A control method and device of a robot hand and the robot hand

Abstract

This invention discloses a control method, device, and robotic hand for a robotic arm. The method includes: acquiring electromyographic (EMG) signals from the surface of a user's forearm when the robotic hand grasps a target object; inputting the EMG signals into a motion intention recognition model to obtain the current motion intention; and controlling the robotic hand to adjust the grasping force or release the target object based on the current motion intention. This invention, by determining whether to adjust the grasping force or release the target object based on the EMG signals from the user's forearm surface when the robotic hand grasps the target object, and based on the user's subjective force adjustment mechanism, can adjust the grasping force in real time according to the characteristics of the target object, thereby improving the stability, safety, and adaptability of the grasping process.

Description

Technical Field

[0001] This application relates to the field of robotic arm technology, and in particular to a control method, device and robotic arm for a robotic arm. Background Technology

[0002] With the continuous development of bionic robot technology, robotic hands, as key execution components in human-computer interaction and assisted movement, have been widely used in fields such as human-computer collaboration, rehabilitation assistance, and intelligent prostheses.

[0003] Existing robotic arms generally employ a constant grip force control strategy. When grasping objects of different weights, shapes, or hardness, they lack a user-defined force adjustment mechanism and cannot adjust the grip force in real time according to the object's characteristics. This constant grip force control strategy leads to a mismatch between the gripping force and actual needs when facing different objects, easily resulting in slippage due to excessive looseness or damage due to excessive tightness. This reduces the stability, safety, and adaptability of the gripping process.

[0004] Therefore, there is still an urgent need for a control method for robotic arms that can improve the stability, safety, and adaptability of grasping. Summary of the Invention

[0005] The main purpose of this application is to propose a control method, device and manipulator for a robotic arm, which aims to solve the problems that the robotic arm may slip or be damaged when gripping, either due to gripping too loosely or gripping too tightly.

[0006] To achieve the above objectives, this application proposes a control method for a robotic arm, the control method comprising: When the robotic arm grasps the target object, electromyographic signals are collected from the surface of the user's forearm. The electromyographic signal is input into the motion intention recognition model to obtain the current motion intention; Based on the current motion intention, the robotic arm is controlled to adjust the force of gripping the target object or release the target object.

[0007] In some embodiments, inputting the electromyographic signal into the motion intent recognition model to obtain the current motion intent includes: The electromyographic signal is input into the motion intention recognition model; The current grasping intention or current releasing intention is obtained by the motion intention recognition model from the electromyographic signals. The current grasping intent or the current releasing intent is taken as the current motion intent.

[0008] In some embodiments, controlling the robotic arm to adjust the force of grasping the target object according to the current motion intention includes: When the current motion intention is the current grasping intention, the current grasping intention is analyzed to obtain the current grasping gesture and the current grasping force; Disable the current grasping gesture and calculate the difference between the current grasping force and the initial grasping force to obtain the grasping force difference; Based on the initial grasping gesture and the difference in grasping force, the robotic arm is controlled to continue grasping the target object, so that the force of the robotic arm grasping the target object is adjusted from the initial grasping force to the current grasping force.

[0009] In some embodiments, after controlling the robotic arm to continue grasping the target object based on the initial grasping gesture and the difference in grasping force, the method further includes: The current vibration parameters are obtained by querying the preset force-vibration relationship table based on the current gripping force. The vibration module of the robotic arm is controlled to output vibration based on the current vibration parameters.

[0010] In some embodiments, controlling the robotic arm to release the target object according to the current motion intention includes: When the current motion intention is the current release intention, the robotic arm is controlled to gradually reduce the gripping force, and the tactile sensor of the robotic arm is controlled to collect the real-time normal force in real time. Detect whether the real-time normal force is zero; When the real-time normal force is zero, it is determined that the robotic arm has released the target object.

[0011] In some embodiments, before acquiring electromyographic signals from the user's forearm surface when the robotic arm grasps the target object, the method further includes: The tactile sensors of the robotic arm are controlled to collect real-time normal force. Determine whether the real-time normal force is greater than the preset normal force; When the real-time normal force is greater than the preset normal force, it is determined that the robotic arm has made contact with the target object.

