A method for determining robot measurement configuration based on multi-station measurement

Abstract

The present invention discloses a robot measurement configuration determination method based on multi-station measurement, comprising the following steps: 1. In an offline simulation environment, a displacement measuring device is placed at one of the stations of a base, the point constraint of the center of a measuring ball is maintained, and the attitude angles of 3D measuring clubs are evenly distributed in the measurement space of the displacement measuring device, and the corresponding robot joint angles are solved based on robot inverse kinematics, and then screened through configuration screening, and finally a large number of feasible and reasonable robot joint angles are obtained, and the configuration screening includes simulation collision detection and kinematic solvability detection; 2. The displacement measuring device is clamped on other stations on the base and the above process is repeated, and finally all feasible and reasonable posture joint angles are obtained; 3. The measurement joint angles are selected based on observable indicators, and the optimal measurement joint angle combination is selected from all feasible and reasonable joint angles. The present invention improves calibration performance and reduces the influence of accidental errors on the final calibration performance.

Description

Technical Field

[0001] The invention relates to a robot measurement configuration determination method based on multi-station measurement. Background Art

[0002] The repeatability of a robot is generally very high, but the absolute positioning accuracy is usually low due to the influence of machining and assembly errors, rod deformation, etc. To improve the absolute positioning accuracy of a robot, kinematic parameter calibration is generally required, and high-precision measuring equipment and appropriate parameter identification methods are required to identify the accurate parameters of the robot model. CN113084798A discloses a robot calibration device based on multi-station measurement, which includes a 3D measuring ball bar, a multi-station base, and a 3D displacement measuring device. The user installs the 3D measuring ball bar at the end of the robot to be calibrated, and then lets the mechanical arm move the 3D measuring ball bar in different postures to the measurement area of ​​the 3D displacement measuring device for measurement, and then switches the 3D displacement measuring device to different stations on the multi-station base to repeat the above process, thereby achieving multi-point measurement. The device has the advantages of high accuracy, simple and reliable structure, but in the process of using the device, it is found that the selection of different measurement postures will also have a greater impact on the calibration accuracy. In order to maximize the calibration performance of the device, it is necessary to design a method to select the measurement posture of the robot. Summary of the invention

[0003] In order to solve the above problems existing in the prior art, the present invention proposes a robot measurement configuration determination method based on multi-station measurement.

[0004] The present invention can be implemented through the following technical solutions:

[0005] A method for determining a robot measurement configuration based on multi-station measurement comprises the following steps:

[0006] (1) In an offline simulation environment, a displacement measuring device is placed at one of the workstations of the base, the point constraint of the center of the measuring ball is maintained, the attitude angles of the 3D measuring club are evenly distributed in the measurement space of the displacement measuring device, the corresponding robot joint angles are solved based on the robot inverse kinematics, and then they are screened through configuration screening to finally obtain a large number of feasible and reasonable robot joint angles, wherein the configuration screening includes simulation collision detection and kinematic solvability detection;

[0007] (2) Clamp the displacement measuring device on other workstations on the base and repeat the above process to finally obtain all feasible and reasonable joint angles;

[0008] (3) The measured joint angles are selected based on observable indicators, and the optimal measured joint angle combination is selected from all feasible and reasonable joint angles.

[0009] Furthermore, the measurement configuration, the position of the base relative to the robot arm, and the relative positions between different workstations are optimized through simulation.

[0010] Furthermore, the measurement space of the displacement measuring device refers to defining the intersection of three orthogonally placed displacement sensor axes as O1, taking point O1 or a point O2 near point O1 as the measuring point, and making three reference surfaces through the measuring point that are parallel to the measuring planes of the three flat probes. The three reference surfaces divide the spherical space with the measuring point as the center into eight equal parts, and taking one-eighth of the open area facing outwards of the three displacement sensors as the measurement space.

[0011] Furthermore, the specific method of measuring the attitude angle of the club in the measurement space of the displacement measuring device is as follows: the coordinates on the spherical surface are defined by the longitude and latitude in geography according to a certain division value, and the direction of the line connecting the intersection of each longitude and latitude line and the center of the sphere is the z-axis direction of the measuring spherical coordinate system of the end of the robot arm in the measured posture, and the x-axis direction is obtained by rotating an initial x-axis around the z-axis by an angle γ, and the angle γ is divided into -180°~180° by a division value d γ By traversing and determining the z-axis and the x-axis, the entire end measurement spherical coordinate system can be determined;

[0012] Alternatively, instead of using longitude and latitude to define the coordinate points on the sphere, the points can be evenly distributed on the sphere, and the direction of the line connecting the evenly distributed points and the center of the sphere is defined as the z-axis direction of the spherical coordinate system measured by the end of the manipulator. The direction of the x-axis is obtained by rotating an initial x-axis around the z-axis by an angle γ, which is divided into a scale value d from -180° to 180°. γ By traversing and determining the z-axis and the x-axis, the entire end measurement spherical coordinate system can be determined.

