CN116740183A - Double-view cabin pose adjusting method - Google Patents
Double-view cabin pose adjusting method Download PDFInfo
- Publication number
- CN116740183A CN116740183A CN202311026182.5A CN202311026182A CN116740183A CN 116740183 A CN116740183 A CN 116740183A CN 202311026182 A CN202311026182 A CN 202311026182A CN 116740183 A CN116740183 A CN 116740183A
- Authority
- CN
- China
- Prior art keywords
- cabin
- pose
- standard
- image
- adjusted
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000001105 regulatory effect Effects 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 16
- 206010034701 Peroneal nerve palsy Diseases 0.000 claims description 8
- 230000000007 visual effect Effects 0.000 claims description 5
- 238000013459 approach Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 230000001276 controlling effect Effects 0.000 claims description 2
- 239000002775 capsule Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D3/00—Control of position or direction
- G05D3/12—Control of position or direction using feedback
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
Abstract
The invention discloses a method for adjusting the body pose of a double-view cabin. The pose motion is carried out on a standard cabin body, the corresponding relation between the five-degree-of-freedom space pose and the characteristic parameters is obtained through shooting by two cameras and image processing calibration, and the different calibration matrixes of the two cameras are obtained; the method comprises the steps that a to-be-adjusted cabin body is used for replacing a standard cabin body to perform pose movement, and characteristic parameters of the to-be-adjusted cabin body under shooting of two cameras are obtained; reversely solving the approximate relative pose and the actual pose; the position and the posture of the cabin body to be adjusted are adjusted in real time according to the actual position and the posture; and continuously repeating the steps to identify the pose of the cabin to be regulated in real time and dynamically regulate the pose until the error is within the threshold range. The invention can quickly and efficiently adjust the position and the posture of the revolving body cabin body, avoids a great amount of operations for solving the direct position and the posture, and can meet the position and the posture adjustment requirement of the revolving body in practical application.
Description
Technical Field
The invention relates to a workpiece pose adjusting method in the fields of computer graphics and computer-aided intelligent assembly, in particular to a double-view cabin pose adjusting method.
Background
The position and posture adjustment of the cabin body is divided into two parts, namely position and posture estimation and position and posture adjustment, and normally the two parts are continuously circulated, the sensor obtains data in real time to calculate the position and posture, and the position and posture adjustment of the executing mechanism is continuously carried out until the position and posture reach the target position and posture.
Pose adjustment by vision has stronger robustness than other sensors, and is used in more and more occasions at present. Usually, a vision sensor acquires an image of the cabin, solves a space pose by matching with template information, and then drives an executing mechanism to adjust the pose. However, the required precision cannot be obtained by directly solving the pose, and the pose settlement in the 3-dimensional space directly from the corresponding relation between the template and the image comprises complex geometric operation, so that the real-time performance is poor.
Disclosure of Invention
The invention provides a revolving body pose adjustment method based on a double-view image, which can be used for adjusting the pose of a revolving body cabin body more quickly and efficiently, avoiding a large amount of operations for solving the direct pose and meeting the requirements for adjusting the pose of the revolving body in practical application.
The invention needs to carry out vision-assisted adjustment on the cabin body, and the pose to be adjusted is called as target pose; the process of adjusting the pose involves 5 degrees of freedom of the cabin in space, namely the X, Y, Z degrees of freedom, and two degrees of rotation about Y, Z are defined herein as the X direction being the same as the axis direction of the target pose.
The technical scheme of the invention comprises the following steps:
step 1: calibrating: the pose motion is carried out on a standard cabin body by utilizing a pose adjusting mechanism, the corresponding relation between the five-degree-of-freedom space pose (without the degree of freedom of rotation around an axis) and the characteristic parameters of end face projection is obtained through shooting by two cameras and image processing calibration, and the different calibration matrixes of the two cameras are obtained;
the end projection is an ellipse in the image, which is calculated as 5 relatively independent parameters. And (5) obtaining the linear corresponding relation between the elliptic parameter and the degree of freedom of the standard cabin 5 through calibration.
Step 2: and replacing the standard cabin with the cabin to be regulated, and performing pose movement by using a pose regulating mechanism to obtain characteristic parameters of the cabin to be regulated under the shooting of the two cameras.
Step 3: reversely solving the approximate relative pose and the actual pose;
step 4: according to the actual pose, the pose adjusting mechanism is utilized to adjust the pose of the cabin body to be adjusted in real time:
step 5: and (3) continuously repeating the steps (2) to (4) to perform real-time pose recognition and dynamic adjustment on the cabin body to be adjusted, continuously collecting images by a camera, continuously repeating the steps at smaller intervals, and continuously performing dynamic adjustment in a servo mode until the error is within a threshold range.