[0012] In some embodiments, after determining that the robotic arm has made contact with the target object, the method further includes: Establish a normal force-acquisition time table based on the real-time normal force and the acquisition time of the real-time normal force; The response characteristics are obtained by analyzing the normal force-acquisition time table, and the response characteristics are input into a preset hardness analysis model to obtain the hardness data of the target object; The initial gripping gesture and initial gripping force are determined based on the hardness data; The robotic arm is controlled to grasp the target object based on the initial grasping gesture and the initial grasping force.

[0013] This application further proposes a control device for a robotic arm, the control device comprising: The acquisition unit is used to acquire electromyographic signals from the surface of the user's forearm when the robotic arm grasps the target object; The recognition unit is used to input the electromyographic signal into the motion intention recognition model to obtain the current motion intention; The control unit is used to control the robotic arm to adjust the force of grasping the target object or to release the target object according to the current motion intention.

[0014] In some embodiments, the identification unit is specifically used for: The electromyographic signal is input into the motion intention recognition model; The current grasping intention or current releasing intention is obtained by the motion intention recognition model from the electromyographic signals. The current grasping intent or the current releasing intent is taken as the current motion intent.

[0015] This application further proposes a robotic arm, the robotic arm comprising: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that are executed by the at least one processor to enable the at least one processor to perform the control method for the robotic arm described above.

[0016] This application's technical solution involves collecting electromyographic (EMG) signals from the user's forearm surface when the robotic arm grasps a target object; inputting the EMG signals into a motion intention recognition model to obtain the current motion intention; and controlling the robotic arm to adjust the grasping force or release the target object based on the current motion intention. By determining whether to adjust the grasping force or release the target object based on the EMG signals from the user's forearm surface when the robotic arm grasps the target object, and based on the user's subjective force adjustment mechanism, the grasping force can be adjusted in real time according to the characteristics of the target object, thereby improving the stability, safety, and adaptability of the grasping process. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an embodiment of the control method for the robotic arm of this application; Figure 2 This is a flowchart illustrating another embodiment of the control method for the robotic arm in this application; Figure 3 This is a flowchart illustrating another embodiment of the control method for the robotic arm in this application; Figure 4This is a flowchart illustrating another embodiment of the control method for the robotic arm in this application; Figure 5 This is a flowchart illustrating another embodiment of the control method for the robotic arm in this application; Figure 6 This is a flowchart illustrating another embodiment of the control method for the robotic arm in this application; Figure 7 This is a flowchart illustrating another embodiment of the control method for the robotic arm in this application; Figure 8 This is a schematic diagram of the control device for the robotic arm according to an embodiment of this application; Figure 9 This is a schematic diagram of the structure of the robotic arm involved in the embodiments of this application. Detailed Implementation

[0018] The solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments in this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0019] It should be noted that all directional indicators (such as up, down, left, right, front, back, etc.) in the embodiments of this application are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicator will also change accordingly.

[0020] It should also be noted that when a component is described as "fixed to" or "set on" another component, it can be directly on the other component or there may be an intervening component present. When a component is described as "connected to" another component, it can be directly connected to the other component or there may be an intervening component present.

[0021] Furthermore, the use of terms such as "first" and "second" in this application is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed in this application.

[0022] This application proposes a control method for a robotic arm, referring to... Figure 1 and Figure 8 , Figure 1 This is a flowchart illustrating an embodiment of the control method for the robotic arm of this application. Figure 8 This is a schematic diagram of the control device for the robotic arm according to an embodiment of the present application. In some embodiments, the control method of the robotic arm includes: Step S110: When the robotic arm grasps the target object, collect electromyographic signals from the surface of the user's forearm. Step S120: Input the electromyographic signal into the motion intention recognition model to obtain the current motion intention; Step S130: Control the robotic arm to adjust the force of gripping the target object or release the target object according to the current motion intention.

[0023] In this embodiment, as Figure 1 and Figure 8 As shown, the control method for the robotic arm can be applied to the control device of the robotic arm. The control device can be configured on the robotic arm, thereby controlling its movements. In this embodiment, the execution entity of the method steps is the control device of the robotic arm, hereinafter referred to as the control device.

