Parameter identification method and device of mechanical arm, electronic equipment and storage medium

Abstract

The invention provides a parameter identification method and device of a mechanical arm, electronic equipment and a storage medium, and relates to the technical field of mechanical arm control. The method comprises the following steps: acquiring a dynamic inertial parameter, a nonlinear friction parameter, joint motion information and a dynamic total torque of the mechanical arm; according to the dynamic inertia parameters, the nonlinear friction parameters and the joint motion information, the dynamic total torque is split, and split inertia torque and split friction torque are obtained; according to the split inertia moment and the split friction moment, performing loop iteration optimization on the dynamic inertia parameter and the nonlinear friction parameter until a preset convergence condition is met; and obtaining a target dynamic inertial parameter and a target nonlinear friction parameter which meet a preset convergence condition, wherein the target dynamic inertial parameter and the target nonlinear friction parameter are used for performing dynamic control on the mechanical arm. According to the method, the dynamic inertial parameters conforming to the physical characteristics and the nonlinear friction parameters better fitting the joint friction force can be identified.

Description

Technical Field

[0001] The present application relates to the technical field of robotic arm control, and in particular, to a method, device, electronic device, and storage medium for parameter identification of a robotic arm. Background Art

[0002] In the field of artificial intelligence, the application of robotic arms is becoming more and more extensive and in-depth. Based on this, the requirements for the control accuracy of robotic arms are also getting higher and higher. An accurate dynamic model is the basis for high-precision control of robotic arms.

[0003] The existing methods for identifying the dynamic parameters of robotic arms are divided into the parameter set method and the neural network training method. Among them, the parameter set method can only identify inertial parameters and friction parameters separately, resulting in large parameter identification errors. The neural network training method has a large demand for training data and is difficult to perform accurate predictive control on robotic arms. Summary of the Invention

[0004] The purpose of the present application is to provide a method, device, electronic device, and storage medium for parameter identification of a robotic arm, so as to identify dynamic inertial parameters that conform to physical characteristics and non-linear friction parameters that better fit joint friction forces, in view of the deficiencies in the above-mentioned existing technologies.

[0005] To achieve the above objective, the technical solutions adopted in the embodiments of the present application are as follows: In a first aspect, an embodiment of the present application provides a method for parameter identification of a robotic arm, the method including: Obtain the dynamic inertial parameters of the robotic arm, the non-linear friction parameters of the robotic arm, the joint motion information of the robotic arm, and the total dynamic torque; According to the dynamic inertial parameters, the non-linear friction parameters, and the joint motion information, split the total dynamic torque to obtain a split inertial torque and a split friction torque; According to the split inertial torque and the split friction torque, perform cyclic iterative optimization on the dynamic inertial parameters and the non-linear friction parameters until a preset convergence condition is met; Obtain the target dynamic inertial parameters and target non-linear friction parameters when the preset convergence condition is met, and the target dynamic inertial parameters and the target non-linear friction parameters are used for dynamic control of the robotic arm.

[0006] Optionally, the step of splitting the total dynamic torque according to the dynamic inertial parameters, the non-linear friction parameters, and the joint motion information to obtain a split inertial torque and a split friction torque includes: Determine the split friction torque from the total dynamic torque according to the dynamic inertial parameters and the joint motion information; Determine the split inertia torque from the total dynamic torque according to the friction regression matrix of the split friction torque, the non-linear friction parameter, and the joint motion information.

[0007] Optionally, the determining the split inertia torque from the total dynamic torque according to the friction regression matrix of the split friction torque, the non-linear friction parameter, and the joint motion information includes: Calculate a first linear friction parameter according to the split friction torque and the friction regression matrix; Determine the split inertia torque from the total dynamic torque according to the first linear friction parameter and the friction regression matrix.

[0008] Optionally, the cyclically iteratively optimizing the dynamic inertia parameter and the non-linear friction parameter according to the split inertia torque and the split friction torque until a preset convergence condition is satisfied includes: Cyclically iteratively optimize the dynamic inertia parameter according to the dynamic inertia parameter, the joint motion information, and the split inertia torque until a first convergence condition is satisfied; According to the dynamically optimized inertia parameter and the joint motion information each time, re-split the total dynamic torque to obtain a new split friction torque; Each time the dynamic inertia parameter is optimized, cyclically iteratively optimize the non-linear friction parameter according to the non-linear friction parameter, the joint motion information, and the new split friction torque until a second convergence condition is satisfied.

