A method for docking a surgical robot and a tunnel needle

Abstract

The present invention discloses a method for docking a surgical robot with a tunnel needle, including coarse positioning and fine positioning, wherein the coarse positioning includes: moving the robot end to the entrance of the tunnel needle and making the robot end consistent with the orientation of the tunnel needle; the fine positioning includes: setting a threshold τ; photographing the interior of the tunnel needle, recording the area S1 of the tunnel needle wall, the area S2 inside the target object, and the inner ring area S=S1+S2 of the tunnel needle, if S1 is greater than the threshold τ, calculating the center point w of area S, moving the robot end to the position w, and calculating the center of gravity w1 and w2 of S1 and S2, and adjusting the direction of the robot end according to the direction of the vector #imgabs0# until S1 is no greater than the threshold τ. The method for docking a surgical robot with a tunnel needle provided by the present invention first performs coarse positioning, and then uses fine positioning to make the robot end continuously adjust its posture until the robot end and the tunnel needle are accurately aligned, and no manual intervention is required when the robot end and the tunnel needle are aligned.

Description

Technical Field

[0001] The present invention relates to the technical field of robot precision positioning, and more specifically, to a method for docking a surgical robot with a tunnel needle. Background Art

[0002] Even experienced surgeons experience hand tremors during surgeries, and physical fatigue from prolonged surgeries can lead to improper operations. Consequently, the use of surgical robots to assist in surgeries is becoming increasingly popular. The use of surgical robots requires an operating system to operate them, and having an excellent operating system to assist can significantly improve the performance of surgical robots, especially for delicate surgeries.

[0003] Currently, when surgical robots are used to assist in surgery, delicate operations such as docking the surgical instruments carried by the end of the surgical robot with the tunnel needle are mainly performed by the doctor manually adjusting the robot's posture to align the surgical instruments and tunnel needle bound to the end. Doctors need to rely on their experience as well as visual and tactile feedback to continuously adjust and align the surgical instruments and tunnel needle, resulting in long surgical times and a significant drain on the doctor's energy. At the same time, doctors also need to spend a lot of time practicing alignment operations, which imposes additional time and cognitive requirements on the doctor. For example, when the end of the robot docks with the tunnel needle in ophthalmic surgery, the eyeball into which the tunnel needle is inserted is a very fragile and delicate structure. If the surgical instrument at the end of the surgical robot can be accurately docked when entering the tunnel needle, the disturbance to the tunnel needle and thus the damage to the eyeball tissue can be reduced, thereby improving the surgical effect and achieving a better prognosis. Summary of the Invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art in which the robot end and the tunnel needle rely on manual judgment, making it difficult to accurately dock, and to provide a method for docking a surgical robot with a tunnel needle. The method for docking a surgical robot with a tunnel needle provided by the present invention first performs coarse positioning so that the robot end is located at the entrance of the tunnel needle and roughly aligned with the tunnel needle. Then, fine positioning is used to continuously adjust the posture of the robot end until the robot end and the tunnel needle are accurately aligned. The docking accuracy is very high. At the same time, when the robot end and the tunnel needle are docked, no manual intervention is required, saving the doctor's learning time and operation time.

[0005] The purpose of the present invention can be achieved by adopting the following technical solutions:

[0006] A method for docking a surgical robot with a tunnel needle includes coarse positioning and fine positioning, wherein the coarse positioning includes the following steps:

[0007] S1: The robot's end is equipped with a camera and surgical instruments. Under the illumination of a light source, the camera is used to photograph the target object with the tunnel needle inserted. The control terminal identifies the location of the tunnel needle based on the color difference between the tunnel needle and the target object, moves the robot's end to the entrance of the tunnel needle, and adjusts the robot's end position so that the axis of the operating part of the surgical instrument on the robot's end is roughly aligned with the axis of the tunnel needle.

[0008] Precision positioning includes the following steps:

[0009] S2.1: set threshold τ;

[0010] S2.2: Use the macro-magnification lens of the camera on the end of the robot to capture the interior of the tunnel needle. Record the area S1 of the tunnel needle wall that the camera on the end of the robot focuses on, the area S2 inside the target object, and the inner ring area S = S1 + S2 of the tunnel needle. Calculate the center point w of area S and move the end point of the robot end to the position w.

