An automatic calibration robot and track mechanism system and automatic calibration method

Abstract

The present invention belongs to the technical field of industrial robots, and discloses an automatically calibrated robot and track mechanism system and an automatic calibration method. A calibration tool is placed in the reachable area of ​​each track section; a global visual sensor collects the point cloud coordinate information of the calibration tool, an information processing computer performs rough positioning, and plans the transition path for the multi-axis robot to reach the location of the positioning tool; plans the non-interference motion system execution path required for calibration and the measurement position of the local visual sensor; the local visual sensor collects the point cloud coordinate information, and the information processing computer performs precise positioning; and determines whether to enter the calibration solution; the automatic calibration process ends through the feature quantity solution set obtained by the solution and the calibration relationship between the multi-axis robot and the track motion mechanism. The present invention obtains the calibration relationship between the robot and the track mechanism by means of automatic planning, automatic execution, and automatic solution, and can accurately and efficiently realize the calibration of the multi-axis robot coordinates corresponding to the target point.

Description

Technical Field

[0001] The invention belongs to the technical field of industrial robots and relates to a fully automatic robot and track mechanism system automatic calibration method. Background Art

[0002] In the existing robot system technology, there is usually a problem that the coordinate system under different track mechanisms and the robot coordinate system do not overlap during use due to installation problems. For the calibration of the installation relationship between the robot and the track mechanism in different positions, the relationship needs to be calibrated. The existing calibration method uses manual data collection for calibration. The robot is manually moved to the target point corresponding to the feature, and then the system records the robot coordinates corresponding to the feature point. This process is not only time-consuming and unstable in accuracy, but also depends on the operator's experience. At the same time, when the system collides or the system installation status changes, it needs to be recalibrated. This process is very cumbersome and requires a very high level of operator operation. Summary of the invention

[0003] In order to solve the above-mentioned defects existing in the prior art, the purpose of the present invention is to overcome the defects of the current calibration that manual aiming is required, point-by-point measurement is time-consuming, inefficient, and requires high operating level of personnel. A method for automatic calibration of a robot and a track mechanism system is provided. By simply setting a continuous or fixed shooting position of a global vision sensor, the system can automatically plan the motion system execution path and shooting position required for calibration, and obtain the calibration relationship between the robot and the track mechanism through automatic planning, automatic execution, and automatic solution, so that the calibration work can be completed concisely, accurately and efficiently.

[0004] The present invention is achieved through the following technical solutions.

[0005] One aspect of the present invention provides an automatic calibration robot and track mechanism system, comprising:

[0006] A motion system, including an orbital motion mechanism and a multi-axis robot, wherein the orbital motion mechanism is configured with a multi-dimensional orthogonal motion system for connecting with a multi-axis robot arm to perform orbital motion within a working area;

[0007] Calibration tooling: at least one calibration tooling is configured for each track section, and is used to determine the structural characteristics of the track according to the number of track sections;

[0008] A global vision sensor, mounted on a track motion mechanism, is used to collect point cloud coordinate information of the calibration tooling under a global field of view through static, continuous or discrete motion;

[0009] The local vision sensor is installed on the actuator of the multi-axis robot to obtain the point cloud coordinate information of the calibration tooling in the local field of view;

[0010] The information processing computer is connected to the motion system, the global vision sensor and the local vision sensor respectively, and is used to receive the point cloud coordinate information of the global vision sensor and the local vision sensor, autonomously identify the initial position and posture of all calibration tools, perform feature solution based on the initial position and posture to obtain feature coordinate information; and plan the execution path of the motion system and the measurement position of the local vision sensor according to the feature coordinate information, and control the motion system to execute the planned path sequence.

[0011] Preferably, the orbital motion mechanism is configured with a multi-dimensional orthogonal motion system including a one-dimensional orthogonal motion system, a two-dimensional orthogonal motion system or a three-dimensional orthogonal motion system.

[0012] Preferably, the multi-axis robot is at least a robot with a six-degree-of-freedom robotic arm.

[0013] Preferably, each track section is equipped with at least one calibration tool which is placed in the working area of ​​the global vision sensor and in the reachable area of ​​the multi-axis robot system; the information processing computer obtains the relative posture relationship between the track motion mechanism and the multi-axis robot based on the characteristic information of the calibration tool, thereby realizing auxiliary positioning of the calibration tool.

