A shape sensing cross-sectional deformation error correction method
By calculating and optimizing the curvature and bending angle of the sensor, the problem of deformation error of the matrix section in shape sensing technology is solved, real-time correction and accuracy improvement are achieved, and the hardness of the matrix is changed.
Patent Information
- Application Number
- CN202211566813.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-12-07
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2042-12-07
AI Technical Summary
The existing shape sensing technology is difficult to effectively solve the error problem when the cross-section of the substrate is deformed, and some sensing methods will damage the softness and hardness of the substrate.
By obtaining strain data of different measuring points in the shape sensor, using curvature and bending angle to calculate the angle and distance from the center between different sensing points, applying formulas to calculate the theoretical strain value, and adjusting the curvature and bending angle through an optimization algorithm To achieve the optimal solution, the correction of cross-section deformation error is achieved.
Real-time automatic correction of cross-section deformation errors during bending during sensing of the base shape is realized, and the shape reconstruction accuracy is improved without changing the softness and hardness of the base.
Smart Images

Figure CN115781239B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of shape sensing and relates to a shape sensing cross-sectional deformation error correction method. Background Art
[0002] In recent years, shape sensing technology has received extensive attention in the fields of industrial machinery, biomedicine, aerospace, etc. Since the technology can well sense and display the deformation of the substrate, it has great application prospects in the fields of exploration, rescue, minimally invasive surgery, etc.
[0003] Among them, most of the shape sensing technology used to perceive the shape and movement of robots or robotic arms is surface-mounted or embedded. Most of the sensing units of robots or robotic arms can be regarded as hollow or solid circular tubes. Under the action of bending, their cross-section is likely to be squeezed into an ellipse, thereby changing the distance from the sensing point to the center of the cross-section. Therefore, the cross-sectional deformation of the substrate is an important factor affecting the accuracy of deformation monitoring. At present, the solutions to the cross-sectional deformation of the substrate are: directly using multi-core optical fibers to embed into the center of the substrate, increasing the strength of the substrate by embedding metal braids in the substrate to reduce elliptical deformation, attaching the sensor to the memory alloy wire, embedding the memory alloy wire into the software, embedding the sensor into a multi-layer polyimide film, and embedding the film into the substrate. These methods have their own advantages and disadvantages and are suitable for different application scenarios.
[0004] The current sensing methods have the following main disadvantages: 1. There is no good solution to the problem of errors caused by deformation of the substrate cross section; 2. Some sensing methods will destroy the hardness of the substrate itself. Summary of the invention
[0005] In view of the above-mentioned prior art, the technical problem to be solved by the present invention is to provide a shape sensing cross-sectional deformation error correction method to solve the error problem caused by cross-sectional deformation of the substrate.
[0006] In order to solve the above technical problems, a shape sensing cross-sectional deformation error correction method of the present invention comprises the following steps:
[0007] Step 1: When the shape sensor is used, the strain data of different sensor arrangement points at the same measuring point under any bending curvature and bending direction are obtained as (ε ij ′、ε ij ′、ε ij ′, …), where i represents different measuring points, and j represents different sensing points at the same measuring point;
[0008] Step 2: Within the range defined by measuring strain Inner curvature k i , where (ε′) max represents (ε ij′、ε ij ′、ε ij ′, …);
[0009] Step 3: Select the bending angle within the bending angle limit range [0°, 360°]
[0010] Step 4: Based on the fact that the arc length between the two sensing points before and after bending remains unchanged, the angles (α1, α2, ...) between different sensing points are calculated by curve integral using formula f1, where f1 is specifically:
[0011]
[0012] Wherein, a represents the major axis of the cross-sectional ellipse, b represents the minor axis of the cross-sectional ellipse, and r represents the cross-sectional radius of the substrate when not bent;
[0013] Step 5: According to the curvature k of the same measuring point i , bending angle The angle between the sensing points (α1, α2, ...) and the distance between the sensing point and the center point are calculated using the formula f2 (d ij d ij d ij , …), where i represents different measuring points, j represents different sensing points at the same measuring point, and f2 is specifically:
[0014]
[0015]
[0016]
[0017]
[0018] Where t represents the wall thickness of the substrate, and v represents the Poisson's ratio of the substrate;
[0019] Step 6: Set the curvature k i , bending angle Distance between sensing point and center point (d i1 d i2 d i3 , …) and the angle between the sensing points (α1, α2, …) are substituted into formula f3 to obtain the theoretical strain value (ε i1 , ε i2 , ε i3 , …), f3 is specifically:
[0020]
[0021] Step 7: Compare the theoretical strain value with the measured strain value to determine whether the difference meets the output condition. If not, return to step 2 and reselect a different curvature k. i and bending angle If the output condition is met, the optimal solution (ε i1 , ε i2 , ε i3 , …), and output the corresponding curvature k i and bending angle
[0022] Furthermore, a coordinate rotation method, genetic algorithm, annealing algorithm, particle swarm algorithm or ant colony algorithm are used to reselect different curvatures k i and bending angle
[0023] Beneficial effects of the invention: The invention provides a shape sensing cross-sectional deformation error correction method, which automatically corrects the cross-sectional deformation error generated during the bending process of the substrate shape sensing in real time. The calibration method does not need to change the hardness of the substrate itself, and does not need to embed other objects in the substrate, thereby improving the accuracy of shape reconstruction. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] Figure 1 It is a flow chart of the present invention;
[0025] Figure 2 The fixing structure of the sensor and the substrate before and after bending of the present invention;
[0026] Figure 3 This is a schematic diagram of the curvature error before and after correction of the present invention;
[0027] Figure 4 This is a schematic diagram of the bending angle error before and after correction according to the present invention. DETAILED DESCRIPTION
[0028] The present invention is further described below in conjunction with the accompanying drawings and embodiments. Based on the examples in the present invention, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present invention.
