Multi-degree-of-freedom forming equipment pose perception data fusion method considering error sensitivity

Abstract

The present application relates to a kind of multi-degree-of-freedom forming equipment pose perception data fusion methods considering error sensitivity, multi-degree-of-freedom forming equipment is coupled motion by multiple kinematic chains to realize multi-degree-of-freedom motion, grating ruler is installed on the active branch of multi-degree-of-freedom forming equipment, comprising the following steps: S1, based on grating ruler calculation multi-degree-of-freedom forming equipment pose;S2, obtain the error sensitivity based on grating ruler calculation pose;S3, based on monocular vision calculation multi-degree-of-freedom forming equipment pose;S4, obtain the error sensitivity based on monocular vision calculation pose;S5, considering the error sensitivity of two pose calculation methods of step S3 and S4, pose data fusion is carried out.The present application can fuse the pose data collected by multiple sensors, improve the perception accuracy of equipment under different poses, avoid the situation that the perception pose error is larger under some special poses, to meet the requirements of multi-degree-of-freedom forming equipment in heavy load working condition to realize complex motion.

Description

Technical Field

[0001] This invention relates to the field of multi-degree-of-freedom forming, and more specifically, to a method for fusion of pose perception data of multi-degree-of-freedom forming equipment that takes into account error sensitivity. Background Technology

[0002] Multi-degree-of-freedom forming is an advanced local incremental metal forming technology. The upper die oscillates at a certain angle around both the machine tool's axis and its own main axis, while the lower die feeds upwards at a certain speed. The blank is forced to undergo local near-net-shape plastic deformation until the target shape of the complex part is achieved. Compared with traditional single-degree-of-freedom integral forming processes, it has technical advantages such as lower forming force, higher part dimensional accuracy, lower vibration and noise, and better flexibility.

[0003] Heavy-duty multi-degree-of-freedom forming machine tools achieve multi-degree-of-freedom motion through the coupling of multiple motion chains. The pose of the multi-degree-of-freedom forming equipment is typically calculated based on information from grating rulers installed on the active chains. This method is effective for parallel machine tools under light load conditions. However, under heavy load conditions, parallel machine tools may experience excessive deformation in certain specific poses, leading to inaccurate pose calculations. Furthermore, a single data source is susceptible to environmental influences, resulting in significant deviations in the perceived data. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method for fusing pose perception data of multi-degree-of-freedom forming equipment that takes into account error sensitivity. This method can fuse pose data collected by multiple sensors to improve the perception accuracy of the equipment under different poses and avoid large pose perception errors under certain special poses, so as to meet the requirements of multi-degree-of-freedom forming equipment to achieve complex motion under heavy load conditions.

[0005] The technical solution adopted by this invention to solve its technical problem is as follows: A method for fusion of pose perception data for a multi-degree-of-freedom forming equipment considering error sensitivity is constructed. The multi-degree-of-freedom forming equipment achieves multi-degree-of-freedom motion through the coupling motion of multiple motion chains. A grating ruler is installed on the active chain of the multi-degree-of-freedom forming equipment. The data fusion method includes the following steps: S1. Calculate the pose of a multi-degree-of-freedom forming equipment based on a grating ruler; S2. Obtain the error sensitivity of pose calculation based on grating ruler; S3. Calculate the pose of multi-degree-of-freedom forming equipment based on monocular vision; S4. Obtain the error sensitivity of pose calculation based on monocular vision; S5. Considering the error sensitivity of the two pose calculation methods in steps S3 and S4, perform pose data fusion.

[0006] According to the above scheme, in step S1, the method for calculating the pose of a multi-degree-of-freedom forming equipment based on a grating ruler includes establishing two coordinate systems: a moving coordinate system and a fixed coordinate system. A fixed coordinate system is constructed on the surface of the static platform of the multi-degree-of-freedom forming equipment. ( ), its origin Located at the geometric center of the static platform, The axis is perpendicular to the static platform, i.e., perpendicular to the sliding direction of the slider; a coordinate system for the moving platform is constructed on the moving platform of the parallel equipment in space. ( ), its origin Located at the geometric center of the circumcircle of the moving platform. The axis is perpendicular to the surface of the moving platform; when the moving platform is in its initial position, i.e., the equilibrium position, the moving coordinate system and the fixed coordinate system... The directions are consistent, and the coordinate system is moving. The axis passes through the origin of the static coordinate system. The six ball joints of the static platform are used To represent, the ball joint of the moving platform is used express. ( =1-6) represents the distribution angle of the ball joint points on the static platform. ( =1-6) represents the distribution angle of the ball joint points on the moving platform.

