Vehicle-mounted self-adaptive multi-view vision pavement rut detection device

Abstract

The utility model relates to a vehicle-mounted self-adaptive multi-view vision pavement rut detection device, which comprises a base and a detection device arranged on the base, and the detection device comprises a first camera, a second camera and a third camera which are respectively arranged on a cradle head frame along the transverse direction of a vehicle body and are arranged at equal intervals, the first camera, the second camera and the third camera realize multi-angle three-dimensional coverage of the same rut area through combination of different depression angles by virtue of the corresponding holders; the base comprises a fixed seat and a sliding rail device, the fixed seat is used for detachably and fixedly connecting the base to a vehicle body, and the sliding rail device comprises a transverse sliding rail, a longitudinal sliding rail and a vertical sliding rail; the sliding rail device is installed on the fixing base and used for adjusting the position of the detection device, and a connecting base is arranged on the sliding rail device and used for being detachably and fixedly connected with the detection device. The track three-dimensional reconstruction system has the effects of high-precision track three-dimensional reconstruction, low cost and strong dynamic adaptability.

Description

Technical Field

[0001] This utility model relates to the field of road surface detection device technology, and in particular to a vehicle-mounted adaptive multi-view vision road rut detection device. Background Technology

[0002] Road surface rutting detection is a key technology for assessing road load-bearing capacity and service condition, directly impacting traffic safety and maintenance decisions. With my country's road mileage exceeding 5.3 million kilometers, traditional manual inspection and single-sensor detection technologies are insufficient to meet the demands of large-scale, automated inspection. There is an urgent need for intelligent, integrated equipment to achieve efficient and accurate detection.

[0003] Existing technologies for rut detection include manual measurement, two-dimensional vision, laser scanning, and binocular vision. However, although existing technologies have achieved preliminary rut detection through different principles, they still have the following drawbacks:

[0004] Firstly, manual inspection technology relies on manpower and is inefficient. The traditional 3m ruler measurement method requires manual point-by-point comparison, and the inspection of a single point takes more than 3 minutes.

[0005] Secondly, the lack of depth information in two-dimensional vision technology leads to insufficient detection capabilities. Although two-dimensional image detection schemes based on monocular cameras can identify surface defects such as road cracks, they can only acquire planar grayscale information and cannot reconstruct the three-dimensional contours of ruts through perspective projection.

[0006] Third, laser scanning technology is expensive and has poor environmental adaptability. The cost of laser point cloud detection equipment per kilometer far exceeds the budget for ordinary highway maintenance, and its high cost makes it difficult to promote on a large scale.

[0007] Fourth, binocular vision technology suffers from dual bottlenecks in accuracy and dynamic performance. Existing binocular vision solutions obtain parallax information through stereo matching of two cameras. Although they can recover some depth data, their baseline distance is usually less than 300mm, making it difficult to achieve centimeter-level accuracy in rut depth detection, which fails to meet industry standard requirements. Utility Model Content

[0008] To address the shortcomings of existing technologies, the purpose of this invention is to provide a vehicle-mounted adaptive multi-view vision road rut detection device, which features high-precision 3D rut reconstruction, low cost, and strong dynamic adaptability.

[0009] The above-mentioned utility model objective is achieved through the following technical solution:

[0010] A vehicle-mounted adaptive multi-view vision road rut detection device includes a base and a detection device mounted on the base.

[0011] The detection device includes a first camera, a second camera, and a third camera, which are respectively mounted on a gimbal frame and arranged at equal intervals along the transverse direction of the vehicle body. The first camera, the second camera, and the third camera achieve multi-angle three-dimensional coverage of the same rut area by using different combinations of downward angles through the corresponding gimbals.

[0012] The base includes a fixed seat and a slide rail device. The fixed seat is used to detachably and fixedly connect the base to the vehicle body. The slide rail device includes a transverse slide rail, a longitudinal slide rail, and a vertical slide rail. The slide rail device is installed on the fixed seat and is used to adjust the spatial position of the detection device. The slide rail device is provided with a connecting seat for detachably and fixedly connecting the detection device.

