Joint torque sensor-based drag teaching method and device and electronic device

Abstract

The application provides a joint torque sensor-based dragging teaching method and device and electronic equipment, an external force torque measured by a torque sensor is used to establish an external force estimation model, the control bandwidth of a current loop is improved, and the increase of engineering cost caused by a force sensor is avoided. Moreover, an inertia shaping method is used to compensate for an inertial torque, and the secondary superposition of the applied external force and the inertial force is avoided. For joint static friction compensation in a speed dead zone stage, an estimated external force torque is obtained according to the external force estimation model of the torque sensor, friction torque compensation is performed in the dead zone range, and the dead zone nonlinear effect of joint friction is reduced. Further, an adaptive damping model is used to constrain the joint running speed and correct the torque instruction, so that the stability of the system in the dragging teaching process is ensured.

Description

Technical Field

[0001] This invention relates to the field of control technology, and more specifically, to a drag teaching method, apparatus, and electronic device based on a joint torque sensor. Background Technology

[0002] With the rapid development of collaborative robot technology, collaborative robots are widely used in fields such as human-machine assembly and flexible assembly. Among them, drag-and-drop teaching technology is one of the core functions of collaborative robots. Drag-and-drop teaching allows operators to freely drag the robot to various points in the workspace, thereby reducing programming difficulty and improving the operating efficiency of collaborative robots.

[0003] In existing technologies, commonly used drag teaching methods for robots mainly include zero-force control methods based on current loops and drag teaching methods based on six-dimensional force sensors.

[0004] In zero-force control methods based on current loops (such as...) Figure 1 As shown, to complete the drag teaching, a robot model and a friction torque model need to be established. Then, the torque constant module converts it into a current value to drive the motor and achieve drag teaching. However, this method heavily relies on a high-bandwidth torque control closed loop and is also highly dependent on the dynamic model. Modeling errors will directly affect the dragging effect. Furthermore, this scheme is susceptible to the friction and anti-drag performance of the reducer, resulting in a poor dragging experience and insufficient overall coordination in high-reduction-ratio collaborative robots.

[0005] In the drag-and-teach method based on a six-dimensional force sensor (such as...) Figure 2 As shown, the external force sensed by the six-dimensional force sensor needs to be converted into a new target position and sent to the motor through the position controller to achieve robot drag teaching. This method also has many drawbacks. Specifically, the cost of force sensors currently on the market is high, significantly increasing engineering application costs. Force sensors can only detect forces at the end effector and cannot accurately reflect the force on each joint. Furthermore, the position controller requires absolute data from the force sensor as input to convert the external force acting on the sensor into the robot's target position for movement. However, when external disturbances occur, the force sensor cannot distinguish between human intervention and external interference, leading to a risk of loss of control. Simultaneously, in rigid contact applications, the force sensor can generate large sudden changes in reverse force, causing the control system to diverge.

[0006] It is evident that existing drag-and-drop teaching methods share a common problem: they cannot directly obtain the actual joint forces at the joint level. To achieve highly sensitive, low-hysteresis, and dynamically adjustable control effects in drag-and-drop teaching, it is necessary to directly perceive the actual forces acting on each joint and drive joint movement through human-machine interaction. Most robot manufacturers install torque sensors at the joints to detect the actual output torque of each joint in real time. However, simply using these sensors for gravity compensation or current calibration does not fundamentally improve drag-and-drop teaching performance. If traditional impedance or admittance structures are still used, merely replacing the external force signal source will result in limited improvement in control performance.

[0007] The existing control structure uses an indirect force-driven architecture, where external force is converted into position deviation, a position controller generates a new position command, and finally, a driver generates a current command for control. The entire process suffers from low control bandwidth and slow response, relying on force or torque sensor measurements for the actual external force, making it susceptible to external interference. Furthermore, it fails to distinguish between external force and inertial force; during dynamic motion, the applied external force and inertial force are simultaneously superimposed, leading to unstable drag and susceptibility to friction and anti-drag performance of the reducer.