[0013] Furthermore, the kinematic solvability detection is as follows: for each posture, the kinematic inverse solution is calculated to obtain the corresponding joint angle; if the posture has no inverse solution, the posture is eliminated.

[0014] Furthermore, the configuration screening also includes dexterity detection, that is, calculating the morbid evaluation index of the corresponding Jacobian matrix of the joint angles inversely solved for the posture, and eliminating those with morbid evaluation index less than the threshold k. limit The joint angle, the pathological evaluation index is one of OI2, OI3, OI4 and OI5:

[0015]

[0016] OI3=σ q (15-3)

[0017]

[0018]

[0019] Among them, σ i (i=1,2,…,q,σ q ≤...≤σ1) are the singular values ​​of the Jacobian matrix, and q is the number of singular values.

[0020] Furthermore, the simulation collision detection includes static collision detection and dynamic collision detection. The static collision detection is that after the robot arm moves to a certain joint angle, it will collide with itself or with the outside world when it is placed statically, and then the joint angle will be eliminated. Through the collision detection judgment in the simulation, the self-collision refers to the collision between the joint links of the robot arm, and the collision with the outside world refers to the collision between the robot arm and the ground, the base, and the measuring device.

[0021] The dynamic collision detection is: if a dynamic collision occurs when the robot moves from the initial posture to the posture, including self-collision and collision with the outside world, the joint angle is eliminated, and the movement mode is one or more combinations of linear interpolation movement, curve interpolation movement, and MOVEJ movement.

[0022] Furthermore, the observable index can characterize the degree of pathological condition of the Jacobian matrix when the kinematic parameters are identified based on the optimization method, specifically one of OI1, OI2, OI3, O14 and OI5:

[0023]

[0024]

[0025] OI3=σ q (15-3)

[0026]

[0027]

[0028] Among them, σ i (i=1,2,…,q,σ q ≤...≤σ1) are the singular values ​​of the Jacobian matrix J, q is the number of singular values, m is the number of data sets, and there are many ways to evaluate the observability index. The larger the OI value, the better the flexibility of the robot under this group of joint angles.

[0029] Furthermore, the measurement of the joint angle is selected using the DETMAX algorithm, and the specific steps are as follows:

[0030] a. Randomly select m groups of data from all feasible and reasonable joint angle data, and calculate the observation index values ​​corresponding to these m groups of data;

[0031] b. In the remaining Traverse the group data to find a group of data so that the observed index value corresponding to the (m+1) group of data after adding this group of data to the originally selected m groups of data is the largest;

[0032] c. Traverse and select a group of data from the (m+1) groups of data so that the observed index value of the m groups of data after deleting the group of data is the largest;

[0033] This cycle is repeated continuously until the group of data deleted is exactly the group of data added, thereby obtaining the m groups of data with the largest observable index values ​​and realizing the selection of the optimal measurement joint angle combination.

[0034] Beneficial Effects

[0035] This method can optimize the measurement posture of the robot, thereby maximizing the calibration performance of the multi-station calibration measurement device and minimizing the impact of accidental errors on the final calibration performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a schematic diagram of a robot calibration system based on multi-station measurement of the present invention;

[0037] Figure 2 It is a schematic diagram of the contact between the 3D measuring ball bar and the flat probe of the displacement sensor in the robot calibration system based on multi-station measurement of the present invention;

[0038] Figure 3 This is a side view of the installation of the base and pedestal of the displacement measuring device in the robot calibration system based on multi-station measurement of the present invention.

[0039] Figure 4 It is a schematic diagram of an angle-adjustable calibration measurement device in a robot calibration system based on multi-station measurement of the present invention.

[0040] In the figure: 2 - base; 4 - positioning pin; 5 - positioning ball; 6 - fastening assembly; 7 - base; 8 - connecting rod; 9 - angle adjustment seat; 10 - measuring head; 11 - electric control box; 12 - measuring ball; 13 - extension rod; 14 - flat probe; 15 - calibration point; 16 - adapter DETAILED DESCRIPTION

[0041] The following describes the implementation of the present invention through specific embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification.

