Motion control simulation method of mechanical arm and electronic device

Abstract

The application relates to the technical field of mechanical arm control, and particularly provides a motion control simulation method of a mechanical arm and an electronic device, and aims to solve the technical problems of insufficient authenticity and large difference between the existing method and the dynamic characteristics of a real mechanical arm. For the purpose, the motion control simulation method of the mechanical arm comprises the following steps: determining an intermediate torque instruction based on a user-given instruction; obtaining a final execution torque based on the intermediate torque instruction; applying the final execution torque and a preset excitation trajectory to a real mechanical arm, and collecting running response data of the real mechanical arm; obtaining an optimal parameter set based on the running response data; and determining final simulation parameters based on the optimal parameter set and the user-given instruction. The consistency between the simulation mechanical arm and the real mechanical arm is significantly improved, and the error between the simulation output and the real response is effectively reduced.

Description

Technical Field

[0001] This invention relates to the field of robotic arm control technology, specifically providing a motion control simulation method and electronic device for a robotic arm. Background Technology

[0002] In robot simulation and practical applications, upper-level controllers typically provide control inputs in various ways, such as outputting target joint position, target joint velocity, or target joint torque. However, existing simulation platforms often employ simplified driving methods, such as providing only a model based on robot structural information or using fixed proportional-derivative (PD) parameters for driving in the physics engine.

[0003] While these solutions enable robots to move in simulated environments, the dynamic response of their actuators, their characteristics under varying loads, and their behavior within the control loop still differ significantly from those of actual robots. Specifically, existing technologies suffer from the following drawbacks: First, the existing structural models lack descriptions of actuator dynamic behaviors such as motor delay, bandwidth, friction, and torque limitations, leading to significant discrepancies between simulation and real control processes. Second, different control input interfaces are not standardized, and simulation schemes are often strongly tied to a particular type of control method, lacking versatility. Third, when the load at the end effector of the robotic arm changes, simulation models relying solely on fixed drive parameters cannot reflect the differences in the response of the real robotic arm under different working conditions, resulting in a severe decrease in simulation accuracy. Therefore, control strategies or parameters debugged based on such simulation environments perform poorly and lack stability when transferred to real systems, exhibiting a significant "sim-to-real gap." Summary of the Invention

[0004] To overcome the aforementioned shortcomings, this application is proposed to provide solutions, or at least partially solutions, to the technical problems of insufficient realism and significant differences from the dynamic characteristics of real robotic arms in existing methods. This application provides a motion control simulation method and electronic device for a robotic arm.

[0005] In a first aspect, this application provides a motion control simulation method for a robotic arm, the method comprising:

[0006] The intermediate torque command is determined based on the user-provided instructions.

[0007] The final execution torque is obtained based on the intermediate torque command;

[0008] The final execution torque and the preset excitation trajectory are applied to the real robotic arm, and the operation response data of the real robotic arm is collected.

[0009] Obtain the optimal parameter set based on the aforementioned operational response data;

[0010] The final simulation parameters are determined based on the optimal parameter set and the user-given instructions.

[0011] In one embodiment of the motion control simulation method for the robotic arm in this application, determining the intermediate torque command based on the user-given command includes:

[0012] Obtain the instruction type of the user-given command, wherein the instruction type includes target joint position, target joint velocity, or target joint torque;

[0013] When the user-given command type is a target joint position, the intermediate torque command is calculated based on the position error between the current joint position and the target joint position.

[0014] When the user-given command type is a target joint speed, the intermediate torque command is calculated based on the speed error between the current joint speed and the target joint speed.

[0015] When the user-given instruction type is target joint torque, the target joint torque is used as the intermediate torque instruction.

[0016] In one embodiment of the motion control simulation method for the robotic arm in this application, the instruction type further includes an end-effector pose instruction;

[0017] The step of determining the intermediate torque command based on the user-given command further includes: when the command type of the user-given command is an end-effector pose command, determining any one of the target joint position, target joint velocity, or target joint torque based on the end-effector pose command, and determining the intermediate torque command based on any one of the target joint position, target joint velocity, or target joint torque.

[0018] In one embodiment of the motion control simulation method for the robotic arm in this application, obtaining the final execution torque based on the intermediate torque command includes:

[0019] Determine the feedforward compensation term, which includes at least one of gravity compensation term, load compensation term, and friction compensation term;

[0020] The intermediate torque command is added to the feedforward compensation term to obtain the compensated torque command;

[0021] Dynamic constraints are applied to the compensated torque command to obtain the final execution torque.