[0007] According to the above scheme, in step S1, the method for calculating the pose of a multi-degree-of-freedom forming equipment based on a grating ruler further includes: The position of the slider is known to be The installation angle of the static platform hinge point is The radius of the moving platform is The hinge point installation angle of the moving platform The ball joint point on the static platform slider In the fixed system The coordinate vector in the figure is: (1) In the formula, This is the distance from the center of the slider on the static platform to the center of the static platform. Furthermore, each hinge point of the moving platform In motion The coordinate vector in the equation is determined by equation (2): (2) In the formula, The distance from the center of the moving platform slider to the center of the moving platform; , , This indicates the translational motion of the moving platform in three directions. , , This represents the rotation angle of the moving platform in three directions; in the multi-degree-of-freedom forming process, the motion equation of the upper mold is: (3) In the formula, It is the swing angle of the upper mold movement. It refers to the rotational speed during the movement of the upper mold. It is the feed rate during the upper die movement. It is the feed distance during the movement of the upper die; thereby Compared to rotation matrix Determined by equation (4): (4) and then, In the global coordinate system position vector Determined by equation (5): (5) In the formula, yes The origin of the coordinate system relative to Position vector; Determined by equation (6): (6) Based on the closed-loop vector relationship, we can obtain: (7) Move the platform position Treating it as the independent variable, we get: (8) In the formula, To obtain a constant value, the slider position is measured using a grating ruler. Substitution , It is also a constant value, and the unknown quantity is the one in the formula. and The pose of the moving platform, represented by the above nonlinear equations, can be obtained by solving the pose calculated based on the grating ruler.

[0008] Transform (8) into , can be obtained at any time position The point Taylor formula is: (9) In the formula, Nonlinear equation system The Jacobian matrix is ​​denoted as:

[0009] Solve The iterative formula for Newton's method is: (10).

[0010] According to the above scheme, step S2 includes the following steps: Transform equation (8) into the form of equation (11): (11) The derivative of pose with respect to input represents the error sensitivity based on the pose calculation method using a grating ruler. The calculation formula is as follows: (12) In the formula,

[0011] At a certain moment The maximum value of the input is chosen to represent the error sensitivity of this pose, expressed as: (13).

[0012] According to the above scheme, step S3 includes the following steps: Set the two-dimensional coordinates of the cooperative target in the camera as... Finally, the position vector of the cooperative target is set as ; Based on the existing measurement points, four virtual points are calculated and determined by equation (14): (14) In the formula, For matrix Eigenvalues , For matrix eigenvectors; The coordinates of the cooperative target in the end coordinate system are represented by virtual control points as shown in equation (15): (15) The coefficients can be obtained from equation (15). As in equation (16): (16) Virtual control points represent the coordinates of the cooperative target in the camera coordinate system. It can be expressed by equation (17): (17) In the formula, Indicates the coordinates of the control points in the camera coordinate system; Based on the pinhole camera model, we can conclude that: (18) In the formula, Focal length For the coordinates of the optical center, ; Expanding equation (18), we get: (19) The matrix form is represented by equation (20): (20) Solution of equation (20) exist From the right null space, we can obtain: (twenty one) In the formula, This represents the matrix. Eigenvectors with zero eigenvalues; for pinhole imaging models, N is usually taken as 1, which yields equation (22): (twenty two) Since the distance between virtual control points does not change with the reference coordinate system, we can conclude that: (twenty three) It can be expressed by equation (24): (twenty four) The coordinates of the four control points in the camera coordinate system can be obtained ; ICP is used to solve the pose relationship between the moving platform coordinate system and the camera coordinate system. The pose of the camera coordinate system relative to the static platform coordinate system is fixed and represented as follows: The pose of the moving coordinate system relative to the static coordinate system can be determined by equation (25): (25) Each pose component can be determined by equation (26): (26) In the formula, .