[0011] S2.3: Determine whether S1 is greater than a threshold value τ. If S1 is greater than the threshold value τ, proceed to step S2.4. If S1 is less than the threshold value τ, determine that the robot end is precisely aligned with the tunnel needle.

[0012] S2.4: If S1 is greater than the threshold τ, the robot terminal needs to be adjusted. When adjusting the posture, calculate the center of gravity position w1 of S1 and the center of gravity position w2 of S2, and connect w1 and w2 to obtain the vector At the same time, the end point of the robot is fixed at w and does not move according to the vector Adjust the direction of the robot end in the direction of the vector The direction is adjusted by a certain angle δ, and step S2.3 is repeated after the posture adjustment is completed.

[0013] The target object is a spherical or nearly spherical, translucent object, such as an eyeball. When the tunnel needle is inserted into the target object, it is aligned with the target's center point and inserted along the target's radius. Since the target object is not necessarily a perfect sphere, and it is difficult to accurately observe the overall shape of a target object such as an eyeball, the axis of the tunnel needle may not completely coincide with the radius of the target object. Therefore, when the robot end enters the tunnel needle, the positioning accuracy is limited by simply locating the target's center point, a certain point on the target object, or a certain direction on the target object. Therefore, the method for docking the surgical robot and the tunnel needle provided in step S1 of the present invention only serves as rough positioning. During rough positioning, the axes of the robot end and the tunnel needle can only roughly align and cannot be precisely aligned. After rough positioning, a more precise fine positioning is required. Thus, before the robot end enters the tunnel needle, the axis of the operating member of the surgical instrument carried by the robot end is more accurately aligned with the tunnel needle, thereby reducing the possibility that the robot end disturbs the tunnel needle due to misalignment, thereby preventing damage to the target object caused by tunnel needle displacement, which is particularly harmful when the eyeball is used as the target object.

[0014] During coarse positioning, the front of the current target object is photographed, and the position of the tunnel needle is identified based on the pixel difference. During the coarse positioning process, the target object is regarded as a standard sphere, and the center of the target object can be converted based on the image taken from the front of the target object. The end point of the robot end is moved to the entrance of the tunnel needle and the orientation of the robot end is adjusted, that is, the posture of the surgical instrument carried on the robot end is adjusted so that the working direction of the surgical instrument is roughly consistent with the tunnel needle. The coarse positioning here can adjust the direction of the robot end according to the recorded insertion direction of the tunnel needle, or point both the tunnel needle and the robot end to the center point of the target object, or calculate the position and orientation of the tunnel needle through the front image of the target object, or other methods are all possible. However, the more accurate the coarse positioning is, the more conducive it is to the subsequent fine positioning.

[0015] During precision positioning, since the target object is translucent, its interior is illuminated by the light source. However, the color inside the tunnel needle differs from that of the target object, and the tunnel needle is not translucent. Therefore, when the camera on the robot end observes the tunnel needle from one end to the other, there is a certain color difference between the observed area S1 of the tunnel needle's inner wall and the area S2 of the target's interior observed from the other end of the tunnel needle. Therefore, by continuously comparing areas S1 and S2, the robot end is guided to rotate around the end point of the robot end. The end point of the robot end is the front-most end point of the surgical instrument carried by the robot end. Each rotation re-determines and compares areas S1 and S2 until the robot end is fully aligned with area S1, at which point the camera cannot focus on the inner wall of the tunnel needle. Here, the center of gravity position w1 of S1 is the coordinate (x1, y1), and the center of gravity position w2 of S2 is the coordinate (x2, y2).

[0016] Based on the relative position of the camera and the robot end-point, a threshold value τ is obtained through statistical analysis of multiple experiments. This threshold τ is set as the condition for determining whether the robot end-point is fully aligned. If S1 is greater than the threshold τ, it indicates that the robot end-point is aligned with the inner wall of the tunnel needle. If S1 is less than the threshold τ, it can be considered that the robot end-point is accurately positioned with the tunnel needle, and the axis of the robot end-point is aligned with the center axis of the tunnel needle. Image recognition algorithms can be used to determine the center point w.