[0022] In one embodiment of the motion control simulation method for the robotic arm in this application, the operational response data includes first operational response data and second operational response data;

[0023] The acquisition of operational response data from the actual robotic arm includes:

[0024] Under no-load conditions, the final execution torque and preset excitation trajectory are applied to the real robotic arm, and the first operation response data is collected;

[0025] Under load conditions, the final execution torque and preset excitation trajectory are applied to the real robotic arm, and second operation response data are collected.

[0026] In one embodiment of the motion control simulation method for the robotic arm of this application, obtaining the optimal parameter set based on the operational response data includes:

[0027] Establish the parameter set for the actuator model;

[0028] A loss function is constructed based on the parameter set and the runtime response data;

[0029] Optimize the loss function until it converges to obtain the optimal parameter set.

[0030] In one embodiment of the motion control simulation method for the robotic arm in this application, the step of constructing a loss function based on the parameter set and the operational response data includes:

[0031] The simulation output is obtained based on the parameter set;

[0032] The loss function is determined based on at least one of the position error term, velocity error term, torque error term, or constraint term between the simulation output and the running response data.

[0033] In one embodiment of the motion control simulation method for the robotic arm of this application, determining the final simulation parameters based on the optimal parameter set and the user-given instructions includes:

[0034] Obtain the joint states of the simulated robotic arm at each simulation time step;

[0035] Based on the optimal parameter set and the user-given instructions, determine the execution torque of the current simulation time step;

[0036] The joint state of the simulated robotic arm is updated based on the execution torque;

[0037] Error terms are determined based on the joint states of the simulated robotic arm;

[0038] The final simulation parameters are determined based on the error term.

[0039] In one embodiment of the motion control simulation method for the robotic arm in this application, the error term includes the maximum error of a single joint and the average error of a single joint.

[0040] The step of determining the final simulation parameters based on the error term includes: when the maximum error of a single joint is less than a first preset accuracy threshold and the average error of a single joint is less than a second preset accuracy threshold, the optimal parameter set is used as the final simulation parameters, wherein the first preset accuracy threshold is greater than or equal to the second preset accuracy threshold.

[0041] In a second aspect, an electronic device is provided, comprising:

[0042] At least one processor;

[0043] And, a memory communicatively connected to the at least one processor;

[0044] The memory stores a computer program, which, when executed by the at least one processor, is the aforementioned motion control simulation method for the robotic arm.

[0045] The above-described technical solutions of this application have at least one or more of the following beneficial effects:

[0046] The motion control simulation method for the robotic arm in this application includes: determining intermediate torque commands based on user-given instructions; obtaining the final execution torque based on the intermediate torque commands; applying the final execution torque and a preset excitation trajectory to the real robotic arm and collecting the operational response data of the real robotic arm; obtaining the optimal parameter set based on the operational response data; and determining the final simulation parameters based on the optimal parameter set and user-given instructions. By establishing a complete conversion link from user-given instructions to intermediate torque commands and then to the final execution torque, and combining the operational response data of the real robotic arm under the preset excitation trajectory to identify the optimal parameter set of the actuator model, and finally determining the final simulation parameters based on the optimal parameter set and user-given instructions, the consistency between the simulated robotic arm and the real robotic arm is significantly improved, and the error between the simulation output and the real response is effectively reduced. Attached Figure Description

[0047] The disclosure of this application will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this application. Furthermore, similar numbers in the drawings are used to denote similar components, wherein:

[0048] Figure 1 This is a schematic diagram of the main flow of the motion control simulation method for a robotic arm in one embodiment of this application;

[0049] Figure 2 This is a schematic diagram of the main structure of a simulation robotic arm actuator modeling system in one embodiment of this application;

[0050] Figure 3 This is a schematic diagram of the main structure of an electronic device in one embodiment of this application. Detailed Implementation

[0051] Some embodiments of this application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of this application and are not intended to limit the scope of protection of this application.

[0052] In the description of this application, "module" and "processor" can include hardware, software, or a combination of both. A module can include hardware circuitry, various suitable sensors, communication ports, memory, and can also include software components, such as program code, or a combination of software and hardware. A processor can be a central processing unit, microprocessor, image processor, digital signal processor, or any other suitable processor. The processor has data and / or signal processing capabilities. The processor can be implemented in software, in hardware, or a combination of both. Non-transitory computer-readable storage media includes any suitable medium capable of storing program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc. The term "A and / or B" means all possible combinations of A and B, such as only A, only B, or A and B. The terms "at least one A or B" or "at least one of A and B" have a similar meaning to "A and / or B" and can include only A, only B, or A and B. The singular terms "a" or "this" can also include plural forms.