[0014] Preferably, the field of view of the global vision sensor is larger than that of the local vision sensor. The global vision sensor collects point cloud coordinate information of the calibration tooling and performs coarse positioning through an information processing computer; the local vision sensor collects precise point cloud coordinate information of the calibration tooling and performs fine positioning through an information processing computer.

[0015] Another aspect of the present invention provides an automatic calibration method for the automatic calibration robot and track mechanism system, comprising the following steps:

[0016] 1) Place at least one calibration tool in the accessible area of ​​each track section;

[0017] 2) Determine the shooting position of the continuous or fixed global vision sensor, the track motion mechanism and the multi-axis robot movement, and at the shooting position, the global vision sensor on the track motion mechanism collects the coordinate information of the calibration tooling;

[0018] 3) The information processing computer obtains the point cloud coordinate information of the calibration tooling collected by the global vision sensor, identifies the location of the calibration tooling through feature information for rough positioning, and automatically plans the transition path for the multi-axis robot to reach the location of the positioning tooling based on the result of the rough positioning;

[0019] 4) The information processing computer calibrates the tooling position, plans the execution path of the motion system that does not interfere with the calibration, and the measurement position of the local vision sensor;

[0020] 5) The motion system moves to the measurement position of the local vision sensor according to the planned execution path, and the local vision sensor collects point cloud coordinate information;

[0021] 6) The information processing computer obtains the point cloud coordinate information collected by the local visual sensor, identifies and calibrates the position of the tooling through feature information, and performs precise positioning to obtain the feature coordinates in the coordinate system of the track motion mechanism and the feature coordinates in the coordinate system of the multi-axis robot;

[0022] 7) If the number of feature coordinates collected meets the solution requirements, then enter feature solution; if not, repeat steps 5) to 6) and repeat the process of collecting point cloud coordinate information with local visual sensors and processing information for computer precise positioning;

[0023] 8) The information processing computer obtains the solution set of characteristic coordinate quantities and the calibration relationship between the multi-axis robot and the track motion mechanism, and the automatic calibration process ends.

[0024] Preferably, the information processing computer obtains the point cloud coordinate information of the calibration tooling collected by the global vision sensor, identifies the position of the calibration tooling through feature information for rough positioning, and automatically plans a transition path for the multi-axis robot to reach the position of the positioning tooling according to the result of the rough positioning, including:

[0025] 3a) Using filtering method to remove interference points in the point cloud coordinate information of the calibration tooling under the global field of view;

[0026] 3b) Using the template matching method, the three-dimensional point cloud data containing only the calibration tooling is segmented and extracted from the point cloud coordinate data of the calibration tooling under the global field of view;

[0027] 3c) Perform feature recognition and extraction on the obtained three-dimensional point cloud data of the calibration tooling to obtain feature coordinate information on the calibration tooling in the global point cloud coordinate system;

[0028] 3d) Convert the three-dimensional coordinates in the global point cloud coordinate system to the three-dimensional coordinates in the multi-axis robot coordinate system;

[0029] 3e) Convert the three-dimensional coordinates in the multi-axis robot coordinate system to the three-dimensional coordinates in the multi-axis robot flange coordinate system;

[0030] 3f) Locate the position of the current calibration tooling according to the position in the multi-axis robot flange coordinate system;

[0031] 3g) Based on the calibrated tooling position, plan the transition path for the multi-axis robot to reach the calibrated tooling position.

[0032] Preferably, the information processing computer obtains the point cloud coordinate information collected by the local visual sensor, and performs precise positioning by identifying and calibrating the position of the tooling through feature information, including:

[0033] 6a) Using filtering method to remove interference points in the point cloud coordinate information of the calibration tooling in the local field of view;

[0034] 6b) Using the template matching method, segment and extract the three-dimensional point cloud data containing only the calibration tooling from the point cloud coordinate information of the calibration tooling in the local field of view;

[0035] 6c) Perform feature recognition and extraction on the obtained three-dimensional point cloud data of the calibration tooling to obtain feature coordinate information on the calibration tooling in the local point cloud coordinate system;

[0036] 6d) Convert the three-dimensional coordinates in the local point cloud coordinate system to the three-dimensional coordinates in the flange coordinate system of the multi-axis robot;

[0037] 6e) Locate the position of the current calibration tooling according to the position of the feature in the multi-axis robot flange coordinate system.