[0013] The above technical solution enables multi-angle three-dimensional coverage detection of rut areas in a vehicle environment. Through quick disassembly and assembly of the base, three-dimensional adjustment of the slide rail, and split gimbal design, it can adapt to different vehicle models and road conditions, solving the problems of single viewing angle and poor installation adaptability of traditional fixed detection devices.

[0014] As a further technical solution of this utility model: the first camera is set at a 90° downward angle in the middle of the gimbal, the second camera is set at a 40° downward angle on the left side of the gimbal, and the third camera is set at a 45° downward angle on the right side of the gimbal.

[0015] The above technical solution uses an asymmetrical combination of a 90° downward angle from the first camera and a 40° downward angle from the second camera, along with a 45° downward angle from the third camera (tilted and complementary). This combination enhances the ability to capture three-dimensional depth information while covering the entire rut. The slight difference between 40° and 45° avoids symmetrical occlusion and improves the imaging integrity of the concave area.

[0016] As a further technical solution of this utility model: the horizontal distance between the first camera and the second camera, and the horizontal distance between the first camera and the third camera are all 500mm, and the installation height of the first camera, the second camera and the third camera on the gimbal is all 1000mm.

[0017] Through the above technical solution, the 500mm equidistant layout forms the optimal field of view overlap rate at an installation height of 1m, balancing the accuracy of 3D reconstruction with the compactness of the equipment, and avoiding insufficient parallax caused by too small a spacing or excessive parallax caused by too large a spacing.

[0018] As a further technical solution of this utility model: the transverse slide rail is composed of an inner slide rail, a middle slide rail and an outer slide rail nested in sequence. The outer slide rail is fixedly connected to the fixed base. The top of the inner slide rail is provided with a first slide rail. The transverse slide rail is used to adjust the left and right position of the detection device along the width direction of the vehicle.

[0019] The longitudinal slide rail is slidably connected to the first slide rail, and the longitudinal slide rail is provided with an L-shaped connecting plate. The longitudinal slide rail is used to adjust the front and rear position of the detection device along the vehicle's travel direction.

[0020] The vertical slide rail is fixedly connected to the L-shaped connecting plate, and the vertical slide rail is used to adjust the vertical position of the detection device along the vehicle height direction.

[0021] The above technical solution, through three-level decoupling adjustment of "lateral-longitudinal-vertical", independently controls the displacement in each direction, solving the problem of full-degree-of-freedom precise positioning under limited vehicle space, while the nested lateral slide rail realizes flexible contraction range.

[0022] As a further technical solution of this utility model: the adjustment range of the vertical slide rail is 300-800mm.

[0023] The above technical solution covers the height difference between cars and off-road vehicles with a travel range of 300-800mm, ensuring that the camera is always at the optimal observation height and avoiding viewpoint obstruction or excessive distortion.

[0024] As a further technical solution of this utility model: the adjustment range of the transverse slide rail is 800-2500mm.

[0025] Through the above technical solution, the three nested sections of the transverse slide rail achieve a transverse movement of 800-2500mm, covering the full range of vehicle types from cars (with ruts in the center) to engineering vehicles (with ruts slightly off the outside), ensuring that the camera is always at the optimal observation height and avoiding viewpoint obstruction or excessive distortion.

[0026] As a further technical solution of this utility model: the transverse slide rail and the longitudinal slide rail are locked with the positioning hole by spring pins, and the vertical slide rail is equipped with anti-reverse pins.

[0027] Through the above technical solution, the transverse and longitudinal slide rails achieve millimeter-level repeatability (±0.5mm) through spring pins and positioning holes. The anti-reverse pin is inserted into the guide groove of the vertical slide rail to ensure that the height position of the detection device remains unchanged and to ensure the stability of the detection parameters during long-distance travel.

[0028] As a further technical solution of this utility model: a second slide rail is provided on the front end of the vertical slide rail and is slidably connected thereto, and the second slide rail is fixedly connected to the connecting seat.

