Torque prediction method and device, robot and computer readable storage medium
By collecting and analyzing the robot's historical motion parameters and establishing a preset torque compensation model, the problem of inaccurate torque prediction in robot motion control is solved, achieving higher prediction accuracy and resistance to environmental interference.
Patent Information
- Application Number
- CN202410280010.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-12
- Publication Date
- 2025-09-12
AI Technical Summary
In the prior art, the accuracy of torque prediction in robot motion control is low, especially when a larger parameter affects a smaller parameter during the fitting process, resulting in inaccurate torque prediction.
By collecting the robot's historical motion parameters, including historical joint torque, historical predicted torque and historical joint speed information, the error value is determined, and based on the historical joint speed information and error value, the compensation parameters in the preset torque compensation model are established, and then the preset torque prediction model is established.
The accuracy of torque prediction is improved, the influence of temperature and environmental factors is avoided, and the accuracy of torque prediction is ensured.
Smart Images
Figure CN120620170A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of robots, and more specifically, to a torque prediction method, device, robot, and computer-readable storage medium. Background Art
[0002] Predicting torque plays a crucial role in robot motion control. Related technologies control the robot's movement along an excitation trajectory, collecting relevant parameters during the movement. These parameters are then fitted to obtain the dynamic parameters and friction parameters used in the dynamics model to establish a dynamics model. However, during the fitting process, larger parameters can affect smaller parameters, resulting in low accuracy in the predicted torque obtained from the dynamics model. Summary of the Invention
[0003] In view of the above problems, the present application proposes a torque prediction method, device, robot and computer-readable storage medium, which can effectively improve the accuracy of predicted torque.
[0004] In the first aspect, the present application provides a torque prediction method applied to a robot, the method comprising: collecting historical motion parameters of the robot; the historical motion parameters include historical joint torque, historical predicted torque and historical joint speed information; determining the error value between the historical joint torque and the historical predicted torque; determining the compensation parameters in a preset torque compensation model based on the historical joint speed information and the error value; and establishing a preset torque prediction model based on the compensation parameters.
[0005] In the second aspect, the present application also provides a torque prediction device, which is applied to a robot, and the device includes: a first acquisition module, which is used to collect historical motion parameters of the robot; the historical motion parameters include historical joint torque, historical predicted torque and historical joint speed information; a first determination module, which is used to determine the error value between the historical joint torque and the historical predicted torque; a second determination module, which is used to determine the compensation parameters in the preset torque compensation model based on the historical joint speed information and the error value; and an establishment module, which is used to establish a preset torque prediction model based on the compensation parameters.
[0006] In a third aspect, the present application also provides a robot comprising a processor, a memory, and one or more applications; the one or more applications are stored in the memory and configured to be executed by the processor to implement the above-mentioned torque prediction method.
[0007] In a fourth aspect, the present application further provides a computer-readable storage medium, in which program code is stored, wherein the above-mentioned torque prediction method is executed when the program code is run by a processor.
[0008] The technical solution provided in this application is applied to a robot. The method includes: collecting the robot's historical motion parameters; the historical motion parameters include historical joint torque, historical predicted torque, and historical joint velocity information; determining the error between the historical joint torque and the historical predicted torque; determining the compensation parameters in a preset torque compensation model based on the historical joint velocity information and the error value; and establishing a preset torque prediction model based on the compensation parameters. Thus, the compensation parameters are determined based on the robot's historical motion parameters to establish a preset torque prediction model, thereby obtaining a more accurate predicted torque, thereby avoiding the influence of factors such as temperature and environment. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] In order to more clearly illustrate the technical solutions in the embodiments of this application, the following is a brief introduction to the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments and drawings obtained by ordinary technicians in this field without creative work are within the scope of protection of this invention.
[0010] Figure 1 It is a structural diagram of a torque prediction method provided in an embodiment of the present application.
[0011] Figure 2 This is a structural diagram of a predicted torque and actual torque provided in an embodiment of the present application.
[0012] Figure 3 This is another structural diagram of predicted torque and actual torque provided in an embodiment of the present application.