[0038] As a preferred method, feature solving, the steps are as follows:

[0039] 7a) The characteristic coordinates in the coordinate system of the track motion mechanism and the characteristic coordinates in the coordinate system of the multi-axis robot collected by the local vision sensor;

[0040] 7b) The number of feature coordinates collected meets the following solution requirements:

[0041]

[0042] Among them, Pc1, Pc2…Pci are the three-dimensional coordinates of the visible calibration features in the multi-axis robot coordinate system; R cb and is the coordinate transformation matrix between the track coordinate system and the multi-axis robot; Pr1, Pr2…Pri are the three-dimensional coordinates of the visible calibration features in the track motion mechanism coordinate system;

[0043] The coordinate conversion matrix between the track coordinate system and the multi-axis robot is obtained by solving the least square method based on the obtained three-dimensional coordinates of the calibration feature in the multi-axis robot coordinate system and the three-dimensional coordinates of the calibration feature in the track motion mechanism coordinate system.

[0044] Preferably, the information processing computer solves the solution set of characteristic coordinate quantities to obtain the rotation relationship Rx and translation relationship Tx of the track mechanism relative to the robot coordinate system at different coordinate positions; and obtains the data conversion relationship between the two coordinate systems.

[0045] The present invention adopts the above technical solution, which has the following beneficial effects:

[0046] 1) The method of manually placing the calibration fixture, automatically identifying the characteristic position of the calibration fixture, automatically planning the transition path, and automatically solving the calibration result replaces the original manual operation point selection calibration process, automates the calibration process operation, improves the automation degree of the calibration process, and reduces the manual time consumption of the process. At the same time, it does not require the high technical level of the operator, and can be simply set up to obtain the calibration results of the track mechanism and the multi-axis robot through the automatic operation of the system.

[0047] 2) By ensuring that the motion paths of the motion system executed during the automatic calibration process are all automatically planned non-interference paths, there is no need to manually set calibration collection points and process path points, which reduces the degree of manual participation and time consumption of the calibration process.

[0048] 3) Through the segmentation and extraction of the global point cloud, the characteristic information of the calibration tooling is identified and extracted, and the characteristics under the global point cloud coordinate system are converted to the multi-axis robot coordinate system. The entire process is automatically executed without human participation and without complex requirements on the operator's operating level.

[0049] 4) By filtering the local point cloud, segmenting and extracting it through template matching, the feature positions of the points with the same name in the point cloud are identified through the feature recognition algorithm, and the features in the local point cloud coordinate system are converted to the multi-axis robot coordinate system. The feature coordinate system information of the automatic calibration process is automatically solved by the algorithm, which improves the accuracy of the coordinate point position and the stability of the coordinates.

[0050] 5) By identifying the calibration tooling features at different ground rail positions, the least squares method is used to calculate the transformation matrix between the multi-axis robot and the track coordinate system for the feature coordinate sequence pairs of the same-name points, thereby improving the accuracy of the feature transformation matrix.

[0051] 4) The number of calibration tools can be flexibly set according to the number of track segments, which improves the simplicity of manual operation during the calibration process.

[0052] The robot automatic calibration algorithm process is simple and efficient. The calibration process is automatically executed without the need for complex manual operations. The process is simple and efficient with stable accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of the present application, and do not constitute an improper limitation of the present invention. In the drawings:

[0054] Figure 1 It is composed of a robot system;

[0055] Figure 2It is a schematic diagram of the system automatic calibration process;

[0056] In the figure, 101 is an information processing computer; 102 is a track motion mechanism; 103 is a multi-axis robot; 104 is a global vision sensor; 105 is a local vision sensor; and 106 is a calibration tool. DETAILED DESCRIPTION

[0057] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments. The exemplary embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0058] Figure 1 The figure shows a three-dimensional schematic diagram of an automatic calibration robot and track mechanism system provided by the present embodiment, including an information processing computer 101, a motion system, a global vision sensor 104, a local vision sensor 105 and a calibration tool 106, wherein:

[0059] The motion system includes an orbital motion mechanism 102 and a multi-axis robot 103. The orbital motion mechanism 102 is configured with a multi-dimensional orthogonal motion system for connecting with a multi-axis robot arm to perform orbital motion within a working area.

[0060] The calibration tool 106 determines the structural characteristics of the track according to the number of track segments, ensuring that there is at least one calibration tool for each track segment, which is used to determine the structural characteristics of the track according to the number of track segments.

[0061] The global vision sensor 104 is installed on the track motion mechanism 102, and is used to collect the point cloud coordinate information of the calibration tooling 106 under the global field of view through static, continuous or discrete motion, and the coordinate conversion relationship between the global sensor and the multi-axis robot arm has been calibrated.

