Dynamics parameter synchronous identification method considering robot base installation inclination angle

Abstract

A kind of dynamics parameter synchronous identification method considering robot base installation inclination, initialization parameter, after extracting the acceleration of each joint by collecting sensor signal, the residual error of base force sensing information about robot installation inclination is calculated according to base model observation matrix and force screw vector Jacobian matrix, then the robot installation inclination, coupling matrix parameter and dynamics model parameter are updated, whether the result difference obtained by judging the result difference of two iterations is less than given threshold value in the way of judging whether the robot installation inclination converges, when converging, the gravity acceleration vector under the robot base coordinate corresponding to the robot installation inclination and its Jacobian matrix are calculated again, otherwise the correctness of the model identified by using different trajectory is verified.The industrial robot system of the present application uses base additional force / torque sensor, expands the installation inclination of the identified robot, calibrates the coupling parameter of force sensor in situ and realizes the dynamics parameter identification of robot under any installation inclination, which significantly improves the whole machine motion control performance of industrial robot based on dynamics model under different installation inclination.

Description

TECHNICAL FIELD

[0001] The application relates to a technology in the field of intelligent manufacturing, and particularly relates to a dynamic parameter synchronous identification method considering a robot base installation inclination angle. BACKGROUND

[0002] The most important theoretical basis in robot dynamic model identification is that the inertia parameters of a machine dynamics model are linearly related, so that the model parameters can be identified by the least square method. However, when a robot has an installation inclination angle and the inclination angle is unknown, the installation inclination angle cannot be identified by the most basic linear parameters, and currently, there is no research on identifying the installation inclination angle of a robot while identifying dynamic parameters. SUMMARY

[0003] The application provides a dynamic parameter synchronous identification method considering a robot base installation inclination angle, and the method is used for an industrial robot system with a base additional force / torque sensor, the installation inclination angle of the robot is identified, the coupling parameters of the force sensor are calibrated in situ, the dynamic parameter identification of the robot under any installation inclination angle is realized, and the motion control performance of the whole robot based on the dynamic model under different installation inclination angles is improved.

[0004] The application is implemented by the following technical scheme:

[0005] The application relates to a dynamic parameter synchronous identification method considering a robot base installation inclination angle, and the method comprises the following steps:

[0006] Step one, initializing parameters, collecting sensor signals and extracting the accelerations of each joint, and specifically comprising the following steps:

[0007] 1.1 Collecting the joint encoder position feedback, joint speed feedback and base force / torque sensor signals of each joint, filtering the feedback signals by using a zero-phase delay filter, and obtaining the accelerations of each joint by numerical differentiation;

[0008] 1.2 Initializing the robot installation inclination angle ζ = [phi rho] T , and setting the installation inclination angle ζ to [0 0] T ;

[0009] 1.3 Initializing the dynamic model parameters pi of the additional base mechanical parameters of the robot and the coupling matrix parameters gamma of the force sensor arranged on the robot base.

[0010] Step two, calculating the Jacobian matrix of the residual error of the base force sensor information about the robot installation inclination angle according to the base model observation matrix and the force screw vector, and specifically comprising the following steps: the Jacobian matrix of the residual error about the robot installation inclination angle Wherein: D is Frechet derivative operator, base model observation matrix composed of each robot motion related data point and g Gravity acceleration vector in robot base coordinate corresponding to robot installation inclination g0 is gravity acceleration, χ is observation matrix of robot single data point based on base information, intermediate value of dynamics model parameter with robot base force sensor information q(i) of the ith data point is Generalized inverse of Force screw matrix composed of each data point of base force sensor o(i) is the external force sensor information of the ith data point, o i The ith element of external force sensor information, E6 is six order unit matrix, intermediate value of coupling matrix parameter of force sensor set in robot base Wherein: λ>0 is regularization constant, E is unit matrix, π is dynamics model parameter, force screw vector composed of each robot motion related data point and robot mass calculation and splicing m0 is the total mass of the robot.

[0011] Step three, after updating the robot installation inclination, coupling matrix parameter and dynamics model parameter, whether the robot installation inclination converges is judged by judging whether the difference between the results obtained by two iterations is less than the given threshold value, when the convergence is returned to step two to recalculate the gravity acceleration vector in robot base coordinate corresponding to robot installation inclination and its Jacobian matrix, otherwise the correctness of the model obtained by identification is verified by using different trajectories, which specifically includes:

[0012] 3.1 update the robot installation inclination ζ new = ζ old - α(J T J+ ρE) -1 D, wherein: α, ρ are constants in optimization algorithm, residual

[0013] 3.2 update the coupling matrix parameter Wherein: And It is calculated by ζ new .

