Optical Positioning Method for Autonomous Underwater Vehicles Based on the Principle of Cone Projection

Abstract

This invention relates to the field of autonomous underwater vehicle (AUV) positioning, specifically based on the principle of cone projection. It addresses the problems of complex and time-consuming positioning processes in existing monocular vision-based target positioning methods. The invention utilizes acoustic signals detected by an acoustic guidance device to navigate to the recovery range of a positioning and recovery device. The AUV's bow is aligned with the cage opening of the device, and the central axes of both the AUV and the device are coplanar and parallel to the horizontal plane. Image information from a guide light source is then acquired, and the yaw angle of the AUV is calculated based on the principle of cone projection. This invention is primarily applied to the autonomous recovery of AUVs.

Description

Technical Field

[0001] This invention relates to the field of autonomous underwater vehicle (AUV) positioning. Background Technology

[0002] Autonomous underwater vehicles (AUVs) have wide applications in marine geosciences and are increasingly being used in scientific research, commerce, and military fields. As a reasonable requirement for unmanned equipment, AUVs need to possess autonomous recovery capabilities. Improving the accuracy and stability of autonomous recovery is crucial for extending the total operation time of AUVs. For the autonomous recovery process of an AUV, a positioning and recovery device is necessary. Given the distance and yaw angle of the AUV relative to the positioning and recovery device, the AUV autonomously recovers to the recovery device. In monocular vision-based target localization methods, for cases where target points are on the same plane, the Perspective-n-Point (PNP) method is typically used to obtain pose information and further obtain the yaw angle. PNP is a method for solving the motion of 3D to 2D point pairs, aiming to solve the pose of the camera coordinate system relative to the world coordinate system. It describes how to estimate the camera pose when the coordinates (relative to the world coordinate system) and pixel coordinates of n 3D points are known, i.e., solving for the rotation matrix R and translation vector t from the world coordinate system to the camera coordinate system. The multiple solutions to the PNP problem result in large computational loads and long computation times. Furthermore, the need to use the camera's intrinsic parameter matrix for camera calibration during the calculation process leads to a complex and time-consuming yaw angle localization process, resulting in high computational costs and a significant waste of computational resources. Existing monocular vision-based target localization methods are not ideal for achieving optical guidance. Therefore, the problems of complex and time-consuming localization processes in existing optical localization methods urgently need to be addressed. Summary of the Invention

[0003] The purpose of this invention is to solve the problems of complex and time-consuming target localization processes in existing monocular vision-based methods. This invention provides an optical localization method for autonomous underwater vehicles based on the principle of cone projection.

[0004] An optical positioning method for autonomous underwater vehicles (AUVs) based on the principle of cone projection is proposed. This method is implemented using an AUV and a positioning and recovery device. The AUV is equipped with an acoustic guidance device and an image acquisition device. The positioning and recovery device has four guide light sources and one acoustic beacon on its cage opening. The first to fourth guide light sources are arranged counterclockwise along the circumferential edge of the cage opening, ensuring that the geometric shape formed by the four guide light sources is rectangular. The first coefficient of the relative positional relationship of the guide light sources is determined based on the horizontal side 'a' and the vertical side 'b' of the rectangle. Optical positioning methods include:

[0005] S1. The acoustic guidance device on the autonomous underwater vehicle determines the location of the positioning and recovery device based on the acoustic signals emitted by the detected acoustic beacon, and navigates to the recovery range of the positioning and recovery device and hovers. When hovering, the bow of the autonomous underwater vehicle faces the cage opening of the positioning and recovery device, the central axis of the autonomous underwater vehicle and the positioning and recovery device are coplanar, and the central axis is parallel to the horizontal plane.

[0006] S2. The image acquisition device on the autonomous underwater vehicle acquires image information from the guide light source;

[0007] S3. Based on the principle of cone projection, calculate the image information to obtain the position information of each guide light source in the image coordinate system, as well as the midpoint A of the line connecting the first guide light source I1 and the second guide light source I2, and the midpoint B of the line connecting the third guide light source I3 and the fourth guide light source I4; obtain the second coefficient k2 of the relative position relationship of the guide light sources according to the line AB and the line I1I2, and obtain the second coefficient k3 of the relative position relationship of the guide light sources according to the line AB and the line I3I4.

[0008] S4. Calculate the yaw angle of the autonomous underwater vehicle based on k1, k2, and k3. Achieve optical positioning of autonomous underwater vehicles;

[0009] Wherein, the yaw angle θ is the angle between the heading of the underwater vehicle and the central axis of the positioning and recovery device, and the positive direction is the direction from the cage opening of the positioning and recovery device to the central axis of its recovery chamber. When θ is less than 0, the underwater vehicle is determined to be to the left of the positioning and recovery device; when θ is greater than 0, the underwater vehicle is determined to be to the right of the positioning and recovery device.