[0013] Figure 4 It is a structural schematic diagram of a torque prediction device provided in an embodiment of the present application.
[0014] Figure 5 This is a schematic structural diagram of a robot provided in an embodiment of the present application.
[0015] Figure 6 It is a structural diagram of a computer-readable storage medium provided in an embodiment of the present application. DETAILED DESCRIPTION
[0016] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0017] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict. In the following description, the term "plurality" refers to at least two.
[0018] In the following description, the terms "first\second" are only used to distinguish similar objects and do not represent a specific order for the objects. It can be understood that "first\second" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.
[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.
[0020] Predicted torque plays a crucial role in robot motion control. For example, during a robot's drag-teaching operation, the predicted torque can be used to derive predicted friction, which in turn can be used to determine friction compensation. Another example is when a robot performs collision detection and the difference between the predicted torque and the feedback torque exceeds a preset value, it can be determined that the robot is experiencing interference from an obstacle.
[0021] In related technologies, the robot is controlled to move along the excitation trajectory, and relevant parameters during the movement are collected, so as to perform fitting processing based on the relevant parameters, and at the same time obtain the power parameters and friction parameters in the dynamic model to establish a dynamic model.
[0022] However, during the fitting process, there are cases where larger parameters affect smaller parameters, resulting in low accuracy of the predicted torque obtained by the dynamic model.
[0023] In order to improve the above problems, the present application provides a torque prediction method, device, robot and computer-readable storage medium. The torque prediction method is applied to the robot, including: collecting historical motion parameters of the robot; the historical motion parameters include historical joint torque, historical predicted torque and historical joint speed information; determining the error value between the historical joint torque and the historical predicted torque; determining the compensation parameters in the preset torque compensation model based on the historical joint speed information and the error value; and establishing a preset torque prediction model based on the compensation parameters.
[0024] Therefore, the compensation parameters are determined according to the historical motion parameters of the robot to establish a preset torque prediction model, thereby obtaining a more accurate predicted torque, thereby avoiding the influence of factors such as temperature and environment.
[0025] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application.
[0026] See also Figure 1 , Figure 1 Schematic diagram of a torque prediction method provided in an embodiment of the present application. The torque prediction method is applied to robots, such as Figure 1 As shown, the torque prediction method includes steps 110 to 140, specifically:
[0027] In step 110 , historical motion parameters of the robot are collected.
[0028] A robot is an intelligent machine capable of semi-autonomous or fully autonomous operation. It can be programmed and automatically controlled to perform tasks such as work or movement. A robot's joints are the connections between two links, allowing for relative motion. Each joint provides the robot with a degree of freedom. The motion provided by each joint allows the robot structure to be positioned in a specific location.
[0029] Historical motion parameters include the historical joint torque, historical predicted torque, and historical joint velocity information corresponding to the joint. The historical joint torque is the joint torque received by the robot during a particular movement of the robot's joint in the first motion mode. The historical predicted torque is the torque predicted by the robot for that joint. The historical joint velocity information is the velocity value corresponding to the joint during the movement.
[0030] In order to avoid the situation in which a larger parameter affects a smaller parameter during the fitting process, resulting in a lower accuracy of the predicted torque obtained by the dynamic model, thereby making the accuracy of the historical predicted torque lower, in some embodiments, the torque prediction method further includes the following steps:
[0031] (1) Determine the motion parameters of the robot.
[0032] In some embodiments, the motion parameters include joint position information, which indicates position coordinate information corresponding to the joint for which torque prediction is required.
[0033] In some embodiments, the motion parameters further include joint velocity information, which is the velocity value corresponding to the movement of the joint for which the torque prediction is required.
[0034] In some embodiments, the motion parameters further include joint acceleration information, which is the acceleration value corresponding to the movement of the joint for which the torque prediction is required.
[0035] In subsequent steps, the robot inputs the joint velocity information into a preset friction prediction model, and can determine the predicted friction force corresponding to the joint based on the output value of the preset friction prediction model. Furthermore, the robot inputs the predicted friction force, joint position information, joint velocity information, and joint acceleration information into a preset torque prediction model to obtain the predicted torque corresponding to the joint output by the preset torque prediction model. Specifically:
[0036] (2) Determine the predicted friction force based on the motion parameters and the preset friction force prediction model.
