A small multi-channel flexible space angle measurement device and method

Abstract

The present invention discloses a small multi-channel flexible space angle measurement device and method, the device includes a robot, a multi-beam laser vision sensor, a slide rail and a data acquisition and processing system; the slide rail is arranged close to the object to be measured; the robot is movably arranged on the slide rail; multiple groups of multi-beam laser vision sensors are arranged at the end of the robot and are electrically connected to the data acquisition and processing system, which are used to collect multiple laser images of the object to be measured and process and calculate, and obtain multi-channel unified measurement results of the relative angles between multiple measured surfaces of the object to be measured. The present invention uses image processing, system calibration and angle calculation, assisted by industrial robots as carriers of multiple groups of multi-beam laser vision sensors, to achieve multi-channel, large-space, flexible, multi-angle synchronous measurement of large heterogeneous objects, and real-time feedback of angle deviation to guide product design and improvement.

Description

Technical Field

[0001] The present invention relates to the field of photoelectric detection technology, and in particular to a small multi-channel flexible space angle measurement device and method. Background Art

[0002] High-precision angle measurement is a key link in precision machining and manufacturing, which directly determines the accuracy and reliability of the product. For a long time, high-precision angle measurement in the production process has been mostly done manually using angle measuring instruments, which has the disadvantages of low efficiency, poor accuracy, and high safety risks, which seriously restricts the accuracy and efficiency of current precision machining and manufacturing.

[0003] At present, photoelectric detection technology is developing rapidly. Due to its advantages of non-contact, high precision, and large range, it is widely used in intelligent manufacturing production lines, effectively improving detection efficiency and accuracy. However, traditional photoelectric measurement devices cannot meet the needs of high-precision and flexible measurement of large-sized heterogeneous objects, and cannot adapt to the unified measurement of large heterogeneous objects. Specifically, traditional photoelectric measurement devices can only measure size and angle information within a limited range, and their measurement accuracy cannot flexibly change and adapt to the volume and complexity of large heterogeneous objects. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a small multi-channel flexible spatial angle measurement device and method, which can perform flexible synchronous measurement of multiple angles of large heterogeneous workpieces and is suitable for precision machining.

[0005] To solve the above technical problems, the present invention is implemented by adopting the following solutions:

[0006] The present invention provides a small multi-channel flexible space angle measuring device, comprising a robot, a multi-beam laser vision sensor, a slide rail and a data acquisition and processing system; the slide rail is arranged close to the object to be measured; the robot is movably arranged on the slide rail; multiple groups of multi-beam laser vision sensors are arranged at the end of the robot and are electrically connected to the data acquisition and processing system, and are used to collect multiple laser images of the object to be measured and process and calculate, so as to obtain a multi-channel unified measurement result of the relative angles between multiple measured surfaces of the object to be measured.

[0007] Preferably, the multi-beam laser vision sensor includes a pixel industrial camera and a 5-beam near-infrared laser head projector; the pixel industrial camera and the 5-beam near-infrared laser projector are fixedly installed, the 5-beam near-infrared laser projector irradiates the laser onto the surface of the object to be measured, and the pixel industrial camera is used to collect laser images.

[0008] Preferably, the resolution of the pixel industrial camera is not less than 5 million pixels, and the frame rate is not less than 30 Hz; the power of the 5-beam near-infrared laser head projector is not less than 1 watt; the pixel industrial camera is provided with a narrow-band filter.

[0009] The present invention also provides a small multi-channel flexible space angle measurement method, comprising:

[0010] According to the spatial distribution of multiple measured surfaces of the measured object, optimize the spatial layout of multiple groups of multi-beam laser vision sensors at the end of the robot;

[0011] Calibrate multiple groups of multi-beam laser vision sensors and the main camera and robot coordinate systems to obtain calibration parameters;

[0012] The robot plans the measurement path according to the shape of the object being measured;

[0013] The robot moves on the slide rail and drives multiple sets of multi-beam laser vision sensors to collect laser images of the measured object along the measurement path. The multiple sets of multi-beam laser vision sensors input the collected laser images into the data acquisition and processing system.