[0009] Optionally, the cyclically iteratively optimizing the non-linear friction parameter according to the non-linear friction parameter, the joint motion information, and the new split friction torque until a second convergence condition is satisfied includes: Calculate a second linear friction parameter according to the friction regression matrix and the new split friction torque; Cyclically iteratively optimize the non-linear friction parameter according to the second linear friction parameter, the friction regression matrix, and the new split friction torque until the second convergence condition is satisfied.

[0010] Optionally, the method further includes: Re-determine the split inertia torque from the total dynamic torque according to the optimized non-linear friction parameter.

[0011] Optionally, the method further includes: Construct a manipulator friction model based on the non-linear friction parameter, the joint motion information, and the linear friction parameter; Determine the friction regression matrix according to the manipulator friction model.

[0012] In a second aspect, an embodiment of the present application further provides a parameter identification device for a manipulator. The device includes: An information acquisition module, configured to acquire the dynamic inertia parameters of the manipulator, the non-linear friction parameters of the manipulator, the joint motion information of the manipulator, and the total dynamic torque. A torque splitting module, configured to split the total dynamic torque according to the dynamic inertia parameters, the non-linear friction parameters, and the joint motion information to obtain a split inertia torque and a split friction torque. An iterative optimization module, configured to perform cyclic iterative optimization on the dynamic inertia parameters and the non-linear friction parameters according to the split inertia torque and the split friction torque until a preset convergence condition is met. A parameter acquisition module, configured to acquire target dynamic inertia parameters and target non-linear friction parameters when the preset convergence condition is met. The target dynamic inertia parameters and the target non-linear friction parameters are used for dynamic control of the manipulator.

[0013] Optionally, the torque splitting module is specifically configured to determine the split friction torque from the total dynamic torque according to the dynamic inertia parameters and the joint motion information; and determine the split inertia torque from the total dynamic torque according to the split friction torque, the non-linear friction parameters, and the friction regression matrix of the joint motion information.

[0014] Optionally, the torque splitting module is further configured to calculate a first linear friction parameter according to the split friction torque and the friction regression matrix; and determine the split inertia torque from the total dynamic torque according to the first linear friction parameter and the friction regression matrix.

[0015] Optionally, the iterative optimization module is specifically configured to perform outer-loop iterative optimization on the dynamic inertia parameters according to the dynamic inertia parameters, the joint motion information, and the split inertia torque until a first convergence condition is met; re-split the total dynamic torque according to the optimized dynamic inertia parameters and the joint motion information each time to obtain a new split friction torque; and perform inner-loop iterative optimization on the non-linear friction parameters according to the non-linear friction parameters, the joint motion information, and the new split friction torque each time the dynamic inertia parameters are optimized until a second convergence condition is met.

[0016] Optionally, the iterative optimization module is further configured to calculate a second linear friction parameter according to the friction regression matrix and the new split frictional torque; perform cyclic iterative optimization on the non-linear friction parameter according to the second linear friction parameter, the friction regression matrix, and the new split frictional torque until the second convergence condition is met.

[0017] Optionally, the torque splitting module is further configured to re-determine the split inertial torque from the total dynamic torque according to the optimized non-linear friction parameter.

[0018] Optionally, the device further includes: A matrix design module, configured to construct a manipulator friction model based on non-linear friction parameters, joint motion information, and linear friction parameters; determine the friction regression matrix according to the manipulator friction model.

[0019] In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, a storage medium, and a bus. The storage medium stores program instructions executable by the processor. When the electronic device runs, the processor communicates with the storage medium through the bus. The processor executes the program instructions to perform the steps of the parameter identification method of the manipulator according to any one of the first aspects.

[0020] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, it performs the steps of the parameter identification method of the manipulator according to any one of the first aspects.

[0021] The beneficial effects of the present application are: The parameter identification method, device, electronic device, and storage medium of the manipulator provided by the present application adopt a two-stage coupled optimization framework to jointly and alternately iteratively optimize the dynamic inertial parameters and non-linear friction parameters. Each time the dynamic inertial parameters are optimized, the non-linear friction parameters are immediately optimized, forming dynamic decoupling between the parameters, avoiding the problem of error accumulation, ensuring the accurate identification of the dynamic inertial parameters and non-linear friction parameters, and improving the accuracy of the dynamic inertial parameters and non-linear friction parameters; the optimized dynamic inertial parameters and non-linear friction parameters are full parameters, which are more convenient to be applied to the dynamic model. Description of the Drawings

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required for the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can be obtained based on these drawings.