The cabin body is a workpiece of the revolving body cabin body.
The step 1 specifically comprises the following steps:
step 1.1: fixing a standard cabin on a pose adjusting platform, and adjusting the standard cabin to a target pose;
the standard cabin body is consistent with or basically similar to the appearance characteristics of the cabin body to be adjusted.
Step 1.2: starting a pose adjustment platform to drive the standard cabin to move, and shooting the standard cabin from two different visual angles in real time by using two cameras to obtain a cabin image of the standard cabin; neither of the two different viewing angles is parallel to the axial direction of the standard capsule nor to the radial direction of the standard capsule.
The image picture shot by the camera contains the end face of the complete standard cabin.
The subsequent calibration steps are basically identical in the two camera calibration methods, taking a picture captured by a camera A as an example.
Step 1.3: for each cabin image shot by each camera, an image coordinate system is established, and five characteristic parameters under the image coordinate system are acquired according to the following modes:
most of the end face circles are elliptical under different visual angles, an ellipse of the end face in the cabin image is extracted by using a Hough ellipse extraction method to serve as the end face of a standard cabin, and the lengths a and b of the long and short axes of the ellipse are recorded;
identifying a center point O of an ellipse, fitting the center point positions of all end-face ellipses in the moving process of a standard cabin to form a straight line m in an image as a central axis of the standard cabin, recording an included angle phi from the long axis of the ellipse to the straight line m in the image of the cabin, the length d of a perpendicular line from the center point O of the ellipse to the straight line m, and the foot-hanging distance l between the foot-hanging distance between the center point O of the ellipse to the straight line m and the center point O;
the image coordinate system is a two-dimensional coordinate system established by taking the center of an image as an origin, taking the horizontal direction of the image as an x axis and taking the vertical direction of the image as a y axis.
In specific implementation, pose parameters of ellipses in the image are defined, and the ellipses are expressed by five characteristic parameters in the image. Drawing a perpendicular to a straight line m from the central point of the ellipse, recording the length d of the perpendicular, and taking the negative number of the length d of the perpendicular if the central point is below the straight line m; the distance l between the foot drop and the O point, the length a of the major axis of the ellipse, the length b of the minor axis of the ellipse, and the included angle phi between the major axis of the ellipse and the straight line.
Step 1.4: the standard cabin is adjusted to return to the target pose, a world coordinate system is established, after the world coordinate system moves to the X direction for delta X distance, five characteristic parameters in the cabin image at the moment are recorded in the same way according to the step 1.3.1l x ,d x , a x , b x , φ x ，l x ,d x , a x , b x , φ x Respectively representing a first foot drop distance, a first vertical line length, a first ellipse major axis length, a first ellipse minor axis length and a first included angle;
returning the standard cabin to the target pose, moving the standard cabin to the Y direction by delta Y distance under the world coordinate system, and shooting by a camera and recording 5 characteristic parameters in the cabin image at the momentl y ,d y , a y , b y , φ y ，l y ,d y , a y , b y , φ y Respectively representing a second foot drop distance, a second vertical line length, a second ellipse major axis length, a second ellipse minor axis length and a second included angle;
returning the standard cabin to the target pose, moving the standard cabin to the Z direction by delta Z distance under the world coordinate system, and shooting by a camera and recording 5 characteristic parameters in the cabin image at the momentl z ,d z , a z , b z , φ z ，l z ,d z , a z , b z , φ z Respectively representing a third foot drop distance, a third vertical line length, a third ellipse major axis length, a third ellipse minor axis length and a third included angle;
returning the standard cabin to the target pose, rotating the standard cabin by delta beta angle with Y axis under the world coordinate system, and shooting by a camera and recording 5 characteristic parameters in the cabin image at the momentl β ,d β , a β , b β , φ β ，l β ,d β , a β , b β , φ β Respectively representing a fourth foot drop distance, a fourth vertical line length, a fourth ellipse major axis length, a fourth ellipse minor axis length and a fourth included angle;
returning the standard cabin to the target pose, rotating the standard cabin by delta gamma angle with Z axis under the world coordinate system, and shooting by a camera and recording 5 characteristic parameters in the cabin image at the momentl γ ,d γ , a γ , b γ , φ γ ，l γ ,d γ , a γ , b γ , φ γ Respectively representing a fifth foot drop distance, a fifth vertical line length, a fifth ellipse major axis length, a fifth ellipse minor axis length and a fifth included angle;
the world coordinate system is a three-dimensional coordinate system which is established by taking the center of an end face close to a camera in a cabin body of a target pose as an origin, taking the axial direction of the cabin body as the X direction and taking the vertical direction of gravity as the Y direction.