[0013] According to the above scheme, step S4 includes the following steps: posture For input The derivative is expressed as: (27) The maximum value of all two-dimensional coordinates is selected to represent the pose error sensitivity at this moment. Its value represents the reliability of the pose data based on monocular vision measurement, and is denoted as the maximum error sensitivity. The calculation formula is as follows: (28).

[0014] According to the above scheme, step S5 includes the following steps: Error sensitivity based on grating ruler measurement using vector This indicates that the error sensitivity based on monocular vision measurement is expressed using vectors. The covariance matrix is ​​expressed as follows: and ; exist At any given time, the equipment pose is set to... Let its first and second derivatives with respect to time be set as state variables, i.e. Each state variable is independent, and the state equations for the two measurement methods are as follows: (29) In the formula, Let the state noise vector be Gaussian noise with zero mean and covariance. It is the state transition matrix; Assuming the sampling time is Then at time and time The relationship between attitude and attitude can be expressed using Taylor expansion: (30) Therefore, the state transition matrix Equation (31) represents: (31) To achieve the fusion of the two types of data, two sub-filters are needed, and the measurement equation for each sub-filter is as follows: (32) In the formula, The pose values ​​of sub-filters 1 and 2 are the measured values ​​obtained from the position of the grating ruler and monocular vision. For the measurement matrix, The error sensitivity of different measurement methods represents the reliability of the measurement data; The two sub-filters receive pose data and measurement noise from the grating ruler-based measurement subsystem and the monocular vision-based measurement subsystem, respectively, to obtain the state vector. and state covariance matrix The main filter first performs a time update to obtain the state vector. and state covariance matrix Then, the main filter merges all the obtained state vectors and state covariance matrices to obtain the globally optimal estimated state vector. and state covariance matrix .

[0015] According to the above scheme, the main filter merges all the obtained state vectors and state covariance matrices to obtain the globally optimal estimated state vector. and state covariance matrix The methods include: (1) The initial values ​​of the three filters are obtained from the optimal estimate of the previous time step: (33) In the formula, For allocation coefficients, ; (2) Three filter state vectors and state covariance matrix : (34) In the formula, , Indicates sub-filters 1 and 2; (3) Optimal estimation of equipment pose: (35)

[0016] The multi-degree-of-freedom forming equipment pose perception data fusion method considering error sensitivity of the present invention has the following beneficial effects: 1. This invention proposes a method for fusing pose perception data of multi-degree-of-freedom forming equipment that considers error sensitivity. First, the equipment pose is calculated based on grating ruler data from the active branch. Next, an error sensitivity model for pose calculation based on the grating ruler is established. Then, the equipment pose is calculated based on monocular vision, and an error sensitivity model for pose calculation based on monocular vision is also established. Finally, the error sensitivities of the two pose calculation methods are considered simultaneously, and the calculation results are fused to obtain pose data with higher accuracy and stronger robustness. This method can be used to calculate the pose data of equipment, thereby guiding the development and control of the equipment.

[0017] 2. The method proposed in this invention can fuse two different pose calculation results, avoiding excessive pose measurement error values ​​caused by the influence of the environment on a single data source. Considering the error sensitivity of the measurement method, an improved federated Kalman filtering method is proposed, thereby realizing real-time monitoring of equipment under different working conditions and laying the foundation for subsequent real-time control of equipment. Attached Figure Description

[0018] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings: Figure 1 Schematic diagram of coordinate system and ball joint point distribution of multi-degree-of-freedom forming equipment; Figure 2 Vector diagram showing the positions of the branch chains in a multi-degree-of-freedom forming equipment; Figure 3 Schematic diagram of a monocular vision measurement system; Figure 4 Schematic diagram of image processing; Figure 5 Block diagram of the Kalman filtering method; Figure 6 The sensitivity of the pose of a multi-degree-of-freedom forming equipment to the error of each input slider; Figure 7 The maximum error sensitivity of the pose of a multi-degree-of-freedom forming equipment to the input slider; Figure 8 The sensitivity of the pose of a multi-degree-of-freedom forming equipment to the coordinates of each cooperative target; Figure 9 The maximum error sensitivity of the pose of a multi-degree-of-freedom forming equipment to the coordinates of the cooperative target; Figure 10 This is a schematic diagram of the experimental platform; Figure 11 This is a comparison chart of the pose error of the moving platform under different sensors and different fusion algorithms. Detailed Implementation