[0023] Figure 1 Flow schematic of the parameter identification method for the robotic arm provided by the embodiments of the present application Figure 1 ; Figure 2 Flow schematic of the parameter identification method for the robotic arm provided by the embodiments of the present application Figure 2 ; Figure 3 Flow schematic of the parameter identification method for the robotic arm provided by the embodiments of the present application Figure 3 ; Figure 4 Flow schematic of the parameter identification method for the robotic arm provided by the embodiments of the present application Figure 4 ; Figure 5 Flow block diagram of the parameter identification method for the robotic arm provided by the embodiments of the present application; Figure 6 Structure schematic diagram of the parameter identification device for the robotic arm provided by the embodiments of the present application; Figure 7 Schematic diagram of the electronic device provided by the embodiments of the present application. Detailed implementation manners

[0024] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments.

[0025] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but merely represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0026] In addition, the terms "first", "second", etc. in the specification and claims of the present application and the above accompanying drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0027] It should be noted that, without conflict, the features in the embodiments of the present application may be combined with each other.

[0028] Figure 1 Flow schematic of the parameter identification method for the robotic arm provided by the embodiment of the present application Figure 1 , such as Figure 1 shown, the method may include: S101. Obtain the dynamic inertia parameters of the robotic arm, the non-linear friction parameters of the robotic arm, the joint motion information of the robotic arm, and the total dynamic torque.

[0029] In this embodiment, the robotic arm is composed of multiple joints, and the multiple joints are rigidly connected by connecting rods. The dynamic inertia parameters of the robotic arm are the parameters of the multiple connecting rods, and the obtained dynamic inertia parameters and non-linear friction parameters are the initial dynamic inertia parameters and the initial non-linear friction parameters .

[0030] Among them, the initial dynamic inertia parameters may include the initial mass of the i-th connecting rod, the initial centroid coordinates of the i-th connecting rod, and the initial inertia tensor matrix of the i-th connecting rod, and the initial dynamic inertia parameters .

[0031] In some embodiments, the initial mass , the initial centroid coordinates , and the initial inertia tensor matrix of each connecting rod may be obtained from a Unified Robot Description Format (URDF) file obtained from 3D software, so as to obtain the initial dynamic inertia parameters . The initial dynamic inertia parameters obtained from the URDF file can largely restore the physical properties of the robotic arm, so as to solve the problem of insufficient physical consistency in parameter optimization from the root cause.

[0032] Furthermore, according to the initial dynamic inertia parameters , set the optimization boundary of the dynamic inertia parameters as , and ensure that the optimization of the dynamic inertia parameters changes within a reasonable range by constraining the search space.

[0033] The initial non-linear friction parameters may also be referred to as shape parameters, and may include the Stribeck curve shape parameters , which is used to control the friction attenuation rate of the Stribeck curves of each joint to adapt to the friction characteristics of different joints. Among them, the Stribeck curve is used to represent the velocity-dependent friction of the joint at low speeds, and the initial non-linear friction parameter .

[0034] In some embodiments, the initial non-linear friction parameter The friction parameters of each joint in can be preset values, for example, it can be 1.

[0035] Control the manipulator to move according to the pre-designed excitation trajectory, and collect the joint motion information and the total dynamic torque of the manipulator moving based on the excitation trajectory . Among them, the joint motion information may include: joint position , speed and acceleration .

[0036] In some embodiments, a fifth-order Fourier series is used to design the excitation trajectory.

[0037] S102. According to the dynamic inertia parameter, the non-linear friction parameter, and the joint motion information, split the total dynamic torque to obtain a split inertia torque and a split friction torque.

[0038] In this embodiment, the total dynamic torque of the manipulator is decomposed to determine that the total dynamic torque of the manipulator consists of two parts: an inertia term and a friction term.

[0039] According to the dynamic inertia parameter and the joint motion information, determine the inertia term, split the inertia term from the total dynamic torque , and the remaining part is used as the split friction torque .

[0040] The friction force received by the joint of the manipulator includes linear friction force and non-linear friction force. According to the non-linear friction parameter, the joint motion information, and the split friction torque , split the friction term from the total dynamic torque , and the remaining part is used as the split inertia torque .

[0041] S103. According to the split inertia torque and the split friction torque, perform cyclic iterative optimization on the dynamic inertia parameter and the non-linear friction parameter until the preset convergence condition is met.

[0042] In this embodiment, optimize the dynamic inertia parameter according to the split inertia torque to obtain the optimized dynamic inertia parameter. According to the optimized dynamic inertia parameter, re-execute the above S102 to determine the new split friction torque , according to the non-linear friction parameters, joint motion information, and the newly split frictional torque , optimize the non-linear friction parameters until the optimized non-linear friction parameters are obtained.