Step 1.5: constructing a calibration matrix, namely constructing the following calibration matrix TA by the 25 parameters:
wherein ,a 0 representing the long axis of the standard cabin under the image coordinate system when the object is in position,b 0 representing the short axis of the standard cabin in the image coordinate system when the object is in pose.
The calibration matrix TA is a calibration matrix of the camera a, and the camera B can be calibrated similarly to obtain a calibration matrix TB of the camera B.
The calibration of the camera A is completed, a calibration matrix of the camera A is obtained, and the space pose and the image feature are linearly corresponding by utilizing the calibration matrix according to the following formula.
[l’, d’, a’/a 0 , b’/b 0 ,φ’] T = TA×[Δx’, Δy’, Δy’, Δβ’, Δγ’] T
Wherein l ', d', a ', b',φ’respectively expressed as the foot hanging distance, the length of a vertical line, the length of an elliptic long axis, the length of an elliptic short axis and an included angle of the cabin body to be adjusted; deltax’Representing the distance delta of the cabin to be regulated moving in the X direction under the world coordinate systemy’Representing the distance delta of the cabin to be regulated moving in the Y direction under the world coordinate systemz’Representing a cabin to be adjustedDistance, delta, of movement in the Z direction under world coordinate systemβ’Representing the rotation angle delta of the cabin body to be regulated around the Y axis under the world coordinate systemγ’The angle of the cabin body to be adjusted rotating around the Z axis under the world coordinate system is represented;a 0 representing the long axis of the standard cabin under the image coordinate system when the object is in position,b 0 representing the short axis of the standard cabin in the image coordinate system when the object is in pose.
In the step 2, two cameras are used to obtain the cabin image of the cabin to be adjusted, the ellipse of the end face is extracted by using Hough ellipse fitting in the same way as in the step 1, 5 characteristic parameters of the cabin to be adjusted at the moment are obtained by processing in the cabin image captured by each camera, and the characteristic parameters are the camera Al A ,d A , a A , b A , φ A In B camera isl B ,d B , a B , b B , φ B 。
In the step 3, the relative pose between the linear estimation under the camera and the target pose is obtained by combining the characteristic parameters obtained by each camera in the step 2 with the calibration matrix obtained by the calibration in the step 1, the relative poses obtained by the two cameras are generally different, and the relative poses obtained by the two cameras are combined through a certain weight to obtain the actual pose of the cabin to be adjusted.
In the step 3, the actual pose of the cabin to be adjusted is obtained by processing in the following manner:
[Δx, Δy, Δy, Δβ, Δγ] T = k × TA -1 × [l A , d A , a A / a 0 , b A / b 0 ,φ A ] T +
(1-k) × TA -1 × [l B , d B , a B / a 0 , b B / b 0 ,φ B ] T
k= 0.5 × (1 - (b A /b 0 +b 0 /b A ) / (b A /b 0 +b 0 /b A +b B /b 0 +b 0 /b B ))+
0.5 × (1 - (a
A
/a
0
+a
0
/a
A
) / (a
A
/a
0
+a
0
/a
A
+a
B
/a
0
+a
0
/a
B
))
where k is the weight coefficient of the A camera, and the coefficient k is the weight, which is related to distance, i.ea/a 0 、b/b 0 In relation, the ratio is approximately close to 1, and the higher the credibility is, the larger the weight is;l A ,d A , a A , b A , φ A representing 5 characteristic parameters of the cabin to be adjusted obtained by shooting by the first camera A,l B ,d B , a B , b B , φ B representing 5 characteristic parameters of the cabin to be adjusted obtained by shooting by the second camera B,a 0 representing the standard cabin body in the target poseThe long axis in the image coordinate system,b 0 representing the short axis of the standard cabin in the image coordinate system when the object is in pose.
In the step 4, the pose adjusting mechanism is controlled to move according to the actual pose of the cabin to be adjusted, so that the cabin to be adjusted approaches to the target pose.
And step 4, controlling a pose adjusting mechanism according to the actual pose of the cabin to be adjusted to drive the cabin to move along the direction of enabling each parameter to be 0, and if the deltax obtained through calculation is a number larger than 0, moving towards the negative X direction.
In the step 5, until the error is within the threshold range, the error between the end ellipse in the captured cabin image and the ellipse corresponding to the target pose is within the threshold range.