[0062] The local vision sensor 105 is installed on the actuator of the multi-axis robot 103 and is used to obtain the point cloud coordinate information of the calibration tool 106 under the local field of view, and the coordinate conversion relationship between the local sensor and the multi-axis robot arm has been calibrated.

[0063] In the current system, the relationship between the robot arm and the track mechanism is calibrated, where the robot arm itself has been calibrated, and the installation relationship between the robot arm and the global vision sensor and the local vision sensor has been calibrated. The accuracy of the global vision sensor is lower than that of the local vision sensor, and the comprehensive accuracy of autonomous recognition, global vision, and autonomous planning is lower than the perception measurement range of the local vision sensor. Through the more accurate local vision sensor data collection, high-precision calibration data is finally obtained.

[0064] The information processing computer 101 is respectively connected to the motion system (track motion mechanism 102 and multi-axis robot 103), the global vision sensor 104 and the local vision sensor 105, and is used to receive the point cloud coordinate information of the global vision sensor 104 and the local vision sensor 105, autonomously identify the initial position and posture of all calibration tools 106, perform feature solution based on the initial position and posture to obtain feature coordinate information, and plan the execution path of the motion system and the measurement position of the local vision sensor 105 according to the feature coordinate information, control the motion system to execute the planned path sequence, and perform calibration result solution after obtaining the feature sequence that meets the quantity requirements.

[0065] The comprehensive recognition accuracy of the global vision sensor 104 is smaller than the perception measurement range of the local vision sensor 105. The comprehensive system execution accuracy automatically plans the precise acquisition motion system execution path and the acquisition position of the local vision sensor 105. The motion system executes the planned path, and the local vision sensor 105 collects precise point cloud coordinate information. The precise feature coordinate information is obtained through the information processing computer 101, and finally the posture transformation matrix of the track mechanism system and the multi-axis robot arm at different positions, the straightness error of the track extension axis, and the angle deviation between the extension axis and the design datum are solved.

[0066] Figure 1 A schematic diagram of a six-axis robot is shown, but the present invention is not limited to the illustrated implementation example, and may be other types of multi-axis robot 103 mechanical arms.

[0067] Figure 1 A schematic diagram of a one-dimensional orthogonal motion system is shown; however, the present invention is not limited to the illustrated implementation example, and may be a two-dimensional or three-dimensional orthogonal motion system.

[0068] Figure 1 The embodiment includes two vision sensors, a global vision sensor 104 and a local vision sensor 105, which are used to obtain data information under the field of view of the vision sensor and transmit it to the information processing computer 101.

[0069] Among them, the global visual sensor has a large field of view and can collect calibration tooling information for rough positioning, while the local visual sensor has a small field of view and can collect calibration tooling information for precise positioning. The calibration tooling 106 has known feature information, and the information processing computer 101 identifies the position coordinate information of the tooling through the known feature information and the data information collected by the visual sensor.

[0070] The calibration tooling includes one or more calibration toolings, which can be placed in the working area of ​​the global visual sensor and in the reachable area of ​​the multi-axis robot system. The calibration tooling features are used to assist in positioning the calibration tooling. Based on the calibration tooling feature information, the relative position and posture relationship between the track mechanism and the robot is calculated.

[0071] The motion of the motion system includes the motion of the orbital motion mechanism, the motion of the multi-axis robot, and the joint motion of the orbital motion mechanism and the multi-axis robot.

[0072] This system collects data through the global vision sensor 104 for coarse positioning, and then the information processing computer 101 automatically plans according to the result of coarse positioning to obtain the motion path for fine positioning, and then performs fine positioning through the local vision sensor 105, and solves the characteristic information of the calibration tooling 106, which is used to solve the calibration relationship between the track mechanism and the multi-axis robot arm.

[0073] like Figure 2 As shown, an embodiment of the present invention provides an automatic calibration method for a robot and a track mechanism system, comprising the following steps:

[0074] Step 1: Place at least one calibration tool 106 within the accessible area of ​​each track section.

[0075] Step 2, determine the shooting position of the continuous or fixed global vision sensor 104, the track motion mechanism 102 and the multi-axis robot 103 of the motion system move, and after moving into place, the global vision sensor 104 installed on the track motion mechanism 102 at the shooting position collects the calibration tooling point cloud coordinate information.