[0024] It is understood that the robotic hand may also include an electromyography (EMG) signal acquisition module, tactile sensors, multiple fingers, and multiple drive motors, with at least one drive motor configured for each finger to drive finger movement. Multiple tactile sensors may also be configured on each finger. In this embodiment, the robotic hand can be worn on the user's forearm.

[0025] When a user wears the robotic arm on their forearm, they can control it to grasp a target object. While the robotic arm is grasping the target object, a control device inside the arm can control the electromyography (EMG) signal acquisition module to collect EMG signals from the surface of the user's forearm.

[0026] The control device can be pre-configured with a motion intention recognition model. This model can be pre-trained by the developer and configured onto the control device. For example, the developer can acquire a large amount of electromyographic (EMG) signals when the user makes a motion intention, associate each motion intention with its corresponding EMG signal, and then train the motion intention recognition model based on these motion intentions and their corresponding EMG signals.

[0027] After acquiring electromyographic (EMG) signals, the control device inputs these signals into the motion intention recognition model. The model then analyzes and identifies the EMG signals to determine the current motion intention, which is then output. At this point, the control device can determine the current motion intention.

[0028] After receiving the current motion intention, the control device can control the robotic arm to adjust the gripping force or release the target object accordingly. For example, if the current motion intention is to adjust the gripping force, the control device can adjust the gripping force accordingly. If the current motion intention is to release the target object, the control device can release the target object accordingly.

[0029] The acquisition of electromyographic signals can be continuous, allowing the control device to continuously control the robotic arm based on the user's electromyographic signals. This enables the user to adjust the gripping force in real time according to the characteristics of the object.

[0030] This application's technical solution involves collecting electromyographic (EMG) signals from the user's forearm surface when the robotic arm grasps a target object; inputting the EMG signals into a motion intention recognition model to obtain the current motion intention; and controlling the robotic arm to adjust the grasping force or release the target object based on the current motion intention. By determining whether to adjust the grasping force or release the target object based on the EMG signals from the user's forearm surface when the robotic arm grasps the target object, and based on the user's subjective force adjustment mechanism, the grasping force can be adjusted in real time according to the characteristics of the target object, thereby improving the stability, safety, and adaptability of the grasping process.

[0031] Reference Figure 2 , Figure 2 This is a flowchart illustrating another embodiment of the control method for the robotic arm of this application. In some embodiments, the aforementioned input of electromyographic signals into a motion intention recognition model to obtain the current motion intention includes: Step S140: Input electromyographic signals into the motion intention recognition model; Step S141: Obtain the current grasping intention or current releasing intention obtained by the motion intention recognition model from the electromyographic signals; Step S142: Take the current grasping intention or the current releasing intention as the current motion intention.

[0032] In this embodiment, as Figure 2 As shown, when the control device executes step S120, it can first input electromyographic (EMG) signals into the motion intention recognition model. The control device can be pre-configured with the motion intention recognition model. This model can be pre-trained and configured on the control device by the developer. For example, the developer can acquire a large number of EMG signals from the user when making motion intentions, associate each motion intention with its corresponding EMG signal, and then train the motion intention recognition model based on these motion intentions and their corresponding EMG signals. The motion intention recognition model can be a binary classification model, specifically designed to distinguish between "grasping" and "releasing" intentions.

[0033] The control device can input electromyography (EMG) signals into the motion intention recognition model. After acquiring the EMG signals, the motion intention recognition model can analyze and identify them to determine the current grasping or releasing intention; finally, it outputs the current grasping or releasing intention. At this point, the control device can obtain the current grasping or releasing intention obtained by the motion intention recognition model from the EMG signals.

[0034] After obtaining the current grasping intention or the current releasing intention, the control device can also use the current grasping intention or the current releasing intention as the current motion intention.

[0035] Reference Figure 3 , Figure 3 This is a flowchart illustrating another embodiment of the control method for the robotic arm of this application. In some embodiments, the aforementioned control of the robotic arm to adjust the force of grasping the target object according to the current motion intention includes: Step S150: When the current motion intention is the current grasping intention, analyze the current grasping intention to obtain the current grasping gesture and the current grasping force; Step S151: Disable the current grasping gesture and calculate the difference between the current grasping force and the initial grasping force to obtain the grasping force difference; Step S152: Based on the initial grasping gesture and the difference in grasping force, control the robotic arm to continue grasping the target object, so that the force of the robotic arm grasping the target object is adjusted from the initial grasping force to the current grasping force.