[0008] Therefore, there is an urgent need for a drag teaching method suitable for collaborative robots with high deceleration ratios that can improve drag sensitivity and stability. Summary of the Invention

[0009] The purpose of this invention is to provide a drag teaching method, device, and electronic device based on a joint torque sensor, so as to improve the sensitivity and stability of the drag algorithm.

[0010] In a first aspect, the present invention provides a drag teaching method based on a joint torque sensor, the method comprising: The raw torque signal is obtained based on torque sensors installed at the joints of the collaborative robot; Based on the original torque signal and combined with the dynamic model, the external torque detected by the torque sensor is obtained; The relative change trend of the external force torque is obtained in order to construct an external force estimation model for the joint space; The friction torque is compensated by combining the estimated external force torque obtained from the external force estimation model and the maximum static friction force of the joint. A virtual inertial matrix is ​​obtained based on the inertial matrix of the collaborative robot. The virtual inertial matrix and the estimated external torque are substituted into the dynamic equation to obtain the corrected torque command. The constructed adaptive damping model is used to constrain the joint running speed, and the torque command is adjusted to obtain the adjusted motor torque command for drag teaching.

[0011] In an optional implementation, the step of obtaining the raw torque signal based on a torque sensor mounted at the joint position of the collaborative robot includes: Obtain the deformation voltage of the torque sensor installed at the joint of the collaborative robot; Based on the deformation voltage and the deformation relationship between voltage and torque, the original torque signal is obtained.

[0012] In an optional implementation, the step of obtaining the external torque detected by the torque sensor based on the original torque signal and in conjunction with the dynamic model includes: The original torque signal is subjected to Kalman filtering to obtain the filtered torque signal; The filtered torque signal is decomposed using a dynamic model, and the inertial, gravity, and Coriolis terms are removed, while the torque caused by external forces is retained.

[0013] In an optional implementation, the step of obtaining the relative change trend of the external torque to construct an external force estimation model for the joint space includes: The torque increment within a set sampling interval is calculated based on the external torque, and the torque increment represents the relative change trend of the external torque; Based on the derivative of the external torque and the torque increment, the trend of external force variation is established to obtain an external force estimation model for the joint space.

[0014] In an optional implementation, the step of compensating for friction torque by combining the estimated external torque obtained from the external force estimation model and the maximum static friction force of the joint includes: If the joint speed is less than the dead zone speed, multiple compensation intervals are defined within the dead zone based on the relationship between the maximum static friction force and friction torque of the joint. Within each of the aforementioned compensation intervals, the estimated external force torque obtained by combining the external force estimation model is used to compensate for the friction torque according to the set compensation rules.

[0015] In an optional implementation, the step of obtaining the virtual inertial matrix based on the inertial matrix of the collaborative robot includes: Obtain the inertia matrix, initial value of the inertia matrix, and scaling factor of the inertia matrix of the collaborative robot; Based on the inertia matrix, the initial value of the inertia matrix, the scaling factor, and the estimated external torque, a virtual inertia matrix is ​​obtained.

[0016] In an optional implementation, the step of constraining the joint running speed based on the constructed adaptive damping model and adjusting the torque command to obtain an adjusted motor torque command for drag teaching includes: An adaptive damping model is constructed to obtain constraint terms, and the torque command is adjusted using the constraint terms. Based on the adjusted torque command, joint speed, and speed threshold, a motor torque command for drive teaching is obtained.

[0017] In an optional implementation, the step of constructing the adaptive damping model to obtain the constraint terms includes: The adjustment coefficient is calculated based on the joint speed and the speed limit of the drag teaching. Based on the adjustment coefficient, estimated external torque, and offset coefficient, the constraint terms are obtained according to the adaptive damping model.

[0018] Secondly, the present invention provides a drag teaching device based on a joint torque sensor, the device comprising: The signal acquisition module is used to acquire raw torque signals based on torque sensors installed at the joints of the collaborative robot; An external force torque acquisition module is used to obtain the external force torque detected by the torque sensor based on the original torque signal and in combination with a dynamic model; A construction module is used to obtain the relative change trend of the external force torque in order to construct an external force estimation model for the joint space; The compensation module is used to compensate for friction torque by combining the estimated external force torque obtained from the external force estimation model and the maximum static friction force of the joint. The correction module is used to obtain a virtual inertia matrix based on the inertia matrix of the collaborative robot, and to substitute the virtual inertia matrix and the estimated external torque into the dynamic equation to obtain the corrected torque command. An adjustment module is used to constrain the joint running speed based on a constructed adaptive damping model and adjust the torque command to obtain an adjusted motor torque command for drag teaching.