[0019] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0020] The method for fusion of pose perception data of multi-degree-of-freedom forming equipment considering error sensitivity of the present invention includes the following steps: S1. Calculate the pose of a multi-degree-of-freedom forming equipment based on a grating ruler; like Figure 1 As shown, two coordinate systems are established: a moving coordinate system and a fixed coordinate system. A fixed coordinate system is constructed on the surface of the static platform of the multi-degree-of-freedom forming equipment. ( ), its origin Located at the geometric center of the static platform, The axis is perpendicular to the stationary platform, i.e., perpendicular to the sliding direction of the slider. A coordinate system for the moving platform is constructed on the moving platform of the spatially parallel equipment. ( ), its origin Located at the geometric center of the circumcircle of the moving platform. The axis is perpendicular to the surface of the moving platform. When the moving platform is in its initial position, i.e., the equilibrium position, the moving coordinate system and the fixed coordinate system... The directions are consistent, and the coordinate system is moving. The axis passes through the origin of the static coordinate system. The six ball joints of the static platform are used To represent, the ball joint of the moving platform is used express. ( =1-6) represents the distribution angle of the ball joint points on the static platform. ( =1-6) represents the distribution angle of the ball joint points on the moving platform.

[0021] The position of the slider is known to be The installation angle of the static platform hinge point is The radius of the moving platform is The hinge point installation angle of the moving platform The ball joint point on the static platform slider In the fixed system The coordinate vector in the figure is: (1) In the formula, This is the distance from the center of the slider on the static platform to the center of the static platform.

[0022] Furthermore, each hinge point of the moving platform In motion The coordinate vector in the equation is determined by equation (2): (2) In the formula, This is the distance from the center of the slider of the moving platform to the center of the moving platform.

[0023] The motion of the equipment's moving platform can be represented by six parameters. , , This indicates the translational motion of the moving platform in three directions. , , This represents the rotation angle of the moving platform in three directions. In the multi-degree-of-freedom forming process, the motion equation of the upper mold is: (3) In the formula, It is the swing angle of the upper mold movement. It refers to the rotational speed during the movement of the upper mold. It is the feed rate during the upper die movement. It is the feed distance during the movement of the upper die.

[0024] thereby Compared to rotation matrix Determined by equation (4): (4) and then, In the global coordinate system position vector Determined by equation (5): (5) In the formula, yes The origin of the coordinate system relative to The position vector. It can be determined by equation (6): (6) No. Vector diagram of the branch locations as shown Figure 2 As shown, based on the closed-loop vector relationship, we can obtain: (7) Move the platform position Treating it as the independent variable, we can obtain: (8) In the formula, To obtain a constant value, the slider position is measured using a grating ruler. Substitution , It is also a constant value, and the unknown quantity is the one in the formula. and The pose of the moving platform, represented by the above nonlinear equations, can be obtained by solving the pose calculated based on the grating ruler.

[0025] Transform (8) into , can be obtained at any time position The point Taylor formula is: (9) In the formula, Nonlinear equation system The Jacobian matrix is ​​denoted as:

[0026] The solution can be obtained using Newton's iteration method. The iterative formula for Newton's method is: (10) For the parallel mechanism of the present invention, the initial value can be determined based on the working space of the moving platform. The range of the initial value from the reasonable solution is small, and Newton's method can quickly converge to the reasonable solution.

[0027] S2. Error sensitivity of pose calculation based on grating ruler; The detection error of the grating ruler on the slider will have different effects on the pose of the multi-degree-of-freedom envelope forming equipment at different positions of the slider. To study its differential motion, it is first necessary to convert equation (8) into the form of equation (11): (11) The derivative of pose with respect to input can represent the error sensitivity under the pose calculation method based on grating ruler measurement. The calculation formula is as follows: (12) In the formula,

[0028] At a certain moment Error sensitivity represents the reliability of the measurement data and is an expected value. Therefore, we choose the maximum value of the six inputs to represent the error sensitivity of this pose, expressed as: (13).