[0010] As a preferred option

[0011] As a preferred option

[0012] Preferably, the retrieval range of the positioning and recovery device is between 1 meter and 10 meters between the autonomous underwater vehicle and the positioning and recovery device.

[0013] Preferably, the image acquisition device is a monocular camera.

[0014] As a preferred choice, a > b.

[0015] Preferably, the image acquisition device is located at the bow of the underwater vehicle and on its central axis.

[0016] The beneficial effects of this invention are:

[0017] This invention presents an optical positioning method for autonomous underwater vehicles based on the principle of cone projection. In calculating the yaw angle, it does not use the camera's intrinsic parameter matrix, thus omitting the calibration step of the image acquisition device and simplifying the computation process. Compared to other existing methods, this invention is simpler and unaffected by calibration errors. Furthermore, it reduces computational load, shortens processing time, and improves computational efficiency.

[0018] Therefore, the optical positioning method for autonomous underwater vehicles based on the principle of cone projection in this invention replaces the PNP series methods, thereby improving the calculation speed and reducing the computational cost while ensuring that the accuracy of yaw angle calculation is not reduced. Attached Figure Description

[0019] Figure 1 This is a diagram showing the positional relationship of an autonomous underwater vehicle relative to a positioning and recovery device during autonomous recovery.

[0020] Figure 2 This is a schematic diagram of the yaw relationship of an autonomous underwater vehicle relative to a positioning and recovery device.

[0021] Figure 3 It is a planar cone diagram of the guiding light source;

[0022] Figure 4 yes Figure 3 Top view;

[0023] Figure 5 This is a diagram illustrating the principle of yaw angle calculation.

[0024] Figure 6 These are images collected during a water tank test.

[0025] Figures 1 to 6 In the diagram, o′uv represents the image coordinate system with o′ as the origin, u and v are two mutually perpendicular axes of the image coordinate system, OXYZ represents the spatial rectangular coordinate system with O as the origin, X, Y and Z are three mutually perpendicular axes of the spatial rectangular coordinate system, w represents the heading of the autonomous underwater vehicle, x1 and y1 are the x and y coordinates of the first guide light source I1, x2 and y2 are the x and y coordinates of the second guide light source I2, x3 and y3 are the x and y coordinates of the third guide light source I3, x4 and y4 are the x and y coordinates of the fourth guide light source I4, x5 and y5 are the x and y coordinates of point A, and x6 and y6 are the x and y coordinates of point B. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0028] The positioning and recovery device selected for recovering the autonomous underwater vehicle (AUV) is a cage-type device, which employs a funnel-shaped cage opening and a cylindrical recovery chamber. One acoustic beacon is mounted on the cage opening, and four guide lights are arranged counter-clockwise around the circumference of the cage opening. Connecting the four guide lights sequentially forms a rectangle, where the distance between adjacent guide lights is fixed. The geometric center of this rectangle is defined as [insert geometric center here]. An image acquisition device is located at the bow of the AUV, on its central axis. Optical guidance-based positioning and recovery devices are as follows... Figure 1 As shown, the yaw relationship of the AUV relative to the positioning and recovery device is described in [reference]. Figure 2 The optical positioning method for autonomous underwater vehicles based on the principle of cone projection described in this embodiment includes:

[0029] The acoustic guidance device on the autonomous underwater vehicle (AUV) determines the location of the positioning and recovery device based on the acoustic signals emitted by the detected acoustic beacons, navigates to the recovery range of the positioning and recovery device, and hovers there. When hovering, the bow of the AUV faces the cage opening of the positioning and recovery device, the central axes of the AUV and the positioning and recovery device are coplanar, and the central axis is parallel to the horizontal plane. Specifically, the recovery range is the distance between the AUV and the positioning and recovery device, which is 1 to 10 meters.

[0030] The image acquisition device on the autonomous underwater vehicle acquires image information from the guiding light source; specifically, the image acquisition device can be implemented using a monocular camera; the image acquisition device can be set at the bow of the underwater vehicle and located on its central axis;

[0031] Based on the principle of cone projection, the image information is calculated to obtain the position information of each guide light source in the image coordinate system, as well as the midpoint A of the line connecting the first guide light source I1 and the second guide light source I2, and the midpoint B of the line connecting the third guide light source I3 and the fourth guide light source I4; the second coefficient k2 of the relative position relationship of the guide light sources is obtained according to the line AB and the line I1I2, and the second coefficient k3 of the relative position relationship of the guide light sources is obtained according to the line AB and the line I3I4.

[0032] Based on k1, k2, and k3, calculate the yaw angle θ of the autonomous underwater vehicle.