[0043] According to the optimized non-linear friction parameters, re-execute the above S102 to determine the new split inertial torque , according to the new split inertial torque , loop and execute S103 until the optimized dynamic inertial parameters and the optimized non-linear friction parameters meet the preset convergence conditions.

[0044] In some embodiments, the dynamic inertial parameters and the non-linear friction parameters meet different convergence conditions respectively.

[0045] In some embodiments, the dynamic inertial parameters are optimized by an outer loop, and the non-linear friction parameters are optimized by an inner loop, that is, for each optimization of the dynamic inertial parameters, the non-linear friction parameters are optimized in a round of loop until the non-linear friction parameters converge, and then the dynamic inertial parameters are optimized for the next time, and the non-linear friction parameters are optimized in the next round of loop.

[0046] S104. Obtain the target dynamic inertial parameters and the target non-linear friction parameters when meeting the preset convergence conditions, and the target dynamic inertial parameters and the target non-linear friction parameters are used for dynamic control of the robotic arm.

[0047] In this embodiment, the dynamic inertial parameters when meeting the preset convergence conditions are determined as the target dynamic inertial parameters, and the non-linear friction parameters when meeting the preset convergence conditions are determined as the target non-linear friction parameters.

[0048] During the process of controlling the robotic arm, according to the target dynamic inertial parameters and the target non-linear friction parameters, use the dynamic model of the robotic arm to control each joint of the robotic arm.

[0049] The parameter identification method of the robotic arm provided in the above embodiment adopts a two-stage coupled optimization framework to jointly and alternately iteratively optimize the dynamic inertial parameters and the non-linear friction parameters. Each time the dynamic inertial parameters are optimized, the non-linear friction parameters are immediately optimized, forming a dynamic decoupling between the parameters, avoiding the problem of error accumulation, ensuring the accurate identification of the dynamic inertial parameters and the non-linear friction parameters, and improving the accuracy of the dynamic inertial parameters and the non-linear friction parameters; the optimized dynamic inertial parameters and non-linear friction parameters are full parameters, which are more convenient to be applied to the dynamic model.

[0050] In a possible implementation manner, Figure 2 is the flow schematic of the parameter identification method of the robotic arm provided in the embodiment of the present application Figure 2 , asFigure 2 As shown, the process of splitting the total dynamic torque into the split inertial torque and the split frictional torque according to the dynamic inertia parameters, the non-linear friction parameters, and the joint motion information in S102 may include: S201. Determine the split frictional torque from the total dynamic torque according to the dynamic inertia parameters and the joint motion information.

[0051] In this embodiment, the expression of decomposing the total dynamic torque of the robotic arm into an inertial term and a frictional term can be expressed as:

[0052] where is the kinematic regression matrix, that is, the observation matrix of the inertial term part, are the dynamic inertia parameters, and the information about each parameter of the frictional term part will be introduced in the following part.

[0053] According to the joint position , velocity and acceleration , model and linearize through the Newton-Euler method to determine the kinematic regression matrix , and calculate the predicted inertial torque according to the initial dynamic inertia parameters and the kinematic regression matrix . Split the predicted inertial torque from the total dynamic torque , and the remaining part is the split frictional torque .

[0054] Specifically, the calculation formula of the split frictional torque is as follows: [[ID=4)),

[0055] S202. Determine the split inertial torque from the total dynamic torque according to the split frictional torque, the non-linear friction parameters, and the friction regression matrix of the joint motion information.

[0056] In this embodiment, decompose the pre-established robotic arm friction model, and design the friction regression matrix of the non-linear friction parameters and the velocity in the joint motion information to convert the non-linear term of the robotic arm friction model into an explicit expression of the linear friction parameter . Among them, the friction regression matrix is a function of the non-linear friction parameter and the joint velocity .

[0057] Determine the predicted frictional torque according to the split frictional torque and the friction regression matrix, and from the total dynamic torque The predicted frictional torque is split out, and the remaining part is the split inertial torque 。

[0058] The parameter identification method for the robotic arm provided in the above embodiment, based on the dynamic inertial parameters, the non-linear friction parameters, and the joint motion information, splits the total dynamic torque into a split inertial torque and a split frictional torque, so as to perform differential optimization for the dynamic inertial parameters and the non-linear friction parameters, decouple the dynamic inertial parameters and the non-linear friction parameters dynamically, and reduce the parameter identification error.

[0059] In some embodiments, the method may further include: Construct a robotic arm friction model based on the non-linear friction parameters, the joint motion information, and the linear friction parameters; determine a friction regression matrix according to the robotic arm friction model.