[0053] See appendix Figure 1 , Figure 1 This is a schematic flowchart of the main steps of a motion control simulation method for a robotic arm according to an embodiment of this application.

[0054] like Figure 1 As shown, the motion control simulation method for the robotic arm in this embodiment mainly includes the following steps S10-S50.

[0055] Step S10: Determine the intermediate torque command based on the user-given instructions.

[0056] Step S20: Obtain the final execution torque based on the intermediate torque command.

[0057] Step S30: Apply the final execution torque and preset excitation trajectory to the real robotic arm and collect the operation response data of the real robotic arm.

[0058] Step S40: Obtain the optimal parameter set based on the running response data.

[0059] Step S50: Determine the final simulation parameters based on the optimal parameter set and user-given instructions.

[0060] Based on steps S10-S50 above, the intermediate torque command is first determined based on the user-given instruction; the final execution torque is obtained based on the intermediate torque command; the final execution torque and the preset excitation trajectory are applied to the real robotic arm, and the operational response data of the real robotic arm is collected; the optimal parameter set is obtained based on the operational response data; and the final simulation parameters are determined based on the optimal parameter set and the user-given instruction. By establishing a complete conversion link from the user-given instruction to the intermediate torque command and then to the final execution torque, and combining the operational response data of the real robotic arm under the preset excitation trajectory to identify the optimal parameter set of the actuator model, and finally determining the final simulation parameters based on the optimal parameter set and the user-given instruction, the consistency between the simulated robotic arm and the real robotic arm is significantly improved, and the error between the simulation output and the real response is effectively reduced.

[0061] Figure 2 This is a schematic diagram of a simulation robotic arm actuator modeling system according to an embodiment of this application. In the diagram, the upper-level controller generates joint control commands (cmd) based on user instructions and inputs these commands to the actuator adapter. After receiving the joint control commands, the actuator adapter calculates intermediate torque commands, performs feedforward compensation and dynamic constraint processing, and outputs the final execution torque to the lower-level static model. The static model includes a robot model and a physics engine. The robot model defines the kinematic and dynamic parameters of the robotic arm, and the physics engine performs dynamic calculations based on the received execution torque, outputs the joint state at each simulation time step, and feeds it back to the actuator adapter, forming a closed-loop execution for each simulation time step.

[0062] The following provides further explanation of steps S10 to S50.

[0063] Specifically, regarding step S10 above, in one specific embodiment of this application, determining the intermediate torque command based on the user-given command includes: obtaining the command type of the user-given command, where the command type includes target joint position, target joint velocity, or target joint torque; when the command type of the user-given command is target joint position, calculating the intermediate torque command based on the position error between the current joint position and the target joint position; when the command type of the user-given command is target joint velocity, calculating the intermediate torque command based on the velocity error between the current joint velocity and the target joint velocity; and when the command type of the user-given command is target joint torque, using the target joint torque as the intermediate torque command.

[0064] Specifically, the actuator adapter receives and processes various types of user-given commands, mapping them uniformly to joint-level intermediate torque commands. First, it needs to obtain the command type of the user-given command, which can include target joint position, target joint velocity, or target joint torque. When the user-given command type is target joint position, the actuator adapter calculates the intermediate torque command based on the position error between the current joint position and the target joint position. For example, the calculation process can be represented as:

[0065] in, The target joint location, This is the current joint position. For the target joint velocity, The current joint velocity, This is an intermediate torque command; , , These are the parameters for the control mode.

[0066] When the user-provided instruction type is target joint speed, the actuator adapter calculates the intermediate torque instruction based on the speed error between the current joint speed and the target joint speed. For example, proportional control can be used for calculation.

[0067] in, This is the intermediate torque command. For the target joint velocity, The current joint velocity, These are the parameters for the control mode.

[0068] When the user-provided instruction type is target joint torque, the actuator adapter directly uses the target joint torque as the intermediate torque instruction, that is:

[0069] in, This is the intermediate torque command. The target joint torque.