[0037] The robot inputs the collected motion parameters into a preset friction prediction model, and the output value of the preset friction prediction model is the predicted friction force. In a subsequent step, the robot inputs the predicted friction force and motion parameters into a preset torque prediction model, so that the preset torque prediction model outputs the predicted torque.
[0038] In some embodiments, each joint corresponds to a preset friction force prediction model. Specifically, the robot controls a joint to operate in a first movement mode, collects movement parameters of the joint during movement, and then determines the friction parameters in the preset friction force prediction model based on the collected movement parameters, thereby determining the preset friction force prediction model. In some embodiments, the first movement mode is reciprocating movement at multiple preset speeds. When the robot establishes the preset friction force prediction model corresponding to a joint, it controls the joint to reciprocate at different speeds and collects the movement parameters corresponding to the joint.
[0039] In some embodiments, the movement parameter includes joint position information corresponding to a joint collected by the robot during the process of the joint moving in the first movement mode.
[0040] In some embodiments, the movement parameter further includes joint velocity information corresponding to a joint collected by the robot during the process of the joint moving in the first movement mode.
[0041] In some embodiments, the movement parameter further includes joint acceleration information corresponding to a joint collected by the robot during the process of the joint moving in the first movement mode.
[0042] In some embodiments, the movement parameter further includes corresponding joint torque information fed back by the robot when a joint moves in the first movement mode.
[0043] In a specific embodiment, the robot controls a joint to move back and forth at different preset speeds, and collects joint position information, joint velocity information, and joint acceleration information during the movement of the joint. The robot determines the friction parameters in a preset friction prediction model based on the collected joint position information, joint velocity information, joint acceleration information, and joint torque information, thereby establishing a preset friction prediction model, so that the robot can obtain the predicted friction by inputting the robot's motion parameters into the preset friction prediction model.
[0044] Specifically, in some embodiments, the torque prediction method further includes the following steps:
[0045] a. Control the robot to work in a first movement mode to determine movement parameters.
[0046] For example, the robot controls joint a to move back and forth at a constant speed of b1. During the process of joint a moving back and forth at a constant speed of b1, the corresponding movement parameter c1 of joint a is collected; the robot controls joint a to move back and forth at a constant speed of b2. During the process of joint a moving back and forth at a constant speed of b2, the corresponding movement parameter c2 is collected.
[0047] b. Determine the joint friction force based on the movement parameters and Newton-Euler equations.
[0048] The robot inputs the movement parameters corresponding to a joint into the Newton-Euler equation to determine the joint friction term in the Newton-Euler equation, and then determines the joint friction force corresponding to the joint.
[0049] For example, the robot controls joint a to work in a first movement mode. During the process of joint a working in the first movement mode, the robot collects the movement parameter b corresponding to joint a. The robot inputs the movement parameter b into the Newton-Euler equation to determine the joint friction term in the Newton-Euler equation.
[0050] Specifically, the steps of determining the joint friction force according to the movement parameters and the Newton-Euler equation include: determining the joint friction force according to the joint position information, joint velocity information, joint acceleration information, joint torque information and the Newton-Euler equation.
[0051] In some embodiments, the Newton-Euler equation is expressed as:
[0052]
[0053] Among them, τ is the joint torque information; q is the joint position information; is the joint velocity information; is the joint acceleration information; τ f is the joint friction force; H(q) is the inertia term; is the Coriolis force term; G(q) is the gravity term.
[0054] c. Determine the friction parameters of the preset friction prediction model based on the joint friction and movement parameters.
[0055] The robot uses the offline least squares fitting method to fit the joint friction force and the collected movement parameters to obtain the friction parameters of the preset friction prediction model.
[0056] In some embodiments, the expression of the preset friction prediction model is:
[0057]
[0058] Among them, τ f is the joint friction force; a, b, c and d are friction force parameters; is the joint velocity information.