[0042] like Figure 1The figure shows a robot calibration device based on multi-station measurement according to the present invention, which includes four main parts: a base 2, a displacement measuring device, a 3D measuring ball bar and a mechanical arm (not shown in this embodiment). The base 2 is provided with five mounting positions for mounting a displacement measuring device for calibration data collection; the displacement measuring device comprises a base, a connecting rod 8, an angle adjustment seat 9, a measuring head 10, an electric control box 11, a flat probe 14, and an adapter 16; the measuring head 10 comprises at least three orthogonally arranged displacement sensors, the displacement sensor is a contact displacement sensor, the head probe of the contact displacement sensor is a flat probe or a ball probe, or the displacement sensor is a laser non-contact displacement sensor; the 3D measuring ball bar is located on the displacement measuring device and detects the position error of the calibration object when it moves around a point through the measuring head 10. In this embodiment, the 3D measuring ball bar comprises a top measuring ball 12 and an extension rod 13, the head probe of the contact displacement sensor is a flat probe 14, the measuring ball 12 and the flat probe 14 are always tangent to each other, so as to realize the measurement of the 3D displacement of the center of the ball; the measuring ball 12 is connected to the mechanical arm (not shown in this embodiment) through the extension rod 13.

[0043] like Figure 2 As shown, the axes of two of the three orthogonally arranged displacement sensors have the same angle with the horizontal plane, and the angle between the plane formed by the two axes and the horizontal plane is recorded as the measuring angle. There are multiple angle adjustment seats 9, and the measuring angle can be adjusted by replacing different angle adjustment seats 9; or an angle adjustment locking mechanism is provided at the bottom of the angle adjustment seat 9, which can adjust the measuring angle and lock it.

[0044] like Figure 3 As shown, a group of first positioning elements are provided on the base 7, and a positioning method of forming a positioning surface by three groups of positioning points is adopted; the first positioning elements are positioning pins or positioning balls arranged in a triangle, or double positioning pins, or double positioning balls; the double positioning pins are two positioning pins arranged in parallel, and the cylindrical surface or conical surface of the positioning pins protrudes from the bottom surface of the base 7, and the axis of the double positioning pins faces the center of the triangular arrangement; the double positioning balls are two positioning balls arranged at a certain distance, and the spherical surface of the positioning balls protrudes from the bottom surface of the base 7, and the perpendicular bisector of the line connecting the centers of the two positioning balls faces the center of the triangular arrangement.

[0045] The base 2 is provided with a plurality of mounting positions, each mounting position corresponding to a group of second positioning elements, each group of second positioning elements corresponding to a group of first positioning elements; the second positioning elements are triangularly arranged positioning pins or positioning balls, or double positioning pins, or double positioning balls; the double positioning pins are two positioning pins arranged in parallel, the cylindrical surface or conical surface of the positioning pins protrudes from the upper surface of the base 2, and the axis of the double positioning pins faces the center of the triangular arrangement; the double positioning balls are two positioning balls arranged at a distance, the spherical surface of the positioning balls protrudes from the upper surface of the base 2, and the perpendicular bisector of the line connecting the centers of the two positioning balls faces the center of the triangular arrangement.

[0046] In this embodiment, the first positioning element on the base 7 is a triangularly arranged positioning pin 4, and the second positioning element on the base 2 is a triangularly arranged double positioning ball 5, that is, a positioning hole that can clamp the positioning pin 4 is formed between the two positioning balls 5, and each group of positioning balls 5 is embedded in the base 2. The three groups of positioning pins 4 have a triangular structure. Each group of positioning pins 4 can be embedded in the corresponding positioning hole and be tangent to each positioning ball 5 to form a positioning point.

[0047] like Figure 4 As shown, in order to obtain a better and more convenient measurement angle, the displacement measuring device of this embodiment can adopt another alternative solution, wherein the angle adjustment seat 9 is multiple, and the measurement angle can be adjusted by replacing different angle adjustment seats 9; or an angle adjustment locking mechanism is provided at the bottom of the angle adjustment seat 9, which can adjust the measurement angle and lock it. This embodiment adopts an adapter 16, and different angle adjustment seats 9 can be replaced by connecting the adapter 16 to adjust the measurement angle, so that the angles of the three mutually perpendicular planes on the angle adjustment seat 9 and the horizontal plane can be adjusted, so that when the mechanical arm drives the calibrated measuring device to move, it can be detected at a more favorable movement angle, laying the foundation for accurate measurement.