[0070] In another embodiment, the instruction type further includes an end-effector pose instruction; determining an intermediate torque instruction based on a user-given instruction further includes: when the instruction type of the user-given instruction is an end-effector pose instruction, determining any one of the target joint position, target joint velocity, or target joint torque based on the end-effector pose instruction, and determining an intermediate torque instruction based on any one of the target joint position, target joint velocity, or target joint torque.

[0071] Specifically, when the user provides an end-effector pose command, the upper-level controller (such as an inverse kinematics (IK) controller or an operational space control (OSC) controller) first performs joint-level solving internally, converting it into any one of the following joint-level target commands: target joint position, target joint velocity, or target joint torque. The converted joint-level target command is then input to the actuator adapter. Subsequently, the actuator adapter determines the intermediate torque command based on the joint-level target command. In this way, regardless of the form of the command output by the upper-level controller, it can ultimately be unified into the joint torque execution channel, achieving unification and decoupling of the control input interface.

[0072] The above is a further explanation of step S10. Step S20 will be further explained below.

[0073] Regarding step S20, in a specific embodiment of this application, obtaining the final execution torque based on the intermediate torque command includes: determining a feedforward compensation term, wherein the feedforward compensation term includes at least one of gravity compensation term, load compensation term, and friction compensation term; adding the intermediate torque command to the feedforward compensation term to obtain a compensated torque command; and applying dynamic constraints to the compensated torque command to obtain the final execution torque.

[0074] Specifically, the feedforward compensation term is first determined. This term can include at least one of gravity compensation, load compensation, and friction compensation. Its calculation formula can be expressed as:

[0075] in, For feedforward compensation term, Indicates the gravity compensation term; This represents the load compensation term, used to reflect the influence of different end load masses and centroid distributions on the joint torque; This indicates friction compensation terms, including but not limited to viscous friction, Coulomb friction, and Stribeck-type friction.

[0076] It is understood that in other embodiments, in addition to gravity compensation, load compensation and friction compensation, the feedforward compensation term may also include compliance compensation, backlash compensation, temperature drift compensation or coupling term compensation.

[0077] Then, the intermediate torque command is added to the feedforward compensation term to obtain the compensated torque command, i.e.:

[0078] in, To compensate for the torque command, This is the intermediate torque command. This is a feedforward compensation term.

[0079] Finally, to simulate the physical characteristics of a real motor and driver, dynamic constraints are applied to the compensated torque command. This means inputting the command into the actuator's dynamic link and outputting the final execution torque. Dynamic constraints may include, but are not limited to: a delay module shifting the torque command backward by a fixed time to simulate drive link latency; a bandwidth module filtering out high-frequency components of the torque command using a low-pass filter to simulate motor response hysteresis or first / second-order dynamic characteristics; a torque limit module clamping the torque value of the command between a preset maximum and minimum value to simulate actuator saturation characteristics; a torque change rate limit module limiting the magnitude of torque change per unit time to prevent sudden torque changes; a speed limit module actively reducing the output torque based on the comparison between the current joint speed and the limit value to avoid overspeeding; and a friction model calculating resistance torques such as Coulomb friction, viscous friction, and the Stribeck effect based on the direction and magnitude of the current speed, subtracting these resistance torques from the command torque to simulate the friction characteristics of a real joint. Ultimately, the torque output after the above link processing is the final execution torque. Compared to directly applying the compensated torque command to the physics engine, this method more realistically reflects the dynamic response process of the actuator.

[0080] The above is a further explanation of step S20. Step S30 will be further explained below.

[0081] In one specific embodiment of this application, the operation response data includes first operation response data and second operation response data; collecting the operation response data of the real robotic arm includes: under no-load conditions, applying the final execution torque and the preset excitation trajectory to the real robotic arm and collecting the first operation response data; under load conditions, applying the final execution torque and the preset excitation trajectory to the real robotic arm and collecting the second operation response data.

[0082] Specifically, by applying preset excitation trajectories to a real robotic arm under at least two working conditions—no-load and loaded—and simultaneously collecting its operational response data, samples covering different working states are provided for subsequent parameter identification. The first operational response data is collected under no-load conditions, where the robotic arm's end effector has no or only a standard light load, primarily used to identify the actuator's basic dynamic parameters, such as delay, bandwidth, friction characteristics, and basic torque / speed limits. The second operational response data is collected under loaded conditions, where the end effector is fitted with tooling of different masses or at different center-of-gravity positions, primarily used to identify load-related compensation parameters (such as load compensation terms) and verify the model's response consistency under different working conditions.