[0014] The data acquisition and processing system processes the collected laser images and obtains a multi-channel unified measurement result of the relative angles between multiple measured surfaces of the measured object according to the calibration parameters.

[0015] Preferably, calibration is performed between multiple groups of multi-beam laser vision sensors and between a main camera and a robot coordinate system, including calibrating the multi-beam laser vision sensors, calibrating master-slave cameras in the multiple groups of multi-beam laser vision sensors, and calibrating between the main camera and a robot coordinate system.

[0016] Preferably, calibration of the multi-beam laser vision sensor includes camera internal parameter calibration and structured light calibration; camera internal parameter calibration performs high-precision calibration based on a calibration method of a plane target; structured light calibration performs high-precision calibration based on a calibration method of a luminous plane target.

[0017] Preferably, calibrating a master-slave camera in a plurality of sets of multi-beam laser vision sensors comprises:

[0018] Select one 100-megapixel industrial camera as the master camera and the remaining 100-megapixel industrial cameras as slave cameras;

[0019] A single camera is used to collect a spatial metal target image and obtain the external parameters of the metal target in the camera coordinate system;

[0020] The robot's articulated arm is used to measure the coordinates of the characteristic points on the space metal target, and the coordinates of the metal target in the articulated arm coordinate system are obtained;

[0021] The external parameters between the master camera and the slave camera are calibrated through spatial coordinate system conversion to unify multi-channel measurement data;

[0022] Among them, the relationship between the spatial coordinate system transformation is:

[0023]

[0024] Where, T cb1 is the transformation matrix from camera 1 to the joint arm coordinate system, T cb2 is the transformation matrix from camera 2 to the joint arm coordinate system, T tb1 is the transformation matrix from metal target 1 to the joint arm coordinate system, T tb2 is the transformation matrix from metal target 2 to the joint arm coordinate system, T tc1 is the transformation matrix from metal target 1 to camera 1 coordinate system, T tc2 is the transformation matrix from metal target 2 to camera 2 coordinate system, T 12 is the transformation matrix from camera 2 to camera 1 coordinate system.

[0025] Preferably, calibration is performed between the main camera and the robot coordinate system, including:

[0026] Keep the position of the spatial plane target, the relative position of the robot and the robot coordinate origin unchanged;

[0027] The main camera collects the image of the planar target, calculates the external parameters of the planar target in the camera coordinate system, and simultaneously records the external parameters of the robot as the origin;

[0028] Repeatedly move the robot end and drive multiple groups of multi-beam laser vision sensors to move synchronously, collect plane target images multiple times, calculate the external parameters of the plane target in the camera coordinate system, and synchronously record the spatial position of the robot end, and obtain the motion changes of the two main cameras relative to the spatial plane target and the motion changes of the robot end;

[0029] According to the hand-eye calibration model, the extrinsic parameters between the main camera and the robot coordinate system are solved.

[0030] Preferably, the data acquisition and processing system processes the collected laser images and obtains a multi-channel unified measurement result of the relative angles between multiple measured surfaces of the measured object according to the calibration parameters, including:

[0031] Extract the center position of the structured light strip in the collected laser image and reconstruct the plane fitting in three dimensions, and calculate the plane normal vectors of multiple measured surfaces;

[0032] The plane normal vectors of multiple measured surfaces are converted to the main camera coordinate system, and the relative angles of the measured surfaces are calculated using the normal vectors of the planes measured by the main camera;

[0033] Among them, the calculation formula of the relative angle of the measured surface is:

[0034]

[0035] In the formula, are the plane normal vectors of the measured surfaces 1 and 2 respectively, is the normal vector of the plane measured by the main camera, and α1 and α2 are the angles between the measured surfaces 1 and 2 and the plane measured by the main camera, respectively.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. The present invention adopts a combined layout of multiple groups of multi-beam laser vision sensors to ensure the synchronous detection of the measured object at multiple angles; an industrial robot is used as the motion carrier of the vision sensor, assisted by a data acquisition and processing system, to achieve multi-angle and flexible measurement of large heterogeneous workpieces, which is suitable for precision machining.