The error refers to the error between each characteristic parameter obtained by shooting by the camera and the same characteristic parameter of the target pose, and then the square sum of the errors of all the characteristic parameters is calculated as the total error shot by each camera; the threshold range is that the total errors shot by the two cameras are smaller than a preset error threshold at the same time and the difference value of the two is smaller than a preset difference threshold.
The specific threshold can be freely set according to actual use conditions, and if the pose requirement is relatively high, the specific threshold can be set more strictly.
The beneficial effects of the invention are as follows:
1) The invention completes the estimation of the pose by a view mode, adjusts the pose by a visual servo mode, avoids the operation of large-scale pose calculation, can complete the pose calculation by only needing a small number of steps such as ellipse fitting, and has higher efficiency.
2) Compared with the traditional pose resolving mode for judging the adjusting result, the method and the device have the advantages that the image information is directly used, the precision is higher, and the robustness is stronger.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a schematic view of a camera and pod arrangement;
FIG. 3 is a diagram illustrating parameters;
FIG. 4 is an exemplary diagram of a cabin and camera arrangement;
fig. 5 is a diagram of a shooting example.
Detailed Description
The invention is further described below with reference to the drawings and examples. The flow chart of the invention is shown in figure 1.
The implementation process of the embodiment of the invention is as follows:
the example selects a cylindrical cabin; the process of adjusting the pose relates to 5 degrees of freedom of the cabin body in space, namely the degree of freedom in the X, Y, Z direction, and two degrees of rotation freedom taking Y, Z as an axis are defined, wherein in the cabin body with the target pose, the center of a circle of an end face close to a camera is taken as an origin, and the X direction is defined to be the same as the axis direction of a cylinder on the target pose.
Step 1: and (5) calibrating. The calibration aims at establishing the corresponding relation between the 5-degree-of-freedom space pose (free of rotation about an axis) of the cabin and the characteristics of the projection of the end face of the cabin in an image through the shooting of the two cameras and the image processing calibration so as to obtain calibration matrixes of the two cameras. The calibration specifically comprises the following steps.
Step 1.1: a standard cabin is fixed on the pose adjustment platform and is manually adjusted to the target pose.
Step 1.2: starting a pose adjustment platform to drive a standard cabin to move, and shooting the standard cabin from two different visual angles in real time by using two cameras to obtain a cabin image of the standard cabin as shown in fig. 2; neither of the two different viewing angles is parallel to the axial direction of the standard capsule nor to the radial direction of the standard capsule. It should be noted that in the camera frame, the end face of the complete standard cabin needs to be included. An example size diagram of a camera and pod arrangement is shown in fig. 4.
Step 1.3: the two camera calibration methods are basically identical in the subsequent calibration steps, so the step 1.3 to the step 1.6 take the image captured by the camera A as an example. The example uses Hough ellipse extraction method to extract ellipse of end face, and marks center point position of ellipse of end face as O point. And starting the pose adjusting platform to move forwards and backwards in the X direction, shooting by a camera in real time, and recording the position of the center point of the ellipse of the end surface to form a straight line in the image, wherein the straight line is marked as a straight line m as a black thick solid line in fig. 3 and 5.
Step 1.4: and defining pose parameters of ellipses in the image, wherein the ellipses are expressed by 5 characteristic parameters in the image. As shown in fig. 3, an O-point is extracted from the position of the center point of the ellipse, a perpendicular to the straight line m is drawn, the length of the perpendicular is denoted as d, and if the center point is below the straight line m, the number of m is negative; the distance between the foot drop and the O point is recorded as l; the length of the major axis of the ellipse is marked as a, and the length of the minor axis of the ellipse is marked as b; the included angle phi between the long axis of the ellipse and the straight line. As in fig. 3.
Step 1.5: returning the cabin to the target pose, moving the cabin to the X direction by delta X distance, and recording 5 characteristic parameters of the ellipse in the image at the momentl x ,d x , a x , b x , φ x The method comprises the steps of carrying out a first treatment on the surface of the Returning the cabin to the target pose, moving the cabin to the Y direction by delta Y distance, and recording 5 characteristic parameters of the ellipse in the image at the momentl y ,d y , a y , b y , φ y The method comprises the steps of carrying out a first treatment on the surface of the Returning the cabin to the target pose, moving the cabin to the Z direction by delta Z distance, and recording 5 characteristic parameters of the ellipse in the image at the momentl z ,d z , a z , b z , φ z The method comprises the steps of carrying out a first treatment on the surface of the Returning the cabin to the target pose, rotating the cabin by delta beta angle along the Y axis, and recording 5 characteristic parameters of ellipse in the image at the momentl β ,d β , a β , b β , φ β The method comprises the steps of carrying out a first treatment on the surface of the Returning the cabin to the target pose, rotating the cabin by delta gamma angle along Z axis, and recording 5 characteristic parameters of ellipse in the image at the momentl γ ,d γ , a γ , b γ , φ γ 。
Step 1.6: constructing a calibration matrix, and constructing a matrix TA by the 25 parameters.