[0019] Thirdly, the present invention provides an electronic device including a memory and a processor, wherein the memory stores computer-readable instructions, which, when executed by the processor, cause the processor to perform the method as described in any of the foregoing embodiments.

[0020] This invention provides a drag teaching method, apparatus, and electronic device based on joint torque sensors. It obtains raw torque signals from torque sensors installed at the joints of a collaborative robot. Based on these raw torque signals and a dynamic model, it obtains the external torque detected by the torque sensors. The relative change trend of the external torque is obtained to construct an external force estimation model for the joint space. Friction torque compensation is performed by combining the estimated external torque obtained from the external force estimation model with the maximum static friction force of the joint. A virtual inertia matrix is ​​obtained based on the inertia matrix of the collaborative robot. The virtual inertia matrix and the estimated external torque are substituted into the dynamic equations to obtain a corrected torque command. An adaptive damping model is used to constrain the joint running speed and adjust the torque command to obtain an adjusted motor torque command for drag teaching.

[0021] In this scheme, an external force estimation model is established using the external torque measured by a torque sensor, which improves the control bandwidth of the current loop and avoids the increased engineering costs introduced by the force sensor. Furthermore, an inertia shaping method is used to compensate for the inertial torque, avoiding the secondary superposition of the applied external force and the inertial force. For joint static friction compensation during the speed dead zone, the estimated external torque is obtained based on the external force estimation model of the torque sensor, and friction torque compensation is performed within the dead zone range, reducing the dead zone nonlinear effect of joint friction. Further, an adaptive damping model is adopted to constrain the joint running speed and correct the torque command, ensuring the stability of the system during the towing teaching process. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments of the present invention will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of a zero-force control structure based on a current loop in the prior art; Figure 2 This is a schematic diagram of a position loop drag teaching structure based on a force sensor in the prior art; Figure 3 A flowchart of a drag teaching method based on a joint torque sensor provided in an embodiment of the present invention; Figure 4 for Figure 3 A flowchart of the sub-steps included in S11; Figure 5 for Figure 3 A flowchart of the sub-steps included in S12; Figure 6 for Figure 3A flowchart of the sub-steps included in S13; Figure 7 for Figure 3 A flowchart of the sub-steps included in S14; Figure 8 for Figure 3 A flowchart of the sub-steps included in S15; Figure 9 A logical structure block diagram of the drag teaching method based on a joint torque sensor provided in an embodiment of the present invention; Figure 10 for Figure 3 A flowchart of the sub-steps included in S16; Figure 11 A functional block diagram of a drag teaching device based on a joint torque sensor provided in an embodiment of the present invention; Figure 12 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0024] The technical solutions of the present invention will now be described with reference to the accompanying drawings in the embodiments of the present invention.

[0025] Please see Figure 3 The following is a flowchart of a drag teaching method based on a joint torque sensor provided in an embodiment of the present invention. The detailed steps of the drag teaching method based on a joint torque sensor are described below.

[0026] S11, obtains the raw torque signal based on torque sensors installed at the joints of the collaborative robot; S12, Based on the original torque signal and combined with the dynamic model, obtain the external torque detected by the torque sensor; S13, obtain the relative change trend of the external force torque to construct an external force estimation model for the joint space; S14, Compensate for friction torque by combining the estimated external torque and the maximum static friction force of the joint obtained from the external force estimation model; S15, Based on the inertia matrix of the collaborative robot, a virtual inertia matrix is ​​obtained, and the virtual inertia matrix and the estimated external torque are substituted into the dynamic equation to obtain the corrected torque command. S16, based on the constructed adaptive damping model to constrain the joint running speed, and adjust the torque command to obtain the adjusted motor torque command for drag teaching.