[0076] The present invention is not limited to point cloud coordinate information, but also includes image coordinate information and video coordinate information; feature coordinates include but are not limited to plane coordinates, vertex coordinates, sphere center coordinates, contour coordinates and irregular feature coordinate structures.

[0077] In step 3, the global vision sensor 104 transmits the collected point cloud coordinate information of the calibration tooling to the information processing computer 101. The information processing computer 101 identifies the position of the calibration tooling 106 through feature information for rough positioning, and automatically plans the transition path for the multi-axis robot 103 to reach the position of the positioning tooling based on the result of the rough positioning.

[0078] In step 3, the information processing computer 101 collects feature information through the global vision sensor 104 to identify the position of the calibration tool 106 for rough positioning, and performs automatic planning based on the result of the rough positioning, including the following steps:

[0079] 3a) Using filtering method to remove interference points in the coordinate point cloud information of the calibration tooling under the global field of view;

[0080] 3b) Using the template matching method, the three-dimensional point cloud data containing only the calibration tooling is segmented and extracted from the coordinate point cloud information of the calibration tooling under the global field of view;

[0081] 3c) Perform feature recognition and extraction on the obtained three-dimensional point cloud data of the calibration tooling to obtain feature coordinate information on the calibration tooling in the global point cloud coordinate system;

[0082] 3d) Convert the three-dimensional coordinates in the global point cloud coordinate system to the three-dimensional coordinates in the multi-axis robot coordinate system;

[0083] 3e) Convert the three-dimensional coordinates in the multi-axis robot coordinate system to the three-dimensional coordinates in the multi-axis robot flange coordinate system;

[0084] 3f) Locate the current positioning tooling according to the position in the multi-axis robot flange coordinate system;

[0085] 3g) According to the position of the calibration tooling, a transition path for the multi-axis robot to reach the position of the calibration tooling is planned, and the path interferes with the calibration tooling and the system.

[0086] The non-interference motion system execution path and the measurement position of the local vision sensor required for planning calibration include n groups, where n is greater than the number of solution sets required to solve the calibration relationship (n≥3), and each group includes multiple (≥3) multi-axis robot execution pose coordinates and 1 track motion mechanism execution coordinate, and multiple (≥3) local vision sensor measurement position coordinates.

[0087] Step 4, the information processing computer 101 plans the non-interfering motion system execution path required for calibration and the measurement position of the local vision sensor 105 by calibrating the position of the tooling 106, and the path interferes with the calibration tooling and the system.

[0088] Step 5: The motion system moves according to the planned execution path and moves to the measurement position of the local vision sensor 105 to collect point cloud coordinate information.

[0089] The motion system moves according to the planned execution path. The execution order is that the track motion mechanism executes the planned point coordinate movement, the multi-axis robot executes multiple execution pose coordinates (≥3) according to the planned sequence, and the multi-axis robot executes the local vision sensor to measure the position coordinates.

[0090] Step 6: The local vision sensor 105 transmits the collected point cloud coordinate information to the information processing computer 101 . The information processing computer 101 identifies and calibrates the position of the tooling 106 through the feature information; and performs precise positioning through the local vision sensor 105 .

[0091] In step 6, the information processing computer performs precise positioning through the local visual sensor 105 by identifying the feature information, including the following steps:

[0092] 6a) Using filtering method to remove interference points in the coordinate point cloud information of the calibration tooling under the global field of view;

[0093] 6b) using a template matching method to segment and extract the three-dimensional point cloud data containing only the calibration tooling from the coordinate point cloud information of the calibration tooling in the local field of view;

[0094] 6c) performing feature recognition and extraction on the obtained three-dimensional point cloud data of the calibration tooling to obtain feature coordinate information on the calibration tooling in the local point cloud coordinate system;

[0095] 6d) converting the three-dimensional coordinates in the local point cloud coordinate system into the three-dimensional coordinates in the multi-axis robot flange coordinate system;

[0096] 6e) Locate the position of the current calibration tooling according to the position of the feature in the multi-axis robot flange coordinate system.

[0097] Step 7: If the number of features collected meets the solution requirements, the calibration solution is entered; if it does not meet the requirements, steps 5 to 6 are repeated to repeat the local visual sensor 105 data collection and the information processing computer 101 precise positioning process.