The calibration matrix TA is as follows:
wherein ,a 0 representing the long axis of the standard cabin under the image coordinate system when the object is in position,b 0 representing the short axis of the standard cabin in the image coordinate system when the object is in pose.
The calibration of the camera A is completed, and the space pose and the image feature are linearly corresponding.
[l, d, a/a 0 , b/b 0 ,φ] T = TA× [Δx, Δy, Δy, Δβ, Δγ] T
The camera B can be calibrated as such, resulting in a calibration matrix TB. The calibration is completed.
Step 2: and calculating characteristic parameters of the cabin to be adjusted. Taking a first cabin workpiece as a standard cabin, and taking the rest rear cabin workpieces as cabins to be adjusted. And acquiring cabin images by using two cameras, extracting end face information of the cabin images, and fitting by using Hough ellipses to obtain end face ellipses. The 5 feature parameters at this time are calculated at each camera captured image, in the A cameral A ,d A , a A , b A , φ A In B camera isl B ,d B , a B , b B , φ B . FIG. 5 shows a frame obtained by the camera A, wherein the parameter values are respectivelyl A =189,d A =78, a A =905, b A =709, φ A =14°
Step 3: and reversely solving the approximate relative pose. And obtaining the relative pose between the linear estimation and the target pose according to the matrix obtained by calibration. The relative poses of the two cameras are usually different and are combined by a certain weight.
[Δx, Δy, Δy, Δβ, Δγ] T = k × TA -1 × [l A , d A , a A /a 0 , b A /b 0 ,φ A ] T +
(1-k) × TA -1 × [l B , d B , a B /a 0 , b B /b 0 ,φ B ] T = [24, 6, 15, 20, 0.1] T
Wherein the coefficient k is the weight coefficient of the A camera, which is related to the distance, namely a/a 0 、b/b 0 In relation, the ratio is approximately 1, and the higher the reliability, the greater the weight. Specifically, k is calculated as follows:
k= 0.5 × (1 - (b A /b 0 +b 0 /b A ) / (b A /b 0 +b 0 /b A +b B /b 0 +b 0 /b B ))+
0.5 × (1 - (a A /a 0 +a 0 /a A ) / (a A /a 0 +a 0 /a A +a B /a 0 +a 0 /a B )) = 0.51
step 4: the pose adjusting mechanism adjusts the pose. According to the calculated pose of the cabin, the adjusting mechanism moves according to the pose to enable the cabin to approach the target pose.
Specifically, according to the actual pose of the cabin body to be adjusted, the pose adjusting mechanism is controlled to drive the cabin body to move along the direction of enabling each parameter to be 0, and if deltax obtained through calculation is a number larger than 0, the cabin body moves towards the negative direction of X.
Step 5: the camera continuously collects images, the steps 2-4 are continuously repeated at smaller intervals, and dynamic adjustment is continuously carried out in a servo mode. Until the error of the ellipse in the captured image corresponding to the target pose is within a certain threshold range. The threshold value can be freely set according to actual use conditions, and if the pose requirement is relatively high, the threshold value can be set more strictly.
In particular, the error for an angle is the difference between the phase value of the angle and 90 degrees in pixels multiplied by the long axis as the angle error.