[0033] During recovery, the acoustic guidance system can provide the relative positional deviation between the AUV and the positioning and recovery device, but it cannot determine the orientation of the funnel-shaped cage opening. At this point, an image acquisition device can provide position and attitude information, working in conjunction with the acoustic guidance system to complete the AUV recovery. Let's define the angle between the underwater vehicle's heading and the centerline of the positioning and recovery device as the yaw angle θ, with a range of values... The central axis of the AUV relative positioning and recovery device can deviate to the left or right. Taking the direction from the cage opening of the positioning and recovery device towards the central axis of its recovery chamber as the positive direction, geometrically, the angle between the funnel-shaped cage opening end face where the guide light source is located and the imaging plane of the image acquisition device is θ. At this time, the cone of view of the guide light source plane is as follows: Figure 3 As shown.

[0034] Figure 3 In the diagram, I1, I2, I3, and I4 are the first to fourth guiding light sources, forming a rectangle within the end face of the positioning and retrieval device. Let the horizontal side of the rectangle be *a* and the vertical side be *b*. The top view of the guiding light source planar cone is shown below. Figure 3 As shown.

[0035] The angle between the plane containing the guiding light source and the visual cone section is θ. The horizontal side length of the four guiding light sources on the far-vision cone section is c, the vertical side length is d, and the distance between the lines connecting the midpoints of the two vertical sides is e. The distance between the lines connecting the midpoints of the two vertical sides on the near-vision cone section is f. (See also...) Figure 4 According to the principles of geometric optics, we can obtain:

[0036]

[0037] Since the myopic cone section is parallel to the hyperopic cone section, then:

[0038]

[0039] Substituting Formula 2 into Formula 1, we get...

[0040]

[0041] Where a and b are the distances between the light sources, and these two values ​​are fixed when the guide light source is installed on the positioning and retrieval device. d and e are the distances between two points on the farsighted cone section, which cannot be measured directly, but the distances between corresponding points on the imaging plane are proportional to the distances on the farsighted cone section, and therefore can be obtained from the images captured by the image acquisition device.

[0042] Image information from the guidance light sources is acquired through the image acquisition device on the autonomous underwater vehicle (AUV). During image acquisition, the bow of the AUV faces the cage opening of the positioning and recovery device, the central axes of the AUV and the positioning and recovery device are coplanar, and the central axes are parallel to the horizontal plane. Acoustic guidance ensures the light sources appear in the center of the image (the geometric center of the geometric shape formed by the AUV and the four guidance light sources is at the same height). The yaw angle calculation principle diagram is shown below. Figure 5 As shown. Based on the principle of cone projection, the image information is calculated, and the position information of each guiding light source under the image system, as well as the midpoint A of the line connecting the first guiding light source I1 and the second guiding light source I2, and the midpoint B of the line connecting the third guiding light source I3 and the fourth guiding light source I4, can all be obtained.

[0043] Let k1 be the ratio of the horizontal side length a to the vertical side length b of the rectangular light source inside the end face of the positioning and recycling device. Figure 3 On the mid-to-far conical section, the ratios of e to b and d are k2 and k3, respectively. Combined with... Figures 3 to 5 Based on the relationship between the view cone section and the imaging plane, we have

[0044]

[0045]

[0046]

[0047] in,

[0048]

[0049]

[0050]

[0051] Substituting formulas 4 through 6, which contain k1, k2, and k3, into formula 3, we have:

[0052]

[0053] Therefore, the angle between the AUV's centerline and the centerline of the positioning and recovery device, which is also the AUV's yaw angle, can be calculated as follows:

[0054]

[0055] The AUV can deflect to the left or right relative to the centerline of the positioning and recovery device. Taking the direction from the cage opening of the positioning and recovery device towards its recovery chamber's centerline as the positive direction, the yaw angle can be determined by the lines I1I2 and I3I4 in the image. The yaw angle θ is the angle between the underwater vehicle's heading and the centerline of the positioning and recovery device. When the centerlines of the autonomous underwater vehicle and the positioning and recovery device are coplanar, the underwater vehicle's heading is also coplanar with the centerline of the positioning and recovery device. When |I1I2| > |I3I4|, the AUV deflects to the right to the left of the positioning and recovery device, and θ < 0; when |I1I2| < |I3I4|, the AUV deflects to the left to the right of the positioning and recovery device, and θ > 0. Therefore, the formula for the yaw angle between the AUV and the centerline of the positioning and recovery device is:

[0056] .

[0057] In determining the yaw angle using the autonomous underwater vehicle optical positioning method based on the principle of frustum projection described in this invention, the camera's intrinsic parameter matrix is ​​not used, thus omitting the camera calibration step. Compared with other methods, the frustum projection method is simpler and not affected by calibration errors. It also reduces computational load, improves computational efficiency, and shortens recovery time.

[0058] An underwater low-light black-and-white camera was used to acquire images in the pool, with a resolution of 704×576. The underwater camera was positioned at different distances and angles relative to the end face of the positioning and recovery device. Figure 6 The images captured under these circumstances are shown as follows: Figure 6 As shown.