[0060] In this embodiment, the expression of the constructed robotic arm friction model can be expressed as:

[0061] Among them, the initial linear friction parameter , is the Coulomb friction coefficient of the j-th joint, which is a constant friction coefficient independent of the velocity direction, is the Stribeck effect coefficient of the j-th joint, which is a velocity-dependent friction coefficient at low speeds, is the viscous friction coefficient of the j-th joint, which is a friction coefficient proportional to the velocity.

[0062] Decompose the above robotic arm friction model, design a friction regression matrix ,convert the non-linear term of the robotic arm friction model into an explicit expression of the linear friction parameter , then the frictional term part in the total dynamic torque of the robotic arm can be expressed as 。

[0063] In some embodiments, Figure 3 is the process flow diagram of the parameter identification method for the robotic arm provided in the embodiment of the present application Figure 3 ,as Figure 3 shown, the process of determining the split inertial torque from the total dynamic torque in S202 according to the split frictional torque, the non-linear friction parameters, and the friction regression matrix of the joint motion information may include: S301. Calculate the first linear friction parameter according to the split frictional torque and the friction regression matrix.

[0064] In this embodiment, since the frictional torque can be expressed based on the linear friction parameter, the non-linear friction parameter, and the joint velocity, when determining the split frictional torque , Friction regression matrix of non-linear friction parameters and joint motion information After that, the first linear friction parameter can be calculated .

[0065] In some embodiments, the least squares method and the initial non-linear friction parameters are used to fit the split frictional torque to obtain the first linear friction parameter . The first linear friction parameter is the initial linear friction parameter. After optimizing the dynamic inertia parameters subsequently, a new split frictional torque is calculated based on the optimized dynamic inertia parameters, and the first linear friction parameter is updated based on the new split frictional torque .

[0066] Exemplarily, the calculation formula of the first linear friction parameter can be expressed as:

[0067] S302. Determine the split inertial torque from the total dynamic torque according to the first linear friction parameter and the friction regression matrix.

[0068] In this embodiment, according to the first linear friction parameter and the friction regression matrix , a predicted friction matrix is calculated, the predicted frictional torque is split from the total dynamic torque, and the remaining part is the split inertial torque .

[0069] Exemplarily, the calculation formula of the split inertial torque can be expressed as:

[0070] The parameter identification method of the robotic arm provided in the above embodiment uses a robotic arm friction model to determine a friction regression matrix, so as to convert non-linear friction parameters into an explicit expression of linear friction parameters, and significantly reduce the error of modeling the friction mutation phenomenon in the low-speed region based on the non-linear friction parameters.

[0071] In a possible implementation manner, Figure 4 is the process schematic of the parameter identification method of the robotic arm provided in the embodiment of the present application Figure 4 , as Figure 4 shown, the above S103 process of cyclically iteratively optimizing the dynamic inertia parameters and non-linear friction parameters according to the split inertial torque and the split frictional torque until the preset convergence condition is met may include: S401. Optimize the dynamic inertial parameters through an outer loop iteration based on the dynamic inertial parameters, joint motion information, and decomposed inertial torque until the first convergence condition is met.

[0072] In this embodiment, first define the outer loop to optimize the dynamic inertial parameters. Based on the joint position , velocity and acceleration , perform modeling through the Newton-Euler method and linearize it to determine the kinematic regression matrix . Based on the current dynamic inertial parameters and the kinematic regression matrix , determine the predicted inertial torque. Based on the difference between the predicted inertial torque and the decomposed inertial torque , use the interior point method to optimize the dynamic inertial parameters to obtain the optimized dynamic inertial parameters .

[0073] Exemplarily, the objective function for optimizing the dynamic parameters can be expressed as:

[0074] where m is the number of links of the robotic arm. The first convergence condition for the dynamic inertial parameters is that the absolute value of the difference between the optimized dynamic inertial parameters and the dynamic inertial parameters before optimization is greater than the first preset threshold. Exemplarily, the first convergence condition can be: .

[0075] After optimizing the dynamic inertial parameters once, optimize the non-linear friction parameters based on the optimized dynamic inertial parameters . After optimizing the non-linear friction parameters once, calculate the new decomposed inertial torque based on the optimized non-linear friction parameters, and perform the next optimization on the dynamic inertial parameters.

[0076] S402. According to the optimized dynamic inertial parameters each time and the joint motion information, re-decompose the total dynamic torque to obtain a new decomposed frictional torque.