[0059] The robot controls a joint to move in a first movement mode and collects joint position information, joint velocity information, joint acceleration information, and joint torque information corresponding to the joint during the movement. The robot determines the joint friction force in the Newton-Euler equation based on the joint position information, joint velocity information, joint acceleration information, and joint torque information and the Newton-Euler equation. Furthermore, the robot uses an offline least squares fitting method to calculate the friction parameters in the preset friction prediction model based on the joint friction force, joint velocity information, and a preset friction prediction model.
[0060] The robot then establishes a preset friction force prediction model corresponding to the joint based on the friction force parameters. After the robot establishes a preset friction force prediction model corresponding to a joint, it determines whether there are joints for which a corresponding preset friction force prediction model has not been established. If the robot determines that there are joints for which a corresponding preset friction force prediction model has not been established, the robot will establish a preset friction force prediction model corresponding to the joint. If the robot determines that all joints have corresponding preset friction force prediction models established, step 130 is executed. Specifically:
[0061] (3) Determine the predicted torque based on the predicted friction force, motion parameters and the preset torque prediction model.
[0062] When determining the predicted torque for a particular joint, the robot collects the corresponding motion parameters (i.e., joint velocity information) and inputs them into a preset friction prediction model to obtain the joint friction output from the preset friction prediction model. The robot then determines the predicted torque based on the obtained joint friction, motion parameters, and the preset torque prediction model.
[0063] The motion parameters of the preset torque prediction model are determined based on the operating parameters obtained when the robot operates in the second movement mode.
[0064] In some embodiments, the operating parameters include joint position information corresponding to a joint collected by the robot during the process of the joint moving in the second movement mode.
[0065] In some embodiments, the operating parameters include joint velocity information corresponding to a joint collected by the robot during the process of the joint moving in the second movement mode.
[0066] In some embodiments, the operating parameters include joint acceleration information corresponding to a joint collected by the robot during the process of the joint moving in the second movement mode.
[0067] In some embodiments, the operating parameters include joint torque information corresponding to a joint collected by the robot during the process of the joint moving in the second movement mode.
[0068] In some embodiments, the second movement mode is movement along an excitation trajectory. The excitation trajectory is a Fourier series joint excitation trajectory. Specifically, the equation of the excitation trajectory is:
[0069] Where i indicates the joint number; L indicates the number of sine and cosine terms; a i,l Indicates the sine amplitude of the i-th joint; b i,l Indicates the cosine amplitude of the i-th joint; w indicates the fundamental frequency.
[0070] The robot uses the optimal algorithm and takes the minimum condition number of the regression matrix as the optimal target to determine the trajectory parameters in the excitation trajectory, and then the trajectory parameters determine the excitation trajectory, and then control a certain joint to move along the excitation trajectory, and then obtain the corresponding operating parameters of the joint during the movement process to determine the motion parameters in the preset torque prediction model, and then establish the preset torque prediction model.
[0071] Specifically, the torque prediction method further includes the following steps: the robot determines the trajectory parameters in the excitation trajectory according to the optimal algorithm, taking the minimum condition number of the regression matrix as the optimization target. The robot determines the excitation trajectory according to the trajectory parameters.
[0072] In some embodiments, the trajectory parameters include the sine amplitude of the i-th joint and the cosine amplitude of the i-th joint. That is, the trajectory parameters are a in the equation of the excitation trajectory. i , item l and item b i ,l item.
[0073] In some embodiments, the base angular frequency w=2πf, and the robot selects the joint number L, b and, under the constraints of the joint position information, joint velocity information, and joint acceleration information, uses the optimal algorithm and takes the minimum condition number of the regression matrix as the optimal target to solve the trajectory parameters.
[0074] After the robot completes the establishment of the excitation trajectory, the robot controls the joint to move according to the excitation trajectory to obtain the corresponding operating parameters of the joint, thereby determining the motion parameters in the preset torque prediction model, and then completing the establishment of the preset torque prediction model. Specifically, in some embodiments, the torque prediction method further includes the following steps:
[0075] a. Control the robot to work in the second movement mode to determine the operating parameters.