Claims (9)
1. The method for adjusting the body pose of the double-view cabin is characterized by comprising the following steps of:
step 1: calibrating: the pose motion is carried out on a standard cabin body, the corresponding relation between the five-degree-of-freedom space pose and the characteristic parameters is obtained through shooting by two cameras and image processing calibration, and the different calibration matrixes of the two cameras are obtained;
step 2: the method comprises the steps that a to-be-adjusted cabin body is used for replacing a standard cabin body to perform pose movement, and characteristic parameters of the to-be-adjusted cabin body under shooting of two cameras are obtained;
step 3: reversely solving the approximate relative pose and the actual pose;
the actual pose of the cabin to be adjusted is obtained by processing in the following manner:
[Δx, Δy, Δy, Δβ, Δγ] T = k × TA -1 × [l A , d A , a A / a 0 , b A / b 0 , φ A ] T +
(1-k) × TA -1 × [l B , d B , a B / a 0 , b B / b 0 , φ B ] T
k = 0.5 × (1 - (b A /b 0 +b 0 /b A ) / (b A /b 0 +b 0 /b A +b B /b 0 +b 0 /b B ))+
0.5 × (1 - (a
A
/a
0
+a
0
/a
A
) / (a
A
/a
0
+a
0
/a
A
+a
B
/a
0
+a
0
/a
B
))
wherein k is the weight coefficient of the A camera;l A , d A , a A , b A , φ A representing 5 characteristic parameters of the cabin to be adjusted obtained by shooting by the first camera A,l B , d B , a B , b B , φ B representing the cabin to be adjusted obtained by shooting by the second camera BThe number of the characteristic parameters of the method is 5,a 0 representing the long axis of the standard cabin under the image coordinate system when the object is in position,b 0 representing the short axis of the standard cabin body under an image coordinate system when the standard cabin body is in the target pose;
step 4: according to the actual pose, carrying out real-time pose adjustment on the cabin body to be adjusted:
step 5: and (3) continuously repeating the steps (2) to (4) to identify the position and the posture of the cabin to be regulated in real time and dynamically regulate the position and the posture until the error is within a threshold range.
2. The method for adjusting the pose of a dual-view cabin according to claim 1, wherein,
the cabin body is a workpiece of the revolving body cabin body.
3. The method for adjusting the pose of a dual-view cabin according to claim 1, wherein,
the step 1 specifically comprises the following steps:
step 1.1: fixing a standard cabin on a pose adjusting platform, and adjusting the standard cabin to a target pose;
step 1.2: starting a pose adjustment platform to drive the standard cabin to move, and shooting the standard cabin from two different visual angles in real time by using two cameras to obtain a cabin image of the standard cabin;
step 1.3: for each camera-captured cabin image, five characteristic parameters under the image coordinate system are collected as follows:
extracting ellipses of the end surfaces in the cabin image by using a Hough ellipse extraction method to serve as the end surfaces of the standard cabin, and recording the lengths a and b of the long and short axes of the ellipses;
identifying a center point O of an ellipse, fitting the center point positions of all end-face ellipses in the moving process of a standard cabin to form a straight line m in an image as the central axis of the standard cabin, and recording an included angle phi from the long axis of the ellipse to the straight line m, the length d of a perpendicular line from the center point O of the ellipse to the straight line m and the foot hanging distance l between the foot hanging of the center point O of the ellipse to the straight line m and the center point O in the cabin image;
step 1.4: the standard cabin is adjusted to return to the target pose, and five characteristic parameters in the cabin image at the moment are recorded in the same way as the step 1.3.1 after the standard cabin is moved to the X direction by delta X distance under the world coordinate systeml x , d x , a x , b x , φ x ，l x , d x , a x , b x , φ x Respectively representing a first foot drop distance, a first vertical line length, a first ellipse major axis length, a first ellipse minor axis length and a first included angle;
returning the standard cabin to the target pose, moving the standard cabin to the Y direction by delta Y distance under the world coordinate system, and shooting by a camera and recording 5 characteristic parameters in the cabin image at the momentl y , d y , a y , b y , φ y ；
Returning the standard cabin to the target pose, moving the standard cabin to the Z direction by delta Z distance under the world coordinate system, and shooting by a camera and recording 5 characteristic parameters in the cabin image at the momentl z , d z , a z , b z , φ z ；
Returning the standard cabin to the target pose, rotating the standard cabin by delta beta angle with Y axis under the world coordinate system, and shooting by a camera and recording 5 characteristic parameters in the cabin image at the momentl β , d β , a β , b β , φ β ；
Returning the standard cabin to the target pose, rotating the standard cabin by delta gamma angle with Z axis under the world coordinate system, and shooting by a camera and recording 5 characteristic parameters in the cabin image at the momentl γ , d γ , a γ , b γ , φ γ ；
Step 1.5: constructing a calibration matrix, namely constructing the following calibration matrix TA by the 25 parameters:
wherein ,a 0 representing the long axis of the standard cabin under the image coordinate system when the object is in position,b 0 representing the short axis of the standard cabin in the image coordinate system when the object is in pose.
4. The method for adjusting the pose of a dual-view cabin according to claim 1, wherein,
in the step 2, two cameras are used to obtain the cabin image of the cabin to be adjusted, and 5 characteristic parameters of the cabin to be adjusted are obtained through processing in the same manner in the step 1.
5. The method for adjusting the pose of a dual-view cabin according to claim 1, wherein,
in the step 3, the characteristic parameters obtained by each camera in the step 2 are combined with the calibration matrix obtained by the calibration in the step 1 to obtain the relative pose of the linear estimation and the target pose, and the relative poses obtained by the two cameras are combined through a certain weight to obtain the actual pose of the cabin body to be adjusted.