[0077] In this embodiment, after optimizing the dynamic inertial parameters, the decomposed frictional torque split from the total dynamic torque will change. Specifically, based on the optimized dynamic inertial parameters and the kinematic regression matrix , calculate the new predicted inertial torque, split the new predicted inertial torque from the total dynamic torque , and the remaining part is the new decomposed frictional torque .

[0078] Exemplarily, the calculation formula of the new split frictional torque can be expressed as:

[0079] S403. Each time the dynamic inertia parameters are optimized, the inner-loop iterative optimization of the non-linear friction parameters is performed according to the non-linear friction parameters, joint motion information, and the new split frictional torque until the second convergence condition is satisfied.

[0080] In this embodiment, the inner-loop optimization of the non-linear friction parameters is defined. According to the non-linear friction parameters and joint motion information, the predicted frictional torque is calculated. According to the difference between the predicted frictional torque and the new split frictional torque , the interior point method is used to optimize the non-linear friction parameters to obtain the optimized non-linear friction parameters .

[0081] Among them, for the outer-loop optimization and the inner-loop optimization, after one round of iterative optimization of the non-linear friction parameters until convergence, the next iterative optimization of the dynamic inertia parameters is performed. That is, each time the dynamic inertia parameters are optimized, the non-linear friction parameters are iteratively optimized for one round until convergence. Finally, when the dynamic inertia parameters converge, the optimization of the dynamic inertia parameters and the non-linear friction parameters is completed.

[0082] In some embodiments, the process of S403 performing the inner-loop iterative optimization of the non-linear friction parameters according to the non-linear friction parameters, joint motion information, and the new split frictional torque until the second convergence condition is satisfied may include: Calculating the second linear friction parameter according to the friction regression matrix and the new split frictional torque; performing cyclic iterative optimization of the non-linear friction parameters according to the second linear friction parameter, the friction regression matrix, and the new split frictional torque until the second convergence condition is satisfied.

[0083] In this embodiment, after calculating the new split frictional torque based on the optimized dynamic inertia parameters , the first linear friction parameter needs to be updated to the second linear friction parameter so as to re-determine the new split inertial torque from the total dynamic torque.

[0084] Exemplarily, the calculation formula of the second linear friction parameter can be expressed as:

[0085] According to the second linear friction parameter and the friction regression matrix, calculate a new predicted friction matrix. Based on the difference between the new predicted friction matrix and the new split frictional torque , use the interior point method to optimize the non-linear friction parameter to obtain the optimized non-linear friction parameter .

[0086] Exemplarily, the objective function for optimizing the non-linear friction parameter can be expressed as:

[0087] wherein, n is the number of joints of the robotic arm, and the second convergence condition of the non-linear friction parameter is that the absolute value of the difference between the optimized non-linear friction parameter and the non-linear friction parameter before optimization is greater than a second preset threshold. Exemplarily, the first convergence condition can be: .

[0088] In some embodiments, the method may further include: According to the optimized non-linear friction parameter, re-determine the split inertial torque from the total dynamic torque.

[0089] In this embodiment, according to the optimized non-linear friction parameter and the second linear friction parameter related to the joint velocity, calculate the predicted frictional torque, split a new predicted frictional torque from the total dynamic torque , and the remaining part is the new split inertial torque .

[0090] Exemplarily, the calculation formula for the new split inertial torque can be expressed as:

[0091] The parameter identification method for the robotic arm provided in the above embodiment respectively identifies the dynamic inertial parameter and the non-linear friction parameter through the double-loop method, ensuring that the dynamic inertial parameter and the non-linear friction parameter can be accurately identified, and facilitating the separate application of the dynamic inertial parameter and the non-linear friction parameter to their respective required scenarios.

[0092] Figure 5 is the flowchart of the parameter identification method for the robotic arm provided in the embodiment of the present application. As Figure 5 shown, the process of the parameter identification method may include: S501. Initialize parameters; wherein, initializing parameters includes obtaining initial dynamic inertial parameters, initial non-linear friction parameters, joint motion information, and total dynamic torque.

[0093] S502. Calculate the initial split inertia torque and split frictional torque.

[0094] S503. Optimize the dynamic inertia parameters using the interior point method.

[0095] S504. Calculate the new split frictional torque and linear friction parameters using the optimized dynamic inertia parameters.

[0096] S505. Optimize the non - linear friction parameters using the interior point method.

[0097] S506. Determine whether the optimized non - linear friction parameters converge. If not, jump to S507; if so, jump to S508.

[0098] S507. Calculate the new linear friction parameters based on the optimized non - linear friction parameters, and then jump to S505.

[0099] S508. Determine whether the optimized dynamic inertia parameters converge. If not, jump to S509; if so, jump to S510.