[0036] In this embodiment, as Figure 3 As shown, when the control device controls the robotic arm to adjust the gripping force of the target object according to the current motion intention in step S130, it can analyze the current gripping intention. In step S110, when the robotic arm grips the target object, it can grip the target object with an initial gripping gesture and an initial gripping force. When the current motion intention is the current gripping intention, the control device can analyze the current gripping intention to determine the gripping gesture and gripping force expressed by the current gripping intention, thus obtaining the current gripping gesture and current gripping force. Here, the current gripping gesture refers to the gripping gesture expressed by the user's current intention; the current gripping force refers to the gripping force expressed by the user's current intention.

[0037] After obtaining the current grasping gesture and the current grasping force, the control device can also disable the current grasping gesture to prevent sudden changes in posture due to misrecognition of the gesture during the grasping process. For example, the initial grasping gesture is a two-finger grasping gesture, and the current grasping gesture is a three-finger grasping gesture; that is, when the robotic arm is performing a two-finger grasping gesture, even if a three-finger grasping gesture is recognized, it will not switch to a three-finger grasping gesture.

[0038] After disabling the current gripping gesture, the control device can calculate the difference between the current gripping force and the initial gripping force, thus obtaining the gripping force difference. For example: Gripping force difference = Current gripping force - Initial gripping force. When the gripping force difference is greater than zero, it indicates that the user expects to increase the gripping force; the larger the difference, the more significant the need for increased gripping force. When the gripping force difference is less than zero, it indicates that the user expects to decrease the gripping force to avoid damaging the target object due to excessive gripping force. When the gripping force difference is equal to zero, it indicates that the user expects to maintain the same gripping force, requiring no adjustment.

[0039] The control device can also control the robotic arm to continue grasping the target object based on the initial grasping gesture and the difference in grasping force, thereby adjusting the grasping force of the robotic arm from the initial grasping force to the current grasping force. That is, the control device continues to control the robotic arm to grasp the target object based on the initial grasping gesture, while simultaneously adjusting the grasping force of the robotic arm based on the difference in grasping force, thus adjusting the grasping force of the robotic arm to the current grasping force.

[0040] Reference Figure 4 , Figure 4 This is a flowchart illustrating another embodiment of the control method for the robotic arm of this application. In some embodiments, after controlling the robotic arm to continue grasping the target object based on the initial grasping gesture and the difference in grasping force, the method further includes: Step S160: Query the preset force-vibration relationship table based on the current gripping force to obtain the current vibration parameters; Step S161: Control the vibration module of the robot arm to output vibration according to the current vibration parameters.

[0041] In this embodiment, as Figure 4 As shown, after executing step S152, the control device can also output vibration. The robotic arm may also include a vibration module. The control device can be pre-configured with a preset force-vibration relationship table, which can be configured by the developer according to actual needs. For example, when the current gripping force is light, the vibration is slight; when the current gripping force is heavy, the vibration is stronger.

[0042] After controlling the robotic arm to continue grasping the target object based on the initial grasping gesture and the difference in grasping force, the control device can also obtain the current vibration parameters by consulting a preset force-vibration relationship table based on the current grasping force. For example, if the current grasping force is light, the control device can obtain a slight vibration by consulting the preset force-vibration relationship table, and then generate the current vibration parameters corresponding to the slight vibration for the vibration module. Similarly, if the current grasping force is heavy, the control device can obtain a stronger vibration by consulting the preset force-vibration relationship table, and then generate the current vibration parameters corresponding to the stronger vibration for the vibration module.

[0043] After obtaining the current vibration parameters, the control device can control the vibration module of the robotic arm to output vibrations based on these parameters. For example, if the current vibration parameters correspond to slight vibration, the control device can control the robotic arm's vibration module to output slight vibrations to indicate to the user that a light-force grasp is being applied. Similarly, if the current vibration parameters correspond to stronger vibrations, the control device can control the robotic arm's vibration module to output stronger vibrations to indicate to the user that a heavy-force grasp is being applied. This allows the control device to provide real-time vibration feedback to the user when the robotic arm switches between different forces. Through this vibration feedback mechanism, the user can promptly and intuitively perceive their own force control status, thereby achieving precise adjustment of the grasping force.