[0048] The data processing device in the present invention includes a data acquisition module and a calibration module. The data acquisition module is used to collect readings of three displacement sensors. The calibration module calculates the robot geometric parameter errors based on the sensor readings and the robot joint angles. The robot joint angles can be directly obtained through the robot controller.

[0049] The measuring device of the present invention is used to calibrate the three-dimensional coordinates of the center of the measuring ball 12 in the measuring coordinate system, and establish the measuring coordinate system 0 for the measuring device. E X E Y E Z E ,like Figure 1 As shown, the plane on the base 2 is used as the reference plane X E O E Y E , with the lower left corner of base 2 as origin O E , the long side of base 2 is X EAxis. Before the measuring device is used to calibrate the robot after processing, it is calibrated using a three-coordinate machine. The calibration content includes: the initial coordinates of the center of the flat probe 14 corresponding to the three displacement sensors in the measurement coordinate system, the linear equations of the three displacement sensors in the measurement coordinate system, and the plane equations of the flat probe corresponding to the three displacement sensors in the measurement coordinate system.

[0050] like Figure 1 As shown, the calibration device is calibrated using a three-coordinate machine before being used to calibrate the robot after processing. The calibration content includes: measuring the diameter of the ball 12.

[0051] A kinematic calibration method based on the above robot comprises the following steps:

[0052] 1) If Figure 1 As shown, the base is fixed near the base of the robot and is always kept relatively still with the robot, the displacement measuring device is installed on one of the stations of the base on the physical device platform, the calibration device (3D measuring ball bar) is installed on the end of the robot, and the measuring device is placed in the robot workspace;

[0053] 2) Initialize the robot control system and the data processing device;

[0054] 3) Using the measuring ball on the calibration device to touch the flat probe with the m1 group of optimal postures and joint angles selected by the calibration posture and joint angle selection method, and using the data acquisition module to record the readings of the d displacement sensors at each touch. i =(l i1 , l i2 , ..., l id ) and the corresponding b-DOF robot joint angle value θ i =(θ i1 ,θ i2 , ..., θ ib ), and get m1 group data.

[0055] Among them, i=1,...,m1.

[0056] 4) Install the displacement measuring device on the remaining (t-1) stations of the base, repeat the above operation, and obtain m2, ..., m respectively. t Group data, where

[0057] 5) The relative posture relationship between different workstations is measured by a high-precision measuring device, and the actual three-dimensional coordinates of the center of the measuring ball in the measuring coordinate system are calculated based on the measurement data.

[0058] A robot measurement configuration determination method based on multi-station measurement of the present invention, that is, the calibration posture joint angle selection method adopted in the above calibration method for screening, comprises the following steps:

[0059] (1) In an offline simulation environment, a displacement measuring device is placed at one of the workstations of the base, the point constraint of the center of the measuring ball is maintained, the attitude angles of the 3D measuring club are evenly distributed in the measurement space of the displacement measuring device, the corresponding robot joint angles are solved based on the robot inverse kinematics, and then they are screened through configuration screening to finally obtain a large number of feasible and reasonable robot joint angles, wherein the configuration screening includes simulation collision detection and kinematic solvability detection;

[0060] (2) Clamp the displacement measuring device on other workstations on the base and repeat the above process to finally obtain all feasible and reasonable joint angles;

[0061] (3) The measured joint angles are selected based on observable indicators, and the optimal measured joint angle combination is selected from all feasible and reasonable joint angles.

[0062] Among them, the measurement configuration, the position of the base relative to the robot arm, and the relative positions between different workstations were optimized through simulation.

[0063] Among them, the measurement space of the displacement measuring device refers to defining the intersection of three orthogonally placed displacement sensor axes as O1, taking point O1 or a point O2 near point O1 as the measuring point, and making three reference surfaces parallel to the measuring planes of the three flat probes through the measuring point. The three reference surfaces divide the spherical space with the measuring point as the center into eight equal parts, and taking one-eighth of the open area facing outward of the three displacement sensors as the measurement space.

