An error mapping method for digital twin systems of manufacturing equipment

Abstract

An error mapping method for a digital twin system of manufacturing equipment is characterized in that it includes the following steps: first, according to the detected equipment TCP point posture, the difference with the theoretical TCP posture is calculated; second, according to the theoretical TCP point posture and the geometric dimensions of the theoretical equipment, the robot kinematics inverse algorithm is used to solve the position of each motion axis; third, the position of the motion axis obtained by the theoretical model is compared with the position of the motion axis actually collected; fourth, the actually collected information is displayed on the digital twin model as required. The present invention can express the working state and posture in the twin system when the physical system contains non-negligible errors. Thereby solving the problem of inconsistency caused by errors between the physical and twin systems. It does not need to adjust the models of products and equipment according to the errors of the physical system, and can be adjusted in real time in the twin system, which greatly improves the workload and simulation effect.

Description

Technical Field

[0001] The present invention relates to a digital manufacturing technology, and in particular to a solution to the problem of inconsistency between a twin system and a physical system caused by errors between actual manufactured products and actual manufacturing equipment and products and equipment in a digital twin model. Specifically, it is an error mapping method for a digital twin system of manufacturing equipment, which can be used for visualization of digital twin systems in production processes of various types of manufacturing equipment, and uses the error mapping method to visualize the manufacturing processes of a physical system and a twin system in real time. Background Art

[0002] Digital twin technology has received widespread attention in recent years, and many companies regard digital twins as one of the technologies that the industry needs to pay the most attention to. The development of Internet technology and the emerging and increasingly booming artificial intelligence and Internet of Things technologies have made more and more people realize the possibility of digital twins, and begin to accept and try to apply them in some fields. Digital twins have gradually become a digital link for the exchange of various data and information, allowing products to communicate and integrate with digital twin models at all stages of their life, and manage the design analysis and production and manufacturing processes of products at all stages.

[0003] At present, most twin systems of manufacturing equipment display the collected data directly on the twin model, but the movement state of the equipment is just a demonstration, and the actual posture state is not paid attention to, and even replaced by an animation. This makes the twin system of the equipment a false twin, which can only roughly reflect the general state of the equipment and cannot reflect the actual production process.

[0004] A common idea is to use the collected data to drive the twin system, and let the digital twin system move in the twin system according to the collected data to solve the twin problem in the equipment manufacturing process. Unfortunately, this idea does not work, or it is difficult to truly reflect the actual situation. There are two reasons: First, the processing object has errors (especially in the assembly field). This error causes the TCP point at the end of the equipment that actually needs to be reached to be inconsistent with the theoretical position in the digital twin system; second, each piece of equipment itself (such as an industrial robot arm) also has errors. This error causes each motion axis to be different from the theoretical model. In this way, the theoretical model of the twin system is driven according to the actual collected motion axis data, and the position reached by the end of the twin system is inconsistent with the actual position of the end of the system.

[0005] Therefore, how to deal with the product and equipment errors between the actual physical system and the digital twin system has become a key problem in solving the problem of ensuring that the digital twin system of manufacturing equipment "truthfully" reflects the actual working conditions. Summary of the invention

[0006] The purpose of the present invention is to invent an error mapping method for a digital twin system of manufacturing equipment to address the problem that the products and equipment of actual physical systems and digital twin systems cannot fully reflect the actual working conditions due to the existence of errors.

[0007] The technical solution of the present invention is:

[0008] An error mapping method for a digital twin system of manufacturing equipment is characterized in that it includes the following steps: first, according to the detected equipment TCP point posture, the difference with the theoretical TCP posture is calculated; second, according to the theoretical TCP point posture and the geometric dimensions of the theoretical equipment, the robot kinematics inverse algorithm is used to solve the position of each motion axis; third, the position of the motion axis calculated by the theoretical model is compared with the position of the motion axis actually collected. If it exceeds the defined error allowable range, feedback information is considered (prompting abnormal status); fourth, the actually collected information, including TCP posture information and equipment status information, is displayed on the digital twin model as required.

[0009] The TCP here refers to the "Tool Center Point", which is generally the origin of the coordinate system of the end effector (tool) fixed at the end of the robot. When we generally use programming or manual form to let the robot approach a certain point in space, its essence is to let the tool center point approach the point. The trajectory movement of the robot is the movement of the tool center point (TCP), so that the TCP point on the end effector reaches the corresponding spatial working position. The TCP point posture described in the present invention refers to the position and posture of the TCP point, wherein the position of TCP refers to the X, Y, and Z coordinates of the TCP point in the Cartesian coordinate system, and the posture of TCP is the direction angle or direction cosine of the end coordinate system established on the TCP point relative to the base coordinate system.