[0044] Reference Figure 5 , Figure 5 This is a flowchart illustrating another embodiment of the control method for the robotic arm of this application. In some embodiments, the aforementioned control of the robotic arm to release the target object according to the current motion intention includes: Step S170: When the current motion intention is the current release intention, control the robotic arm to gradually reduce the gripping force, and control the tactile sensor of the robotic arm to collect the real-time normal force in real time. Step S171: Detect whether the real-time normal force is zero; Step S172: When the real-time normal force is zero, determine that the robot arm has released the target object.

[0045] In this embodiment, as Figure 5As shown, when the control device executes step S130, which involves controlling the robotic arm to release the target object based on the current motion intention, it can control the robotic arm to gradually reduce its gripping force. The robotic arm may also include a tactile sensor. When the current motion intention is to release, the control device can control the robotic arm to gradually reduce its gripping force. Simultaneously, the control device also controls the tactile sensor of the robotic arm to collect real-time normal force. For example, the control device can be pre-configured with a force reduction gradient table, which includes the correspondence between each force reduction speed and each hardness data and weight. For instance, the force reduction speed is positively correlated with the hardness data and weight of the currently gripped target object. For hard, heavy objects, the force reduction speed is slower to prevent the object from slipping or being damaged due to its large inertia and rapid force reduction. For soft, light objects, the force reduction speed can be appropriately increased, as soft objects have good deformation buffering properties and do not require excessively long force reduction times. The control device can first acquire the hardness data and weight of the target object, then look up the force reduction gradient table based on the hardness data and weight to obtain the corresponding force reduction speed, and then control the robot arm to gradually reduce the gripping force based on the force reduction speed. While gradually reducing the gripping force, the control device also controls the robot arm's tactile sensors to collect real-time normal force. The control device can determine whether the actual force reduction speed of the robot arm is consistent with the force reduction speed from the force reduction gradient table based on the real-time normal force. If they are inconsistent, the actual force reduction speed of the robot arm can be dynamically adjusted.

[0046] The control device can also detect whether the real-time normal force is zero. When the real-time normal force is zero, the control device can determine that the robotic arm has released the target object. In this embodiment, during the release phase, the control device can gradually reduce the output gripping force based on the feedback signal from the tactile sensor until the real-time normal force is detected to have dropped to zero, confirming that the target object has been safely placed and completely removed from contact, ensuring a smooth and safe release action. At this point, after the release action is completed, the control device can also automatically clear and reset all tactile state quantities and control parameters, preparing for the recognition and execution of the next gripping task.

[0047] Reference Figure 6 , Figure 6 This is a flowchart illustrating another embodiment of the control method for the robotic arm of this application. In some embodiments, before collecting electromyographic signals from the user's forearm surface when the robotic arm grasps the target object, the method further includes: Step S180: Control the tactile sensors of the robotic arm to collect real-time normal force. Step S181: Determine whether the real-time normal force is greater than the preset normal force; Step S182: When the real-time normal force is greater than the preset normal force, it is determined that the robot arm has made contact with the target object.

[0048] In this embodiment, as Figure 6As shown, before executing step S110, the control device can first control the tactile sensor of the robotic arm to collect the real-time normal force. When the robotic arm is about to grasp the target object, the control device can first make contact between the robotic arm and the target object. The control device can control the tactile sensor of the robotic arm to collect the real-time normal force. After collecting the real-time normal force, the control device can determine whether the real-time normal force is greater than the preset normal force. When the real-time normal force is greater than the preset normal force, the control device can determine that the robotic arm has made contact with the target object. In this embodiment, the contact time between the robotic arm and the target object is accurately determined through the normal force feedback of the tactile sensor, providing a key triggering condition for subsequently determining the hardness data of the target object and avoiding misoperation in the non-contact state.