[0100] S509. Recalculate the split inertia torque based on the optimized non - linear friction parameters and the new linear friction parameters, and then jump to S503.

[0101] S510. End, and output the target dynamic inertia parameters and target non - linear friction parameters.

[0102] Based on the above method embodiments, an embodiment of the present application further provides a parameter identification device for a robotic arm. Figure 6 is a structural schematic diagram of the parameter identification device for the robotic arm provided by the embodiment of the present application. As Figure 6 shown, the device may include: An information acquisition module 601, configured to acquire the dynamic inertia parameters of the robotic arm, the non - linear friction parameters of the robotic arm, the joint motion information of the robotic arm, and the total dynamic torque; A torque splitting module 602, configured to split the total dynamic torque into a split inertia torque and a split frictional torque according to the dynamic inertia parameters, the non - linear friction parameters, and the joint motion information; An iterative optimization module 603, configured to perform cyclic iterative optimization on the dynamic inertia parameters and the non - linear friction parameters according to the split inertia torque and the split frictional torque until a preset convergence condition is met; A parameter acquisition module 604, configured to acquire the target dynamic inertia parameters and the target non - linear friction parameters when the preset convergence condition is met, and the target dynamic inertia parameters and the target non - linear friction parameters are used for dynamic control of the robotic arm.

[0103] Optionally, the torque splitting module 602 is specifically configured to determine a split frictional torque from the total dynamic torque according to the dynamic inertia parameters and the joint motion information; and determine a split inertial torque from the total dynamic torque according to the split frictional torque, the non-linear friction parameters, and the friction regression matrix of the joint motion information.

[0104] Optionally, the torque splitting module 602 is further configured to calculate a first linear friction parameter according to the split frictional torque and the friction regression matrix; and determine a split inertial torque from the total dynamic torque according to the first linear friction parameter and the friction regression matrix.

[0105] Optionally, the iterative optimization module 603 is specifically configured to perform an outer-loop iterative optimization on the dynamic inertia parameters according to the dynamic inertia parameters, the joint motion information, and the split inertial torque until a first convergence condition is satisfied; re-split the total dynamic torque according to the optimized dynamic inertia parameters and the joint motion information to obtain a new split frictional torque; and perform an inner-loop iterative optimization on the non-linear friction parameters according to the non-linear friction parameters, the joint motion information, and the new split frictional torque until a second convergence condition is satisfied.

[0106] Optionally, the iterative optimization module 603 is further configured to calculate a second linear friction parameter according to the friction regression matrix and the new split frictional torque; and perform a cyclic iterative optimization on the non-linear friction parameters according to the second linear friction parameter, the friction regression matrix, and the new split frictional torque until the second convergence condition is satisfied.

[0107] Optionally, the torque splitting module 602 is further configured to re-determine a split inertial torque from the total dynamic torque according to the optimized non-linear friction parameters.

[0108] Optionally, the device may further include: a matrix design module, configured to construct a manipulator friction model based on the non-linear friction parameters, the joint motion information, and the linear friction parameters; and determine a friction regression matrix according to the manipulator friction model.

[0109] The above device is used to execute the method provided in the foregoing embodiment, and its implementation principle and technical effects are similar, which will not be elaborated here.

[0110] The above-mentioned modules may be one or more integrated circuits configured to implement the above methods. For example: one or more Application Specific Integrated Circuits (ASICs), or, one or more microprocessors, or, one or more Field Programmable Gate Arrays (FPGAs), etc. Again, when a certain module above is implemented in the form of a processing element scheduler code, the processing element may be a general-purpose processor, such as a Central Processing Unit (CPU) or other processors that can call program code. Again, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0111] Figure 7 The figure is a schematic diagram of the electronic device provided by the embodiment of the present application. The electronic device 700 includes: a processor 701, a storage medium 702, and a bus. The storage medium 702 stores program instructions executable by the processor 701. When the electronic device 700 runs, the processor 701 communicates with the storage medium 702 through the bus, and the processor 701 executes the program instructions to execute the above method embodiment. The specific implementation manners and technical effects are similar and will not be elaborated here.

[0112] Optionally, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, it executes the above method embodiment.

[0113] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other may be through some interfaces. The indirect couplings or communication connections of the devices or units may be electrical, mechanical or other forms.

[0114] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0115] In addition, in each embodiment of the present application, each functional unit can be integrated into a processing unit, can exist physically alone for each unit, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware, or in the form of a hardware plus software functional unit.