[0017] Furthermore, in step S2.2, when performing area recognition on the captured image, the inner ring area S of the tunnel needle is calculated based on the pixel difference, the pixel coordinate range of the inner ring area is obtained, the pixel points within the inner ring coordinate area are clustered to obtain an area with two clustering centers, and the number of pixel points in the two categories is counted and updated respectively, and S1 and S2 are updated.

[0018] When distinguishing between areas S1 and S2, since the pixels at their boundaries are close, and the camera equipment carried by the robot end is constantly adjusted as the position and posture of the robot end are adjusted, the light intensity will also change. If the threshold method is used to divide areas S1 and S2, the threshold is difficult to determine and the value will change continuously. Therefore, it is more accurate to use a clustering algorithm to divide areas S1 and S2.

[0019] Furthermore, in step S2.3, when calculating the center of gravity position w1 of S1 and the center of gravity position w2 of S2, the S1 and S2 regions are binarized respectively, and the center of gravity positions w1 and w2 of S1 and S2 are obtained by calculating the first-order moments of the regions S1 and S2.

[0020] Furthermore, in step S2.3, when the posture of the robot end is adjusted, the fixed point movement algorithm is used according to the vector Adjust the direction of the robot end.

[0021] Furthermore, the step S2.1 further includes the following steps: recording the initial state A of the robot end;

[0022] In step S2.3, after each posture adjustment of the robot end, the translation matrix and rotation matrix of the robot end are recorded at this time, and the current state A is obtained by multiplying the translation matrix and rotation matrix with the initial state A. k .

[0023] Different surgical instruments may be replaced during the same operation. Therefore, the current spatial posture information of the robot end is recorded through the state matrix. When the surgical instrument carried on the robot end is replaced, the robot end can be quickly positioned at one time according to the recorded matrix information, without the need to redo the coarse and fine positioning process.

[0024] Furthermore, the step S1 specifically includes the following steps:

[0025] S1.1: The tunnel needle is pointed at the center point O of the target object and inserted into the target object. Under the illumination of the light source, the front of the target object with the tunnel needle inserted is photographed using a camera. The initial position of the robot end is that the endpoint of the robot end is located in front of the center point O1 of the front of the target object and the axis of the operating part of the surgical instrument at the robot end is on the same straight line as the line connecting the center point O1 of the front of the target object and the center point O of the target object.

[0026] S1.2: In the frontal image of the target captured by the camera, the lateral diameter of the target is AB. The color difference between the tunnel needle and the target is used to determine the position C of the tunnel needle within the target. The distance from point C to the horizontal plane containing line AB is CD, where point D is the foot of the perpendicular. The distance from point D to line OO1 is D O2, where O2 is the foot of the perpendicular. The diameter AB is determined based on the average size of the target. The lengths of D O2 and CD are calculated. The end of the robot is translated in the vertical plane parallel to line AB toward the end closest to point C by a distance |DO2| and parallel to line CD toward the end closest to point C by a distance |CD|. At this point, the end point of the robot is located at the entrance of the tunnel needle.

[0027] S1.3: Based on the radius CO, the lengths of line segment CD, and line segment D O2, calculate the length of line segment O1 O2, and translate the robot end parallel to line O O1 by a distance |O1O2| toward the end closest to point O. S1.4: Calculate the angle α of ∠COD and the angle β of ∠DO O2, and rotate the robot end around line CD by an angle β toward point C. Then, rotate the robot end around a line perpendicular to plane COD by an angle α toward point C. At this point, the robot end and the tunnel needle are aligned in the same direction.

[0028] Because the true distance between objects cannot be calculated directly from an image, a reference object is needed. By capturing the front of the target object, we can determine the target's diameter AB. Since the first step only requires coarse positioning, a standard model of the target is assumed during coarse positioning, and the average size of the standard model is calculated. Therefore, the diameter of the standard model is assigned to the diameter AB for calculation.

[0029] After the camera equipment takes a picture of the front of the target object, the captured image is identified to obtain the center O1, diameter AB, and the position of the tunnel needle in the front image of the target object. In the coarse positioning, the target object is regarded as a sphere and the average value of the target object's diameter is assigned to the diameter AB. Therefore, the radius R=OC=OO1, CO2, DO2 and CD of the target object can all be obtained by analyzing the front image of the target object.