[0027] The drag teaching method based on a joint torque sensor provided in this embodiment utilizes the external torque measured by the torque sensor to establish an external force estimation model, thereby improving the control bandwidth of the current loop and avoiding the increased engineering costs introduced by the force sensor. Furthermore, it employs an inertia shaping method to compensate for inertial torque, avoiding the secondary superposition of applied external force and inertial force. For joint static friction compensation during the speed dead zone, the estimated external torque is obtained based on the external force estimation model of the torque sensor, and friction torque compensation is performed within the dead zone range, reducing the dead zone nonlinear effect of joint friction. An adaptive damping model is adopted to constrain the joint running speed and the torque command, ensuring the stability of the system during drag teaching.

[0028] The specific implementation methods of each of the above steps will be explained in detail below.

[0029] The drag-and-drop teaching method provided in this embodiment can be applied to six-degree-of-freedom collaborative robots and can also be extended to humanoid robots.

[0030] In this embodiment, torque sensors are installed at the joints of the collaborative robot to obtain raw torque signals. For details, please refer to [link to documentation]. Figure 4 This can be achieved in the following ways: S111, obtain the deformation voltage of the torque sensor installed at the joint position of the collaborative robot; S112, based on the deformation voltage and according to the deformation relationship between voltage and torque, the original torque signal is obtained.

[0031] The deformation relationship between voltage and torque can be expressed as follows:

[0032] In the formula, k The deformation coefficient, U The voltage of the torque sensor. For torque.

[0033] Based on the deformation relationship between voltage and torque shown above, the original torque signal can be obtained according to the measured deformation voltage.

[0034] Based on this, the external torque detected by the torque sensor is obtained from the original torque signal and combined with the dynamic model. For details, please refer to [link to relevant documentation]. Figure 5 This step can be achieved in the following way: S121, Perform Kalman filtering on the original torque signal to obtain the filtered torque signal; S122, combined with the dynamic model, decomposes the filtered torque signal, removes the inertial term, gravity term and Coriolis term, and retains the external torque caused by external force.

[0035] To eliminate random noise and high-frequency interference in the torque sensor deformation voltage signal and improve torque estimation accuracy, this embodiment first employs an Extended Kalman Filter (EKF) to filter the obtained raw torque signal. Compared to other filtering algorithms, the Extended Kalman Filter can preserve the true low-frequency torque changes.

[0036] Then, combining the dynamic model of the collaborative robot, the filtered torque signal was processed. The process involves decomposition, removing inertia, Coriolis, and gravity terms, and extracting the external torque caused by external forces. It can be characterized as follows:

[0037] in, It is the sum of the inertial term, the Coriolis term, and the gravity term.

[0038] Based on this, an external force estimation model for the joint space is established by utilizing the relative change trend of the torque sensor.

[0039] Errors in dynamic modeling can lead to deviations in the data regarding changes caused by external forces. Using absolute external force data for algorithmic control can easily cause control system malfunctions. Furthermore, the material of the torque sensor itself can cause a zero-point deviation in the torque signal when no force is applied after a continuous external force is removed, affecting the algorithm's performance. Because torque sensors have good sensing performance, within their normal operating range, the torque signal can quickly detect the presence of external forces, avoiding reliance on high-precision dynamic and friction model parameters. Therefore, in this embodiment, the external force estimation model is constructed using the changing trend of torque data.

[0040] In this embodiment, the relative changing trend of the external torque is obtained to construct an external force estimation model for the joint space. Specifically, please refer to [link to relevant documentation]. Figure 6 This step can be achieved in the following way: S131, calculate the torque increment within a set sampling interval based on the external torque, whereby the torque increment characterizes the relative change trend of the external torque; S132, Based on the derivative of the external torque and the torque increment, establish the trend of external force variation to obtain an external force estimation model for the joint space.

[0041] In this embodiment, the torque increment within the set sampling interval is first calculated, and the calculation formula is as follows:

[0042] in, The torque represents the external force at time t. This represents the external torque at time tT, where T represents the set sampling interval. This indicates the torque increment.

[0043] Based on the principle of smoothing filtering, and using the derivative of the external torque and the torque increment, an external force estimation model for the joint space is constructed, specifically representing the changing trend of the external force, as shown below:

[0044] in, , Let represent the derivatives of two adjacent external torques, respectively. Represents the coefficient.