[0098] In step 7, feature solving, the steps are as follows:

[0099] 7a) according to the characteristic coordinates in the coordinate system of the track motion mechanism and the characteristic coordinates in the coordinate system of the multi-axis robot collected by the local vision sensor 105;

[0100] 7b) The number of feature coordinates collected meets the following solution requirements:

[0101]

[0102] Among them, Pc1, Pc2…Pci are the three-dimensional coordinates of the visible calibration features in the multi-axis robot coordinate system; R cb and is the coordinate transformation matrix between the track coordinate system and the multi-axis robot; Pr1, Pr2…Pri are the three-dimensional coordinates of the visible calibration features in the track motion mechanism coordinate system;

[0103] The coordinate conversion matrix between the track coordinate system and the multi-axis robot is obtained by solving the least square method based on the obtained three-dimensional coordinates of the calibration feature in the multi-axis robot coordinate system and the three-dimensional coordinates of the calibration feature in the track motion mechanism coordinate system.

[0104] Step 8: The information processing computer 101 obtains the characteristic quantity solution set and the calibration relationship between the multi-axis robot and the track mechanism through the above steps, and the automatic calibration process ends.

[0105] In step 8, the calibration relationship between the multi-axis robot and the track mechanism is as follows:

[0106] The information processing computer obtains the characteristic quantity solution set through the above steps, and the number of solution sets is ≥3, and the rotation relationship Rx and translation relationship Tx of the track mechanism relative to the robot coordinate system at different coordinate positions are obtained; wherein the data conversion relationship between the two coordinate systems is shown in the following formula:

[0107]

[0108] in is the point in the robot coordinate system, is the coordinate of a point in the multi-axis robot coordinate system in the sensor coordinate system.

[0109] The present invention is further illustrated by a specific embodiment below.

[0110] The track coordinate system XbaseYbaseZbase is located at the zero point of the track. The XbaseYbase plane formed by the Xbase axis and the Ybase axis is affected by the horizontality of the multi-axis robot base installation and is approximately located in the horizontal plane. The robot's coordinate system is XrobotYrobotZrobot. The Zrobot axis is perpendicular to the XrobotYrobot plane formed by the Xrobot axis and the Yrobot axis. The multi-axis robot can move in the Ybase axis direction of the track. The movement value of the multi-axis robot in the track direction is recorded as bp.

[0111] The calibration tooling is placed manually, and the shooting position of the tooling is set to [0,2000mm]. The track motion mechanism and multi-axis robot of the motion system move. After the movement is in place, the global vision sensor installed on the track motion mechanism collects the coordinates and feature information of the calibration tooling. The filtering method is adopted, and the number of adjacent point query points is set to 50 by the filtering parameter. Then the filtered result is template matched. The point cloud information of the calibration tooling is extracted from the point cloud data by template matching. The algorithm automatically recognizes and solves. At this time, the feature point sequence on the tooling is identified and converted to the multi-axis robot coordinate system. According to the feature results of the solution, 6 feature coordinate positions are identified, and each feature coordinate is shown in the following table.

[0112] Table 1 Three-dimensional coordinates of the characteristic vertices used for calibration in the multi-axis robot coordinate system

[0113]

[0114] According to the coordinate position of each feature, the shooting position and transition path are planned under the position of the ground track to be measured. This path does not interfere with the calibration tooling and multi-axis robot. The track zero position is equivalent to the origin of the track coordinate system, and the movement of the robot on the track during shooting is bp=0mm. The shooting position and transition path are planned under the current track position. At the same time, the shooting position and transition path of the robot are planned under the track position bp=1200mm to be measured. The motion system moves according to the planned execution path, and moves to the measurement position of the local vision sensor to collect data. The local vision sensor transmits the collected data to the information processing computer, and the feature information of the calibration tooling is recognized through feature information for precise positioning.

[0115] After the local camera obtains the local point cloud, the filtering method is used, and the filtering parameters set the number of adjacent point query points to 50. Then the filtered results are template matched. The template matching extracts the point cloud information of the calibration tooling from the point cloud data. The algorithm automatically recognizes and solves. At this time, the feature point sequence on the tooling is identified and converted to the multi-axis robot coordinate system. At this time, the feature point sequence on the tooling is identified and converted to the multi-axis robot coordinate system and the track mechanism coordinate system. Each feature coordinate is shown in the following table:

[0116] Table 2 Three-dimensional coordinates of the characteristic vertices used for calibration that are visible during precise positioning

[0117]