[0029] S3. Calculate the pose of multi-degree-of-freedom forming equipment based on monocular vision; This invention employs a monocular vision measurement system based on cooperative targets. The system's hardware mainly consists of an industrial camera and cooperative target markers. In the measurement system, such as... Figure 5 As shown, the cooperative target is attached to the end of a multi-degree-of-freedom forming machine and moves with the platform. An industrial camera is fixed to the machine tool to acquire images of the cooperative target. Based on the captured images, the pose of the multi-degree-of-freedom forming machine is solved using the EPNP method.

[0030] like Figure 6 As shown, the image is processed. To eliminate noise and highlight the features of the cooperative target, the image is first denoised. In this work, the camera has a large field of view and there are obvious local gray-level variations. This may cause some features to be located in areas with low gray-level values. Direct processing would result in the loss of information in these areas. To address this issue, feature enhancement is performed to increase the intensity of the feature signals and improve the signal-to-noise ratio. After image preprocessing, adaptive thresholding is performed, followed by edge detection using the Canny operator. The detected edges are tracked, and edge point sets are extracted. Ellipse fitting is applied to the edge point sets, and the ellipse centers are extracted, representing the two-dimensional image coordinates of the cooperative target. The EPnP algorithm is used for end-effector pose estimation relative to the camera. This method only requires optimization of four control points, is computationally fast, and has high accuracy. Figure 5 As shown, the two-dimensional coordinates of the cooperative target in the camera are set as follows: Finally, the position vector of the cooperative target is set as .

[0031] First, based on the existing measurement points, four virtual points are calculated and determined by equation (14): (14) In the formula, For matrix Eigenvalues , For matrix eigenvectors.

[0032] The coordinates of the cooperative target in the end coordinate system are represented by virtual control points as shown in equation (15): (15) The coefficients can be obtained from equation (15). As in equation (16): (16) Virtual control points represent the coordinates of the cooperative target in the camera coordinate system. It can be expressed by equation (17): (17) In the formula, This represents the coordinates of the control point in the camera coordinate system.

[0033] Based on the pinhole camera model, we can conclude that: (18) In the formula, Focal length For the coordinates of the optical center, .

[0034] Expanding equation (18), we get: (19) Represented in matrix form by equation (20): (20) Solution of equation (20) exist From the right null space, we can obtain: (twenty one) In the formula, This represents the matrix. Eigenvectors with zero eigenvalues. For the pinhole imaging model, N is usually taken as 1, which yields equation (22): (twenty two) Since the distance between virtual control points does not change with the reference coordinate system, we can conclude that: (twenty three) It can be expressed by equation (24): (twenty four) The coordinates of the four control points in the camera coordinate system can be obtained .

[0035] The above steps transform the PnP problem into a 3D-3D pose estimation problem. Then, ICP (Iterative Nearest Point) is used to solve for the pose relationship between the moving platform coordinate system and the camera coordinate system. The pose of the camera coordinate system relative to the static platform coordinate system is fixed and represented as follows: The pose of the moving coordinate system relative to the static coordinate system can be determined by equation (25): (25) Each pose component can be determined by equation (26): (26) In the formula, .

[0036] S4. Error sensitivity of pose calculation based on monocular vision; A visual estimation algorithm for the pose of a moving platform takes the two-dimensional coordinates of the cooperative target as input. These two-dimensional coordinates may contain errors compared to the true coordinates. Under different equipment poses, the same coordinate error will lead to different end pose errors after being processed by the pose estimation algorithm.

[0037] Therefore, pose For input The derivative can be expressed as: (27) Analogous to the measurement method based on grating rulers, we select the maximum value of all two-dimensional coordinates to represent the pose error sensitivity at this moment. This value represents the reliability of the pose data measured based on monocular vision, and is denoted as the maximum error sensitivity. The calculation formula is as follows: (28).