[0030] Furthermore, in step S1.3 and step S1.4, the lengths of line segments O1 and O2 and the angles α and β are calculated as follows:

[0031]

[0032] OD=OCcos(α)=Rcos(α)

[0033]

[0034] OO2=ODcos(β)

[0035] O1O2=OO1-OO2=R-OO2

[0036] Furthermore, in step S1.4, when the robot end rotates, the end point of the robot end is fixed and the robot end rotates around the end point.

[0037] During the coarse positioning process, when the end point of the robot moves according to the calculation result, the end point of the robot moves to the corresponding position. When the robot is adjusted, the end point of the robot is rotated around it.

[0038] Furthermore, the tunnel needle adopts a color with a color difference ΔE greater than 1.5 from the target object.

[0039] The greater the difference in color between the tunnel needle and the target object, the easier it is for the computer to analyze the image captured by the camera.

[0040] Furthermore, the irradiation direction of the light source is not parallel to the axis of the tunnel needle.

[0041] The irradiation direction of the light source is not parallel to the axis of the tunnel needle, so that the interior of the tunnel needle is overexposed, affecting the shooting of the camera equipment.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] (1) When the robot end is docked with the tunnel needle, the entire positioning process is automatically aligned with high accuracy and does not require manual intervention, saving the doctor's learning time and operation time.

[0044] (2) When the robot end is docked with the tunnel needle, coarse positioning is first performed so that the robot end is at the entrance of the tunnel needle and roughly aligned with the tunnel needle. Then, fine positioning is used to continuously adjust the posture of the robot end until the robot end is accurately aligned with the tunnel needle. The docking accuracy is very high.

[0045] (3) In the process of precise positioning, different areas are divided by clustering algorithms, and different categories are automatically found, which can avoid the need to constantly adjust the threshold due to changes in light sources.

[0046] (4) The saved rotation matrix and translation matrix can be used to achieve one-time alignment when replacing the instrument carried by the robot end during the same operation on the same target object.

[0047] (5) The coarse positioning process only requires a single shot of the front of the target object to convert the moving direction, distance, and rotation angle. It does not require a lot of calculations and is very fast. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic diagram of the overall structure of the target object of the present invention;

[0049] Figure 2 A schematic diagram of the front view of a target object photographed by the camera device of the present invention;

[0050] Figure 3 Schematic diagram of the overall structure of the tunnel needle;

[0051] Figure 4 This is a schematic diagram of the overall structure of the tunnel needle from another perspective;

[0052] Figure 5 This is a schematic diagram of the interior of the tunnel needle captured by a camera during the docking process of the present invention;

[0053] Figure 6 A schematic diagram of the interior of the tunnel needle captured by the camera device after docking in the present invention;

[0054] Figure 7 Flow chart of the method of the present invention.

[0055] The icon markings are explained as follows:

[0056] 1-Target, 2-Tunnel needle, 3-Light source. DETAILED DESCRIPTION

[0057] The present invention is further described below with reference to specific embodiments. The accompanying drawings are for illustrative purposes only and are schematic, not actual, representations. They should not be construed as limiting this patent. To better illustrate the embodiments of the present invention, some components in the drawings may be omitted, enlarged, or reduced, and do not represent the actual dimensions of the product. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted from the drawings.

[0058] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right" and the like indicate directions or positional relationships based on the directions or positional relationships shown in the drawings, it is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting this patent. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0059] Example 1

[0060] like Figures 1 to 5 As shown, a method for docking a surgical robot with a tunnel needle includes coarse positioning and fine positioning, wherein the coarse positioning includes the following steps:

[0061] S1: The end of the robot is equipped with a camera and surgical instruments. Under the illumination of light source 3, the camera is used to photograph the target object 1 with the tunnel needle 2 inserted therein. The control terminal identifies the position of the tunnel needle 2 based on the color difference between the tunnel needle 2 and the target object 1, moves the end of the robot to the entrance of the tunnel needle 2, and adjusts the posture of the end of the robot so that the axis of the operating part of the surgical instrument at the end of the robot is roughly aligned with the axis of the tunnel needle 2.