6. The method for adjusting the pose of a dual-view cabin according to claim 1, wherein,
in the step 4, the pose adjusting mechanism is controlled to move according to the actual pose of the cabin to be adjusted, so that the cabin to be adjusted approaches to the target pose.
7. The method for adjusting the pose of a dual-view cabin according to claim 1, wherein,
and step 4, controlling a pose adjusting mechanism according to the actual pose of the cabin to be adjusted to drive the cabin to be adjusted to move along the direction of enabling each parameter to be 0.
8. The method for adjusting the pose of a dual-view cabin according to claim 1, wherein,
in the step 5, until the error is within the threshold range, the error between the end ellipse in the captured cabin image and the ellipse corresponding to the target pose is within the threshold range.
9. The method for adjusting the pose of a dual-view cabin according to claim 1, wherein,
the error refers to the error between each characteristic parameter obtained by shooting by the camera and the same characteristic parameter of the target pose for each camera, and then the square sum of the errors of all the characteristic parameters is calculated as the total error shot by the camera; the threshold range is that the total errors shot by the two cameras are smaller than a preset error threshold at the same time and the difference value of the two is smaller than a preset difference threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311026182.5A CN116740183B (en) | 2023-08-15 | 2023-08-15 | Double-view cabin pose adjusting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311026182.5A CN116740183B (en) | 2023-08-15 | 2023-08-15 | Double-view cabin pose adjusting method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116740183A true CN116740183A (en) | 2023-09-12 |
| CN116740183B CN116740183B (en) | 2023-11-07 |
Family
ID=87911888
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311026182.5A Active CN116740183B (en) | 2023-08-15 | 2023-08-15 | Double-view cabin pose adjusting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116740183B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117102856A (en) * | 2023-10-23 | 2023-11-24 | 浙江大学 | Large cabin double-platform five-degree-of-freedom pose recognition and adjustment method |
| CN117589064A (en) * | 2024-01-19 | 2024-02-23 | 中理检验有限公司 | Automatic wood measuring method and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006350553A (en) * | 2005-06-14 | 2006-12-28 | Kyoto Univ | Corresponding point retrieval method, mutual location method, three-dimensional image measurement method, corresponding point retrieval device, mutual location device, three-dimensional image measurement device, corresponding point retrieval program and computer-readable recording medium with its program recorded |
| CN107292927A (en) * | 2017-06-13 | 2017-10-24 | 厦门大学 | A kind of symmetric motion platform's position and pose measuring method based on binocular vision |
| CN108416791A (en) * | 2018-03-01 | 2018-08-17 | 燕山大学 | A kind of monitoring of parallel institution moving platform pose and tracking based on binocular vision |
| CN112833786A (en) * | 2021-01-05 | 2021-05-25 | 西安电子科技大学 | Cabin attitude and pose measurement and alignment system, control method and application |
| CN113751981A (en) * | 2021-08-19 | 2021-12-07 | 哈尔滨工业大学（深圳） | Space high-precision assembling method and system based on binocular vision servo |
| CN113781576A (en) * | 2021-09-03 | 2021-12-10 | 北京理工大学 | Binocular vision detection system, method and device for multi-degree-of-freedom pose real-time adjustment |
| CN115674171A (en) * | 2021-07-30 | 2023-02-03 | 中国科学院微电子研究所 | Robot pose measuring and compensating method and system, control device and storage medium |
| WO2023060964A1 (en) * | 2021-10-14 | 2023-04-20 | 上海商汤智能科技有限公司 | Calibration method and related apparatus, device, storage medium and computer program product |
| WO2023141963A1 (en) * | 2022-01-28 | 2023-08-03 | 深圳市大疆创新科技有限公司 | Pose estimation method for movable platform, movable platform, and storage medium |
-
2023
- 2023-08-15 CN CN202311026182.5A patent/CN116740183B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006350553A (en) * | 2005-06-14 | 2006-12-28 | Kyoto Univ | Corresponding point retrieval method, mutual location method, three-dimensional image measurement method, corresponding point retrieval device, mutual location device, three-dimensional image measurement device, corresponding point retrieval program and computer-readable recording medium with its program recorded |
| CN107292927A (en) * | 2017-06-13 | 2017-10-24 | 厦门大学 | A kind of symmetric motion platform's position and pose measuring method based on binocular vision |