[0059] To evaluate the accuracy and speed of the cone projection method in this invention, the yaw angle was calculated from image sequences acquired at the same position but different angles using both the method of this invention and the PNP reprojection method. The measurement data and calculation results are shown in Table 1.

[0060] Table 1. Actual measurement data and calculation results of yaw angle.

[0061]

[0062] As shown in Table 1, the yaw angle calculation results of this invention are very close to those of the PNP reprojection method, with minimal error compared to actual measurements. This verifies the effectiveness of the frustum projection method, providing effective attitude information for the AUV. Furthermore, it is noted that when the yaw angle is small, the calculation results of both the present invention and the PNP reprojection method are relatively accurate and not significantly different. However, when the yaw angle is large, the calculation error gradually increases, and the present invention performs better in this case. This is because when the yaw angle is large, the shape of the guide light source on the imaging plane of the PNP method changes due to the influence of angle, light, and the underwater environment, leading to errors in the extracted light source coordinates. In contrast, this invention calculates the yaw angle based on the relationship between the distance between guide light sources in the image and the positioning and recovery device in the frustum section. The coordinate error of the light source on the imaging plane has a smaller impact, resulting in better robustness.

[0063] Table 2 shows the yaw angles and computation time calculated using the method of this invention and the PNP reprojection method for image sequences acquired at different distances and angles. The average difference in yaw angle calculation results between the two methods is 0.13 degrees, indicating that the method of this invention can replace the PNP reprojection method for calculating the yaw angle of AUVs. Furthermore, the average speed of yaw angle calculation based on the principle of frustum projection in this invention is more than 100 times that of the PNP reprojection method. Although different equipment may affect the computation time, the computation time and computation cost of this invention are far less than those of the PNP reprojection method, demonstrating excellent speed.

[0064] Table 2. Yaw angle calculation results of the present invention method and the PNP method

[0065]

[0066] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. An optical positioning method for autonomous underwater vehicles based on the principle of cone projection, characterized in that, This method is based on an autonomous underwater vehicle (AUV) and a positioning and recovery device. The AUV is equipped with an acoustic guidance device and an image acquisition device. The positioning and recovery device has four guide light sources and one acoustic beacon on its cage opening. The first to fourth guide light sources are arranged counterclockwise along the circumferential edge of the cage opening, ensuring that the geometric shape formed by the four guide light sources is rectangular. The method is based on the horizontal side of the rectangle... and longitudinal side The first coefficient determines the relative positional relationship of the guiding light source. Optical positioning methods include: S1. The acoustic guidance device on the autonomous underwater vehicle determines the location of the positioning and recovery device based on the acoustic signals emitted by the detected acoustic beacon, and navigates to the recovery range of the positioning and recovery device and hovers. When hovering, the bow of the autonomous underwater vehicle faces the cage opening of the positioning and recovery device, the central axis of the autonomous underwater vehicle and the positioning and recovery device are coplanar, and the central axis is parallel to the horizontal plane. S2. The image acquisition device on the autonomous underwater vehicle acquires image information from the guide light source; S3. Based on the principle of cone projection, calculate the image information to obtain the position information of each guide light source in the image coordinate system, as well as the first guide light source. With the second guiding light source Midpoint of the line and the third guiding light source With the 4th guiding light source Midpoint of the line According to the straight line With a straight line Obtain the second coefficient of the relative positional relationship of the guiding light source. According to the straight line With a straight line Obtain the third coefficient of the relative positional relationship of the guiding light source. ; S4, according to , and Calculate the yaw angle of the autonomous underwater vehicle. To achieve optical positioning of autonomous underwater vehicles; Among them, yaw angle The angle between the underwater vehicle's heading and the central axis of the positioning and recovery device is defined as the positive direction, with the direction from the cage opening of the positioning and recovery device to the central axis of its recovery chamber. ,when If the value is less than 0, the underwater vehicle is determined to be to the left of the positioning and recovery device; when... If the value is greater than 0, the underwater vehicle is determined to be to the right of the positioning and recovery device.

2. The optical positioning method for autonomous underwater vehicles based on the principle of cone projection according to claim 1, characterized in that, The retrieval range of the positioning and recovery device is 1 to 10 meters between the autonomous underwater vehicle and the positioning and recovery device.

3. The optical positioning method for autonomous underwater vehicles based on the principle of cone projection according to claim 1, characterized in that, The image acquisition device is a monocular camera.

4. The optical positioning method for autonomous underwater vehicles based on the principle of cone projection according to claim 1, characterized in that, 。 5. The optical positioning method for autonomous underwater vehicles based on the principle of cone projection according to claim 1, characterized in that, The image acquisition device is located at the bow of the underwater vehicle and on its central axis.

Images