[0045] Based on this, friction torque compensation is performed by combining the estimated external torque obtained from the external force estimation model and the maximum static friction force of the joint. For details, please refer to [link to relevant documentation]. Figure 7 This step can be achieved in the following ways: S141, detect whether the joint speed is less than the dead zone speed. If the joint speed is less than the dead zone speed, then multiple compensation intervals are set within the dead zone based on the relationship between the maximum static friction force and friction torque of the joint. S142, within each of the compensation intervals, the estimated external force torque obtained by combining the external force estimation model is used to compensate the friction torque according to the set compensation rules.

[0046] In this embodiment, experiments show that when no external force is applied, the torque increment is similar to the derivative of a function, which contains noise signals. By establishing a noise threshold, it can be determined whether an external force is applied, as characterized below:

[0047] in, This indicates an estimate of the external torque. The noise threshold representing the change in torque sensor readings. This represents the proportionality coefficient.

[0048] The estimated external torque obtained above can be used to determine the direction of the applied external force or the trend of joint movement. Due to the existence of joint static friction, the joint will only move when the applied external force exceeds the joint static friction; otherwise, the joint is in a dead zone.

[0049] In this embodiment, a speed threshold is used to distinguish dead zones. Simultaneously, within the dead zone, friction torque compensation is performed in stages. Specifically, it detects whether the joint speed is less than the dead zone speed. If the joint speed is less than the dead zone speed, friction torque compensation is performed; if the joint speed is not less than the dead zone speed, subsequent operations are directly executed.

[0050] When the joint speed is less than the dead zone speed, multiple compensation intervals can be defined based on the relationship between the frictional torque and the maximum static friction force of the joint, for example, three compensation intervals. Within each compensation interval, compensation is performed according to different compensation rules. For example, the compensation rules for each compensation interval are as follows:

[0051] in, , These represent the friction torque before and after compensation, respectively. This represents the maximum static friction force of the joint. v Indicates joint velocity. v dead Indicates the dead zone speed.

[0052] The above methods are used to compensate for frictional torque.

[0053] Based on this, an equivalent inertia shaping structure is constructed in this embodiment to perform inertial torque compensation.

[0054] In this embodiment, a virtual inertial matrix is ​​obtained based on the inertial matrix of the collaborative robot. The virtual inertial matrix and the estimated external torque are substituted into the dynamic equation to obtain the corrected torque command.

[0055] Please refer to Figure 8 The virtual inertial matrix can be obtained in the following ways: S151, obtain the inertia matrix, initial value of the inertia matrix, and scaling factor of the inertia matrix of the collaborative robot; S152, based on the inertia matrix, the initial value of the inertia matrix, the scaling factor, and the estimated external torque, a virtual inertia matrix is ​​obtained.

[0056] In this embodiment, it is assumed that the inertia matrix of the collaborative robot is... The virtual inertia matrix is The relationship between the two is as follows:

[0057] in, This represents the initial value of the inertia matrix, used to ensure system stability. The scaling factor represents the inertia matrix. .

[0058] Substituting the aforementioned virtual inertia matrix and estimated external torque into the dynamic equations yields the corrected torque command, as shown below:

[0059] in, This indicates the corrected torque command. , , , Let represent the current position, current velocity, Christoffel matrix, and gravity term vector of each joint of the collaborative robot, respectively. This indicates the additional joint torque caused by Coriolis force and centrifugal force.

[0060] It should be noted that, in order to improve the dragging feel during the drag-and-drop teaching process, it is generally necessary to ensure... .

[0061] In summary, the schematic diagram of the logic structure of the torque closed-loop drag teaching based on the joint torque sensor provided in this embodiment is as follows: Figure 9 As shown, in the case of a closed-loop torque, where and If it is usually set to 0, the output motor torque command will be as follows:

[0062] in, K p Indicates position gain. K v Indicates speed gain. K t This represents the motor torque constant.

[0063] To ensure the stability of the system during drag-and-drop teaching, this embodiment also employs an adaptive damping model to achieve speed constraints and damping adjustments.