[0062] Precision positioning includes the following steps:

[0063] S2.1: set threshold τ;

[0064] S2.2: Use the macro-magnification lens of the camera on the end of the robot to capture the interior of the tunnel needle 2. Record the area S1 of the tunnel needle 2 wall that the camera on the end of the robot focuses on, the area S2 inside the target 1, and the inner ring area S = S1 + S2 of the tunnel needle 2. Calculate the center point w of area S and move the end point of the robot end to the position w.

[0065] S2.3: Determine whether S1 is greater than a threshold value τ. If S1 is greater than the threshold value τ, proceed to step S2.4. If S1 is less than the threshold value τ, determine that the robot end is precisely aligned with the tunnel needle 2.

[0066] S2.4: Adjust the posture of the robot end. When adjusting the posture, calculate the center of gravity position w1 of S1 and the center of gravity position w2 of S2, and connect w1 and w2 to obtain the vector At the same time, the end point of the robot is fixed at w and does not move according to the vector Adjust the direction of the robot end in the direction of the vector The direction is adjusted by a certain angle δ, and step S2.3 is repeated after the posture adjustment is completed.

[0067] The target object 1 is a spherical or nearly spherical translucent object, such as an eyeball. When the tunnel needle 2 is inserted into the target object 1, it is aligned with the center point of the target object 1 and enters along the radius of the target object 1. Since the target object 1 is not necessarily a standard sphere and it is difficult to accurately observe the overall shape of the target object 1 such as the eyeball, the axis of the tunnel needle 2 does not necessarily completely coincide with the radius of the target object 1. Therefore, when the robot end enters the tunnel needle 2, the positioning accuracy is limited only by positioning the center point of the target object 1 or a certain point of the target object 1 or a certain direction of the target object 1. Therefore, the docking method of the surgical robot and the tunnel needle provided in step S1 of the present invention only regards it as rough positioning. In the rough positioning, the axis directions of the robot end and the tunnel needle can only be roughly consistent and cannot be accurately aligned. After the rough positioning, more accurate fine positioning is required. In this way, before the robot end enters the tunnel needle 2, the axis of the operating part of the surgical instrument carried on the robot end is more accurately aligned with the tunnel needle 2, thereby reducing the disturbance of the tunnel needle 2 by the robot end due to misalignment, thereby reducing the displacement of the tunnel needle 2 and avoiding damage to the target object 1 caused by the displacement of the tunnel needle 2, especially when the eyeball is used as the target object 1 and is very vulnerable to damage.

[0068] During coarse positioning, the front of the current target object is photographed, and the position of the tunnel needle 2 is identified based on the pixel difference. During the coarse positioning process, the target object 1 is regarded as a standard sphere, and the center of the target object 1 can be converted based on the image taken from the front of the target object 1. The end point of the robot end is moved to the entrance of the tunnel needle 2 and the orientation of the robot end is adjusted, that is, the posture of the surgical instrument carried on the robot end is adjusted so that the working direction of the surgical instrument is roughly consistent with the tunnel needle 2. The coarse positioning here can adjust the direction of the robot end according to the recorded insertion direction of the tunnel needle 2, or it can point both the tunnel needle 2 and the robot end to the center point of the target object 1, or it can calculate the position and orientation of the tunnel needle 2 by taking the front image of the target object 1, or other methods are all possible. However, the more accurate the coarse positioning is, the more conducive it is to the subsequent fine positioning.

[0069] During precision positioning, since target 1 is translucent, its interior is illuminated by light source 3. However, the color inside tunnel needle 2 differs from that of target 1, and since tunnel needle 2 is not translucent, when the camera on the robot end observes from one end of tunnel needle 2 to the other, there is a certain color difference between area S1 of the inner wall of tunnel needle 2 and area S2 of the interior of target 1 as seen from the other end of tunnel needle 2. Therefore, by continuously comparing areas S1 and S2, the robot end is guided to rotate around its endpoint (the endpoint of the distal end of the surgical instrument carried by the robot end). Areas S1 and S2 are re-determined and compared with each rotation until the robot end is fully aligned with area S1, at which point the camera loses focus on the inner wall of tunnel needle 2. Here, the center of gravity position w1 of S1 is represented by the coordinates (x1, y1), and the center of gravity position w2 of S2 is represented by the coordinates (x2, y2).