[0029] The present invention defines the curvature k at the same measuring point during shape measurement: i , bending angle Distance from the sensing point to the center point (d i1 d i2 d i3 , …); define the curvature k at the same measuring point when measuring shape i , bending angle Distance between sensing point and center point (d i1 di2 d i3 , …) and the angle (α1, α2, …) between the sensing point and the strain ε ij The strain data of different sensing points on the same cross section at the measuring point can be measured under any shape change state (ε i1 ′、ε i2 ′、ε i3 ′, …); select the curvature k within a certain range i and bending angle The angles (α1, α2, ...) between different sensing points are calculated based on the arc length curve integral. The angles (α1, α2, ...) are substituted into the strain ε ij The relationship between the two is used to calculate the strain data (ε i1 , ε i2 , ε i3 ,…); using the optimization algorithm to determine (ε i1 , ε i2 , ε i3 , ...) is the optimal solution, and the curvature k corresponding to the optimal solution i and bending angle That is the corrected curvature and bending angle.
[0030] Combination Figure 1 , the present invention specifically comprises the following steps:
[0031] Step 1: When using the shape sensor, the strain data of different sensor arrangement points at the same measuring point can be obtained under any bending curvature and bending direction (ε ij ′、ε ij ′、ε ij ′, …), where i represents different measuring points, and j represents different sensing points at the same measuring point;
[0032] Step 2: Select curvature k within a certain range i , this range is limited by measuring strain Where (ε′) max represents (ε ij ′、ε ij ′、ε ij ′, …);
[0033] Step 3: Select a bending angle within a certain range The bending angle range is limited to [0°, 360°];
[0034] Step 4: Based on the fact that the arc length between the two sensing points before and after bending remains unchanged, the angles (α1, α2, ...) between different sensing points are calculated by curve integral using formula f1, where f1 is specifically:
[0035]
[0036] Wherein a represents the major axis of the cross-sectional ellipse, b represents the minor axis of the cross-sectional ellipse, and r represents the cross-sectional radius of the substrate when not bent.
[0037] Step 5: According to the curvature k of the same measuring point i , bending angle The angle between the sensing points (α1, α2, ...) and the distance between the sensing point and the center point are calculated using the formula f2 (d ij d ij d ij , …), where i represents different measuring points, j represents different sensing points at the same measuring point, and f2 is specifically:
[0038]
[0039]
[0040]
[0041]
[0042] Where t represents the wall thickness of the substrate, and ν represents the Poisson's ratio of the substrate;
[0043] Step 6: Set the curvature k i , bending angle Distance between sensing point and center point (d i1 d i2 d i3 , …) and the angle between the sensing points (α1, α2, …) are substituted into formula f3 to obtain the theoretical strain value (ε i1 , ε i2 , ε i3 , …), f3 is specifically:
[0044]
[0045] Step 7: Compare the theoretical strain value with the measured strain value to see if the output condition is met. If not, repeat steps 2 to 7. If the output condition is met, the optimal solution (ε i1 , ε i2 , ε i3 , …), and output the corresponding curvature k i and bending angle
[0046] When the substrate to which the shape sensor is attached in step 1 is bent, the cross section will be deformed.
[0047] In step 4, the angle between the sensing points on the same measuring point section is no longer uniformly divided into 360°. Based on the fact that the arc length remains unchanged before and after bending, the angle between the sensing points is calculated using curve integration.