[0064] Based on this, and on the basis of the above, the constructed adaptive damping model is used to constrain the joint running speed, and the torque command is adjusted to obtain the adjusted motor torque command for drag teaching.

[0065] Specifically, please refer to Figure 10 The above steps can be achieved in the following ways: S161, Construct an adaptive damping model to obtain constraint terms, and use the constraint terms to adjust the torque command; S162, based on the adjusted torque command, joint speed and speed threshold, obtains the motor torque command for drag teaching.

[0066] The step of constructing an adaptive damping model to obtain the constraint terms can be achieved in the following way: The adjustment coefficient is calculated based on the joint speed and the speed limit of the drag teaching; the constraint term is obtained according to the adjustment coefficient, the estimated external torque and the offset coefficient, and the adaptive damping model.

[0067] Specifically, the adjustment coefficient is calculated based on the joint speed and the speed limit of the drag teaching according to the following formula:

[0068] in:

[0069] In the above formula, B Indicates the adjustment factor. b 0 indicates the offset coefficient. v 1 indicates the speed limit for dragging and teaching.

[0070] Based on the above adjustment coefficients, the adjusted torque command is expressed as follows:

[0071] As can be seen from the above formula, at low speeds, the compensation torque is relatively large, and the external force used by the operator to drag and teach is relatively small. At medium and high speeds, the damping force will increase. If the external force is constant, the compensation torque will decrease, thus limiting the output torque.

[0072] Based on this, the adjusted motor torque command is as follows:

[0073] in, This indicates the speed threshold.

[0074] In the case of joint torque closed loop, a velocity closed loop is introduced, that is, when the joint velocity exceeds the velocity threshold. After that, the output of the motor torque command will be limited.

[0075] The drag teaching method based on a joint torque sensor provided in this embodiment constructs a complete control framework including external force estimation, inertia shaping, dynamic reconstruction, and torque closed-loop compensation. At the start-up phase, this method uses sliding window filtering to perform staged compensation for static friction, reducing its impact and avoiding sudden torque changes. At low speeds, the external force estimation model from the torque sensor is used to generate a virtual inertia matrix through inertia shaping, which is then substituted into the dynamic model to form a new torque command.

[0076] Furthermore, a dual-layer safety control method combining dynamic damping adjustment and velocity outer loop constraints was designed to achieve active safety limits during the dragging process, avoiding excessively fast movement or excessive output torque problems that may occur during the dragging of the collaborative robot. The damping is dynamically adjusted according to the amplitude of the external force to limit the output torque command. Simultaneously, a speed outer loop constraint control strategy is introduced; when the actual speed approaches a threshold, the motor torque command is dynamically reduced to avoid continuous acceleration, thereby limiting the dragging speed.

[0077] Compared to the zero-force control method of the current loop and the position controller algorithm of the force sensor, this scheme utilizes the joint torque sensor to measure the applied relative external force data and establishes an external force estimation model, thereby improving the control bandwidth of the current loop and avoiding the increased engineering costs introduced by the force sensor. Furthermore, compared to the gravitational torque compensation or current calibration control of the torque sensor, this invention utilizes an inertia shaping method to compensate for the inertial torque, avoiding the secondary superposition of the applied external force and the inertial force. Moreover, it does not require obtaining the absolute torque data of the torque sensor for compensation.

[0078] For joint static friction compensation in the speed dead zone, this scheme obtains external force data based on the external force estimation model of the torque sensor, and uses the sliding window filtering method to set the stage compensation force within the dead zone range to reduce the dead zone nonlinear effect of joint friction.

[0079] To ensure system stability during the teaching process, an adaptive damping model and a speed constraint model were employed. Damping was dynamically adjusted based on the magnitude of the external force to limit the output torque command. Simultaneously, a speed outer loop constraint was introduced to reduce the motor's torque command if the speed exceeded the limit, thus preventing continuous acceleration.

[0080] This solution can effectively improve the sensitivity and stability of the dragging algorithm, thereby enhancing the user experience of collaborative robots for operators.