[0070] Based on the relative position of the camera and the robot end-point, a threshold value τ is obtained through statistical analysis of multiple experiments. This threshold τ is set as the condition for determining whether the robot end-point is fully aligned. If S1 is greater than the threshold τ, it indicates that the robot end-point is aligned with the inner wall of the tunnel needle. If S1 is less than the threshold τ, it can be considered that the robot end-point is accurately positioned with the tunnel needle, and the axis of the robot end-point is aligned with the center axis of the tunnel needle. Image recognition algorithms can be used to determine the center point w.

[0071] In step S2.2, when performing region recognition on the captured image, the inner ring area S of the tunnel needle 2 is calculated based on the pixel difference, and the pixel coordinate range x of the inner ring area is obtained. min ~x max ,y min ~y max , cluster the pixels within the inner ring coordinate area to obtain an area with two cluster centers, count and update the number of pixels in the two categories respectively, and update S1 and S2.

[0072] When distinguishing between areas S1 and S2, since the pixels at their boundaries are close, and the camera equipment carried by the robot end is constantly adjusted as the position and posture of the robot end are adjusted, the light intensity will also change. If the threshold method is used to divide areas S1 and S2, the threshold is difficult to determine and the value will change continuously. Therefore, it is more accurate to use a clustering algorithm to divide areas S1 and S2.

[0073] In step S2.3, when calculating the center of gravity position w1 of S1 and the center of gravity position w2 of S2, the S1 and S2 regions are binarized respectively, and the center of gravity positions w1 and w2 of S1 and S2 are obtained by calculating the first-order moments of the regions S1 and S2.

[0074] In step S2.3, when the robot end position is adjusted, the fixed point movement algorithm RCM is used according to the vector Adjust the direction of the robot end.

[0075] Step S1 specifically includes the following steps:

[0076] S1.1: The tunnel needle 2 points to the center point O of the target object 1 and is inserted into the target object 1. Under the illumination of the light source 3, the front of the target object 1 with the tunnel needle 2 inserted is photographed using a camera. The initial position of the robot end is that the endpoint of the robot end is located in front of the center point O1 of the front of the target object 1 and the axis of the operating part of the surgical instrument at the robot end is on the same straight line as the line connecting the center point O1 of the front of the target object 1 and the center point O of the target object 1.

[0077] S1.2: In the frontal image of target object 1 captured by the camera, the transverse diameter of target object 1 is AB. The color difference between tunnel needle 2 and target object 1 is used to determine the position C of tunnel needle 2 in target object 1. The distance from point C to the horizontal plane where line AB is located is CD, where point D is the foot of the perpendicular. The distance from point D to line OO1 is DO2, where O2 is the foot of the perpendicular. The size of diameter AB is determined based on the average size of target object 1. The lengths of DO2 and CD are calculated. The end point of the robot is translated in the vertical plane parallel to line AB toward the end closest to point C by a distance |DO2| and parallel to line CD toward the end closest to point C by a distance |CD|. At this point, the end point of the robot is located at the entrance of tunnel needle 2.

[0078] S1.3: Based on the radius CO, the lengths of line segment CD, and line segment D O2, calculate the length of line segment O1 O2, and translate the robot end parallel to line O O1 by a distance |O1O2| toward the end closest to point O. S1.4: Calculate the angle α of ∠COD and the angle β of ∠DO O2, and rotate the robot end around line CD by an angle β toward point C. Then, rotate the robot end around a line perpendicular to plane COD by an angle α toward point C. At this point, the robot end and tunnel needle 2 are aligned in the same direction.

[0079] Because the true distance between objects cannot be calculated directly from an image, a reference object is needed. By capturing the front view of object 1, we can determine its diameter AB. Since the first step only requires coarse positioning, a standard model of object 1 is assumed during coarse positioning, and the average size of this standard model is calculated. Therefore, the diameter of the standard model is assigned to diameter AB for calculation.

