A method for acquiring a reference point of a robot base

Abstract

The application discloses a kind of reference point acquisition methods of robot base, comprising: the first 3D image of robot base with multiple marks on outer wall is collected;Multiple first feature points of multiple marks in the first 3D image are acquired one by one;The first reference point of base is positioned according to multiple first feature points.This application effectively calculates the position relationship of actual robot and tool after landing deviation;The position of equipment is calibrated by the way of increasing calibration round hole of robot base, which avoids manual teaching of robot and reduces the risk of collision;The problem of robot calibration process lag caused by factors such as power-on, system configuration, cross construction of robot in project site construction process is solved;Compared with previous calibration method, the relative position relationship of robot base and vehicle system is more reliable.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robots, in particular to a method for acquiring a reference point of a robot base. BACKGROUND

[0002] With the increasing application of industrial robots in the field of automobile production, automobile manufacturers have increasing demand for high-precision factories and "power-on automatic" debugging. The workpiece coordinate accuracy and calibration of industrial robots have become a crucial problem. There is a positioning error between the robot and the tooling, which leads to low utilization of offline programs, increases the cost of labor and equipment, and increases the risk of equipment collision. The calibration methods of the robot workpiece coordinate mainly include calibration point method, four-point method, global measurement method and visual calibration. These operation processes are relatively complex, the robot needs to be powered on in advance, and the engineer needs to manually demonstrate, so the project cost is relatively high. SUMMARY

[0003] To solve the technical problems in the background art, the present application provides a method for acquiring a reference point of a robot base.

[0004] The method for acquiring a reference point of a robot base provided by the present application comprises the following steps:

[0005] S1, acquiring a first 3D image of a robot base having a plurality of marks on the outer wall;

[0006] S2, acquiring a plurality of first feature points corresponding to the plurality of marks in the first 3D image;

[0007] S3, positioning a first reference point of the base according to the plurality of first feature points.

[0008] Preferably, step S1 specifically comprises:

[0009] S11, acquiring a first 3D simulation image of the base;

[0010] S12, acquiring a second reference point of the base according to the first 3D simulation image, and selecting a plurality of second feature points on the outer wall of the base in the first 3D simulation image, which can be positioned to the second reference point;

[0011] S13, machining a plurality of calibration round holes as the plurality of marks on the base, and the hole center positions of the plurality of calibration round holes correspond to and coincide with the plurality of second feature points one by one;

[0012] S14, acquiring a first 3D image of a robot base having a plurality of marks on the outer wall.

[0013] Preferably, the mark is specifically a calibration round hole opened on the base, and the first feature point is specifically a hole center position of the calibration round hole.

[0014] Preferably, the plurality of first feature points are located on the same circumference and arranged at equal intervals, and the plurality of calibration holes are equal in diameter and depth and the orifices are located on the first plane.

[0015] Preferably, the first 3D image further comprises a robot assembly in assembly relationship with the base, and the acquisition method further comprises:

[0016] S4, acquiring a first reference point on the robot assembly in the first 3D image based on a first preset strategy, and obtaining a first positional relationship between the first reference point and a first datum point;

[0017] S5, acquiring a second 3D simulation image containing the robot assembly in assembly relationship and the base, acquiring a second datum point of the base in the second 3D simulation image through the second 3D simulation image, acquiring a second reference point of the robot assembly based on the first preset strategy, and obtaining a second positional relationship between the second datum point and the second reference point;

[0018] S6, comparing the first positional relationship with the second positional relationship to verify whether the acquired first datum point is accurate.

[0019] Preferably, the robot assembly specifically comprises a robot body that can be assembled onto the base and an execution tooling that can be assembled onto the robot body.

[0020] The first reference point is acquired based on the first preset strategy from the first 3D image, specifically, a third datum point of the execution tooling is acquired from the first 3D image as the first reference point.

[0021] The second reference point of the robot assembly is acquired based on the first preset strategy, specifically, a fourth datum point of the execution tooling is acquired from the second 3D simulation image as the second reference point.

[0022] Preferably, step S6 specifically comprises:

[0023] An average calibration point error value is calculated based on the first positional relationship and the second positional relationship;

[0024] A coordinate item difference value is calculated from the first positional relationship and the second positional relationship item by item;

[0025] When each coordinate item difference value meets a preset difference value and the average calibration point error value meets a preset threshold value, the acquired first datum point is accurate.

[0026] A storage medium has a computer readable program stored therein, and the computer readable program can execute the steps of the datum point acquisition method of the robot base when invoked by a controller.

[0027] In the present application, the reference point acquisition method of the robot base acquires a first 3D image of the robot base with multiple marks on the outer wall; acquires multiple first feature points of the multiple marks in the first 3D image one by one; locates the first reference point of the base according to the multiple first feature points; acquires the first reference point from the robot assembly in the first 3D image based on the first preset strategy, obtains the first positional relationship between the first reference point and the first reference point; acquires a second 3D simulation image containing the robot assembly and the base forming an assembly relationship, acquires a second reference point of the robot assembly based on the first preset strategy through the second 3D simulation image, and acquires a second positional relationship between the second reference point and the second reference point; compare the first positional relationship with the second positional relationship to verify whether the acquired first reference point is accurate. The positional relationship between the actual robot and the tooling after the positioning deviation is effectively calculated; the device position calibration is carried out by increasing the calibration round hole of the robot base, which avoids manual teaching of the robot and reduces the collision risk; solve the problem of robot calibration process lag caused by factors such as power-on, system configuration, cross construction and other factors in the project site construction process; compared with the previous calibration method, the relative position relationship between the robot base and the vehicle system is more reliable. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 The working flow structure diagram of the reference point acquisition method of the robot base proposed in the present application Figure One ;

[0029] Figure 2 The working flow structure diagram of the reference point acquisition method of the robot base proposed in the present application Figure Two . DETAILED DESCRIPTION

[0030] With reference to Figure 1 and Figure 2 , the reference point acquisition method of the robot base proposed in the present application comprises the following steps:

[0031] Step S1, acquiring a first 3D image of the robot base with multiple marks on the outer wall.

[0032] In this embodiment, step S1 specifically comprises:

[0033] Step S11, acquiring a first 3D simulation image of the base;

[0034] Step S12, acquiring a second reference point of the base according to the first 3D simulation image, selecting multiple second feature points on the outer wall of the base in the first 3D simulation image which can locate the second reference point;

[0035] S13, a plurality of calibration round holes as a plurality of marks are processed on the base, and the hole center positions of the plurality of calibration round holes correspond to the plurality of second feature points one by one.

[0036] S14, a first 3D image of the robot base with a plurality of marks on the outer wall is collected.

[0037] Specifically, the first 3D simulation image is obtained according to the robot base 3D data and the planned layout simulation. In the simulation process, the robot base is selected according to the robot brand and model to ensure that the robot body and the base hole position match each other. The first 3D image is an actual scene 3D image of the robot base containing a plurality of calibration round holes collected by a three-coordinate measuring device.

[0038] In this embodiment, the mark is specifically a calibration round hole opened on the base, and the first feature point is specifically the hole center position of the calibration round hole.

[0039] In this embodiment, the plurality of first feature points are located on the same circumference and arranged at equal intervals, and the plurality of calibration round holes are equal in diameter and depth, and the hole openings are located on the first plane.

[0040] Specifically, the plurality of calibration round holes are equal in diameter and depth, and the hole openings are located on the first plane, so as to facilitate the collection of the calibration round holes.

[0041] S2, a plurality of first feature points of a plurality of marks in the first 3D image are one by one correspondingly acquired.

[0042] S3, a first reference point of the base is positioned according to the plurality of first feature points.

[0043] In this embodiment, the first reference point is a reference point for defining the position and attitude of the robot in space in the first 3D image, which is a standard rectangular coordinate system, i.e. the origin of the robot world coordinate system.

[0044] In this embodiment, the first 3D image also includes a robot assembly in assembly relationship with the base, and the acquisition method further includes:

[0045] S4, a first reference point is acquired from the robot assembly in the first 3D image based on a first preset strategy, and a first positional relationship between the first reference point and the first reference point is acquired.

[0046] S5, a second 3D simulation image containing the robot assembly in assembly relationship and the base is acquired, a second reference point of the robot assembly is acquired based on the first preset strategy through the second 3D simulation image, and a second positional relationship between the second reference point and the second reference point is acquired.

[0047] Specifically, the second 3D simulation image is a 3D simulation image simulated according to 3D data of the robot assembly and the base forming the assembly relationship.

[0048] In the embodiment, the second reference point is a reference point for defining the position and pose of the robot in space in the second 3D simulation image, and is a standard rectangular coordinate system, i.e., a robot world coordinate system origin.

[0049] In the embodiment, the robot assembly specifically includes a robot body that can be assembled to the base and an execution tooling that can be assembled to the robot body.

[0050] The first reference point is obtained from the robot assembly in the first 3D image based on a first preset strategy, specifically, a third reference point of the execution tooling is obtained from the first 3D image as the first reference point.

[0051] The second reference point of the robot assembly is obtained based on the first preset strategy, specifically, a fourth reference point of the execution tooling is obtained from the second 3D simulation image as the second reference point.

[0052] In the embodiment, the first reference point is an origin of a coordinate system adopted in automobile body design in the first 3D image, which is a virtual origin, a reference point for determining and maintaining the relative spatial position of the panel and the tooling fixture, and the fixture origin and the body origin should be consistent in the definition, i.e., the origin of the vehicle coordinate system. The first reference point is an origin of a coordinate system adopted in automobile body design in the second 3D simulation image, which is a virtual origin, a reference point for determining and maintaining the relative spatial position of the panel and the tooling fixture, and the fixture origin and the body origin should be consistent in the definition, i.e., the origin of the vehicle coordinate system.

[0053] In step S6, the first position relationship and the second position relationship are compared to verify whether the obtained first reference point is accurate.

[0054] In the embodiment, step S6 specifically includes:

[0055] An average calibration point error value is calculated based on the first position relationship and the second position relationship.

[0056] A coordinate item difference value is calculated for each coordinate item of the first position relationship and the second position relationship.

[0057] When each coordinate item difference value meets a preset difference value and the average calibration point error value meets a preset threshold value, the obtained first reference point is accurate.

[0058] Specifically, the first position relationship is specifically , and the second position relationship is specifically The coordinate item difference value of each coordinate item specifically includes:

[0059]

[0060] wherein, when are all less than a preset difference value, are all less than a preset angle value and the average calibration point error is less than a preset threshold value, the obtained first reference point is accurate.

[0061] Specifically, the preset difference value of the first reference point is 50mm; the corresponding preset angle value is 1°, and the preset threshold value corresponding to the average calibration point error is 0.3mm.

[0062] In the embodiment, the calculation process of the average calibration point error value is as follows:

[0063] ;

[0064] wherein, is the first reference point coordinate; is the first reference point coordinate; is the coordinate corresponding to the first position relationship; n is the number of identification; and RMSE is the average calibration point error value.

[0065] In the embodiment, when the first reference point is verified to be accurate, the data corresponding to the first reference point collected is used to replace the data corresponding to the second reference point.

[0066] A storage medium has a computer readable program stored therein, and the computer readable program can execute the steps of the reference point acquisition method of the robot base of any one of the above when called by a controller.

[0067] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can make equivalent replacements or changes to the technical solution and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A method of acquiring a reference point of a robot base, characterized by, The method comprises the following steps: S1, collecting a first 3D image of a robot base having a plurality of marks on the outer wall; S2, one-to-one acquiring a plurality of first feature points of the plurality of marks in the first 3D image; S3, positioning a first reference point of the base according to the plurality of first feature points; The first 3D image further comprises a robot assembly in assembly relationship with the base, and the acquisition method further comprises: S4, acquiring a first reference point from the robot assembly in the first 3D image based on a first preset strategy, and acquiring a first positional relationship between the first reference point and the first reference point; S5, acquiring a second 3D simulation image containing the robot assembly in assembly relationship and the base, acquiring a second reference point of the robot assembly based on the first preset strategy, and acquiring a second positional relationship between the second reference point and the second reference point; S6, comparing the first positional relationship with the second positional relationship to verify whether the acquired first reference point is accurate; The robot assembly specifically comprises: a robot body that can be assembled to the base, and an execution tooling that can be assembled to the robot body; The first reference point is acquired from the third reference point of the execution tooling in the first 3D image based on the first preset strategy; The second reference point is acquired from the fourth reference point of the execution tooling in the second 3D simulation image based on the first preset strategy; Step S6 specifically comprises: Based on the first positional relationship and the second positional relationship, an average calibration point error value is calculated; The first positional relationship and the second positional relationship are calculated to obtain a coordinate item difference value; When each coordinate item difference value meets the preset difference value and the average calibration point error value meets the preset threshold value, the acquired first reference point is accurate.

2. The method of claim 1, wherein Step S1 specifically comprises: S11, acquiring a first 3D simulation image of the base; S12, acquiring a second reference point of the base according to the first 3D simulation image, and selecting a plurality of second feature points on the outer wall of the base in the first 3D simulation image, which can position the second reference point; S13, machining a plurality of calibration round holes as a plurality of marks on the base, and the hole center positions of the plurality of calibration round holes correspond to and coincide with the plurality of second feature points one-to-one; S14, collecting a first 3D image of a robot base having a plurality of marks on the outer wall; 3. The method of claim 2, wherein The mark is specifically a calibration round hole opened on the base, and the first feature point is specifically the hole center position of the calibration round hole.

4. The method of claim 3, wherein, The plurality of first feature points are located on the same circumference and arranged at equal intervals, and the plurality of calibration round holes are equal in diameter and depth, and the hole openings are located on the first plane.

5. A storage medium, characterized by The computer readable program is stored in the memory and can be called by the controller to execute the steps of the reference point acquisition method of the robot base according to any one of claims 1 to 4.

Images