[0049] Reference Figure 7 , Figure 7 This is a flowchart illustrating another embodiment of the control method for the robotic arm of this application. In some embodiments, after determining that the robotic arm has made contact with the target object, the method further includes: Step S190: Establish a normal force-acquisition time table based on the real-time normal force and the real-time normal force acquisition time. Step S191: Analyze the normal force-acquisition time table to obtain the response characteristics, and input the response characteristics into the preset hardness analysis model to obtain the hardness data of the target object; Step S192: Determine the initial gripping gesture and initial gripping force based on the hardness data; Step S193: Control the robotic arm to grasp the target object based on the initial grasping gesture and initial grasping force.

[0050] In this embodiment, as Figure 7 As shown, after executing step S182, the control device can first establish a normal force-acquisition timetable. After the robotic arm contacts the target object, the control device will not directly control the robotic arm to grasp the target object; instead, it will first establish a normal force-acquisition timetable based on the real-time normal force and the real-time normal force acquisition time. For example, the control device can first establish a two-dimensional data table, where the rows of the two-dimensional data table represent the real-time normal force acquisition time, and the columns of the two-dimensional data table represent the real-time normal force. Then, based on the acquisition time of each real-time normal force, the real-time normal force is filled into the two-dimensional data table, thereby obtaining the normal force-acquisition timetable.

[0051] The control device can also obtain response characteristics by analyzing the normal force-acquisition time table. For example, by analyzing the normal force-acquisition time table, the control device can extract the normal force rise rate, peak normal force, and normal force stability, thus using these three parameters as response characteristics. Specifically, the normal force rise rate characterizes how quickly the normal force rises from its initial value to its first peak after the robot arm contacts the target object, reflecting the initial stiffness of the target object against deformation—the normal force rises quickly for hard objects and slowly for soft objects. The peak normal force characterizes the maximum value of the normal force within a preset time window (e.g., 500ms) and the corresponding time point when the robot arm contacts the target object, reflecting the upper limit of the initial contact force that the target object can withstand during the contact phase, and can also help determine the deformation limit of the target object. The normal force stability characterizes the degree of fluctuation of the normal force over time when the robot arm contacts the target object, reflecting the stability of the mechanical response after deformation of the target object—the normal force is stable for hard objects, while it fluctuates easily for soft objects.

[0052] The control device can be pre-configured with a preset hardness analysis model, which can be pre-trained and configured by the developer. For example, the developer can first control the robotic arm to grasp objects with different hardness data. During grasping, the tactile sensor is controlled to collect real-time normal force, thereby obtaining the response characteristics corresponding to objects with different hardness data. Finally, the preset hardness analysis model is trained based on the objects with different hardness data and their corresponding response characteristics.

[0053] The control device can also input response characteristics into a preset hardness analysis model to obtain the hardness data of the target object. For example, the control device inputs the normal force rise rate, peak normal force, and normal force stability into the preset hardness analysis model. The preset hardness analysis model can then analyze the normal force rise rate, peak normal force, and normal force stability to determine the hardness data of the target object, and finally output the hardness data of the target object. At this point, the control device can obtain the hardness data of the target object.

[0054] After obtaining the hardness data of the target object, the control device can determine the initial gripping gesture and initial gripping force based on the hardness data. For example, the control device can also be configured with a preset hardness-gesture and force association library. This library can be configured by the developer based on actual conditions (such as determining, through the gripping of a large number of objects, what kind of object hardness requires what kind of gesture and force). The control device can query the preset hardness-gesture and force association library based on the hardness data to obtain the gesture and force associated with that hardness data, and thus determine the initial gripping gesture and initial gripping force based on the gesture and force associated with that hardness data.

[0055] After receiving the initial grasping gesture and initial grasping force, the control device can control the robotic arm to grasp the target object based on the initial grasping gesture and initial grasping force, thereby performing a preliminary grasping of the target object.

[0056] This application's technical solution involves collecting electromyographic (EMG) signals from the user's forearm surface when the robotic arm grasps a target object; inputting the EMG signals into a motion intention recognition model to obtain the current motion intention; and controlling the robotic arm to adjust the grasping force or release the target object based on the current motion intention. By determining whether to adjust the grasping force or release the target object based on the EMG signals from the user's forearm surface when the robotic arm grasps the target object, and based on the user's subjective force adjustment mechanism, the grasping force can be adjusted in real time according to the characteristics of the target object, thereby improving the stability, safety, and adaptability of the grasping process.