| CN108416791A (en) * | 2018-03-01 | 2018-08-17 | 燕山大学 | A kind of monitoring of parallel institution moving platform pose and tracking based on binocular vision |
| CN112833786A (en) * | 2021-01-05 | 2021-05-25 | 西安电子科技大学 | Cabin attitude and pose measurement and alignment system, control method and application |
| CN115674171A (en) * | 2021-07-30 | 2023-02-03 | 中国科学院微电子研究所 | Robot pose measuring and compensating method and system, control device and storage medium |
| CN113751981A (en) * | 2021-08-19 | 2021-12-07 | 哈尔滨工业大学（深圳） | Space high-precision assembling method and system based on binocular vision servo |
| CN113781576A (en) * | 2021-09-03 | 2021-12-10 | 北京理工大学 | Binocular vision detection system, method and device for multi-degree-of-freedom pose real-time adjustment |
| WO2023060964A1 (en) * | 2021-10-14 | 2023-04-20 | 上海商汤智能科技有限公司 | Calibration method and related apparatus, device, storage medium and computer program product |
| WO2023141963A1 (en) * | 2022-01-28 | 2023-08-03 | 深圳市大疆创新科技有限公司 | Pose estimation method for movable platform, movable platform, and storage medium |
Non-Patent Citations (4)
| Title |
|---|
| CAI MENG等: "Satellite Pose Estimation via Single Perspective Circle and Line", IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS * |
| ZAIXING HE等: "Sparse Template-Based 6-D Pose Estimation of Metal Parts Using a Monocular Camera", IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS * |
| 陈冠宇;成群林;张解语;洪海波;何军;: "基于多传感器测量的航天器舱段自动对接位姿调整方法", 北京航空航天大学学报, no. 06 * |
| 黄华等: "基于LSTM与牛顿迭代的两轴系统轮廓误差控制", 浙江大学学报(工学版) * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117102856A (en) * | 2023-10-23 | 2023-11-24 | 浙江大学 | Large cabin double-platform five-degree-of-freedom pose recognition and adjustment method |
| CN117102856B (en) * | 2023-10-23 | 2024-02-13 | 浙江大学 | Large cabin double-platform five-degree-of-freedom pose recognition and adjustment method |
| CN117589064A (en) * | 2024-01-19 | 2024-02-23 | 中理检验有限公司 | Automatic wood measuring method and system |
| CN117589064B (en) * | 2024-01-19 | 2024-03-29 | 中理检验有限公司 | Automatic wood measuring method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116740183B (en) | 2023-11-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116740183B (en) | Double-view cabin pose adjusting method | |
| CN109308693B (en) | Single-binocular vision system for target detection and pose measurement constructed by one PTZ camera | |
| CN107471218B (en) | Binocular vision-based hand-eye coordination method for double-arm robot | |
| CN109102525B (en) | Mobile robot following control method based on self-adaptive posture estimation | |
| CN111801198B (en) | Hand-eye calibration method, system and computer storage medium | |
| CN109064516B (en) | Camera self-calibration method based on absolute quadratic curve image | |
| CN112669385B (en) | Industrial robot part identification and pose estimation method based on three-dimensional point cloud features | |
| CN110060304B (en) | Method for acquiring three-dimensional information of organism | |
| CN111897349A (en) | Underwater robot autonomous obstacle avoidance method based on binocular vision | |
| CN113910219A (en) | Exercise arm system and control method | |
| CN112419429B (en) | Large-scale workpiece surface defect detection calibration method based on multiple viewing angles | |
| CN112847334A (en) | Mechanical arm target tracking method based on visual servo | |
| CN107292956B (en) | Scene reconstruction method based on Manhattan hypothesis | |
| CN108154536A (en) | The camera calibration method of two dimensional surface iteration | |
| CN110706285A (en) | Object pose prediction method based on CAD model | |
| CN111993422B (en) | Robot axis and hole alignment control method based on uncalibrated vision | |
| CN113658266A (en) | Moving axis rotation angle visual measurement method based on fixed camera and single target | |
| CN106996769B (en) | Active pose rapid relocation method without camera calibration | |
| CN116766194A (en) | Binocular vision-based disc workpiece positioning and grabbing system and method | |
| CN113172632A (en) | Simplified robot vision servo control method based on images | |
| CN111523547A (en) | 3D semantic segmentation method and terminal | |
| CN109785393B (en) | Camera self-calibration method based on plane motion constraint | |
| CN113240751B (en) | Calibration method for robot tail end camera | |
| Seo et al. | 3D Hole center and surface normal estimation in robot vision systems | |
| CN111696158B (en) | Affine matching point pair-based multi-camera system relative pose estimation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |