grasping device, the grasping device including: an electrical signal detection component, the electrical signal detection component being used to collect bioelectrical signals of the grasping object targeted by the grasping device, the bioelectrical signals being used to adjust the torque of the grasping device on the grasping object.

[0046] The aforementioned gripping device is used to grip an object, and the gripping device includes, but is not limited to, a robotic hand, a robotic arm, a clamp, a suction cup, etc.

[0047] Accordingly, the object currently being grasped by the grasping device is taken as the grasping object.

[0048] The aforementioned bioelectric signal is a signal collected by detecting the grasped object through an electrical signal detection component. Information such as pressure, tangential force, and pH value of the grasped object can be determined through this bioelectric signal.

[0049] The torque mentioned above refers to the torque applied by the gripping device when gripping the object. By adjusting the torque, the force and method applied by the gripping device when gripping the object can be changed.

[0050] Optionally, the electrical signal detection component includes: a microelectrode array, the microelectrode array being arranged in a preset shape, and / or the microelectrode array being composed of a plurality of preset electrodes arranged at preset intervals, wherein the spatial resolution corresponding to the preset electrodes is greater than or equal to a preset threshold.

[0051] The preset shape can be a square, rectangle, or other rectangle, or a circle, triangle, or other irregular shape. The specific shape can be set according to the requirements and is not limited here.

[0052] The aforementioned preset electrodes include, but are not limited to, electrodes made of materials such as platinum microelectrodes and biodegradable magnesium alloy electrodes. Biocompatibility can be improved by using biodegradable magnesium alloy electrodes.

[0053] The aforementioned microelectrode array can be a two-dimensional microelectrode array, such as an integrated 256-channel electrochemical / biopotential sensor. Bioelectric signals are collected through this microelectrode array, so as to simultaneously obtain multimodal information such as force information, chemical information, and electrical information through the bioelectric signals.

[0054] For example, in a scenario where the microelectrode array integrates a 256-channel electrochemical sensor, the electrochemical sensor is a 16×16 gold microelectrode array fabricated on a flexible PI substrate using photolithography.

[0055] For example, in a scenario where the gripping device is a robotic arm, the microelectrode array can cover the curved surface of the robotic arm's fingertips to achieve charge mapping of the pressure distribution detected by the microelectrodes. Due to the shape of the fingertips and the shape of the object being gripped, different electrodes in the microelectrode array experience different pressures, thereby generating specific charges under specific pressure distributions.

[0056] Understandably, in order for the microelectrode array to accurately detect minute changes in force when a robotic arm grasps an object, the spatial resolution corresponding to the preset electrode must be greater than or equal to a preset threshold.

[0057] As can be seen from the above, by setting a microelectrode array in the gripping device, the bioelectric signals of the object to be gripped are collected through the microelectrode array. The bioelectric signals are used to adjust the torque of the gripping device on the object, thereby greatly improving the response speed of the gripper control. Based on the real-time acquisition of bioelectric signals, the torque on the object to be gripped is dynamically adjusted to achieve dynamic impedance matching, so that the contact impedance is stabilized within a certain range (e.g., 1-10kΩ). This solves the problem of high gripping failure rate caused by the lack of a closed-loop adjustment mechanism when the object is sliding. It also solves the problems of limited spatial resolution, difficulty in texture recognition, and signal drift when using piezoresistive sensors. When the object to be gripped experiences a sudden situation, the control delay of the gripper is reduced, thereby improving the response speed to the sudden situation of the object to be gripped and increasing the success rate of gripping the corresponding object.

[0058] For example, when grasping objects containing water (such as jellyfish), specific substances and biological information can be identified to greatly improve the accuracy of real-time detection of the grasped object. This reduces the control delay of the gripper when the grasped object slips or breaks, thereby improving the response speed to sudden situations of the grasped object.

[0059] It is understandable that by collecting the bioelectrical signals of the object being grasped, it is possible to simultaneously perceive the chemical signals of the object being grasped, such as tumor markers in medical settings or corrosive liquids in industrial settings. This bioelectrical signal can then be used to assist in material identification of the object being grasped, improving biocompatibility and enabling contact with human tissue in medical settings.

[0060] Optionally, the electrodes in the microelectrode array can be evaluated based on the bioelectric signal to obtain electrode evaluation results. These results can be used to indicate the degree of aging of the electrodes in the microelectrode array or whether the electrodes in the microelectrode array are aging, so that the stimulation current intensity can be automatically reduced based on the electrode evaluation results to extend the service life of the microelectrode array.

[0061] Optionally, the microelectrode array can be divided into multiple microelectrode subarrays, and the spatial resolution corresponding to the preset electrode in each microelectrode subarray is greater than or equal to a preset threshold. The bioelectric signal collected by the microelectrode array is composed of bioelectronic signals collected by multiple microelectrode subarrays.

[0062] Then, based on the bioelectronic signals collected by each microelectrode subarray, the electrodes in the corresponding microelectrode subarray can be evaluated to obtain the evaluation results of each electrode in the microelectrode subarray. If the evaluation results indicate that an electrode in the microelectrode subarray has failed, the sensing-control circuit, i.e. the circuit in the microelectrode subarray, can be redesigned in the software layer to divide the remaining non-failed electrodes in the microelectrode subarray into adjacent microelectrode subarrays. That is, the bioelectronic signals collected by the remaining non-failed electrodes in the microelectrode subarray are packaged together with the bioelectronic signals collected by the adjacent microelectrode subarrays to realize the routing of the signal to the backup electrode array and maintain the accuracy of material identification.

[0063] The device for redesigning the sensing-control circuitry, i.e., the circuitry in the microelectrode subarray, can be implemented by a device that performs the grasping control method according to the embodiments of this application.

[0064] Please see Figure 1 , Figure 1 This is a flowchart illustrating a capture control method according to an embodiment of this application. For ease of description, this embodiment uses a terminal device for illustration; that is, the above-described capture control method may include:

[0065] S101. Based on the electrical signal detection component on the grasping device, collect the bioelectrical signal of the grasping object targeted by the grasping device.

[0066] S102. Based on the bioelectric signal of the grasped object, adjust the torque of the grasping device on the grasped object.

[0067] Specifically, the terminal device can adjust the torque of the grasping device on the grasped object by driving the drive mechanism of the grasping device based on the bioelectric signal of the grasped object.

[0068] As can be seen from the above, by setting a microelectrode array in the gripping device, the bioelectric signals of the object to be gripped are collected through the microelectrode array. The bioelectric signals are used to adjust the torque of the gripping device on the object, thereby greatly improving the response speed of the gripper control. Based on the real-time acquisition of bioelectric signals, the torque on the object to be gripped is dynamically adjusted to achieve dynamic impedance matching, so that the contact impedance is stabilized within a certain range (e.g., 1-10kΩ). This solves the problem of high gripping failure rate caused by the lack of a closed-loop adjustment mechanism when the object is sliding. It also solves the problems of limited spatial resolution, difficulty in texture recognition, and signal drift when using piezoresistive sensors. When the object to be gripped experiences a sudden situation, the control delay of the gripper is reduced, thereby improving the response speed to the sudden situation of the object to be gripped and increasing the success rate of gripping the corresponding object.

[0069] Furthermore, the improved bioelectric signal transmission reduces feedback delay, significantly decreasing the error rate in fine-tuning operations.

[0070] It is understandable that this embodiment solves the limitation of current gripping devices that can only perform torque feedback gripping on some objects, and cannot detect information such as tangential force and temperature, which may lead to the risk of slippage or breakage.

[0071] Optionally, adjusting the torque of the grasping device on the grasped object based on the bioelectric signal of the grasped object includes: determining object control parameters for the grasped object based on the bioelectric signal of the grasped object; and controlling the grasping device based on the object control parameters to adjust the torque of the grasping device on the grasped object.

[0072] Optionally, adjusting the torque of the grasping device on the grasped object based on the bioelectric signal of the grasped object includes: determining object control parameters for the grasped object based on the bioelectric signal of the grasped object and environmental information of the grasped object; and controlling the grasping device based on the object control parameters to adjust the torque of the grasping device on the grasped object.

[0073] Among them, the environmental information of the object being crawled refers to the environmental information of the environment in which the object is located, including but not limited to temperature information, humidity information, etc.

[0074] Among them, the above-mentioned object control parameters are parameters used to control the gripping device to grip the object. Different object control parameters can control the torque of the gripping device on the object.

[0075] For example, the control parameters of the aforementioned object can be PID controller parameters. By determining these PID controller parameters, the PID controller can be used to control the drive mechanism to drive the robot to perform a response action. Here, PID is used to indicate proportional-integral-derivative control.

[0076] It is understandable that by collecting the characteristic values ​​of the signal, the control parameters of the object driven by the grasping device will be adjusted in a targeted manner to ensure the accuracy of adjusting the torque of the grasping device on the grasped object.