[0010] If the actual equipment does not collect the TCP point pose, the steps are as follows: first, calculate the pose of the actual TCP point of the equipment based on the position of the actually collected motion axis; second, calculate the difference between the actual TCP and theoretical TCP poses; third, according to the theoretical TCP point pose and the geometric dimensions of the theoretical equipment, use the robot kinematics inverse algorithm to solve the position of each motion axis; fourth, compare the position of the motion axis calculated by the theoretical model with the position of the actually collected motion axis. If it exceeds the defined error allowable range, consider feedback information (prompting abnormal status); fifth, display the actually collected information, including TCP pose information and equipment status information, on the digital twin model as needed.

[0011] The above steps are based on the calibrated geometry of the actual equipment. If the calibrated geometry is unknown, the TCP point calculated based on the theoretical geometry will be different from the actual TCP pose. In this case, a geometry calibration step needs to be added to calculate using the calibrated geometry. The steps are as follows:

[0012] First, according to the detected equipment TCP point pose P, the theoretical TCP pose P is calculated. v The difference ΔP;

[0013] Second, according to the theoretical TCP point pose P v The reverse bias difference ΔP is (P v -ΔP) and the geometric dimensions L of the theoretical equipment v , the robot kinematics inverse algorithm is used to solve the position X of each motion axis v ′;

[0014] Third, compare the position X of the motion axis obtained by the theoretical model v If the actual acquired position X of the motion axis exceeds the defined error tolerance range ΔX, feedback information is considered (prompting abnormal status);

[0015] Fourth, the actually collected information, including TCP posture information P and device status information X, is displayed on the digital twin model as required; that is, although the TCP in the twin system is at position P v -ΔP, the state of each motion axis of the equipment is at X v ′, but the displayed data is based on the actual collected physical data TCP posture P and equipment status X.

[0016] The beneficial effects of the present invention are:

[0017] Aiming at the product and equipment error problems of actual physical systems and digital twin systems, the present invention invents an error mapping method for manufacturing equipment digital twin systems, which maps the error of the product to be processed (reflected in the actual TCP point position) to the theoretical position of the digital twin system, and uses the kinematic algorithm of the manufacturing equipment twin system to establish the mapping of the position of the drive shaft, thereby mapping the posture relationship of the actual manufacturing equipment in the digital twin system. This method has the following significant advantages:

[0018] First, this method provides a posture mapping method between the physical and twin systems of the working state of manufacturing equipment, which can "accurately" express the working state and posture in the twin system when the physical system contains non-negligible errors, thereby solving the problem of inconsistency caused by errors between the physical and twin systems.

[0019] Secondly, this method does not need to adjust the product and equipment models according to the errors of the physical system, but can be adjusted in real time in the twin system, which greatly improves the workload and simulation effect. If the motion axis data collected by the physical system is directly used to drive the equipment, it will often appear that the original equipment is a smooth straight line or arc movement, but it becomes a zigzag movement in the twin system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 Schematic diagram of the error mapping method for the digital twin system of manufacturing equipment. DETAILED DESCRIPTION

[0021] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0022] like Figure 1 shown.

[0023] For equipment with end TCP posture detection, or equipment with equivalent TCP posture detection, such as manufacturing equipment with laser tracker measurement, or manufacturing equipment with visual inspection, the following steps are included:

[0024] First, according to the detected equipment TCP point pose P, the theoretical TCP pose P is calculated. v The difference ΔP;

[0025]

[0026]

[0027] Second, according to the theoretical TCP point pose P v and the geometric dimensions L of the theoretical equipment v , use the robot kinematics inverse algorithm similar to formula (5) to solve the position X of each motion axis v ;

[0028] Where m is the number of equipment geometry parameters (3)

[0029] Where n is the number of axes of motion of the equipment (4)

[0030] X v =G(L v ,P v ) (5)

[0031] Third, compare the position X of the motion axis obtained by the theoretical model v If the actual acquired position X of the motion axis exceeds the defined error tolerance range ΔX, feedback information is considered (prompting abnormal status);

[0032] Where n is the number of axes of motion of the equipment (6)

[0033] Where n is the number of axes of motion of the equipment (6)

[0034] Fourth, the actually collected information, including TCP position information P and device status information X, is displayed on the digital twin model as required. v At X, the states of the various motion axes of the equipment are v However, the displayed data is based on the actual collected physical data TCP posture P and device status X.