[0083] The preset excitation trajectory can be selected from one or more of the following: single-joint sweep trajectory, multi-joint linkage trajectory, step response trajectory, sine wave, pseudo-random wave, chirp wave, combined trajectory, or typical task trajectory, to ensure that the dynamic characteristics of the robotic arm are fully excited.

[0084] The collected data can include joint position, velocity, acceleration, controller target commands, motor current or estimated torque, actual output torque, load information, and timestamps. By comparing the no-load and loaded data sets, the load-independent basic parameters and load-related compensation parameters in the actuator model can be optimized respectively, thus enabling the same simulation model to maintain a high-fidelity approximation of the real robotic arm under various working conditions.

[0085] It is understandable that, in addition to the no-load / loaded trajectory, the excitation trajectory can also be further supplemented with sampling data from different speed gears, different temperature conditions, different reducer wear conditions, or different end-tool conditions.

[0086] The above is a further explanation of step S30. Step S40 will be further explained below.

[0087] In one specific embodiment of this application, obtaining the optimal parameter set based on operational response data includes: establishing a parameter set for the actuator model; constructing a loss function based on the parameter set and operational response data; and optimizing the loss function until the loss function converges to obtain the optimal parameter set. Specifically, constructing the loss function based on the parameter set and operational response data includes: obtaining simulation output based on the parameter set; and determining the loss function based on at least one of the position error term, velocity error term, torque error term, or constraint term between the simulation output and the operational response data.

[0088] The actuator model can be an analytical physical model, a gray box model, a neural network model, a lookup table model, or a combination thereof, as long as it can characterize the joint response characteristics of a real robotic arm. No specific limitations are imposed on this.

[0089] Specifically, the first step is to establish a parameter set for the actuator model containing the parameters to be identified. This parameter set It may include at least one of the following: delay parameter, bandwidth parameter, proportional / integral / derivative parameters (Kp, Ki, Kd), torque saturation parameter, torque change rate parameter, velocity saturation parameter, gravity compensation parameter, load compensation parameter, friction parameter, and equivalent dynamic parameters of the motor.

[0090] Next, the actuator model configured with the parameter set is embedded into the complete simulation pipeline and run, outputting simulation results such as joint position, velocity, and torque (i.e., simulation output). Then, a loss function is constructed based on at least one of the position error term, velocity error term, torque error term, or constraint term between the simulation output and the running response data. For example, the loss function can be expressed as:

[0091] in, For loss function, For the actuator model parameter set, , , Based on the parameter set The simulation output, , , These are the measured values ​​corresponding to a real robotic arm. These are regularization terms or physical feasibility constraints. , , , These are the weights corresponding to each item.

[0092] Finally, the parameter set is continuously adjusted using optimization methods such as least squares, Bayesian optimization, genetic algorithms, or gradient optimization, so that the loss function value gradually decreases until convergence. The parameter set at this point is the optimal parameter set. This optimal parameter set enables the simulated robotic arm to match the dynamic response of the real robotic arm to the greatest extent under the same excitation trajectory, thus providing accurate model parameters for subsequent high-fidelity simulations.

[0093] The above is a further explanation of step S40. Step S50 will be further explained below.

[0094] In one embodiment of this application, determining the final simulation parameters based on the optimal parameter set and user-given instructions includes: reading the joint state of the current simulated robotic arm at each simulation time step; determining the execution torque of the current simulation time step according to the optimal parameter set and user-given instructions; updating the joint state of the simulated robotic arm based on the execution torque; determining error terms based on the joint state of the simulated robotic arm; and determining the final simulation parameters based on the error terms. The error terms include the maximum error per joint and the average error per joint. Determining the final simulation parameters based on the error terms includes: when the maximum error per joint is less than a first preset accuracy threshold and the average error per joint is less than a second preset accuracy threshold, using the optimal parameter set as the final simulation parameters, wherein the first preset accuracy threshold is greater than or equal to the second preset accuracy threshold.

[0095] The simulation time step is a fixed time interval (such as 1 millisecond, 2 milliseconds, etc.) that the physics engine advances in each iteration of calculation. Within each step, the actuator adapter recalculates the execution torque according to the current joint state and inputs it into the physics engine to form a closed-loop simulation.