[0077] Specifically, the temperature information can be detected by a temperature sensor, and the humidity information can be detected by a humidity sensor. By combining bioelectrical signals with environmental information, multimodal fusion of data detected by multiple sensors can be achieved, thereby improving the accuracy of adjusting the torque of the grasping device on the grasped object. This avoids the current situation where the grasping device is not dynamic enough to grasp the object due to the use of only the normal pressure detected by the piezoresistive sensor to generate control parameters for the grasping device.

[0078] For example, such as Figure 2 As shown, in Figure 2 The gripping control system shown includes three sensors: sensor 201 is a PT100 temperature sensor, sensor 202 is an electrochemical sensor (16x16 microelectrode array), and sensor 203 is a SnO2 nanowire humidity sensor. The gripping control system also includes a main control chip 204 and a drive mechanism 205. The main control chip 204 is used to connect the sensors and the drive mechanism 205 to send the signals collected by the three sensors to the main control chip 204. The main control chip 204 can be used to execute the gripping control method described above to control the drive mechanism 205 to drive the gripping device to grip the object.

[0079] Among them, sensor 202 is mainly responsible for collecting the material characteristic signals of the grasped object, namely the aforementioned bioelectric signals, while sensors 201 and 202 are mainly responsible for collecting the environmental signals of the grasped object to obtain the corresponding environmental information, so as to perform certain feature matching by combining the environmental information with the bioelectric signals, thereby enhancing the accuracy of adjusting the torque of the grasping device on the grasped object.

[0080] Optionally, adjusting the torque of the grasping device on the grasped object based on the bioelectric signal of the grasped object includes: identifying the bioelectric signal through a preset neural network model, or identifying the bioelectric signal and environmental information of the grasped object, to determine at least one object tag of the grasped object; determining object control parameters for the grasped object based on the object tag; and controlling the grasping device based on the object control parameters to adjust the torque of the grasping device on the grasped object.

[0081] The aforementioned object label is used to indicate the result of the neural network model's recognition of the input signal or information. That is, the neural network model identifies which label the input signal belongs to and outputs that label. For example, if the temperature value in the temperature information belongs to the low temperature label, the neural network model outputs the low temperature label. Or, if the tangential force information in the bioelectric signal belongs to the sliding label, the neural network model outputs the sliding label to indicate that the grasped object is currently sliding.

[0082] For example, the neural network model described above can be a long short-term memory artificial neural network model (denoted as LSTM).

[0083] In this embodiment, a closed-loop control is constructed by dynamically adjusting the data collected from the grasping object using a neural network model, thereby improving the success rate of the grasping device in grasping the aforementioned object.

[0084] It is understandable that, during the neural network recognition process, by combining the bioelectrical signals and the environmental information of the grasped object, feature matching can be achieved between the two, thereby improving the accuracy of the neural network model in extracting information features.

[0085] It is understandable that a closed-loop control system can be constructed using a neural network model. This system dynamically adjusts the torque of the grasping device on the grasped object based on the collected bioelectric signals. In the tactile feedback and motion control of the grasping device, high-frequency disturbances can be compensated. The neural network model calculates bioelectric signals, i.e., the time-series signals generated by touch, to predict high-frequency disturbances in the grasping process of the grasping device. High-frequency disturbances are indicated by tags, thereby allowing the design of object control parameters for a PID controller. These object control parameters then drive the PID controller to adjust the fingertip torque of the grasping device in real time, thereby improving both the response speed and grasping stability of the grasping device.

[0086] Furthermore, by leveraging the nonlinear feature extraction capability of neural network models for bioelectrical signals, a dynamic model of the interaction between the grasping device and the environment can be constructed within a closed-loop control system. This enables the grasping device to effectively grasp objects in various situations within the grasping environment, thus resolving the current control failure issues in grasping organisms with varying curvature.

[0087] Specifically, the step of identifying the bioelectric signal or identifying the bioelectric signal and the environmental information of the grasped object using a preset neural network model to determine at least one object label of the grasped object includes: extracting features from the bioelectric signal using a preset neural network model, and identifying the grasped object based on the extracted features to determine at least one object label of the grasped object; or, extracting features from the bioelectric signal and the environmental information of the grasped object respectively using a preset neural network model, and identifying the grasped object based on the extracted features of the bioelectric signal and the features of the environmental information to determine at least one object label of the grasped object.