[0076] b. Determine the motion parameters of the preset torque prediction model based on the operating parameters, joint friction and motion equation.
[0077] In some embodiments, the operating parameters include joint position information corresponding to a joint during the process of the joint moving in the second movement manner.
[0078] In some embodiments, the operating parameters further include joint speed information corresponding to a joint during the process of the joint moving in the second movement manner.
[0079] In some embodiments, the operating parameter includes joint acceleration information corresponding to a joint during the process of the joint moving in the second movement manner.
[0080] In some embodiments, the operating parameters include joint torque information corresponding to a joint during the process of the joint moving in the second movement manner.
[0081] After the robot collects the operating parameters corresponding to the joints, it inputs the collected operating parameters into the dynamic equation obtained by linear change based on the Newton-Euler equation. Through the offline least squares fitting method, the dynamic parameters in the dynamic equation are calculated, and then a preset torque prediction model is established based on the dynamic parameters.
[0082] In some embodiments, the kinetic equation is expressed as: Among them, Y is the regression matrix; P is the motion parameter; τ f The friction force of the joint.
[0083] When predicting the predicted torque for a joint, the robot collects the corresponding joint velocity information, joint acceleration information, and joint position information. It then inputs the joint velocity information into a preset friction prediction model to obtain the predicted friction output by the preset friction prediction model. The robot then inputs the joint velocity information, joint acceleration information, joint position information, and predicted friction into a preset torque prediction model to obtain the predicted torque output by the preset torque prediction model, thereby determining the predicted torque for the joint.
[0084] However, due to factors such as temperature and the environment, the properties of robot joints can change. For example, the corresponding viscous friction coefficient and Coulomb friction coefficient can change. This makes the friction parameters in the preset friction prediction model unsuitable for all robot motion conditions.
[0085] In order to further improve the accuracy of the predicted torque, in some embodiments, the torque prediction method further includes the following steps:
[0086] a. Determine the compensation torque based on the motion parameters and the preset torque compensation model.
[0087] b. Determine the final predicted torque based on the predicted torque and the compensation torque.
[0088] The robot outputs the motion parameters to the preset torque compensation model, obtains the compensation torque output by the preset torque compensation model, and then determines the final predicted torque based on the compensation torque and the predicted torque, thereby avoiding the influence of factors such as temperature and environment, and further improving the accuracy of the predicted torque.
[0089] By separately identifying the friction and motion parameters, we establish a preset friction prediction model and a preset torque prediction model to determine the robot's preset torque. This prevents large friction and motion parameters from influencing smaller parameters, thereby improving the accuracy of the predicted torque and, in turn, the accuracy of the historically predicted torque.
[0090] Therefore, the historical predicted torque and the corresponding historical joint torque are used to further determine whether it is necessary to update and establish a preset torque prediction model, so that a more accurate predicted torque can be predicted through the preset torque prediction model. Specifically:
[0091] In step 120 , the error value between the historical joint torque and the historical predicted torque is determined.
[0092] The robot determines the error value by using the difference between the predicted torque determined by the preset friction prediction model and the preset torque prediction model and the fed-back joint torque.
[0093] In some embodiments, when the error value is less than a preset threshold, it indicates that the robot's predicted torque is less affected by temperature and environment, and the robot uses the predicted torque determined by the preset friction prediction model and the preset torque prediction model as the final predicted torque corresponding to the joint.
[0094] In step 130 , compensation parameters in a preset torque compensation model are determined based on the historical joint velocity information and the error value.
[0095] In some embodiments, the expression of the preset torque compensation model is: Among them, τ fa is the compensation torque; a a To compensate for the Coulomb friction coefficient; d a To compensate for the viscous friction coefficient; is the joint velocity information.
[0096] The compensated Coulomb friction coefficient and the compensated viscous friction coefficient are the compensation parameters. The error value is the compensation torque. The robot inputs historical joint velocity information and the error value into the preset torque compensation model to determine the compensated Coulomb friction coefficient and the compensated viscous friction coefficient in the preset torque compensation model equation, thereby obtaining the compensation parameters.