[0096] The first and second preset precision thresholds can be pre-set values, which can be adaptively adjusted according to the actual scenario. For example, 5%, 6%, etc. can be used as examples of the first preset precision threshold, and 3%, 4%, etc. can be used as examples of the second preset precision threshold.

[0097] Specifically, the purpose of this step is to deploy the identified optimal parameter set into the closed-loop simulation environment and determine whether to use it as the final simulation parameters through accuracy verification. Unlike traditional simulation methods that write the target values ​​into the physics engine all at once, this application adopts an execution method that performs closed-loop calculations in each simulation time step. The specific process is as follows: In each simulation time step, the joint states of the current simulated robotic arm (such as the current joint position q and joint velocity dq) are first read. Then, based on the identified optimal parameter set and the target command given by the user (e.g., q_des), combined with the current joint state, the execution torque to be applied in the current simulation time step is determined through the complete calculation chain of the aforementioned steps S10 and S20. Subsequently, the execution torque is input into the physics engine to update the joint states of the simulated robotic arm. This process is repeated in each simulation time step, forming a tight closed-loop control, thereby accurately simulating the closed-loop dynamics, response lag, and the impact of changes in operating conditions present in the real system.

[0098] To quantitatively evaluate and confirm the realism of the simulation model, the final simulation parameters are determined based on error terms. First, error terms are determined based on the differences between the joint states of the simulated robotic arm and the response data of the real robotic arm. These error terms can include the maximum error and the average error of a single joint. The maximum error of a single joint refers to the peak value of the normalized error between the simulated and real responses throughout the entire test trajectory; the average error of a single joint refers to the average value of this normalized error over the entire time domain. Specifically, the instantaneous error at a single time point can be calculated first. Take all time points The maximum value can be used to obtain the maximum error of a single joint, by taking the maximum value at all time points. The average value of these values ​​can be used to obtain the average error of a single joint. This includes the instantaneous error at a single time point. It can be defined as:

[0099] in, This indicates the joint states of the simulated robotic arm. For the response data of a real robotic arm, This is the normalization scale.

[0100] Then, the final simulation parameters are determined based on the error term. In a preferred embodiment, when the maximum error of a single joint is less than a first preset accuracy threshold (e.g., 5%) and the average error of a single joint is less than a second preset accuracy threshold (e.g., 3%), the current optimal parameter set is confirmed to meet the realism requirements and is used as the final simulation parameters for subsequent practical simulation applications, such as high-fidelity simulation testing, control algorithm development, or digital twin deployment. If the maximum error and average error of a single joint do not meet the conditions, the weights or constraints of the loss function can be adjusted, the optimization algorithm can be changed, and the parameters can be re-identified based on the existing running response data to obtain a new optimal parameter set; or the data acquisition step can be returned to supplement the collection of more types or longer-term excitation trajectory data to enrich the training samples until the error indicators simultaneously meet the two preset accuracy thresholds. At this point, the optimal parameter set obtained can be used as the final simulation parameters.

[0101] This application can also construct a layered delivery system. For example, the first layer is a robot structural model (USD Only) that only delivers geometry, collision, inertia, joint hierarchy, and limit parameters. This layer only supports basic visualization and simple motion simulation, with limited control realism. The second layer adds actuator model parameter files and actuator adapters to the robot structural model, which can significantly improve the realism of joint-level simulation without binding to the upper-level controller. The third layer further adds a general controller library (such as IK controllers, joint position / velocity / torque controllers) to the second layer to meet the plug-and-play requirement. However, the controller library itself is not a necessary component of this application. The core is still the actuator adaptation and parameter identification mechanism.

[0102] It should be noted that although the steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effect of this application, different steps do not necessarily have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders, and these variations are all within the scope of protection of this application.

[0103] Those skilled in the art will understand that all or part of the processes in the method of the above-described embodiment can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium can include any entity or device capable of carrying computer program code, a medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0104] Furthermore, this application also provides an electronic device, which may include at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program, and when the computer program is executed by the at least one processor, it implements the motion control simulation method for the robotic arm of any of the above embodiments. See also Figure 3 As shown, Figure 3 The structure of an electronic device, including a processor 100 and a memory 200, is illustrated by way of example.

[0105] Furthermore, this application also provides a computer-readable storage medium. In one embodiment of the computer-readable storage medium according to this application, the computer-readable storage medium can be configured to store a program for performing the motion control simulation method of the robotic arm in the above-described method embodiments. This program can be loaded and run by a processor to implement the motion control simulation method of the robotic arm. For ease of explanation, only the parts related to the embodiments of this application are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of this application. The computer-readable storage medium can be a memory device formed by various electronic devices. Optionally, in the embodiments of this application, the computer-readable storage medium is a non-transitory computer-readable storage medium.