[0029] Through the above technical solution, the second slide rail can realize the vertical positioning of the detection device, and the detection device is fixed by the connecting seat. The connecting seat and the second slide rail adopt a dovetail groove + quick-release buckle, so the detection device can be quickly installed and removed, which is convenient for maintenance or equipment replacement.

[0030] As a further technical solution of this utility model: the fixing base includes an electromagnetic adsorption module and a vacuum negative pressure adsorption module;

[0031] The electromagnetic adsorption module uses the magnetic force generated by energizing to adsorb the metal surface of the vehicle body.

[0032] The vacuum negative pressure adsorption module creates negative pressure by drawing a vacuum to adsorb non-metallic surfaces of the vehicle body.

[0033] The above technical solutions enable the electromagnetic adsorption module (with suction force ≥500N) to be used on steel roofs and the vacuum module (with negative pressure ≥80kPa) to be used on aluminum alloy / carbon fiber roofs, thus solving the installation compatibility problem for different vehicle models.

[0034] In summary, this utility model has at least one of the following beneficial technical effects:

[0035] 1. This utility model discloses a vehicle-mounted adaptive multi-view vision road rut detection device, which is equipped with a detection device including a first camera, a second camera and a third camera on a base. The three cameras cover the same rut area in different combinations of downward angles through corresponding pan-tilt units. At the same time, the position of the detection device is adjusted by a sliding rail device to realize multi-angle three-dimensional detection and position adaptive adjustment of the road rut area.

[0036] 2. This utility model discloses a vehicle-mounted adaptive multi-view vision road rut detection device, which achieves a detection range of at least 1.8m×1.2m by setting the horizontal distance between the first camera and the second camera, and between the first camera and the third camera to 500mm, and setting the installation height of the three cameras on the gimbal to 1000mm.

[0037] 3. This utility model discloses a vehicle-mounted adaptive multi-view vision road rut detection device. By setting the lateral slide rail adjustment range to 800-2500mm and the vertical slide rail adjustment range to 300-800mm, the detection device can achieve a wide range of position adjustment along the width and height directions of the vehicle to adapt to different vehicle models and road detection scenarios, and ensure that the size of the overlapping area meets the detection standards. Attached Figure Description

[0038] Figure 1 This is a front view of a vehicle-mounted adaptive multi-view vision road rut detection device according to the present invention.

[0039] Figure 2 This is a schematic diagram of the overall structure of the slide rail device in the vehicle-mounted adaptive multi-view vision road rut detection device of this utility model.

[0040] Reference numerals: 1. Base; 11. Fixed base; 12. Slide rail device; 121. Transverse slide rail; 1211. Inner slide rail; 1212. Middle slide rail; 1213. Outer slide rail; 122. Longitudinal slide rail; 123. Vertical slide rail; 2. Detection device; 3. Gimbal frame; 4. First camera; 5. Second camera; 6. Third camera; 7. Connecting base; 8. First slide rail; 9. L-shaped connecting plate; 10. Second slide rail; 13. Gimbal. Detailed Implementation

[0041] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0042] In the description of this application, it should be noted that the terms "upper," "lower," "inner," "outer," "top / bottom," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0043] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installed," "equipped with," "sleeved / connected," "connected," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0044] Example 1:

[0045] Reference Figure 1 and Figure 2 This utility model discloses a vehicle-mounted adaptive multi-view vision road rut detection device, which includes a base 1 and a detection device 2. The detection device 2 consists of a first camera 4, a second camera 5 and a third camera 6 arranged at equal intervals along the transverse side of the vehicle body. They are respectively installed on the independent gimbal 13 of the gimbal frame 3, and multi-angle coverage of the rut area is achieved by different combinations of downward angles.