[0081] Based on the same inventive concept, please refer to Figure 11 This invention also provides a functional module diagram of a drag teaching device based on a joint torque sensor. This embodiment divides the drag teaching device based on the joint torque sensor into functional modules according to the above method embodiment. For example, each function can be divided into its own functional modules, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. It should be noted that the module division in this embodiment is illustrative and only represents a logical functional division; other division methods may be used in actual implementation.

[0082] For example, when dividing functional modules according to their respective functions, Figure 11 The drag teaching device based on a joint torque sensor shown is only a schematic diagram. This drag teaching device may include a signal acquisition module, an external force torque acquisition module, a construction module, a compensation module, a correction module, and an adjustment module. The functions of each module of this drag teaching device based on a joint torque sensor will be described in detail below.

[0083] The signal acquisition module is used to acquire raw torque signals based on torque sensors installed at the joints of the collaborative robot; An external force torque acquisition module is used to obtain the external force torque detected by the torque sensor based on the original torque signal and in combination with a dynamic model; A construction module is used to obtain the relative change trend of the external force torque in order to construct an external force estimation model for the joint space; The compensation module is used to compensate for friction torque by combining the estimated external force torque obtained from the external force estimation model and the maximum static friction force of the joint. The correction module is used to obtain a virtual inertia matrix based on the inertia matrix of the collaborative robot, and to substitute the virtual inertia matrix and the estimated external torque into the dynamic equation to obtain the corrected torque command. An adjustment module is used to constrain the joint running speed based on a constructed adaptive damping model and adjust the torque command to obtain an adjusted motor torque command for drag teaching.

[0084] The drag teaching device based on a joint torque sensor provided in this embodiment can be used to execute the drag teaching method based on a joint torque sensor under any of the above embodiments. For details not covered in this embodiment, please refer to the corresponding descriptions in the above embodiments. This embodiment will not be elaborated here.

[0085] Please see Figure 12 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. The electronic device can be a control device on a collaborative robot, or a computer device, server, etc., communicating with the collaborative robot. The electronic device includes a memory, a processor, and a communication module. The memory, processor, and communication module are electrically connected directly or indirectly to each other to achieve data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines.

[0086] The memory is used to store computer programs or data. Memory can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc.

[0087] The processor is used to read / write data or programs stored in the memory and execute the drag teaching method based on the joint torque sensor provided in any embodiment of the present invention.

[0088] The communication module is used to establish communication connections between electronic devices and other communication terminals via a network, and to send and receive data via the network.

[0089] It should be understood that, Figure 12 The structure shown is only a schematic diagram of an electronic device; the electronic device may also include components that are larger than those shown. Figure 12 The more or fewer components shown, or having the same Figure 12 The different configurations shown.

[0090] Furthermore, embodiments of the present invention also provide a computer-readable storage medium storing machine-executable instructions, which, when executed, implement the drag teaching method based on a joint torque sensor provided in the above embodiments.

[0091] Specifically, the computer-readable storage medium can be a general-purpose storage medium, such as a removable disk or hard disk. When the computer program on the computer-readable storage medium is executed, it can perform the aforementioned drag-and-drop teaching method based on a joint torque sensor. The processes involved in the execution of the executable instructions on the computer-readable storage medium can be referred to the relevant descriptions in the above method embodiments, and will not be detailed here.

[0092] In the embodiments provided by this invention, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0093] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0094] Furthermore, the functional modules in the various embodiments of the present invention can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0095] It should be noted that if the functionality is implemented as a software module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0096] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0097] The above description is merely an embodiment of the present invention and is not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A drag teaching method based on a joint torque sensor, characterized in that, The method includes: The raw torque signal is obtained based on torque sensors installed at the joints of the collaborative robot; Based on the original torque signal and combined with the dynamic model, the external torque detected by the torque sensor is obtained; The relative change trend of the external force torque is obtained in order to construct an external force estimation model for the joint space; The friction torque is compensated by combining the estimated external force torque obtained from the external force estimation model and the maximum static friction force of the joint. A virtual inertial matrix is ​​obtained based on the inertial matrix of the collaborative robot. The virtual inertial matrix and the estimated external torque are substituted into the dynamic equation to obtain the corrected torque command. The constructed adaptive damping model is used to constrain the joint running speed, and the torque command is adjusted to obtain the adjusted motor torque command for drag teaching.