[0048] In step 5, when bending, the distance between each sensing point and the center point of the cross section is no longer constant to the radius r, and needs to be calculated according to formula f2 (d i1 d i2 d i3 , …).
[0049] Select the curvature k in step 2 and step 3 i and bending angle Optimization algorithms such as coordinate rotation method, genetic algorithm, annealing algorithm, particle swarm algorithm or ant colony algorithm can be used. The comparison between the theoretical value and the measured value in step seven is determined according to the optimization formula.
[0050] Figure 2 The dashed line is a schematic cross-sectional view at a certain measuring point when the substrate is not bent, wherein the three sensors are spaced 120° apart. Figure 2 The solid line is a schematic diagram of the cross section at a certain measuring point when the substrate is bent, in which the intervals between the three sensors and the distance from the center of the cross section have changed. In the example, a fiber Bragg grating is used as a sensor surface mounted on the substrate. After the substrate is bent, the offset of the fiber Bragg grating spectrum can be measured in real time by an optical demodulator. The offset is proportional to the strain information of the grating position, so the strain information at the three fiber Bragg gratings can be solved, as shown in formula f1. When the substrate is bent and deformed, and the external temperature does not change, the relationship between the strain values measured by the three sensors of the same cross section and the curvature and bending angle at this measuring point can be determined according to the geometric model of the sensor as formula f3.
[0051] When the substrate is bent, the intervals between the sensors at the same measuring point change, but the perimeter of the same measuring point before and after bending remains unchanged. It can be considered that the arc length between two adjacent sensors at the same measuring point remains unchanged, as shown in formula f1. When the substrate is bent at any curvature and angle, the relationship between the wavelength change of the three fiber gratings at each measuring point and the measured strain value is shown in formula f4:
[0052]
[0053] In order to verify the effect of the algorithm, a single-side point is selected for simulation experiments. In the simulation experiment: r = 5.79mm, t = 1.73mm, ν = 0.3, k = [11.2, 11.6, 12.0, 12.4, 12.8], according to Figure 1 The curvature error diagram before and after correction is shown in the flowchart. Figure 3 The schematic diagram of the bending angle error before and after correction is shown in Figure 4As shown in the figure, the final curvature error can be reduced from 14% to less than 2%, and the angle error can be reduced from 9° to less than 2°. It can be seen that the algorithm has a good correction effect on the curvature and bending angle errors caused by the deformation of the curved section of the substrate, and improves the accuracy of 3D shape reconstruction.
[0054] The above is only one example of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made on the basis of the present invention shall be included in the protection scope of the present invention.
Claims
1. A shape sensing cross-sectional deformation error correction method, It is characterized in that The following steps are involved: Step 1: When the shape sensor is used, the strain data of different sensor arrangement points at the same measuring point under any bending curvature and bending direction are obtained as (ε ij ′、ε ij ′、ε ij ′, …), where i represents different measuring points, and j represents different sensing points at the same measuring point; Step 2: Within the range defined by measuring strain Inner curvature k i , where (ε′) max Represents (ε ij ′、ε ij ′、ε ij ′, …); Step 3: Select the bending angle within the bending angle limit range [0°, 360°] Step 4: Based on the fact that the arc length between the two sensing points before and after bending remains unchanged, apply formula f 1 Curve integral calculation of the angle (α) between different sensing points 1 , α 2 , …), f 1 Specifically: Wherein, a represents the major axis of the cross-sectional ellipse, b represents the minor axis of the cross-sectional ellipse, and r represents the cross-sectional radius of the substrate when not bent; Step 5: According to the curvature k of the same measuring point i , bending angle and the angle between the sensing point (α 1 , α 2 , …) and the formula for the distance between the sensing point and the center point f 2 Calculate (d ij ,d ij ,d ij , …), where i represents different measuring points, j represents different sensing points at the same measuring point, and f 2 Specifically: Where t represents the wall thickness of the substrate, and v represents the Poisson's ratio of the substrate; Step 6: Set the curvature k i , bending angle Distance between sensing point and center point (d i1 d i2 d i3 , …) and the angle between the sensing points (α 1 , α 2 , …) into the formula f 3 , the theoretical strain value is (ε i1 , ε i2 , ε i3 , …), f 3 Specifically: Step 7: Compare the theoretical strain value with the measured strain value to determine whether the difference meets the output condition. If not, return to step 2 and reselect a different curvature k. i and bending angle If the output condition is met, the optimal solution (ε i1 , ε i2 , ε i3 , …), and output the corresponding curvature k i and bending angle 2. A shape sensing cross-sectional deformation error correction method according to claim 1, Features: Use coordinate rotation method, genetic algorithm, annealing algorithm, particle swarm algorithm or ant colony algorithm to reselect different curvature k i and bending angle