[0038] 2. The present invention proposes an angle measurement method based on a multi-beam laser vision sensor. By collecting multiple laser images, a three-dimensional point cloud of the measured surface is obtained through image processing and three-dimensional reconstruction, and then the plane equation and normal vector of the measured surface are calculated, and finally the relative angle of the measured surface is obtained.

[0039] 3. The present invention proposes a multi-sensor flexible calibration method, adopts a "primary-secondary" multi-channel camera calibration mode to achieve the unification of multi-channel visual sensor data; adopts a hand-eye calibration mode to achieve the coordinate unification of multi-channel measurement angles, and realizes flexible angle measurement of large objects.

[0040] 4. The speed measuring device and method proposed in the present invention have a certain degree of versatility. By designing different visual sensor sizes, matching different sensor shape layouts, and using robots of different sizes, multi-angle synchronous measurement requirements in different scenarios can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a flow chart of a small multi-channel flexible space angle measurement method provided by an embodiment of the present invention;

[0042] Figure 2 It is a structural schematic diagram of a small multi-channel flexible space angle measurement device provided by an embodiment of the present invention;

[0043] In the figure: 1. Robot; 2. Multi-beam laser vision sensor; 3. Slide rail; 4. Unit under test; 5. Large component; 6. Multi-channel probe. DETAILED DESCRIPTION

[0044] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0045] Embodiment 1:

[0046] This embodiment provides a small multi-channel flexible space angle measurement device, such as Figure 2 As shown, it is composed of a robot 1, a multi-beam laser vision sensor 2, a slide rail 3 and a data acquisition and processing system. The slide rail 3 is set close to the object to be measured, and the object to be measured is a large component 5, which includes multiple heterogeneous measured units 4; the robot 1 is movably set on the slide rail 3; multiple groups of multi-beam laser vision sensors 2 are set at the end of the robot 1 and are electrically connected to the data acquisition and processing system. Multiple groups of multi-beam laser vision sensors 2 are used to collect multiple laser images of the object to be measured. Each group of multi-beam laser vision sensors 2 constitutes a measurement channel. Each group of multi-beam laser vision sensors 2 can measure an angle value. Synchronous multiple groups of measurements are multi-channel; the data acquisition and processing system is essentially a host computer software, which is responsible for processing laser images and performing data calculations, obtaining multi-channel unified measurement results of the relative angles between multiple measured surfaces of the object to be measured and reporting and displaying them in real time, and real-time feedback of angle deviations to guide product design and improvement.

[0047] Specifically, the multi-beam laser vision sensor 2 includes a pixel industrial camera, a 5-beam near-infrared laser projector, a filter and a mechanical mounting part. The whole constitutes a multi-channel probe 6 installed at the end of the robot 1. The emitted measuring laser is shaped like a probe to accurately collect images and measure angles. The "probe" is small in size and suitable for multi-angle synchronous measurement of large heterogeneous objects (large in size and complex in structure). The pixel industrial camera and the 5-beam near-infrared laser projector are fixedly installed together. The relative positions of the two remain unchanged, and the structural parameters are obtained through offline calibration. The 5-beam near-infrared laser projector irradiates the laser onto the surface of the object to be measured. The pixel industrial camera is supplemented with power supply, trigger and data acquisition cables to collect laser images and input the collected laser images into the data acquisition and processing system for processing and spatial three-dimensional point cloud acquisition, providing a data basis for subsequent angle calculations, and realizing the measurement of the plane three-dimensional point cloud of the measured workpiece and the measurement of local angles. Among them, the resolution of the pixel industrial camera is not less than 5 million pixels, the frame rate is not less than 30HZ, and it has high photosensitivity and high sensitivity; the pixel industrial camera is equipped with a narrow-band filter to realize imaging of near-infrared lasers; the power of the 5-beam near-infrared laser head projector is not less than 1 watt, the light output is uniform and stable, and it can work continuously for a long time.

[0048] Embodiment 2:

[0049] This embodiment provides a small multi-channel flexible spatial angle measurement method, such as Figure 1 As shown, a small multi-channel flexible spatial angle measurement device described in Example 1 is used to perform multi-channel unified measurement of relative angles between multiple measured surfaces of a measured object, including the following steps:

[0050] Step 1: Layout of multiple groups of multi-beam laser vision sensors

[0051] According to the spatial distribution of multiple measured surfaces of the measured object, the spatial layout of multiple groups of multi-beam laser vision sensors 2 at the end of the robot 1 is optimized to ensure that each group of multi-beam laser vision sensors 2 measures the measured surface of the workpiece. Then, according to the distribution of the measured surfaces of the workpiece in space, the minimum synchronous multi-channel measurement sensor module can be designed. It has modular design, flexible matching and certain versatility.

[0052] Step 2: Calibrate multiple sets of multi-beam laser vision sensors and the main camera and robot coordinate system

[0053] The calibration process involves calibrating the multi-beam laser vision sensor, calibrating the master-slave camera in multiple sets of multi-beam laser vision sensors, and calibrating between the master camera and the robot coordinate system.

[0054] The calibration of multi-beam laser vision sensors includes camera internal parameter calibration and structured light calibration. Among them, the camera internal parameter calibration adopts a calibration method based on a plane target, which can achieve high-precision calibration on site; the structured light calibration adopts a calibration method based on a luminous plane target to meet the on-site calibration accuracy requirements.

[0055] The master-slave camera is calibrated in multiple groups of multi-beam laser vision sensors, including: selecting one pixel industrial camera as the master camera and the other cameras as slave cameras; respectively collecting images of a spatial metal target from a single camera to obtain the external parameters of the metal target in the camera coordinate system; using the high-precision articulated arm of robot 1 to measure the coordinates of feature points on the spatial metal target to obtain the coordinates of the metal target point in the articulated arm coordinate system; completing the unification of the coordinate systems of each camera and the articulated arm in turn through the spatial coordinate system conversion, and completing the calibration of the external parameters between the master camera and the slave camera to unify the multi-channel measurement data, and finally all the three-dimensional point coordinates are unified to the master camera coordinate system; wherein the relationship of the spatial coordinate system conversion is:

[0056]

[0057] Where, T cb1 is the transformation matrix from camera 1 to the joint arm coordinate system, T cb2 is the transformation matrix from camera 2 to the joint arm coordinate system, T tb1 is the transformation matrix from metal target 1 to the joint arm coordinate system, T tb2 is the transformation matrix from metal target 2 to the joint arm coordinate system, T tc1 is the transformation matrix from metal target 1 to camera 1 coordinate system, T tc2 is the transformation matrix from metal target 2 to camera 2 coordinate system, T 12 is the transformation matrix from camera 2 to camera 1 coordinate system.

[0058] The calibration between the main camera and the robot coordinate system is performed using the hand-eye calibration method. First, the spatial plane target is placed fixedly, and the relative position between the robot 1 and its coordinate origin o is kept unchanged; the main camera collects the plane target image, calculates the external parameters of the plane target in the camera coordinate system at this time, and synchronously records the external parameters of the robot 1 at this time as the origin; the end of the robot 1 is moved and drives multiple groups of multi-beam laser vision sensors to move synchronously, and the plane target image is collected again, the external parameters of the plane target in the camera coordinate system are calculated, and the spatial position of the end of the robot 1 is synchronously recorded; and so on, repeat the above actions, move the end of the robot 1 again, calculate the external parameters of the plane target at the third position in the camera coordinate system, and synchronously record the spatial position of the end of the robot 1. At this point, the motion changes of the two groups of main cameras relative to the spatial plane target and the motion changes of the end of the robot 1 are obtained respectively, and then the motion change data is substituted into the hand-eye calibration model, and the external parameters between the main camera coordinate system and the robot coordinate system can be solved.

[0059] Step 3: Flexible angle measurement of large heterogeneous workpieces

[0060] Based on the calibration results in step 2, the robot 1 is returned to its coordinate origin o; first, the robot 1 teaches and plans the measurement path according to the morphology of the object to be measured (the spatial distribution of multiple measured units 4 of the object to be measured); secondly, the robot 1 moves and drives the multi-channel probe 6 to image the measured unit 4 along the planned measurement path, controls the laser to project laser stripes onto the measured surface, and the pixel industrial camera collects the image and inputs it into the data acquisition and processing system. The data acquisition and processing system completes the unit calibration according to the calibration results of the above-mentioned camera internal parameter calibration and structural light calibration after image processing. A group of laser vision sensors measure the three-dimensional point cloud of the side surface, and obtain the plane equation and normal vector of the measured surface; according to the relative external parameters between multiple groups of multi-beam laser vision sensors 2, the data of the point cloud measured by multiple groups of laser vision sensors 2 is unified, and then the relative angles between each measured surface are obtained, which specifically includes: two groups of multi-beam laser vision sensors 2 project multiple beams of parallel light to the two measured surfaces of the first and second channels, and the pixel industrial camera captures the image with structured light strips, extracts the center position of the light strips and reconstructs the plane fitting in three dimensions, and solves the plane normal vectors of the measured surfaces 1 and 2. and The plane normal vector and The quantity is converted to the main camera coordinate system, and the normal vector of the plane measured by the main camera is Calculate the relative angles of the measured surfaces 1 and 2, that is, the angles α1 and α2 between the measured surfaces 1 and 2 and the plane measured by the main camera. The other channels can be calculated by analogy. The calculation formula is:

[0061]

[0062] According to the external parameters between the main camera and the robot coordinate system, the transformation of the plane three-dimensional point clouds and angles of multiple measured units 4 to the robot coordinate system is realized, and the unification of multi-channel detection data is completed. Specifically, when the measurement of the first measured unit 4 is completed, the robot 1 retracts the multi-channel probe 6, and the robot 1 automatically moves to the next detection position through the slide rail 3, and repeats the above steps to complete the next detection position until the measurement of all the angles of the measured units 4 of the large component 5 is completed and converted to the coordinate origin o. At this point, the dynamic measurement of the large component 5 based on the small multi-channel flexible space angle measurement device is completed.

[0063] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A small multi-channel flexible space angle measurement method, characterized in that: include: According to the spatial distribution of multiple measured surfaces of the measured object, optimize the spatial layout of multiple groups of multi-beam laser vision sensors at the end of the robot; Calibrate multiple groups of multi-beam laser vision sensors and the main camera and robot coordinate systems to obtain calibration parameters; The robot plans the measurement path according to the shape of the object being measured; The robot moves on the slide rail and drives multiple sets of multi-beam laser vision sensors to collect laser images of the measured object along the measurement path. The multiple sets of multi-beam laser vision sensors input the collected laser images into the data acquisition and processing system. The data acquisition and processing system processes the collected laser images and obtains the multi-channel unified measurement results of the relative angles between multiple measured surfaces of the measured object according to the calibration parameters; Among them, calibration is performed between multiple groups of multi-beam laser vision sensors and between a main camera and a robot coordinate system, including calibration of the multi-beam laser vision sensors, calibration of master-slave cameras in multiple groups of multi-beam laser vision sensors, and calibration between the main camera and a robot coordinate system; The data acquisition and processing system processes the collected laser images and obtains the multi-channel unified measurement results of the relative angles between multiple measured surfaces of the measured object according to the calibration parameters, including: Extract the center position of the structured light strip in the collected laser image and reconstruct the plane fitting in three dimensions, and calculate the plane normal vectors of multiple measured surfaces; The plane normal vectors of multiple measured surfaces are converted to the main camera coordinate system, and the relative angles of the measured surfaces are calculated using the normal vectors of the planes measured by the main camera; The calculation formula of the relative angle of the measured surface is: In the formula, are the plane normal vectors of the measured surfaces 1 and 2 respectively, is the normal vector of the plane measured by the main camera, and α1 and α2 are the angles between the measured surfaces 1 and 2 and the plane measured by the main camera, respectively.