[0057] This application further proposes a control device for a robotic arm, referring to... Figure 8 , Figure 8 This is a schematic diagram of the control device for a robotic arm according to an embodiment of the present application. In some embodiments, the control device for the robotic arm includes: The acquisition unit 300 is used to acquire electromyographic signals on the surface of the user's forearm when the robotic arm grasps the target object; The recognition unit 310 is used to input electromyographic signals into the motion intention recognition model to obtain the current motion intention; The control unit 320 is used to control the robotic arm to adjust the force of grasping the target object or to release the target object according to the current motion intention.

[0058] In some embodiments, the identification unit 310 is specifically used for: Input electromyography (EMG) signals into the motion intention recognition model; The current grasping intent or current releasing intent is obtained by the motion intent recognition model from the electromyographic signals. Use the current grasping intention or the current releasing intention as the current motion intention.

[0059] In some embodiments, when the control unit 320 performs the action of controlling the robotic arm to adjust the force of grasping the target object according to the current motion intention, it is specifically used for: When the current motion intent is the current grasping intent, the current grasping intent is analyzed to obtain the current grasping gesture and the current grasping force; Disable the current grab gesture and calculate the difference between the current grab force and the initial grab force to obtain the grab force difference; Based on the initial grasping gesture and the difference in grasping force, the robotic arm is controlled to continue grasping the target object, so that the grasping force of the robotic arm is adjusted from the initial grasping force to the current grasping force.

[0060] In some embodiments, after executing the command to control the robotic arm to continue grasping the target object based on the initial grasping gesture and the difference in grasping force, the control unit 320 is further specifically used for: The current vibration parameters are obtained by querying the preset force-vibration relationship table based on the current gripping force. The vibration module of the robotic arm is controlled to output vibration based on the current vibration parameters.

[0061] In some embodiments, when the control unit 320 executes the action of controlling the robotic arm to release the target object according to the current motion intention, it is specifically used for: When the current motion intention is to release, control the robotic arm to gradually reduce the gripping force, and control the tactile sensors of the robotic arm to collect the real-time normal force. Detect whether the real-time normal force is zero; When the real-time normal force is zero, it is determined that the robot arm has released the target object.

[0062] In some embodiments, the control unit 320 is further configured to: The tactile sensors controlling the robotic arm collect real-time normal force. Determine whether the real-time normal force is greater than the preset normal force; When the real-time normal force is greater than the preset normal force, it is determined that the robot has made contact with the target object.

[0063] In some embodiments, after determining that the robotic arm has made contact with the target object, the control unit 320 is further configured to: Establish a normal force-acquisition timetable based on the real-time normal force and the real-time normal force acquisition time; The response characteristics are obtained by analyzing the normal force-acquisition time table, and the response characteristics are input into the preset hardness analysis model to obtain the hardness data of the target object; The initial gripping gesture and initial gripping force are determined based on the hardness data; The robotic arm is controlled to grasp the target object based on the initial grasping gesture and initial grasping force.

[0064] This application further proposes a robotic arm, referring to... Figure 9 , Figure 9 This is a schematic diagram of the structure of the robotic arm according to the embodiments of this application. In some embodiments, the robotic arm includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that are executed by the at least one processor to enable the at least one processor to perform the control method for the robotic arm described above.

[0065] In this embodiment, refer to Figure 9 The robotic arm may include: a processor 1001 (e.g., a CPU), a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. At least one processor 1001 is included; the communication bus 1002 is used to establish communication between these components. The user interface 1003 may include a display screen or an input unit, such as a keyboard. Optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0066] Those skilled in the art will understand that Figure 9 The robotic arm structure shown does not constitute a limitation on the robotic arm and may include more or fewer parts than shown, or combine certain parts, or have different arrangements of parts.

[0067] like Figure 9 As shown, the memory 1005, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and computer programs.

[0068] exist Figure 9 In the robot shown, the network interface 1004 is mainly used to connect to the backend server and communicate data with the backend server; the user interface 1003 is mainly used to connect to the client (user end) and communicate data with the client; and the processor 1001 can be used to call the computer program stored in the memory 1005. When the computer program is called and executed by the processor 1001, it implements the steps of the above-mentioned robot control method.