[0097] In step 140 , a preset torque prediction model is established according to the compensation parameters.
[0098] When the robot predicts the predicted torque for a joint, it determines the predicted torque based on the motion parameters, a preset friction prediction model, and a preset torque prediction model. It then inputs the joint velocity information into the preset torque prediction model to obtain the compensation torque. The predicted joint torque and the compensation torque are then combined to determine the final predicted torque for the joint.
[0099] In addition, in order to enable the preset torque prediction model to obtain a more accurate compensation torque, in some embodiments, the torque prediction method further includes the following steps:
[0100] (1) Determine the average error between the historical joint torque and the corresponding historical predicted torque within a preset time period.
[0101] (2) If the average error value is greater than a preset threshold, the step of determining the error value between the historical joint torque and the historical predicted torque is executed.
[0102] The robot determines the average error value between the historical joint torque and the corresponding historical predicted torque within a preset time period. When the robot determines that the average error value is greater than a preset threshold, it will trigger the robot to re-establish the preset torque prediction model corresponding to the joint, that is, the robot will re-determine more accurate compensation parameters to establish a more accurate preset torque prediction model.
[0103] For example, see Figure 2 and Figure 3 , Figure 2 This is a structural diagram of a predicted torque and actual torque provided in an embodiment of the present application. Figure 3 This is another structural diagram of predicted torque and actual torque provided by the embodiment of the present application. Figure 2 and Figure 3 As shown, Figure 2 It includes the predicted torque A, the joint torque B fed back in working condition 1, and the joint torque C fed back in working condition 2. The predicted torque A is determined by the preset friction prediction model and the preset torque prediction model.
[0104] Figure 3 The predicted torque D is determined by the preset friction prediction model, the preset torque prediction model and the preset torque compensation model, which is equivalent to Figure 2 The predicted torque A in the calculation is improved by more than 10%.
[0105] In addition, from Figure 2 In the figure, it can be seen intuitively from the joint torque B and the joint torque C that the joint torques fed back by the joints under different working conditions are different.
[0106] See also Figure 4 , Figure 4 : is a structural diagram of a torque prediction device provided in an embodiment of the present application, which is applied to a robot. The torque prediction device 200 includes: a first acquisition module 210, a first determination module 220, a second determination module 230, and an establishment module 240. Specifically:
[0107] The first acquisition module 210 is used to collect historical motion parameters of the robot; the historical motion parameters include historical joint torque, historical predicted torque and historical joint velocity information;
[0108] A first determination module 220 is configured to determine an error between a historical joint torque and a historical predicted torque;
[0109] A second determination module 230 is configured to determine compensation parameters in a preset torque compensation model based on the historical joint velocity information and the error value;
[0110] The establishing module 240 is used to establish a preset torque prediction model according to the compensation parameters.
[0111] In some embodiments, the torque prediction device 200 further includes:
[0112] A third determination module is used to determine an average error value between the historical joint torque and the corresponding historical predicted torque within a preset time period;
[0113] The first execution module is configured to execute a step of determining an error value between the historical joint torque and the historical predicted torque if the average error value is greater than a preset threshold.
[0114] In some embodiments, the torque prediction device 200 further includes a fourth determination module, a fifth determination module, and a sixth determination module, wherein:
[0115] A fourth determination module is used to determine motion parameters of the robot;
[0116] a fifth determining module, configured to determine a predicted friction force based on the motion parameters and a preset friction force prediction model; wherein the friction force parameters of the preset friction force prediction model are determined based on movement parameters obtained when the robot operates in the first movement mode;
[0117] The sixth determination module is used to determine the predicted torque based on the predicted friction force, motion parameters and the preset torque prediction model; the motion parameters of the preset torque prediction model are determined based on the operating parameters obtained when the robot works in the second movement mode. The first movement mode is different from the second movement mode, and the historical predicted torque is determined by the predicted torque.
[0118] In some embodiments, the torque prediction device 200 further includes a seventh determination module and an eighth determination module, wherein:
[0119] a seventh determination module, configured to determine a compensation torque based on the motion parameters and a preset torque compensation model;
[0120] The eighth determination module is used to determine the final predicted torque according to the predicted torque and the compensation torque.