[0118] The three-dimensional coordinates of multiple pairs of visible calibration feature vertices in the track coordinate system and the multi-axis robot coordinate system are solved using the least squares method to obtain the coordinate transformation matrix between the track coordinate system and the multi-axis robot. The Rx matrix is ​​and the Tx matrix is:

[0119]

[0120] It can be seen from the above embodiments that the present invention identifies the coordinates of feature points by filtering and template matching the point cloud data, thereby solving the problem that the current calibration measurement is time-consuming and inefficient, and that manual aiming is required to locate the feature coordinates during calibration. By automatically planning the non-interference measurement path of the measurement and calibration tooling, the manual teaching of the transition path to the aiming point is replaced, thereby improving the efficiency and accuracy of the calibration process. The entire process is automatically planned, executed, and solved, without manual participation, and the final calibration matrix can be obtained, overcoming the shortcomings of manual aiming, which is time-consuming and requires high operating skills of personnel.

[0121] The present invention is not limited to the above-mentioned embodiments. On the basis of the technical solution disclosed in the present invention, technicians in this field can make some substitutions and deformations to some technical features therein according to the disclosed technical content without creative labor, and these substitutions and deformations are all within the protection scope of the present invention.

Claims

1. An automatic calibration robot and track mechanism system, characterized in that: include: A motion system, including an orbital motion mechanism and a multi-axis robot, wherein the orbital motion mechanism is configured with a multi-dimensional orthogonal motion system for connecting with a multi-axis robot arm to perform orbital motion within a working area; Calibration tooling: at least one calibration tooling is configured for each track section, and is used to determine the structural characteristics of the track according to the number of track sections; A global vision sensor, mounted on a track motion mechanism, is used to collect point cloud coordinate information of the calibration tooling under a global field of view through static, continuous or discrete motion; The local vision sensor is installed on the actuator of the multi-axis robot to obtain the point cloud coordinate information of the calibration tooling in the local field of view; The information processing computer is connected to the motion system, the global vision sensor and the local vision sensor respectively, and is used to receive the point cloud coordinate information of the global vision sensor and the local vision sensor, autonomously identify the initial position and posture of all calibration tools, perform feature solution based on the initial position and posture to obtain feature coordinate information; and plan the execution path of the motion system and the measurement position of the local vision sensor according to the feature coordinate information, and control the motion system to execute the planned path sequence.

2. The automatic calibration robot and track mechanism system according to claim 1, characterized in that: The orbital motion mechanism is configured with a multi-dimensional orthogonal motion system including a two-dimensional orthogonal motion system or a three-dimensional orthogonal motion system.

3. The automatic calibration robot and track mechanism system according to claim 1, characterized in that: The multi-axis robot is at least a robot with a six-degree-of-freedom manipulator.

4. The automatic calibration robot and track mechanism system according to claim 1, characterized in that: Each track section is equipped with at least one calibration tool placed in the working area of ​​the global vision sensor and within the reachable area of ​​the multi-axis robot system; the information processing computer obtains the relative posture relationship between the track motion mechanism and the multi-axis robot based on the characteristic coordinate information of the calibration tool, thereby realizing auxiliary positioning of the calibration tool.

5. The automatic calibration robot and track mechanism system according to claim 1, characterized in that: The field of view of the global vision sensor is larger than that of the local vision sensor. The global vision sensor collects the point cloud coordinate information of the calibration tooling and performs coarse positioning through an information processing computer; the local vision sensor collects the precise point cloud coordinate information of the calibration tooling and performs fine positioning through an information processing computer.

6. An automatic calibration method for a robot and a track mechanism system according to any one of claims 1 to 5, characterized in that: The steps include: 1) Place at least one calibration tool in the accessible area of ​​each track section; 2) Determine the shooting position of the continuous or fixed global vision sensor, the track motion mechanism and the multi-axis robot movement, and at the shooting position, the global vision sensor on the track motion mechanism collects the point cloud coordinate information of the calibration tooling; 3) The information processing computer obtains the point cloud coordinate information of the calibration tooling collected by the global vision sensor, identifies the location of the calibration tooling through feature information for rough positioning, and automatically plans the transition path for the multi-axis robot to reach the location of the positioning tooling based on the result of the rough positioning; 4) The information processing computer calibrates the tooling position, plans the execution path of the motion system that does not interfere with the calibration, and the measurement position of the local vision sensor; 5) The motion system moves to the measurement position of the local vision sensor according to the planned execution path, and the local vision sensor collects point cloud coordinate information; 6) The information processing computer obtains the point cloud coordinate information collected by the local visual sensor, identifies and calibrates the position of the tooling through feature information, and performs precise positioning to obtain the feature coordinates in the coordinate system of the track motion mechanism and the feature coordinates in the coordinate system of the multi-axis robot; 7) If the number of feature coordinates collected meets the solution requirements, then enter feature solution; if not, repeat steps 5) to 6) and repeat the process of collecting point cloud coordinate information with local visual sensors and processing information for computer precise positioning; 8) The information processing computer obtains the solution set of characteristic coordinate quantities and the calibration relationship between the multi-axis robot and the track motion mechanism, and the automatic calibration process ends.