[0064] The specific method of measuring the attitude angle of the club in a uniformly distributed 3D measurement space of the displacement measuring device is as follows: the coordinates on the spherical surface are defined by the longitude and latitude in geography according to a certain division value, and the direction of the line connecting the intersection of each longitude and latitude line and the center of the sphere is the z-axis direction of the measuring spherical coordinate system of the end of the robot arm in the measured posture, and the direction of the x-axis is obtained by rotating an initial x-axis around the z-axis by an angle γ, and the angle γ is divided into -180°~180° by a division value d γ By traversing and determining the z-axis and the x-axis, the entire end measurement spherical coordinate system can be determined;

[0065] Alternatively, instead of using longitude and latitude to define the coordinate points on the sphere, the points can be evenly distributed on the sphere, and the direction of the line connecting the evenly distributed points and the center of the sphere is defined as the z-axis direction of the spherical coordinate system measured by the end of the manipulator. The direction of the x-axis is obtained by rotating an initial x-axis around the z-axis by an angle γ, which is divided into a scale value d from -180° to 180°. γBy traversing and determining the z-axis and the x-axis, the entire end measurement spherical coordinate system can be determined.

[0066] The kinematic solvability test is as follows: for each posture, the kinematic inverse solution is calculated to obtain the corresponding joint angle; if the posture has no inverse solution, the posture is eliminated.

[0067] The configuration screening also includes dexterity detection, that is, calculating the morbid evaluation index of the corresponding Jacobian matrix of the joint angles obtained by the inverse solution of the posture, and eliminating the morbid evaluation index less than the threshold k. limit The joint angle, the pathological evaluation index is one of OI2, OI3, OI4 and OI5:

[0068]

[0069] OI3=σ q (15-3)

[0070]

[0071]

[0072] Among them, σ i (i=1,2,…,q,σ q ≤...≤σ1) are the singular values ​​of the Jacobian matrix, and q is the number of singular values.

[0073] Among them, the simulation collision detection includes static collision detection and dynamic collision detection. The static collision detection is that after the robot arm moves to a certain joint angle, it will collide with itself or collide with the outside world when it is statically placed, then the joint angle will be eliminated. Through the collision detection judgment in the simulation, the self-collision refers to the collision between the joint links of the robot arm, and the collision with the outside world refers to the collision between the robot arm and the ground, base, and measuring device; the dynamic collision detection is: if a dynamic collision occurs when the robot moves from the initial posture to this posture, including self-collision and collision with the outside world, then the joint angle will be eliminated. The movement mode is one or more combinations of linear interpolation movement, curve interpolation movement, and MOVEJ movement.

[0074] The observable index can characterize the degree of pathological condition of the Jacobian matrix when the kinematic parameters are identified based on the optimization method, and is specifically one of OI1, OI2, OI3, OI4 and OI5:

[0075]

[0076]

[0077] OI3=σ q (15-3)

[0078]

[0079]

[0080] Among them, σ i (i=1,2,…,q,σ q ≤...≤σ1) are the singular values ​​of the Jacobian matrix J, q is the number of singular values, m is the number of data sets, and there are many ways to evaluate the observability index. The larger the OI value, the better the flexibility of the robot under this group of joint angles.

[0081] The measurement of the joint angle is carried out by using the DETMAX algorithm, and the specific steps are as follows:

[0082] a. Randomly select m groups of data from all feasible and reasonable joint angle data, and calculate the observation index values ​​corresponding to these m groups of data;

[0083] b. In the remaining Traverse the group data to find a group of data so that the observed index value corresponding to the (m+1) group of data after adding this group of data to the originally selected m groups of data is the largest;

[0084] c. Traverse and select a group of data from the (m+1) groups of data so that the observed index value of the m groups of data after deleting the group of data is the largest;

[0085] This cycle is repeated continuously until the group of data deleted is exactly the group of data added, thereby obtaining the m groups of data with the largest observable index values ​​and realizing the selection of the optimal measurement joint angle combination.

[0086] The present invention iteratively screens calibration data through the observability index of the manipulator, selects the optimal posture joint angle combination, and reduces the morbidity of the calibration equation group.