[0080] After the camera-related equipment takes a picture of the front of the target object 1, the captured image is identified to obtain the center O1, diameter AB, and the position of the tunnel needle in the front image of the target object 1. In the coarse positioning, the target object 1 is regarded as a sphere and the average value of the diameter of the target object 1 is assigned to the diameter AB. Therefore, the radius R=OC=OO1 of the target object 1, CO2, DO2 and CD can all be obtained by analyzing the front image of the target object 1.

[0081] In steps S1.3 and S1.4, the lengths of line segments O1 and O2 and the angles α and α are calculated as follows:

[0082]

[0083] OD=OCcosα=Rcosα

[0084]

[0085] OO2=ODcos(β)

[0086] O1O2=OO1-OO2=R-OO2

[0087] In step S1.4, when the robot end rotates, the end point of the robot end is fixed and the robot end rotates around the end point.

[0088] During the coarse positioning process, when the end point of the robot moves according to the calculation result, the end point of the robot moves to the corresponding position. When the robot is adjusted, the end point of the robot is rotated around it.

[0089] Example 2

[0090] In addition to the features described in Example 1, this embodiment also includes the following features:

[0091] Step S2.1 also includes the following steps: recording the initial state A of the robot end;

[0092] In step S2.3, after each posture adjustment of the robot end, the translation matrix T of the robot end is recorded. k and the rotation matrix R k , the translation matrix T k and the rotation matrix R k Multiply it by the initial state A to get the current state A k .

[0093] Different surgical instruments may be replaced during the same operation. Therefore, the current spatial posture information of the robot end is recorded through the state matrix. When the surgical instrument carried on the robot end is replaced, the robot end can be quickly positioned at one time according to the recorded matrix information, without the need to redo the coarse and fine positioning process.

[0094] Example 3

[0095] In addition to the features described in Example 1 or 2, this embodiment also includes the following features:

[0096] The irradiation direction of the light source 3 is not parallel to the axis of the tunnel needle 2 , which causes the interior of the tunnel needle 2 to be overexposed, affecting the shooting of the camera equipment.

[0097] The tunnel needle 2 is a color with a color difference ΔE greater than 1.5 from the target object 1. The greater the color difference between the tunnel needle 2 and the target object 1, the easier it is for the computer to analyze the image captured by the camera.

[0098] Obviously, the above embodiments of the present invention are merely examples for the purpose of clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the claims of the present invention.

Claims

1. A method for docking a surgical robot with a tunnel needle, characterized in that: It includes coarse positioning and fine positioning, where coarse positioning includes the following steps: S1: The end of the robot is equipped with a camera and a surgical instrument. Under the illumination of a light source (3), the camera is used to photograph a target object (1) having a tunnel needle (2) inserted therein. The control terminal determines the position of the tunnel needle (2) based on the color difference between the tunnel needle (2) and the target object (1), moves the end of the robot to the entrance of the tunnel needle (2), and adjusts the posture of the end of the robot so that the axis of the operating part of the surgical instrument at the end of the robot is roughly aligned with the direction of the axis of the tunnel needle (2); Precision positioning includes the following steps: S2.1: set threshold τ; S2.2: Use the macro magnification lens of the camera device located at the end of the robot to shoot the interior of the tunnel needle (2), record the area S1 of the tunnel needle (2) tube wall that the camera device at the end of the robot focuses on, the area S2 inside the target object (1), and the inner ring area S = S1 + S2 of the tunnel needle (2), calculate the center point w of the area S, and move the end point of the robot end to the position w; S2.3: Determine whether S1 is greater than a threshold value τ. If S1 is greater than the threshold value τ, proceed to step S2.

4. If S1 is less than the threshold value τ, determine that the robot end is precisely aligned with the tunnel needle (2); S2.4: Adjust the posture of the robot end. When adjusting the posture, calculate the center of gravity position w1 of S1 and the center of gravity position w2 of S2, and connect w1 and w2 to obtain the vector At the same time, the end point of the robot is fixed at w and does not move according to the vector Adjust the direction of the robot end in the direction of the vector The direction is adjusted by a certain angle δ, and step S2.3 is repeated after the posture adjustment is completed.