[0069] The above description is only a part or preferred embodiment of this application. Neither the text nor the drawings should limit the scope of protection of this application. All equivalent structural transformations made using the content of this application's specification and drawings under the overall concept of this application, or direct / indirect applications in other related technical fields, are included within the scope of protection of this application.

Claims

1. A control method of a robot characterized by comprising: The control method for the robotic arm includes: When the robotic arm grasps the target object, electromyographic signals are collected from the surface of the user's forearm. The electromyographic signal is input into the motion intention recognition model to obtain the current motion intention; Based on the current motion intention, the robotic arm is controlled to adjust the force of gripping the target object or release the target object.

2. The control method of a robot according to claim 1, characterized by, The step of inputting the electromyographic signal into the motion intent recognition model to obtain the current motion intent includes: The electromyographic signal is input into the motion intention recognition model; The current grasping intention or current releasing intention is obtained by the motion intention recognition model from the electromyographic signals. The current grasping intent or the current releasing intent is taken as the current motion intent.

3. The control method of the robot according to claim 2, characterized by, The step of controlling the robotic arm to adjust the force of grasping the target object according to the current motion intention includes: When the current motion intention is the current grasping intention, the current grasping intention is analyzed to obtain the current grasping gesture and the current grasping force; Disable the current grasping gesture and calculate the difference between the current grasping force and the initial grasping force to obtain the grasping force difference; Based on the initial grasping gesture and the difference in grasping force, the robotic arm is controlled to continue grasping the target object, so that the force of the robotic arm grasping the target object is adjusted from the initial grasping force to the current grasping force.

4. The control method of the robot according to claim 3, characterized by, After controlling the robotic arm to continue grasping the target object based on the initial grasping gesture and the difference in grasping force, the method further includes: The current vibration parameters are obtained by querying the preset force-vibration relationship table based on the current gripping force. The vibration module of the robotic arm is controlled to output vibration based on the current vibration parameters.

5. The control method of a robot according to claim 2, wherein The step of controlling the robotic arm to release the target object according to the current motion intention includes: When the current motion intention is the current release intention, the robotic arm is controlled to gradually reduce the gripping force, and the tactile sensor of the robotic arm is controlled to collect the real-time normal force in real time. Detect whether the real-time normal force is zero; When the real-time normal force is zero, it is determined that the robotic arm has released the target object.

6. The control method of a robot hand according to claim 1, wherein Before collecting electromyographic signals from the user's forearm surface when the robotic arm grasps the target object, the procedure also includes: The tactile sensors of the robotic arm are controlled to collect real-time normal force. Determine whether the real-time normal force is greater than the preset normal force; When the real-time normal force is greater than the preset normal force, it is determined that the robotic arm has made contact with the target object.

7. The control method of a robot according to claim 6, wherein After determining that the robotic arm has made contact with the target object, the process further includes: Establish a normal force-acquisition time table based on the real-time normal force and the acquisition time of the real-time normal force; The response characteristics are obtained by analyzing the normal force-acquisition time table, and the response characteristics are input into a preset hardness analysis model to obtain the hardness data of the target object; The initial gripping gesture and initial gripping force are determined based on the hardness data; The robotic arm is controlled to grasp the target object based on the initial grasping gesture and the initial grasping force.

8. A control device for a robot arm, characterized in that The control device for the robotic arm includes: The acquisition unit is used to acquire electromyographic signals from the surface of the user's forearm when the robotic arm grasps the target object; The recognition unit is used to input the electromyographic signal into the motion intention recognition model to obtain the current motion intention; The control unit is used to control the robotic arm to adjust the force of grasping the target object or to release the target object according to the current motion intention.

9. The control device of the robot manipulator according to claim 8, characterized in that, The identification unit is specifically used for: The electromyographic signal is input into the motion intention recognition model; The current grasping intention or current releasing intention is obtained by the motion intention recognition model from the electromyographic signals. The current grasping intent or the current releasing intent is taken as the current motion intent.

10. A robot, characterized in that The robotic arm includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that are executed by the at least one processor to enable the at least one processor to perform the control method of the robotic arm according to any one of claims 1 to 7.

Images