[0038] S5. Pose data fusion based on error sensitivity; The preceding text discussed two methods for measuring pose: the grating ruler-based method and the monocular vision-based method. To obtain highly reliable and accurate results, it is necessary to fuse the two sets of data. Kalman filtering is a good method for fusing data from different sources. However, Kalman filtering is based on the assumption that different data sources have the same error sensitivity. But for multi-degree-of-freedom forming equipment, the error sensitivity varies with the pose of the moving platform, regardless of the measurement method used. Therefore, this invention proposes an enhanced Kalman filtering method that considers the measurement error sensitivity under different poses.

[0039] Error sensitivity based on grating ruler measurement using vector This indicates that the error sensitivity based on monocular vision measurement is expressed using vectors. It is indicated that its covariance matrix can be expressed as and .

[0040] exist At any given time, the equipment pose is set to... Let its first and second derivatives with respect to time be set as state variables, i.e. Each state variable is independent, and the state equations for the two measurement methods are as follows: (29) In the formula, Let be the state noise vector, assumed to be Gaussian noise with zero mean and covariance. It is the state transition matrix.

[0041] Assuming the sampling time is Then at time and time The relationship between attitude and attitude can be expressed using Taylor expansion: (30) Therefore, the state transition matrix It can be expressed by equation (31): (31) To achieve the fusion of the two types of data, two sub-filters are needed, and the measurement equation for each sub-filter is as follows: (32) In the formula, The pose values ​​of sub-filters 1 and 2 are the measured values ​​obtained from the position of the grating ruler and monocular vision. For the measurement matrix, The error sensitivity of different measurement methods represents the reliability of the measurement data.

[0042] Figure 9This describes the overall flow of the proposed enhanced Kalman filter method. Sub-filters 1 and 2 receive pose data and measurement noise from the grating ruler-based measurement subsystem and the monocular vision-based measurement subsystem, respectively, to obtain state vectors. and state covariance matrix The main filter first performs a time update to obtain the state vector. and state covariance matrix Then, the main filter merges all the obtained state vectors and state covariance matrices to obtain the globally optimal estimated state vector. and state covariance matrix .

[0043] The specific implementation steps are as follows: (1) The initial values ​​of the three filters are obtained from the optimal estimate of the previous time step: (33) In the formula, For allocation coefficients, .

[0044] (2) Three filter state vectors and state covariance matrix : (34) In the formula, , These represent sub-filters 1 and 2; the main filter only has the first two steps.

[0045] (3) Optimal estimation of equipment pose: (35) Thus, we have obtained the optimal solution based on the fusion of the two pose data.

[0046] Example Based on equations (12) and (13) and the equipment configuration parameters given in Table 1, the error sensitivity of each slider and the maximum error sensitivity of the pose calculation based on the grating ruler can be calculated, such as Figure 6 , Figure 7 As shown. Based on equations (27) and (28), the error sensitivity of each cooperative target and the maximum error sensitivity for pose calculation based on monocular vision can be calculated, such as... Figure 8 , Figure 9 As shown.

[0047] Table 1 Equipment Configuration Parameters

[0048] Based on equations (10), (26), and (35) and Figure 10The experimental platform shown yielded the pose errors of the moving platform under different sensors and fusion algorithms within a complete motion cycle, as follows: Figure 11 As shown. The X-axis represents the motion time of the equipment in one cycle, and the Y-axis represents the error scale. Black represents the error measured by the grating ruler, blue represents the error measured by monocular vision, green represents the error of the traditional federated Kalman filter method, and red represents the error of the improved federated Kalman filter method considering error sensitivity. Overall, compared with the traditional federated Kalman filter method, the average root mean square error of the equipment's position in the x, y, and z directions decreased from 0.569 mm to 0.293 mm, a reduction of 48%, and the average maximum error decreased from 2.09 mm to 0.79 mm, a reduction of 62%. Regarding the angular error of the equipment, The average root mean square error decreased from 0.251° to 0.144°, a reduction of 42%, and the average maximum error decreased from 0.7° to 0.42°, a reduction of 40%. In all directions, the average error of the improved federated Kalman filter method is closer to zero. Therefore, the improved federated Kalman filter method proposed in this invention has higher measurement accuracy and better stability.