[0088] Optionally, determining the object control parameters for the crawled object based on the object tag includes: determining the target scene where the crawled object is located based on the object tag, and determining the object control parameters for the crawled object based on the target scene.

[0089] The system identifies whether the object being grasped is a living organism based on the object tag. The target scene is then determined based on this identification result. If the object being grasped is a living organism, the target scene is a living scene; if the object being grasped is not a living organism, the target scene is a non-living scene. By identifying the action scene grasped by the grasping device, adaptive decisions are made for different types of grasped objects and states in the scene. The algorithm logic of the main control chip makes adaptive decisions based on the collected signals and the object control parameters of the teammates, thereby improving the accuracy of the grasping device control.

[0090] In the case where the target scene is a biological scene, a first parameter value table corresponding to the biological scene is determined, and the object control parameters of the crawling object are determined from the first parameter value table based on the object tag.

[0091] Specifically, when the target scene is a non-biological scene, a second parameter value table corresponding to the non-biological scene is determined; based on the force information corresponding to the bioelectric signal, the object control parameters of the grasping object are determined from the second parameter value table.

[0092] The force information includes, but is not limited to, normal force and tangential force.

[0093] It is understandable that by distinguishing between biological and non-biological scenarios (such as industrial scenarios involving the handling of goods), and based on the algorithms corresponding to different scenarios, hardware that collects data in non-corresponding scenarios can be shut down, and / or the system can be prompted to optimize the algorithm control of the software layer. Through multi-level solutions corresponding to different scenarios, content access can be optimized while reducing hardware wear. By distinguishing different scenarios, it is easier to apply the tactile motion of the grasping device in more fields.

[0094] For example, in the case of a biological scene, changes in electrical signals such as electromyography and electroencephalography obtained from bioelectrical signals can be used to determine the control parameters of the object. In the case of a non-biological scene, since the cargo has no biological characteristics, an emergency evacuation can be triggered in a short time.

[0095] Specifically, determining the object control parameters of the crawled object from the first parameter value table based on the object tag may include: determining the value range or specific value result of the object control parameter corresponding to the object tag from the first parameter value table, so as to determine the corresponding object control parameter based on the value range or specific value result of the object control parameter.

[0096] Understandably, since no biological matter is involved in the non-biological scenario, the torque output control of the gripping device does not require frequent parameter adjustments to save data memory space. Furthermore, when controlling the gripping device, a slow triggering of disengagement can be performed first, i.e., the torque can be increased or decreased slowly to prevent sudden torque changes. Then, the torque for gripping the object can be configured to ensure correct gripping. Therefore, the parameters in the second parameter value table are set based on the characteristics of the non-biological scenario, which is different from the parameters in the first parameter value table.

[0097] Specifically, determining the object control parameters of the grasping object from the second parameter value table based on the force information corresponding to the bioelectric signal may include: determining the value range or specific value result of the object control parameters corresponding to the force information from the second parameter value table, so as to determine the corresponding object control parameters based on the value range or specific value result of the object control parameters.

[0098] Optionally, before identifying the bioelectric signal through a preset neural network model, or identifying the bioelectric signal and the environmental information of the grasped object to determine at least one object label of the grasped object, the method further includes: separating the signal characteristics of the bioelectric signal through a signal classification algorithm to obtain separated static and dynamic signals, so as to input the static and dynamic signals as processed bioelectric signals into the neural network model.

[0099] In this embodiment, static signals are used to indicate signals generated when the grasped object and the grasping device are relatively stationary, while dynamic signals are used to indicate signals generated when the grasped object slides or undergoes other dynamic changes during the grasping process. By distinguishing between static and dynamic signals, the neural network model can more accurately identify the static scene indicated by the static signal and the dynamic scene indicated by the dynamic signal, thus making the output results of the neural network model more accurate.

[0100] Optionally, before identifying the bioelectric signal through a preset neural network model, or identifying the bioelectric signal and the environmental information of the grasped object to determine at least one object label of the grasped object, the method further includes: performing at least one processing on the bioelectric signal.

[0101] The above-mentioned processing methods include, but are not limited to, at least one of the following: applying a safety threshold bias voltage to the micro-electrode array using a potentiostat, converting the bioelectric signal into a voltage signal using a signal amplifier (such as a transimpedance amplifier), filtering the bioelectric signal, and fusing the bioelectric signal with environmental information.

[0102] The processing order of different preprocessing methods can be set according to requirements and is not limited here.