2. The drag teaching method based on a joint torque sensor according to claim 1, characterized in that, The step of obtaining the raw torque signal based on the torque sensor installed at the joint position of the collaborative robot includes: Obtain the deformation voltage of the torque sensor installed at the joint of the collaborative robot; Based on the deformation voltage and the deformation relationship between voltage and torque, the original torque signal is obtained.

3. The drag teaching method based on a joint torque sensor according to claim 1, characterized in that, The step of obtaining the external torque detected by the torque sensor based on the original torque signal and in conjunction with the dynamic model includes: The original torque signal is subjected to Kalman filtering to obtain the filtered torque signal; The filtered torque signal is decomposed using a dynamic model, and the inertial, gravity, and Coriolis terms are removed, while the torque caused by external forces is retained.

4. The drag teaching method based on a joint torque sensor according to claim 1, characterized in that, The step of obtaining the relative change trend of the external force torque to construct an external force estimation model for the joint space includes: The torque increment within a set sampling interval is calculated based on the external torque, and the torque increment represents the relative change trend of the external torque; Based on the derivative of the external torque and the torque increment, the trend of external force variation is established to obtain an external force estimation model for the joint space.

5. The drag teaching method based on a joint torque sensor according to claim 1, characterized in that, The step of compensating for friction torque by combining the estimated external torque obtained from the external force estimation model and the maximum static friction force of the joint includes: If the joint speed is less than the dead zone speed, multiple compensation intervals are defined within the dead zone based on the relationship between the maximum static friction force and friction torque of the joint. Within each of the aforementioned compensation intervals, the estimated external force torque obtained by combining the external force estimation model is used to compensate for the friction torque according to the set compensation rules.

6. The drag teaching method based on a joint torque sensor according to claim 1, characterized in that, The step of obtaining the virtual inertial matrix based on the inertial matrix of the collaborative robot includes: Obtain the inertia matrix, initial value of the inertia matrix, and scaling factor of the inertia matrix of the collaborative robot; Based on the inertia matrix, the initial value of the inertia matrix, the scaling factor, and the estimated external torque, a virtual inertia matrix is ​​obtained.

7. The drag teaching method based on a joint torque sensor according to claim 1, characterized in that, The step of constraining the joint running speed based on the constructed adaptive damping model and adjusting the torque command to obtain the adjusted motor torque command for drive teaching includes: An adaptive damping model is constructed to obtain constraint terms, and the torque command is adjusted using the constraint terms. Based on the adjusted torque command, joint speed, and speed threshold, a motor torque command for drive teaching is obtained.

8. The drag teaching method based on a joint torque sensor according to claim 7, characterized in that, The step of constructing an adaptive damping model to obtain constraint terms includes: The adjustment coefficient is calculated based on the joint speed and the speed limit of the drag teaching. Based on the adjustment coefficient, estimated external torque, and offset coefficient, the constraint terms are obtained according to the adaptive damping model.

9. A drag teaching device based on a joint torque sensor, characterized in that, The device includes: The signal acquisition module is used to obtain the raw torque signal based on the torque sensor installed at the joint position of the collaborative robot; An external force torque acquisition module is used to obtain the external force torque detected by the torque sensor based on the original torque signal and in combination with a dynamic model; A construction module is used to obtain the relative change trend of the external force torque in order to construct an external force estimation model for the joint space; The compensation module is used to compensate for friction torque by combining the estimated external force torque obtained from the external force estimation model and the maximum static friction force of the joint. The correction module is used to obtain a virtual inertia matrix based on the inertia matrix of the collaborative robot, and to substitute the virtual inertia matrix and the estimated external torque into the dynamic equation to obtain the corrected torque command. An adjustment module is used to constrain the joint running speed based on a constructed adaptive damping model and adjust the torque command to obtain an adjusted motor torque command for drag teaching.

10. An electronic device, characterized in that, The device includes a memory and a processor, wherein the memory stores computer-readable instructions that, when executed by the processor, cause the processor to perform the method as described in any one of claims 1 to 8.

Images