[0121] In some embodiments, the torque prediction device 200 further includes a ninth determination module, a tenth determination module, and an eleventh determination module, wherein:
[0122] a ninth determining module, configured to control the robot to operate in a first movement mode to determine movement parameters;
[0123] a tenth determination module, for determining joint friction force based on movement parameters and Newton-Euler equations;
[0124] The eleventh determination module is used to determine the friction parameters of the preset friction prediction model according to the joint friction and the movement parameters.
[0125] In some embodiments, the movement parameters in the ninth determination module include joint position information, joint velocity information, joint acceleration information, and joint torque information. The torque prediction device 200 further includes a first determination unit, wherein:
[0126] The first determining unit is used to determine the joint friction force according to the joint position information, the joint velocity information, the joint acceleration information, the joint torque information and the Newton-Euler equation.
[0127] In some embodiments, the torque prediction device 200 further includes a twelfth determination module and a thirteenth determination module, wherein:
[0128] a twelfth determination module, configured to control the robot to operate in a second movement mode to determine an operating parameter;
[0129] The thirteenth determination module is used to determine the motion parameters of the preset torque prediction model according to the operating parameters, joint friction and motion equation.
[0130] In some embodiments, the torque prediction device 200 further includes a fourteenth determination module and a fifteenth determination module, wherein:
[0131] a fourteenth determination module, configured to determine trajectory parameters in the excitation trajectory according to an optimal algorithm, taking the minimum condition number of the regression matrix as an optimization target;
[0132] The fifteenth determination module is used to determine the excitation trajectory according to the trajectory parameters.
[0133] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0134] In several embodiments provided in this application, the coupling or direct coupling or communication connection between the modules shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.
[0135] In addition, the functional modules in the various embodiments of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into a single module. The above-mentioned integrated modules may be implemented in the form of hardware or software functional modules.
[0136] See also Figure 5 , Figure 5 It is a structural diagram of a robot provided in an embodiment of the present application. The robot 300 in the present application may include one or more of the following components: a processor 310, a memory 320, and one or more applications, wherein the one or more applications may be stored in the memory 320 and configured to be executed by one or more processors 310, and the one or more programs are configured to execute the torque prediction method as described in the aforementioned method embodiment.
[0137] The processor 310 may include one or more processing cores. The processor 310 utilizes various interfaces and circuits to connect various components within the robot 300. It executes instructions, programs, code sets, or instruction sets stored in the memory 320 and accesses data stored in the memory 520 to perform various functions and process data for the robot 300. Optionally, the processor 310 may be implemented using at least one of the following hardware forms: a digital signal processing (DSP), a field-programmable gate array (FPGA), or a programmable logic array (PLA). The processor 310 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU primarily handles the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem handles wireless communications. It is understood that the modem may not be integrated into the processor 310 and may be implemented separately via a communication chip.
[0138] The memory 320 may include random access memory (RAM) or read-only memory (ROM). The memory 520 may be used to store instructions, programs, codes, code sets, or instruction sets. The memory 320 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function, instructions for implementing the various method embodiments described below, and the like. The data storage area may also store data created by the robot 300 during use.
[0139] See also Figure 6 , Figure 6 It is a structural diagram of a computer-readable storage medium provided in an embodiment of the present application. The computer-readable medium 400 stores program code, and the program code can be called by a processor to execute the torque prediction method described in the above method embodiment.
[0140] The computer-readable storage medium 400 can be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), an EPROM, a hard disk, or a ROM. Alternatively, the computer-readable storage medium 400 includes a non-transitory computer-readable storage medium. The computer-readable storage medium 400 has storage space for program code 410 for executing any of the method steps in the above method. These program codes can be read from or written to one or more computer program devices. The program code 410 can be compressed, for example, in a suitable form.