[0106] The technical solution of this application has been described in conjunction with the specific embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. Without departing from the principles of this application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of this application.

Claims

1. A motion control simulation method for a robotic arm, characterized in that, The method includes: The intermediate torque command is determined based on the user-provided instructions. The final execution torque is obtained based on the intermediate torque command; The final execution torque and the preset excitation trajectory are applied to the real robotic arm, and the operation response data of the real robotic arm is collected. Obtain the optimal parameter set based on the aforementioned operational response data; The final simulation parameters are determined based on the optimal parameter set and the user-given instructions.

2. The motion control simulation method for a robotic arm according to claim 1, characterized in that, The step of determining the intermediate torque command based on the user-given instruction includes: Obtain the instruction type of the user-given command, wherein the instruction type includes target joint position, target joint velocity, or target joint torque; When the user-given command type is a target joint position, the intermediate torque command is calculated based on the position error between the current joint position and the target joint position. When the user-given command type is a target joint speed, the intermediate torque command is calculated based on the speed error between the current joint speed and the target joint speed. When the user-given instruction type is target joint torque, the target joint torque is used as the intermediate torque instruction.

3. The motion control simulation method for a robotic arm according to claim 2, characterized in that, The instruction type also includes end-effector pose instructions; The step of determining the intermediate torque command based on the user-given command further includes: when the command type of the user-given command is an end-effector pose command, determining any one of the target joint position, target joint velocity, or target joint torque based on the end-effector pose command, and determining the intermediate torque command based on any one of the target joint position, target joint velocity, or target joint torque.

4. The motion control simulation method for a robotic arm according to claim 1, characterized in that, The step of obtaining the final execution torque based on the intermediate torque command includes: Obtain feedforward compensation terms, wherein the feedforward compensation terms include at least one of gravity compensation terms, load compensation terms, and friction compensation terms; The intermediate torque command is added to the feedforward compensation term to obtain the compensated torque command; Dynamic constraints are applied to the compensated torque command to obtain the final execution torque.

5. The motion control simulation method for a robotic arm according to claim 1, characterized in that, The operational response data includes first operational response data and second operational response data; The acquisition of operational response data from the actual robotic arm includes: Under no-load conditions, the final execution torque and preset excitation trajectory are applied to the real robotic arm, and the first operation response data is collected; Under load conditions, the final execution torque and preset excitation trajectory are applied to the real robotic arm, and second operation response data are collected.

6. The motion control simulation method for a robotic arm according to claim 1, characterized in that, The process of obtaining the optimal parameter set based on the runtime response data includes: Establish the parameter set for the actuator model; A loss function is constructed based on the parameter set and the runtime response data; Optimize the loss function until it converges to obtain the optimal parameter set.

7. The motion control simulation method for a robotic arm according to claim 6, characterized in that, The construction of the loss function based on the parameter set and the runtime response data includes: The simulation output is obtained based on the parameter set; The loss function is determined based on at least one of the position error term, velocity error term, torque error term, or constraint term between the simulation output and the running response data.

8. The motion control simulation method for a robotic arm according to claim 1, characterized in that, The process of determining the final simulation parameters based on the optimal parameter set and the user-given instructions includes: Obtain the joint states of the simulated robotic arm at each simulation time step; Based on the optimal parameter set and the user-given instructions, determine the execution torque for each simulation time step; The joint state of the simulated robotic arm is updated based on the execution torque; Error terms are determined based on the joint states of the simulated robotic arm; The final simulation parameters are determined based on the error term.

9. The motion control simulation method for a robotic arm according to claim 8, characterized in that, The error term includes the maximum error of a single joint and the average error of a single joint; The step of determining the final simulation parameters based on the error term includes: when the maximum error of a single joint is less than a first preset accuracy threshold and the average error of a single joint is less than a second preset accuracy threshold, the optimal parameter set is used as the final simulation parameters, wherein the first preset accuracy threshold is greater than or equal to the second preset accuracy threshold.

10. An electronic device, characterized in that, include: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores a computer program, which, when executed by the at least one processor, implements the motion control simulation method for the robotic arm as described in any one of claims 1 to 9.

Images