[0035] For equipment without terminal TCP pose detection (or equivalent TCP pose detection), the following steps are included:

[0036] First, according to the actual acquired position X of the motion axis and the actual geometric size L of the equipment, a robot kinematics forward solution algorithm similar to formula (8) is used to calculate the position P of the actual TCP point of the equipment;

[0037] Where m is the number of equipment geometry parameters (7)

[0038] P=F(L,X) (8)

[0039] Second, by calculating the actual equipment TCP point pose P, calculate the difference with the theoretical TCP pose P v The difference ΔP;

[0040] Third, according to the theoretical TCP point pose P v and the geometric dimensions L of the theoretical equipment v , use the robot kinematics inverse algorithm similar to formula (5) to solve the position X of each motion axis v .

[0041] Fourth, compare the position X of the motion axis obtained by the theoretical model v If the actual acquired position X of the motion axis exceeds the defined error tolerance range ΔX, feedback information is considered (prompting abnormal status)

[0042] Fifth, the actually collected information, including TCP posture information P and device status information X, is displayed on the digital twin model as required.

[0043] In specific implementation, it may happen that after the device reaches the detection position, it is detected that the target point is offset, that is, the position relationship between the target TCP point and the device is detected after the actual deviation. If it is necessary to express this deviation in the twin environment, the above implementation steps are as follows:

[0044] First, according to the detected equipment TCP point pose P, the theoretical TCP pose P is calculated. v The difference ΔP;

[0045]

[0046] Second, according to the theoretical TCP point pose P v Reverse bias difference ΔP (i.e. P v -ΔP) and the geometric dimensions L of the theoretical equipment v , use the robot kinematics inverse algorithm similar to formula (10) to solve the position X of each motion axis v ′;

[0047] X v ′=G(L v , P v -ΔP) (10)

[0048] Third, compare the position X of the motion axis obtained by the theoretical model v If the actual acquired position X of the motion axis exceeds the defined error tolerance range ΔX, feedback information is considered (prompting abnormal status);

[0049] Fourth, the actually collected information, including TCP position information P and device status information X, is displayed on the digital twin model as required. v -ΔP, the states of the various motion axes of the equipment are at X v ′, but the displayed data is based on the actual collected physical data TCP posture P and equipment status X.

[0050] In particular, in the specific implementation, for the industrial robot arm system commonly used in industry, the position inverse solution algorithm X v =G(L v , P v ) can be solved by common geometric methods, and the position positive solution algorithm P = F (L, X) can be solved by the common DH model. Its theoretical geometric parameter L v And the actual geometric parameters L are the corresponding geometric parameters in the DH model.

[0051] The present invention does not involve parts that are the same as the prior art or can be implemented by using the prior art.

Claims

1. An error mapping method for a digital twin system of manufacturing equipment, characterized by: It includes the following steps: first, according to the detected equipment TCP point posture, the difference with the theoretical TCP posture is calculated; second, according to the theoretical TCP point posture and the geometric dimensions of the theoretical equipment, the robot kinematics inverse algorithm is used to solve the position of each motion axis; third, the position of the motion axis calculated by the theoretical model and the actual collected position of the motion axis are compared. If it exceeds the defined error allowable range, feedback information is considered to prompt an abnormal status; fourth, the actually collected information, including TCP posture information and equipment status information, is displayed on the digital twin model as required.

2. According to the method of claim 1, if the actual equipment does not collect the TCP point pose, the steps are as follows: first, the pose of the actual TCP point of the equipment is calculated according to the position of the motion axis actually collected; Second, calculate the difference between the actual TCP and theoretical TCP poses; third, use the robot kinematics inverse algorithm to solve the position of each motion axis based on the theoretical TCP point pose and the geometric dimensions of the theoretical equipment; fourth, compare the position of the motion axis obtained by the theoretical model with the position of the actual acquired motion axis. If it exceeds the defined error tolerance range, consider feedback information to indicate an abnormal state; Fifth, the actual collected information, including TCP posture information and device status information, is displayed on the digital twin model as needed.

3. The method according to claim 1, characterized in that: If the deviation of the actual environment terminal TCP needs to be expressed in the twin environment, the steps are as follows: First, based on the detected equipment TCP point pose P, calculate the deviation from the theoretical TCP pose P v The difference ΔP; second, according to the theoretical TCP point pose P v The reverse bias difference ΔP is (P v -ΔP) and the geometric dimensions L of the theoretical equipment v , the robot kinematics inverse algorithm is used to solve the position X of each motion axis v ′ ; Third, compare the position X of the motion axis obtained by the theoretical model v If the actual acquired position X of the motion axis exceeds the defined error tolerance range ΔX, the feedback information is considered to indicate an abnormal state; Fourth, the actually collected information, including TCP posture information P and device status information X, is displayed on the digital twin model as required; that is, although the TCP in the twin system is at position P v -ΔP, the state of each motion axis of the equipment is at X v ′ However, the displayed data is displayed according to the actual collected physical data TCP posture P and device status X.

Images