[0049] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A method for fusion of pose perception data for a multi-degree-of-freedom forming equipment considering error sensitivity, wherein the multi-degree-of-freedom forming equipment achieves multi-degree-of-freedom motion through the coupling motion of multiple kinematic chains, and an optical grating ruler is installed on the active chain of the multi-degree-of-freedom forming equipment, characterized in that, Includes the following steps: S1. Calculate the pose of a multi-degree-of-freedom forming equipment based on a grating ruler; S2. Obtain the error sensitivity of pose calculation based on grating ruler; S3. Calculate the pose of multi-degree-of-freedom forming equipment based on monocular vision; S4. Obtain the error sensitivity of pose calculation based on monocular vision; S5. Considering the error sensitivity of the two pose calculation methods in steps S3 and S4, perform pose data fusion. In step S1, the method for calculating the pose of a multi-degree-of-freedom forming equipment based on a grating ruler includes establishing two coordinate systems: a moving coordinate system and a fixed coordinate system. A fixed coordinate system is constructed on the surface of the static platform of the multi-degree-of-freedom forming equipment. ( ), its origin Located at the geometric center of the static platform, The axis is perpendicular to the static platform, i.e., perpendicular to the sliding direction of the slider; a coordinate system for the moving platform is constructed on the moving platform of the parallel equipment in space. ( ), its origin Located at the geometric center of the circumcircle of the moving platform. The axis is perpendicular to the surface of the moving platform; when the moving platform is in its initial position, i.e., the equilibrium position, the moving coordinate system and the fixed coordinate system... The directions are consistent, and the coordinate system is moving. The axis passes through the origin of the static coordinate system. The six ball joints of the static platform are used To represent, the ball joint of the moving platform is used express; ( =1-6) represents the distribution angle of the ball joint points on the static platform. ( =1-6) represents the distribution angle of the ball joint points on the moving platform; In step S1, the method for calculating the pose of a multi-degree-of-freedom forming equipment based on a grating ruler further includes: The position of the slider is known to be The installation angle of the static platform hinge point is The radius of the moving platform is The hinge point installation angle of the moving platform The ball joint point on the static platform slider In the fixed system The coordinate vector in the figure is: (1) In the formula, This is the distance from the center of the slider on the static platform to the center of the static platform. Furthermore, each hinge point of the moving platform In motion The coordinate vector in the equation is determined by equation (2): (2) In the formula, The distance from the center of the moving platform slider to the center of the moving platform; , , This indicates the translational motion of the moving platform in three directions. , , This represents the rotation angle of the moving platform in three directions; in the multi-degree-of-freedom forming process, the motion equation of the upper mold is: (3) In the formula, It is the swing angle of the upper mold movement. It refers to the rotational speed during the movement of the upper mold. It is the feed rate during the upper die movement. It is the feed distance during the movement of the upper die; thereby Compared to rotation matrix Determined by equation (4): (4) and then, In the global coordinate system position vector Determined by equation (5): (5) In the formula, yes The origin of the coordinate system relative to Position vector; Determined by equation (6): (6) Based on the closed-loop vector relationship, we can obtain: (7) Position of the moving platform Treating it as the independent variable, we get: (8) In the formula, To obtain a constant value, the slider position is measured using a grating ruler. Substitution , It is also a constant value, and the unknown quantity is the one in the formula. and The pose of the moving platform, represented by the nonlinear equations, can be obtained by solving the nonlinear equations. Transform equation (8) into , can be obtained at any time position The point Taylor formula is: (9) In the formula, Nonlinear equation system The Jacobian matrix is ​​denoted as: Solve The iterative formula for Newton's method is: (10)。 2. The method for fusing pose perception data of multi-degree-of-freedom forming equipment considering error sensitivity according to claim 1, characterized in that, Step S2 includes the following steps: Transform equation (8) into the form of equation (11): (11) The derivative of pose with respect to input represents the error sensitivity based on the pose calculation method using a grating ruler. The calculation formula is as follows: (12) In the formula, At a certain moment The maximum value of the input is chosen to represent the error sensitivity of this pose, expressed as: (13)。 3. The method for fusing pose perception data of multi-degree-of-freedom forming equipment considering error sensitivity according to claim 2, characterized in that, Step S3 includes the following steps: Set the two-dimensional coordinates of the cooperative target in the camera as Finally, the position vector of the cooperative target is set as ; Based on the existing measurement points, four virtual points are calculated and determined by equation (14): (14) In the formula, For matrix Eigenvalues , For matrix eigenvectors; The coordinates of the cooperative target in the end coordinate system are represented by virtual control points as shown in equation (15): (15) The coefficients can be obtained from equation (15). As in equation (16): (16) Virtual control points represent the coordinates of the cooperative target in the camera coordinate system. It can be expressed by equation (17): (17) In the formula, Indicates the coordinates of the control points in the camera coordinate system; Based on the pinhole camera model, we can conclude that: (18) In the formula, Focal length For the coordinates of the optical center, ; Expanding equation (18), we get: (19) The matrix form is represented by equation (20): (20) Solution of equation (20) exist From the right null space, we can obtain: (21) In the formula, This represents the matrix. Eigenvectors with zero eigenvalues; for pinhole imaging models, N is usually taken as 1, which yields equation (22): (22) Since the distance between virtual control points does not change with the reference coordinate system, we can conclude that: (23) It can be expressed by equation (24): (24) The coordinates of the four control points in the camera coordinate system can be obtained ; ICP is used to solve the pose relationship between the moving platform coordinate system and the camera coordinate system. The pose of the camera coordinate system relative to the static platform coordinate system is fixed and represented as follows: The pose of the moving coordinate system relative to the static coordinate system can be determined by equation (25): (25) Each pose component can be determined by equation (26): (26) In the formula, .