[0087] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining a robot measurement configuration based on multi-station measurement, characterized in that: The steps include: (1) In an offline simulation environment, a displacement measuring device is placed at one of the workstations of the base, the point constraint of the center of the measuring ball is maintained, the attitude angles of the 3D measuring club are evenly distributed in the measurement space of the displacement measuring device, the corresponding robot joint angles are solved based on the robot inverse kinematics, and then they are screened through configuration screening to finally obtain a large number of feasible and reasonable robot joint angles. The configuration screening includes simulation collision detection and kinematic solvability detection; (2) Clamp the displacement measurement device on other workstations on the base and repeat the above process to finally obtain all feasible and reasonable joint angles; (3) Select the measured joint angles based on the observational indicators and select the optimal measurement joint angle combination from all feasible and reasonable joint angles; The kinematic solvability test is as follows: for each posture, the kinematic inverse solution is calculated to obtain the corresponding joint angle; if the posture has no inverse solution, the posture is eliminated; The simulation collision detection includes static collision detection and dynamic collision detection. The static collision detection is that after the robot arm moves to a certain joint angle, it will collide with itself or with the outside world when it is placed statically. In this case, the joint angle will be eliminated. Through the collision detection judgment in the simulation, the self-collision refers to the collision between the joint links of the robot arm, and the collision with the outside world refers to the collision between the robot arm and the ground, the base, and the measuring device. The dynamic collision detection is: if a dynamic collision occurs when the robot moves from the initial posture to the posture, including self-collision and collision with the outside world, then the joint angle is eliminated, and the movement mode is one or more combinations of linear interpolation movement, curve interpolation movement, and MOVEJ movement; The observability index can characterize the degree of pathological condition of the Jacobian matrix when the kinematic parameters are identified based on the optimization method, specifically one of OI1, OI2, OI3, OI4 and OI5: in, is the Jacobian matrix The singular values ​​of is the number of singular values, m is the number of data groups, and there are multiple evaluation methods for observational indicators. The larger the OI value, the better the flexibility of the robot under this group of joint angles.

2. The method for determining a robot measurement configuration based on multi-station measurement according to claim 1, characterized in that: The measurement configuration, the position of the base relative to the robot arm, and the relative positions between different workstations were optimized through simulation.

3. The method for determining a robot measurement configuration based on multi-station measurement according to claim 1, characterized in that: The measurement space of the displacement measuring device refers to defining the intersection of three orthogonally placed displacement sensor axes as O1, taking point O1 or a point O2 near point O1 as the measurement point, and making three reference surfaces through the measurement point that are parallel to the measurement planes of the three flat probes. The three reference surfaces divide the spherical space with the measurement point as the center into eight equal parts, and taking one-eighth of the open area facing outward of the three displacement sensors as the measurement space.

4. The method for determining a robot measurement configuration based on multi-station measurement according to claim 1, characterized in that: The specific method of measuring the attitude angle of the ball club in the measurement space of the displacement measurement device is as follows: the coordinates on the spherical surface are defined by the longitude and latitude in geography according to a certain division value, and the direction of the line connecting the intersection of each longitude and latitude line and the center of the sphere is the z-axis direction of the measurement spherical coordinate system of the end of the robot arm in the measured posture, and the x-axis direction is the rotation angle of a certain initial x-axis around the z-axis Get the angle Depend on By graduation value Traverse and determine the z-axis and x-axis, that is, determine the entire end measurement spherical coordinate system; Alternatively, instead of using longitude and latitude to define the coordinate points on the sphere, evenly distribute the points on the sphere directly, and define the direction of the line connecting the evenly distributed points and the center of the sphere as the z-axis direction of the spherical coordinate system measured by the end of the robot arm, and the x-axis direction is the rotation angle of a certain initial x-axis around the z-axis. Get the angle Depend on By graduation value Traverse and determine the z-axis and x-axis, that is, determine the entire end measurement spherical coordinate system.

5. The method for determining a robot measurement configuration based on multi-station measurement according to claim 1, characterized in that: The configuration screening also includes dexterity detection, that is, calculating the morbid evaluation index of the corresponding Jacobian matrix of the joint angles obtained by the inverse solution of the posture, and eliminating the morbid evaluation index less than the threshold value. The joint angle, the pathological evaluation index is one of OI2, OI3, OI4 and OI5: in, are the singular values ​​of the Jacobian matrix, is the number of singular values.

6. The method for determining a robot measurement configuration based on multi-station measurement according to claim 1, characterized in that: The measurement joint angle selection adopts the DETMAX algorithm, and the specific steps are as follows: a. Randomly select from all feasible and reasonable joint angle data set of data and calculate this The observed indicator value corresponding to the group data; b. In the remaining Traverse the group data to find a group of data so that the originally selected Group data after adding this group data The observed index value corresponding to the group data is the largest; c. Here Traverse and filter out a group of data from the group data so that after deleting the group of data The observed index value of the group data is the largest; d. Continue to loop in this way until the deleted set of data is exactly the added set of data, thus obtaining the maximum observation index value The optimal combination of measured joint angles can be selected by combining the data sets.

Images