[0116] The above integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above software functional unit is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (English: processor) to execute some steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (English: Read-Only Memory, abbreviated as: ROM), random access memories (English: Random Access Memory, abbreviated as: RAM), magnetic disks, or optical discs that can store program codes.

[0117] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A parameter identification method for a robotic arm, characterized in that, The method includes: Obtaining the dynamic inertia parameters of the robotic arm, the non-linear friction parameters of the robotic arm, the joint motion information of the robotic arm, and the total dynamic torque; According to the dynamic inertia parameters, the non-linear friction parameters, and the joint motion information, splitting the total dynamic torque to obtain a split inertia torque and a split friction torque; According to the split inertia torque and the split friction torque, performing cyclic iterative optimization on the dynamic inertia parameters and the non-linear friction parameters until a preset convergence condition is satisfied; Obtaining the target dynamic inertia parameters and the target non-linear friction parameters when the preset convergence condition is satisfied, where the target dynamic inertia parameters and the target non-linear friction parameters are used for dynamic control of the robotic arm.

2. The method according to claim 1, wherein The splitting the total dynamic torque according to the dynamic inertia parameters, the non-linear friction parameters, and the joint motion information to obtain a split inertia torque and a split friction torque includes: Determining the split friction torque from the total dynamic torque according to the dynamic inertia parameters and the joint motion information; Determining the split inertia torque from the total dynamic torque according to the split friction torque, the non-linear friction parameters, and the friction regression matrix of the joint motion information.

3. The method according to claim 2, wherein The determining the split inertia torque from the total dynamic torque according to the split friction torque, the non-linear friction parameters, and the friction regression matrix of the joint motion information includes: Calculating a first linear friction parameter according to the split friction torque and the friction regression matrix; Determining the split inertia torque from the total dynamic torque according to the first linear friction parameter and the friction regression matrix.

4. The method according to claim 2, wherein The performing cyclic iterative optimization on the dynamic inertia parameters and the non-linear friction parameters according to the split inertia torque and the split friction torque until a preset convergence condition is satisfied includes: Performing outer-loop cyclic iterative optimization on the dynamic inertia parameters according to the dynamic inertia parameters, the joint motion information, and the split inertia torque until a first convergence condition is satisfied; According to the optimized dynamic inertia parameters each time and the joint motion information, re-splitting the total dynamic torque to obtain a new split friction torque; Each time the dynamic inertia parameters are optimized, performing inner-loop cyclic iterative optimization on the non-linear friction parameters according to the non-linear friction parameters, the joint motion information, and the new split friction torque until a second convergence condition is satisfied.

5. The method according to claim 4, wherein The performing inner-loop cyclic iterative optimization on the non-linear friction parameters according to the non-linear friction parameters, the joint motion information, and the new split friction torque until a second convergence condition is satisfied includes: Calculating a second linear friction parameter according to the friction regression matrix and the new split friction torque; Performing cyclic iterative optimization on the non-linear friction parameters according to the second linear friction parameter, the friction regression matrix, and the new split friction torque until the second convergence condition is satisfied.

6. The method according to claim 4, wherein The method further includes: Redetermine the split inertia torque from the total dynamic torque according to the optimized non-linear friction parameter.

7. The method according to claim 2, wherein The method further includes: Constructing a manipulator friction model based on the non-linear friction parameter, joint motion information and linear friction parameter; Determining the friction regression matrix according to the manipulator friction model.

8. A parameter identification device for a robotic arm, characterized in that, The device includes: An information acquisition module, configured to acquire the dynamic inertia parameter of the manipulator, the non-linear friction parameter of the manipulator, the joint motion information of the manipulator, and the total dynamic torque; A torque splitting module, configured to split the total dynamic torque according to the dynamic inertia parameter, the non-linear friction parameter, and the joint motion information to obtain a split inertia torque and a split friction torque; An iterative optimization module, configured to perform cyclic iterative optimization on the dynamic inertia parameter and the non-linear friction parameter according to the split inertia torque and the split friction torque until a preset convergence condition is met; A parameter acquisition module, configured to acquire a target dynamic inertia parameter and a target non-linear friction parameter when the preset convergence condition is met, where the target dynamic inertia parameter and the target non-linear friction parameter are used for dynamic control of the manipulator.

9. An electronic device, characterized in that, including: A processor, a storage medium and a bus, where the storage medium stores program instructions executable by the processor. When the electronic device runs, the processor communicates with the storage medium through the bus, and the processor executes the program instructions to perform the steps of the parameter identification method of the manipulator according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the storage medium, and when the computer program is run by the processor, it performs the steps of the parameter identification method of the manipulator according to any one of claims 1 to 7.

Images