2. The small multi-channel flexible spatial angle measurement method according to claim 1 is characterized in that: The calibration of multi-beam laser vision sensors includes camera internal parameter calibration and structured light calibration; the camera internal parameter calibration is based on the calibration method of planar targets for high-precision calibration; the structured light calibration is based on the calibration method of luminous planar targets for high-precision calibration.

3. The small multi-channel flexible spatial angle measurement method according to claim 1, characterized in that: Calibrate master-slave cameras in multiple sets of multi-beam laser vision sensors, including: Select one 100-megapixel industrial camera as the master camera and the remaining 100-megapixel industrial cameras as slave cameras; A single camera is used to collect a spatial metal target image and obtain the external parameters of the metal target in the camera coordinate system; The robot's articulated arm is used to measure the coordinates of the characteristic points on the space metal target, and the coordinates of the metal target in the articulated arm coordinate system are obtained; The external parameters between the master camera and the slave camera are calibrated through spatial coordinate system conversion to unify multi-channel measurement data; Among them, the relationship between the spatial coordinate system transformation is: Where, T cb1 is the transformation matrix from camera 1 to the joint arm coordinate system, T cb2 is the transformation matrix from camera 2 to the joint arm coordinate system, T tb1 is the transformation matrix from metal target 1 to the joint arm coordinate system, T tb2 is the transformation matrix from metal target 2 to the joint arm coordinate system, T tc1 is the transformation matrix from metal target 1 to camera 1 coordinate system, T tc2 is the transformation matrix from metal target 2 to camera 2 coordinate system, T 12 is the transformation matrix from camera 2 to camera 1 coordinate system.

4. The small multi-channel flexible spatial angle measurement method according to claim 1, characterized in that: Calibrate the main camera and the robot coordinate system, including: Keep the position of the spatial plane target, the relative position of the robot and the robot coordinate origin unchanged; The main camera collects the image of the planar target, calculates the external parameters of the planar target in the camera coordinate system, and simultaneously records the external parameters of the robot as the origin; Repeatedly move the robot end and drive multiple groups of multi-beam laser vision sensors to move synchronously, collect plane target images multiple times, calculate the external parameters of the plane target in the camera coordinate system, and synchronously record the spatial position of the robot end, and obtain the motion changes of the two main cameras relative to the spatial plane target and the motion changes of the robot end; According to the hand-eye calibration model, the extrinsic parameters between the main camera and the robot coordinate system are solved.

5. A small multi-channel flexible space angle measurement device, used to implement the small multi-channel flexible space angle measurement method according to any one of claims 1 to 4, characterized in that: The invention comprises a robot, a multi-beam laser vision sensor, a slide rail and a data acquisition and processing system; the slide rail is arranged close to the object to be measured; the robot is movably arranged on the slide rail; a plurality of groups of multi-beam laser vision sensors are arranged at the end of the robot and are electrically connected to the data acquisition and processing system, so as to collect a plurality of laser images of the object to be measured and process and calculate them, so as to obtain a multi-channel unified measurement result of the relative angles between a plurality of measured surfaces of the object to be measured.

6. The small multi-channel flexible spatial angle measuring device according to claim 5, characterized in that: The multi-beam laser vision sensor includes a pixel industrial camera and a 5-beam near-infrared laser head projector; the pixel industrial camera and the 5-beam near-infrared laser projector are fixedly installed, the 5-beam near-infrared laser projector irradiates the laser onto the surface of the object to be measured, and the pixel industrial camera is used to collect laser images.

7. The small multi-channel flexible spatial angle measuring device according to claim 6, characterized in that: The resolution of the pixel industrial camera shall not be less than 5 million pixels, and the frame rate shall not be less than 30hz; the power of the 5-beam near-infrared laser head projector shall not be less than 1 watt; the pixel industrial camera shall be equipped with a narrow-band filter.

Images