[0141] The embodiments of the present application provide a torque prediction method, device, robot, and computer-readable storage medium. The torque prediction method, applied to a robot, comprises: collecting the robot's historical motion parameters; the historical motion parameters include historical joint torques, historical predicted torques, and historical joint velocity information; determining the error value between the historical joint torques and the historical predicted torques; determining compensation parameters in a preset torque compensation model based on the historical joint velocity information and the error value; and establishing a preset torque prediction model based on the compensation parameters. Thus, the compensation parameters are determined based on the robot's historical motion parameters to establish a preset torque prediction model, thereby obtaining a more accurate predicted torque, thereby avoiding the influence of factors such as temperature and environment.
[0142] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims
1. A torque prediction method, characterized in that: Applied to a robot, the method comprises: Collecting historical motion parameters of the robot; the historical motion parameters include historical joint torque, historical predicted torque and historical joint speed information; Determining an error value between the historical joint torque and the historical predicted torque; determining compensation parameters in a preset torque compensation model according to the historical joint velocity information and the error value; A preset torque prediction model is established based on the compensation parameters.
2. The torque prediction method according to claim 1, characterized in that: The method further comprises: Determining an average error value between the historical joint torque and the corresponding historical predicted torque within a preset time period; If the average error value is greater than a preset threshold, a step is executed to determine an error value between the historical joint torque and the historical predicted torque.
3. The torque prediction method according to any one of claims 1-2, characterized in that: The method further comprises: determining motion parameters of the robot; Determining a predicted friction force based on the motion parameters and a preset friction force prediction model; the friction force parameters of the preset friction force prediction model are determined based on movement parameters obtained when the robot operates in a first movement mode; The predicted torque is determined based on the predicted friction force, the motion parameters and the preset torque prediction model; the motion parameters of the preset torque prediction model are determined based on the operating parameters obtained when the robot works in a second movement mode, the first movement mode is different from the second movement mode, and the historical predicted torque is determined by the predicted torque.
4. The torque prediction method according to claim 3, characterized in that: The method further comprises: Determining a compensation torque according to the motion parameters and the preset torque compensation model; A final predicted torque is determined according to the predicted torque and the compensation torque.
5. The torque prediction method according to claim 3, characterized in that: The method further comprises: controlling the robot to work in the first movement mode to determine the movement parameters; determining the joint friction force based on the movement parameters and the Newton-Euler equation; The friction force parameters of the preset friction force prediction model are determined according to the joint friction force and the movement parameter.
6. The torque prediction method according to claim 5, characterized in that: The movement parameters include joint position information, joint velocity information, joint acceleration information and joint torque information; Determining the joint friction force according to the movement parameters and the Newton-Euler equation includes: The joint friction force is determined according to the joint position information, the joint velocity information, the joint acceleration information, the joint torque information and the Newton-Euler equation.
7. The torque prediction method according to any one of claims 4 to 6, characterized in that: The first movement mode is reciprocating movement at a plurality of preset speeds.
8. The torque prediction method according to claim 5, characterized in that: The method further comprises: controlling the robot to operate in the second movement mode to determine the operating parameters; The motion parameters of a preset torque prediction model are determined according to the operating parameters, the joint friction force and the motion equation.
9. The torque prediction method according to claim 8, characterized in that: The second movement mode is moving along an excitation trajectory.
10. The torque prediction method according to claim 8, characterized in that: The method further comprises: According to the optimal algorithm, the minimum condition number of the regression matrix is used as the optimization target to determine the trajectory parameters in the excitation trajectory; The excitation trajectory is determined according to the trajectory parameters.
11. A torque prediction device, characterized in that: Applied to a robot, the device comprises: A first acquisition module is used to acquire historical motion parameters of the robot; the historical motion parameters include historical joint torque, historical predicted torque and historical joint speed information; A first determining module is used to determine an error value between the historical joint torque and the historical predicted torque; a second determining module, configured to determine compensation parameters in a preset torque compensation model according to the historical joint velocity information and the error value; An establishing module is used to establish the preset torque prediction model according to the compensation parameters.
12. A robot, characterized in that: include: one or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and are configured to execute the torque prediction method according to any one of claims 1 to 10 by the one or more processors.
13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program code, and the program code can be called by a processor to execute the torque prediction method according to any one of claims 1 to 10.