2. The method for docking a surgical robot and a tunnel needle according to claim 1, characterized in that: In the step S2.2, when performing region recognition on the captured image, the inner ring region S of the tunnel needle (2) is calculated based on the pixel difference, and the pixel coordinate range (x min ~x max ,y min ~y max ), cluster the pixels within the inner ring coordinate area to obtain an area with two cluster centers, count and update the number of pixels in the two categories respectively, and update S1 and S2.

3. The method for docking a surgical robot with a tunnel needle according to claim 2, characterized in that: In step S2.3, when calculating the center of gravity position w1 of S1 and the center of gravity position w2 of S2, the regions S1 and S2 are binarized respectively, and the center of gravity positions w1 and w2 of S1 and S2 are obtained by calculating the first-order moments of the regions S1 and S2.

4. The method for docking a surgical robot with a tunnel needle according to claim 3, characterized in that: In step S2.3, when the posture of the robot end is adjusted, the fixed point movement algorithm (RCM) is used according to the vector Adjust the direction of the robot end.

5. The method for docking a surgical robot with a tunnel needle according to claim 4, characterized in that: Said step S2.1 further includes the following steps: recording the initial state A of the robot end; In step S2.3, after each posture adjustment of the robot end, the translation matrix (T k ) and the rotation matrix (R k ), the translation matrix (T k ) and the rotation matrix (R k ) is multiplied by the initial state A to get the current state A k .

6. The method for docking a surgical robot and a tunnel needle according to any one of claims 1 to 5, characterized in that: The step S1 specifically includes the following steps: S1.1: The tunnel needle (2) is inserted into the target object (1) with its direction pointing to the center point O of the target object (1). Under the illumination of the light source (3), the front of the target object (1) with the tunnel needle (2) inserted is photographed using a camera. The initial position of the robot end is such that the end point of the robot end is located in front of the center point O1 of the front of the target object (1) and the axis of the operating part of the surgical instrument at the robot end is in the same straight line as the line connecting the center point O1 of the front of the target object (1) and the center point O of the target object (1); S1.2: In the front image of the target object (1) taken by the camera device, the transverse diameter of the target object (1) is AB. The color difference between the tunnel needle (2) and the target object (1) is used to determine the position C of the tunnel needle (2) in the target object (1). The distance from point C to the horizontal plane where the straight line AB is located is CD, where point D is the foot of the perpendicular. The distance from point D to the straight line OO1 is D O2, where O2 is the foot of the perpendicular. The size of the diameter AB is determined according to the average size of the target object (1), and the lengths of D O2 and CD are calculated. The end of the robot is translated in the vertical plane along the direction parallel to the straight line AB toward the end close to point C by a distance |DO2| and along the direction parallel to the straight line CD toward the end close to point C by a distance |CD|. At this time, the end point of the robot is located at the entrance of the tunnel needle (2); S1.3: Based on the radius CO, the length of the line segment CD and the length of the line segment D O2, calculate the length of the line segment O1 O2, and move the end of the robot parallel to the line O O1 toward the end close to point O by a distance |O1O2|; S1.4: Calculate the angle α of ∠COD and the angle β of ∠DOO2, and first rotate the end of the robot around the line CD toward the side close to point C by an angle β, and then rotate the end of the robot around the line perpendicular to the plane COD toward the side close to point C by an angle α. At this time, the end of the robot is in the same direction as the tunnel needle (2).

7. The method for docking a surgical robot with a tunnel needle according to claim 6, characterized in that: In steps S1.3 and S1.4, the lengths of line segments O1 and O2 and the angles α and β are calculated as follows: OD=OC cos(α)=R cos(α) OO2=ODcos(β) O1O2=OO1-OO2=R-OO 2。 8. The method for docking a surgical robot and a tunnel needle according to claim 6, characterized in that: In step S1.4, when the robot end rotates, the end point of the robot end is fixed and the robot end rotates around the end point.

9. The method for docking a surgical robot and a tunnel needle according to claim 1, characterized in that: The tunnel needle (2) adopts a color with a color difference value ΔE greater than 1.5 with the target object (1).

10. The method for docking a surgical robot and a tunnel needle according to claim 1, characterized in that: The irradiation direction of the light source (3) is not parallel to the axis of the tunnel needle (2).

Images