[0103] Optionally, since there are multiple electrodes in the microelectrode array that collect bioelectric signals, the bioelectric signals of each channel are collected by polling through a multiplexer, and the humidity sensor module and the temperature sensor transmit environmental information synchronously through a bus.

[0104] Optionally, the bioelectric signal may be processed in at least one way, including: applying a safety threshold bias voltage to the micro-electrode array through a potentiostat; when the bioelectric signal is acquired through the micro-electrode array, converting the acquired bioelectric signal into a voltage signal through a transimpedance amplifier; and then polling the acquired signal using a multiplexer.

[0105] Furthermore, after the main control chip receives the bioelectric signals collected by the multiplexer, it first performs primary filtering to eliminate high-frequency noise, then extracts features, and separates the static pressure and dynamic tangential force components using a discrete wavelet transform algorithm to obtain static and dynamic signals. Then, the separated static signals, dynamic signals, and environmental information are fused into a multimodal signal, i.e., it is spliced ​​and packaged as a data sequence. Then, the Kalman filter algorithm is used to filter the multimodal signal to obtain filtered data. The filtered data is used as the processed input signal to be input into the neural network model. The processed input signal includes bioelectric signals and environmental information.

[0106] For example, such as Figure 3 As shown, the method for acquiring and transmitting bioelectrical signals includes the following steps S301-S304:

[0107] Step S301: The microelectrode array arranged on the grasping device contacts the grasping object and collects the initial bioelectrical signal as sensor input.

[0108] Step S302: Since the micro-electrode collects charge signals, it needs to be converted. That is, the initial bioelectric signal is converted into current through a high-impedance electrode, and the current signal is converted into voltage signal through the action of an integrator circuit. Then, it is amplified and noise suppressed to output a standard voltage signal, which is the processed bioelectric signal.

[0109] Step S303: The processed bioelectric signal is sampled at a high Hertz frequency using a 24ADC and input to the main control chip.

[0110] Step S304: The main control chip uses wavelet transform to separate the static and dynamic signals of the bioelectric signal, so as to input the static and dynamic signals into the neural network model, causing the neural network model to output the recognition result based on the static and dynamic signals.

[0111] Optionally, before identifying the bioelectric signal or identifying the bioelectric signal and the environmental information of the grasped object using a preset neural network model to determine at least one object label of the grasped object, the method further includes: performing electrochemical feature extraction on the bioelectric signal to obtain an extracted signal; performing feature recognition on the extracted signal to obtain a feature recognition result, so as to input the feature recognition result as a processed bioelectric signal into the neural network model; and / or performing biopotential decoding on the bioelectric signal to obtain decoding information; performing biological characteristic extraction on the decoding information to obtain target decoding information, so as to input the target decoding information as a processed bioelectric signal into the neural network model.

[0112] Specifically, before identifying the bioelectric signal or identifying the bioelectric signal and the environmental information of the grasped object through a preset neural network model to determine at least one object label of the grasped object, the method further includes: performing electrochemical feature extraction on the bioelectric signal to obtain the extracted signal; performing feature recognition on the extracted signal to obtain the feature recognition result, so as to input the feature recognition result as the processed bioelectric signal into the neural network model.

[0113] Specifically, before identifying the bioelectric signal or identifying the bioelectric signal and the environmental information of the grasped object through a preset neural network model to determine at least one object label of the grasped object, the method further includes: performing bioelectric potential decoding on the bioelectric signal to obtain decoding information; extracting biological characteristics from the decoding information to obtain target decoding information, so as to input the target decoding information as the processed bioelectric signal into the neural network model.

[0114] Specifically, before identifying the bioelectric signal or identifying the bioelectric signal and the environmental information of the grasped object using a preset neural network model to determine at least one object label of the grasped object, the method further includes: extracting electrochemical features from the bioelectric signal to obtain an extracted signal; performing feature recognition on the extracted signal to obtain a feature recognition result; performing biopotential decoding on the bioelectric signal to obtain decoding information; and extracting biological characteristics from the decoding information to obtain target decoding information, so as to input the feature recognition result and the target decoding information as processed bioelectric signals into the neural network model.

[0115] It should be noted that the above-mentioned electrochemical feature extraction of the bioelectric signal to obtain the extracted signal refers to the signal that separates the features of the chemical signal from the bioelectric signal, such as a signal that can reflect the pH value. Correspondingly, the above-mentioned feature recognition of the extracted signal to obtain the feature recognition result refers to the chemical signal feature recognition of the extracted signal with relevant chemical signal features, that is, the recognition of the pH value corresponding to the signal that can reflect the pH value, so as to reflect the component information of the target object through the recognized pH value.