7. The automatic calibration method of the robot and track mechanism system according to claim 6, characterized in that: The information processing computer obtains the point cloud coordinate information of the calibration tooling collected by the global vision sensor, identifies the location of the calibration tooling through feature information for rough positioning, and automatically plans the transition path for the multi-axis robot to reach the location of the positioning tooling based on the result of the rough positioning, including: 3a) Using filtering method to remove interference noise in the point cloud coordinate information of the calibration tooling under the global field of view; 3b) Using the template matching method, the three-dimensional point cloud data containing only the calibration tooling is segmented and extracted from the point cloud coordinate information of the calibration tooling under the global field of view; 3c) Perform feature recognition and extraction on the obtained three-dimensional point cloud data of the calibration tooling to obtain feature coordinate information on the calibration tooling in the global point cloud coordinate system; 3d) Convert the three-dimensional coordinates in the global point cloud coordinate system to the three-dimensional coordinates in the multi-axis robot coordinate system; 3e) Convert the three-dimensional coordinates in the multi-axis robot coordinate system to the three-dimensional coordinates in the multi-axis robot flange coordinate system; 3f) Locate the position of the current calibration tooling according to the position in the multi-axis robot flange coordinate system; 3g) Based on the position of the calibration tooling, plan the transition path for the multi-axis robot to reach the calibration tooling position.

8. The automatic calibration method for the robot and track mechanism system according to claim 6, characterized in that: The information processing computer obtains the point cloud coordinate information collected by the local visual sensor, and uses the feature information to identify and calibrate the position of the tooling for precise positioning, including: 6a) Using filtering method to remove interference points in the point cloud coordinate information of the calibration tooling in the local field of view; 6b) Using the template matching method, segment and extract the three-dimensional point cloud data containing only the calibration tooling from the point cloud coordinate information of the calibration tooling in the local field of view; 6c) Perform feature recognition and extraction on the obtained three-dimensional point cloud data of the calibration tooling to obtain feature coordinate information on the calibration tooling in the local point cloud coordinate system; 6d) Convert the three-dimensional coordinates in the local point cloud coordinate system to the three-dimensional coordinates in the flange coordinate system of the multi-axis robot; 6e) Locate the position of the current calibration tooling according to the position of the feature in the multi-axis robot flange coordinate system.

9. The automatic calibration method of the robot and track mechanism system according to claim 6, characterized in that: Feature solving, the steps are as follows: 7a) The characteristic coordinates in the coordinate system of the track motion mechanism and the characteristic coordinates in the coordinate system of the multi-axis robot collected by the local vision sensor; 7b) The number of feature coordinates collected meets the following solution requirements: Among them, Pc1, Pc2…Pci are the three-dimensional coordinates of the visible calibration features in the multi-axis robot coordinate system; R cb and is the coordinate transformation matrix between the track coordinate system and the multi-axis robot; Pr1, Pr2…Pri are the three-dimensional coordinates of the visible calibration features in the track motion mechanism coordinate system; The coordinate conversion matrix between the track coordinate system and the multi-axis robot is obtained by solving the least square method based on the obtained three-dimensional coordinates of the calibration feature in the multi-axis robot coordinate system and the three-dimensional coordinates of the calibration feature in the track motion mechanism coordinate system.

10. The automatic calibration method of the robot and track mechanism system according to claim 6, characterized in that: The information processing computer obtains the characteristic coordinate quantity solution set by solving the obtained solution, and obtains the rotation relationship Rx and translation relationship Tx of the track motion mechanism relative to the multi-axis robot coordinate system at different coordinate positions; wherein the data conversion relationship between the two coordinate systems is as follows: in, is a point in the multi-axis robot coordinate system, is the coordinate of a point in the multi-axis robot coordinate system in the track coordinate system.

Images