[0014] 3.3 update the dynamics model parameter

[0015] The application relates to a system for implementing the above method, comprising a controller unit, an industrial robot unit, a six-dimensional base force sensor unit and an identification unit, wherein: the controller unit sends a target track to the industrial robot unit according to input instruction information, the industrial robot unit performs corresponding movement according to the obtained controller-issued instruction information, outputs corresponding movement data, the six-dimensional base force sensor unit outputs corresponding base force data according to the movement of the robot, and the identification unit obtains an identification result according to the robot movement information and the base force information.

[0016] Technical effects

[0017] The application can identify the robot with an installation inclination angle during installation through a six-dimensional force sensor, and can identify most of the dynamic parameters through the base six-dimensional force sensor without being affected by a joint friction force model; in the face of actual specific tasks, the coupling matrix of the six-dimensional force sensor at the bottom of the robot is calibrated in situ while the dynamic parameters of the robot are identified. Compared with the prior art, the application significantly improves the accuracy of the dynamic model parameter identification of the robot and the generality of the dynamic parameter identification under different installation postures. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 It is a flowchart of the application;

[0019] Figure 2 It is an application schematic diagram of the embodiment;

[0020] Figure 3 It is a trajectory diagram of the embodiment;

[0021] Figures 4 to 5 It is a convergence process diagram of the embodiment;

[0022] Figures 6 to 11 It is a model cross-validation error data diagram of the embodiment;

[0023] Figure 12 It is a force sensor linear calibration result diagram of the embodiment;

[0024] Figures 13 to 14 It is a robot installation inclination angle calibration result diagram of the embodiment. DETAILED DESCRIPTION

[0025] As shown in the figure, the embodiment relates to a dynamic parameter synchronous identification method for an industrial robot equipped with a rotary joint, comprising: Figure 1 Figure 2 As shown in the figure, the embodiment relates to a dynamic parameter synchronous identification method for an industrial robot equipped with a rotary joint, comprising:

[0026] ​Step one, prepare data: collect each external sensor data and joint encoder feedback position; let the robot track the pre-optimized excitation curve, record the data collected by the external sensor and the encoder feedback position, then filter and difference the feedback position by using zero-phase delay filter and median difference filter to get joint speed and joint acceleration, and get 50000 data points in total;

[0027] Step two, initialize the robot installation inclination ζ = [φ ρ] T , the installation inclination ζ is initialized as [0 0] T ;

[0028] Step three, initialize the dynamic model parameter π of the additional robot base mechanical parameters and set the coupling matrix parameter γ of the force sensor on the robot base to zero vector;

[0029] Step four, calculate the force screw matrix wherein: o(i) is the external force sensor information of the i th data point, o i is the i th element of the external force sensor information, and E6 is a six-order unit matrix;

[0030] The force screw matrix is a matrix with a size of 3000 rows and 27 columns.

[0031] Step five, calculate the gravity acceleration vector corresponding to the robot installation inclination in the robot base coordinate wherein: g0 is the size of the gravity acceleration, which is taken as 9.81 in the embodiment.

[0032] Step six, calculate the base model observation matrix wherein: χ is the observation matrix of the robot single data point based on the base information, and q(i) is the external force sensor information of the i th data point;

[0033] The base model observation matrix is a matrix with a size of 3000 rows and 43 columns.

[0034] Step seven, calculate the force screw vector wherein: m0 is the total mass of the robot, which is taken as 37.2 in the embodiment.

[0035] The force screw vector is a vector with a length of 3000.

[0036] Step eight, calculate the intermediate value of the coupling matrix parameter of the force sensor arranged on the robot base Wherein:λ>0 is a regularization constant, which is 100 in the embodiment, and E is a unit matrix. Step nine, calculate the intermediate value of the dynamics model parameter with the robot base mechanical parameter

[0037] Wherein:

[0038] Step ten, calculate the residual of the base force sensor information under the parameters calculated in steps eight and nine

[0039] Step eleven, calculate the Jacobian matrix of the residual with respect to the robot installation inclination angle Wherein: D is the Frechet derivative operator.

[0040] Step twelve, update the robot installation inclination angle ζ new = ζ old - α (J T J+ ρE) -1 d, wherein: α, ρ are constants in the optimization algorithm, which are 1 and 100 in the embodiment.