[0116] It is understandable that by identifying the feature recognition results, the weights of each object label in the neural network model can be defined. That is, the weight of the corresponding object label will increase depending on which object the feature recognition result belongs to.

[0117] It should be noted that the above-mentioned bioelectric signal biopotential decoding to obtain decoded information refers to converting the bioelectric signal into identifiable biological attribute information, such as biological physiological information (ECG, EEG, etc.) or behavioral information. Correspondingly, the above-mentioned extraction of biological characteristics from the decoded information to obtain target decoded information refers to extracting characteristics from the converted biological attribute information, such as extracting peak information from ECG or EEG information.

[0118] In some embodiments, such as Figure 4 As shown, the above-mentioned grasping control method may further include the following steps S401-S413:

[0119] Step S401: Collect bioelectric signals through a microelectrode array to complete the input of the desired information.

[0120] In addition, in step S401, the system can also interact with the bioelectric signals collected by the micro-electro-array based on the current environment, such as temperature and humidity, that is, collect bioelectric signals and environmental information to identify the bioelectric signals of the grasped object in a specific environment.

[0121] Step S402: In the signal preprocessing stage, at least one type of processing may be performed on the bioelectric signal.

[0122] For example, a micro-electrode array applies a safety threshold bias voltage via a potentiostat. When acquiring bioelectrical signals through the micro-electrode array, the acquired bioelectrical signals are converted into voltage signals by a transimpedance amplifier, and then polled using a multiplexer. Finally, after the main control chip receives the bioelectrical signals polled by the multiplexer, it first performs primary filtering to eliminate high-frequency noise, then performs feature extraction, and separates the static pressure and dynamic tangential force components using a discrete wavelet transform algorithm to obtain static and dynamic signals. Then, the separated static signals, dynamic signals, and environmental information are fused into a multimodal signal, i.e., it is spliced ​​and packaged as a data sequence. Then, the Kalman filter algorithm is used to filter the multimodal signal to obtain filtered data, and subsequent processes are executed based on the filtered data.

[0123] Step S403: Extract the electrochemical features of the bioelectric signals in the processed signals to obtain the extracted signals.

[0124] Step S404: Perform feature recognition on the extracted signal to obtain the feature recognition result.

[0125] Step S405: Perform biopotential decoding on the bioelectric signal to obtain decoded information.

[0126] Step S406: Extract biological characteristics from the decoded information to obtain target decoded information, so as to realize the expected feedback of dynamic force indicated by the target decoded information, such as the muscle dynamic feedback of the target organism.

[0127] Step S407: Using a neural network model, based on the feature recognition results, target decoding information, and other information in the processed signal, the corresponding object label is output.

[0128] Specifically, the neural network model implements the function of a "dynamic differential equation solver" through a gating mechanism, transforming the differential constraints of the electrochemical system into a data-driven implicit mapping, thereby determining the object label corresponding to the input information under specific constraints.

[0129] Step S408: Make adaptive decisions based on object labels, i.e., determine whether it is a biological scene. If yes, proceed to step S409; otherwise, proceed to step S410.

[0130] Step S409: Based on the object tag, determine the object control parameters of the object to be captured from the first parameter value table corresponding to the biological scene.

[0131] Step S410: Based on the force information corresponding to the bioelectric signal, determine the object control parameters of the grasping object from the second parameter value table corresponding to the non-biological scene.

[0132] Step S411: Input the object control parameters into the PID controller, so that the PID controller can control the actuator to drive the motor of the gripping device to work based on the object control parameters.

[0133] Step S412: Control the gripping device to perform actions, that is, control the robotic arm to grip the aforementioned object.

[0134] Step S413: After the grasping device performs its action, it interacts with the grasped object in the environment to generate environmental interaction data. That is, it can receive the signal collected by the microelectrode array after the action to achieve closed loop.

[0135] To facilitate better implementation of the grasping control method provided in the embodiments of this application, the embodiments of this application also provide an apparatus based on the above-described grasping control method. The meanings of the terms used are the same as in the grasping control method described above, and specific implementation details can be found in the descriptions in the method embodiments.

[0136] For example, such as Figure 5 As shown, the grasping control device may include:

[0137] The acquisition module 501 is used to acquire the bioelectrical signal of the object to be grasped by the grasping device based on the electrical signal detection component on the grasping device;

[0138] The adjustment module 502 is used to adjust the torque of the grasping device on the grasped object based on the bioelectric signal of the grasped object.