4. The method for fusing pose perception data of multi-degree-of-freedom forming equipment considering error sensitivity according to claim 3, characterized in that, Step S4 includes the following steps: posture For input The derivative is expressed as: (27) The maximum value of all two-dimensional coordinates is selected to represent the pose error sensitivity at this moment. Its value represents the reliability of the pose data based on monocular vision measurement, and is denoted as the maximum error sensitivity. The calculation formula is as follows: (28)。 5. The method for fusing pose perception data of multi-degree-of-freedom forming equipment considering error sensitivity according to claim 4, characterized in that, Step S5 includes the following steps: Error sensitivity based on grating ruler measurement using vector This indicates that the error sensitivity based on monocular vision measurement is expressed using vectors. The covariance matrix is ​​expressed as follows: and ; exist At any given time, the equipment pose is set to... Let its first and second derivatives with respect to time be set as state variables, i.e. Each state variable is independent, and the state equations for the two measurement methods are as follows: (29) In the formula, Let the state noise vector be Gaussian noise with zero mean and covariance. It is the state transition matrix; Assuming the sampling time is Then at time and time The relationship between attitude and attitude can be expressed using Taylor expansion: (30) Therefore, the state transition matrix Equation (31) represents: (31) To achieve the fusion of the two types of data, two sub-filters are needed, and the measurement equation for each sub-filter is as follows: (32) In the formula, The pose values ​​of sub-filters 1 and 2 are the measured values ​​obtained from the grating ruler position and monocular vision measurements. For the measurement matrix, The error sensitivity of different measurement methods represents the reliability of the measurement data; The two sub-filters receive pose data and measurement noise from the grating ruler-based measurement subsystem and the monocular vision-based measurement subsystem, respectively, to obtain the state vector. and state covariance matrix The main filter first performs a time update to obtain the state vector. and state covariance matrix Then, the main filter merges all the obtained state vectors and state covariance matrices to obtain the globally optimal estimated state vector. and state covariance matrix .

6. The method for fusing pose perception data of multi-degree-of-freedom forming equipment considering error sensitivity according to claim 5, characterized in that, The main filter combines all the obtained state vectors and state covariance matrices to obtain the globally optimal estimated state vector. and state covariance matrix The methods include: (1) The initial values ​​of the three filters are obtained from the optimal estimate of the previous time step: (33) In the formula, For allocation coefficients, ; (2) Three filter state vectors and state covariance matrix : (34) In the formula, , Indicates sub-filters 1 and 2; (3) Optimal estimation of equipment pose: (35)。

Images