[0041] Step thirteen, update the coupling matrix parameter of the force sensor arranged on the robot base Wherein: And It is calculated through ζ new

[0042] Step fourteen, update the dynamics model parameter with the robot base mechanical parameter

[0043] Step fifteen, judge whether the robot installation inclination angle ζ converges, when it converges, go to step five, otherwise go to step sixteen, the convergence judgment is whether the difference between the results of two iterations is less than a given threshold 0.001.

[0044] Step sixteen, verify the identification parameters by the installation inclination angle obtained by identification and the model base six-dimensional force wrench, the specific operation is: compare the installation inclination angle obtained by identification with the actually measured installation inclination angle, calculate the base six-dimensional force wrench of the robot on the verification trajectory by using the identification parameters, and compare it with the actually measured base six-dimensional force wrench after the robot moves on the verification trajectory.

[0045] ​Through specific experiments, the experimental data obtained by running the above method with the above parameters is that the installation inclination recognition error is less than 0.5 degrees, and the base six-dimensional force wrench prediction relative error on the verification trajectory is less than 3%.

[0046] The above specific embodiments can be adjusted in different ways by those skilled in the art without departing from the principles and purposes of the present application, the protection scope of the present application is subject to the claims and is not limited by the above specific embodiments, and each implementation scheme within the scope is subject to the present application.

Claims

1. A method for synchronous identification of dynamic parameters considering the inclination angle of the robot base installation, characterized in that: include: Step 1: Prepare data: collect data from various external sensors and joint encoder feedback positions; The robot was instructed to track a pre-optimized excitation curve while recording data from external sensors and encoder feedback positions. The feedback positions were then filtered and differentiated using a zero-phase-delay filter and a median-difference filter to obtain joint velocities and accelerations, yielding a total of 50,000 data points. Step 2: Initialize the robot installation inclination , installation angle Initialized to ; Step 3: Initialize the dynamic model parameters with additional robot base mechanical parameters And the coupling matrix parameters of the force sensor set on the robot base is the zero vector; Step 4: Calculate the force spin matrix formed by combining the data points of the base force sensor ,in: , For the External force sensor information of data points, The external force sensor information elements, is the sixth-order identity matrix; Step 5: Calculate the gravity acceleration vector in the robot base coordinates corresponding to the robot installation inclination angle ,in: is the magnitude of the acceleration due to gravity; Step 6: Calculate the relevant data points of each robot movement and The combined base model observation matrix is the observation matrix of a single data point of the robot based on the base information, For the External force sensor information of data points; Step 7: Calculate the force spin vector of each robot motion-related data point and the robot mass calculation and splice is the total mass of the robot; Step 8: Calculate the intermediate value of the coupling matrix parameters of the force sensor set on the robot base ,in: , is the regularization constant, is the identity matrix; Step 9: Calculate the intermediate values ​​of the dynamic model parameters with the mechanical parameters of the robot base ,in: ; Step 10: Calculate the residual error of the base force sensing information under the parameters calculated in steps 8 and 9. ; Step 11: Calculate the Jacobian matrix of the residual with respect to the robot installation inclination angle ,in: is the Frechet derivative operator; Step 12: Update the robot installation inclination ,in: is a constant in the optimization algorithm; Step 13: Update the coupling matrix parameters of the force sensor set on the robot base ,in: and pass Calculated; Step 14: Update the dynamic model parameters with the robot base mechanical parameters ; Step 15: Determine the robot installation angle Convergence: If converged, go to step 5, otherwise go to step 16. Convergence is determined by whether the difference between the results of the two iterations is less than a given threshold of 0.

001. Step 16: Verify the identification parameters through the identified installation inclination angle and the six-dimensional force spin of the model base. The specific operation is: compare the identified installation inclination angle with the actually measured installation inclination angle, use the identification parameters to calculate the six-dimensional force spin of the base of the robot on the verification trajectory, and compare it with the six-dimensional force spin of the base actually measured after the robot moves along the verification trajectory.

2. A system for synchronously identifying dynamic parameters of a robot base taking into account the inclination angle of the robot base installation to implement the method of claim 1, characterized in that: include: A controller unit, an industrial robot unit, a six-dimensional base force sensor unit and an identification unit, wherein: the controller unit sends the target trajectory to the industrial robot unit according to the input instruction information; the industrial robot unit performs corresponding movement according to the instruction information issued by the controller, and outputs corresponding movement data; the six-dimensional base force sensor unit outputs corresponding base force data according to the movement of the robot; and the identification unit obtains the identification result according to the robot movement information and the base force information.

Images