[0139] Optionally, the adjustment module 502 is specifically used for:

[0140] The bioelectric signal is identified by a preset neural network model, or the bioelectric signal and the environmental information of the grasped object are identified to determine at least one object label of the grasped object;

[0141] Based on the object tag, determine the object control parameters for the crawled object;

[0142] The grasping device is controlled based on the object control parameters to adjust the torque of the grasping device on the grasped object.

[0143] Optionally, the adjustment module 502 is specifically used for:

[0144] The target scene where the crawled object is located is determined based on the object label, and the object control parameters for the crawled object are determined based on the target scene;

[0145] In the case where the target scene is a biological scene, a first parameter value table corresponding to the biological scene is determined, and the object control parameters of the crawling object are determined from the first parameter value table based on the object tag.

[0146] Specifically, when the target scene is a non-biological scene, a second parameter value table corresponding to the non-biological scene is determined; based on the force information corresponding to the bioelectric signal, the object control parameters of the grasping object are determined from the second parameter value table.

[0147] Optionally, the grasping control device further includes a signal separation module, which is specifically used for:

[0148] The bioelectric signal is separated by a signal classification algorithm to obtain a static signal and a dynamic signal. The static signal and the dynamic signal are then used as the processed bioelectric signal and input into the neural network model.

[0149] Optionally, the grasping control device further includes a signal processing module, which is specifically used for:

[0150] Electrochemical feature extraction is performed on the bioelectric signal to obtain the extracted signal;

[0151] The extracted signal is subjected to feature recognition to obtain feature recognition results, which are then used as processed bioelectrical signals input into the neural network model; and / or

[0152] The bioelectric signal is subjected to biopotential decoding to obtain decoded information;

[0153] Biological characteristics are extracted from the decoded information to obtain target decoded information, which is then input into the neural network model as a processed bioelectrical signal.

[0154] The grasping control device proposed in this application includes an electrical signal detection component. This component is used to collect the bioelectrical signals of the object being grasped by the grasping device. The bioelectrical signals are used to adjust the torque of the grasping device on the object, thereby reducing the control delay of the gripper when the object being grasped experiences a sudden situation, and thus improving the response speed to sudden situations of the object being grasped.

[0155] In practice, each of the above modules can be implemented as an independent entity or can be combined arbitrarily to be implemented as the same or several entities. For the specific implementation methods and corresponding beneficial effects of each of the above modules, please refer to the previous method embodiments, which will not be repeated here.

[0156] This application also provides an electronic device, such as... Figure 6 As shown, it illustrates a structural schematic diagram of the electronic device involved in the embodiments of this application, specifically:

[0157] The electronic device may include components such as a processor 601 with one or more processing cores, a memory 602 with one or more computer-readable storage media, a power supply 603, and an input unit 604. Those skilled in the art will understand that... Figure 6 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein:

[0158] The processor 601 is the control center of the electronic device, connecting various parts of the device via various interfaces and lines. It executes computer programs and / or modules stored in the memory 602, and calls data stored in the memory 602, to perform various functions and process data. Optionally, the processor 601 may include one or more processing cores; preferably, the processor 601 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 601.

[0159] The memory 602 can be used to store computer programs and modules. The processor 601 executes various functional applications and data processing by running the computer programs and modules stored in the memory 602. The memory 602 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, computer programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 602 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 602 may also include a memory controller to provide the processor 601 with access to the memory 602.

[0160] The electronic device also includes a power supply 603 that supplies power to the various components. Preferably, the power supply 603 can be logically connected to the processor 601 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 603 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0161] The electronic device may also include an input unit 604, which can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0162] Although not shown, the electronic device may also include a display unit, a grasping device, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 601 in the electronic device loads the executable files corresponding to the processes of one or more computer programs into the memory 602 according to the following instructions, and the processor 601 runs the computer programs stored in the memory 602 to realize various functions, such as:

[0163] Based on the electrical signal detection component on the grasping device, the bioelectrical signal of the grasping object targeted by the grasping device is collected;

[0164] Based on the bioelectric signal of the object being grasped, the torque of the grasping device on the object being grasped is adjusted.

[0165] For details on the specific implementation methods and corresponding beneficial effects of each of the above operations, please refer to the detailed description of the grasping control method above, which will not be repeated here.

[0166] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by a computer program, or by a computer program controlling related hardware. The computer program can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0167] Therefore, embodiments of this application provide a computer-readable storage medium storing a computer program that can be loaded by a processor to execute steps in any of the grasping control methods provided in embodiments of this application. For example, the computer program can execute the following steps:

[0168] Based on the electrical signal detection component on the grasping device, the bioelectrical signal of the grasping object targeted by the grasping device is collected;

[0169] Based on the bioelectric signal of the object being grasped, the torque of the grasping device on the object being grasped is adjusted.

[0170] For details on the specific implementation methods and corresponding beneficial effects of the above operations, please refer to the previous embodiments, which will not be repeated here.

[0171] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0172] Since the computer program stored in the computer-readable storage medium can execute the steps of any of the grasping control methods provided in the embodiments of this application, the beneficial effects that any of the grasping control methods provided in the embodiments of this application can achieve can be realized, as detailed in the preceding embodiments, and will not be repeated here.

[0173] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the aforementioned grasping control method.

[0174] The foregoing has provided a detailed description of a gripping device, gripping control method, electronic device, storage medium, and product provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A gripping device, characterized in that, The grasping device includes an electrical signal detection component, which is used to collect the bioelectrical signal of the grasping object targeted by the grasping device, and the bioelectrical signal is used to adjust the torque of the grasping device on the grasping object.

2. The gripping device according to claim 1, characterized in that, The electrical signal detection component includes: a microelectrode array, the microelectrode array being arranged in a preset shape, and / or the microelectrode array being composed of multiple preset electrodes arranged at preset intervals, the spatial resolution corresponding to the preset electrodes being greater than or equal to a preset threshold.

3. A grasping control method, characterized in that, The method includes: Based on the electrical signal detection component on the grasping device, the bioelectrical signal of the grasping object targeted by the grasping device is collected; Based on the bioelectric signal of the object being grasped, the torque of the grasping device on the object being grasped is adjusted.

4. The grasping control method according to claim 3, characterized in that, Adjusting the torque of the grasping device on the grasped object based on the bioelectrical signal of the grasped object includes: The bioelectric signal is identified by a preset neural network model, or the bioelectric signal and the environmental information of the grasped object are identified to determine at least one object label of the grasped object; Based on the object tag, determine the object control parameters for the crawled object; The grasping device is controlled based on the object control parameters to adjust the torque of the grasping device on the grasped object.

5. The grasping control method according to claim 4, characterized in that, The step of determining the object control parameters for the crawled object based on the object tag includes: The target scene where the crawled object is located is determined based on the object label, and the object control parameters for the crawled object are determined based on the target scene; In the case where the target scene is a biological scene, a first parameter value table corresponding to the biological scene is determined, and the object control parameters of the crawling object are determined from the first parameter value table based on the object tag. Specifically, when the target scene is a non-biological scene, a second parameter value table corresponding to the non-biological scene is determined; based on the force information corresponding to the bioelectric signal, the object control parameters of the grasping object are determined from the second parameter value table.

6. The grasping control method according to claim 4, characterized in that, Before identifying the bioelectric signal using a preset neural network model, or identifying the bioelectric signal and environmental information of the grasped object, and determining at least one object label for the grasped object, the method further includes: The bioelectric signal is separated by a signal classification algorithm to obtain a static signal and a dynamic signal. The static signal and the dynamic signal are then used as the processed bioelectric signal and input into the neural network model.

7. The grasping control method according to claim 4, characterized in that, Before identifying the bioelectric signal using a preset neural network model, or identifying the bioelectric signal and environmental information of the grasped object, and determining at least one object label for the grasped object, the method further includes: Electrochemical feature extraction is performed on the bioelectric signal to obtain the extracted signal; The extracted signal is subjected to feature recognition to obtain feature recognition results, which are then used as processed bioelectrical signals input into the neural network model; and / or The bioelectric signal is subjected to biopotential decoding to obtain decoded information; Biological characteristics are extracted from the decoded information to obtain target decoded information, which is then input into the neural network model as a processed bioelectrical signal.

8. An electronic device, characterized in that, It includes one or more processors and a memory, and a gripping device as described in claim 1 or 2, wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of the gripping control method as described in any one of claims 3 to 7.

9. A storage medium, characterized in that, Includes a computer program, which, when run on a controller, causes the controller to perform the steps of the grasping control method according to any one of claims 3 to 7.

10. A computer program product, characterized in that, Includes a computer program or instructions that, when executed by a processor, implement the steps of the grasping control method according to any one of claims 3 to 7.

Images