[0046] The first camera 4 (left) has a downward angle of 40°±0.5°, the second camera 5 (middle) has a downward angle of 90°±0.2° (vertically downward), and the third camera 6 (right) has a downward angle of 45°±0.5°. The horizontal distance between any two adjacent cameras is 500mm. The installation height of the first camera 4, the second camera 5, and the third camera 6 on the gimbal 3 is 1000mm. The horizontal projection of the optical axes of the three cameras is parallel to the direction of travel. Any tire track point can be captured by the three cameras simultaneously. The downward angle and horizontal distance (500mm) of the three cameras ensure a lateral overlap width of 1.8m and a longitudinal overlap depth of 1.2m, with an overlap rate η≥55% (overlap area / field of view of a single camera).

[0047] The base 1 includes a fixed seat 11 and a slide rail device 12. The fixed seat 11 adopts a composite design of an electromagnetic adsorption module and a vacuum negative pressure adsorption module. The former adsorbs metal car bodies by electromagnetic force (adsorption force ≥ 500N), while the latter adsorbs non-metallic car bodies by vacuuming (negative pressure ≥ 80kPa), supporting quick installation and disassembly.

[0048] Reference Figure 2 The slide rail device 12 includes a three-way adjustment mechanism to achieve three-dimensional positioning of the detection device. The transverse slide rail 121 is composed of an inner slide rail 1211, a middle slide rail 1212, and an outer slide rail 1213 nested in sequence, with a retracted length of 800mm and a fully extended length of 2500mm, covering an adjustment range of 800-2500mm. The outer slide rail 1213 is fixedly connected to the fixed base 11. The top of the inner slide rail 1211 is provided with a first slide rail 8. The transverse slide rail 121 is used to adjust the left and right position of the detection device 2 along the width direction of the vehicle. The longitudinal slide rail 122 is slidably connected to the first slide rail 8 and is provided with an L-shaped connecting plate 9. The longitudinal slide rail 122 is used to adjust the front and rear position of the detection device along the driving direction of the vehicle. The vertical slide rail 123 is fixedly connected to the L-shaped connecting plate 9 and is used to adjust the up and down position of the detection device along the height direction of the vehicle.

[0049] The transverse slide rail 121 and the longitudinal slide rail 122 are locked to the positioning hole by spring pins with a tolerance fit of H7 / g6, ensuring automatic locking after extension into place, with a positioning accuracy of ±0.5mm. The vertical slide rail 123 is equipped with an anti-reverse pin, which is inserted into the guide groove of the vertical slide rail 123 to counteract the inertial force during vehicle braking / acceleration, ensuring that the height remains unchanged and guaranteeing the stability of the detection parameters during long-distance driving.

[0050] The highway rut detection process of a vehicle-mounted adaptive multi-view vision road rut detection device is as follows:

[0051] First, installation and calibration are performed. Adaptive parameters are adjusted for different vehicle models. For SUVs, the lateral slide rail is extended from 121 to 2200mm, and the vertical slide rail is adjusted from 123 to 700mm, so that the center of the camera group is aligned with the rut area.

[0052] Then, accuracy calibration is performed by projecting a 500mm spacing reference grid onto the ground using the detection device 2, adjusting the camera pitch angle to the target value (error ≤ 0.1°), and receiving real-time feedback from the tilt sensor.

[0053] Image acquisition and processing are performed. The exposure time of the first camera 4, the second camera 5, and the third camera 6 is less than 0.5ms. Acquisition is synchronously triggered by the vehicle controller to suppress high-speed motion blur (clear imaging is still achieved at a vehicle speed of 120km / h). Using epipolar constraints and feature point matching algorithms (such as SIFT+RANSAC), the correspondence between the rut points P in the three cameras (P1 / P2 / P3) is established. The three-dimensional coordinates are calculated by triangulation to reconstruct the rut depth and contour.

[0054] Finally, accuracy verification was conducted. On dry asphalt pavement: the measured rut depth was 5.0 mm, the measured value was 5.1 mm, and the error was 2.0%. On wet and reflective pavement: the measured depth was 8.3 mm, the measured value was 8.2 mm, and the error was 1.2%. In a high-speed scenario, at a vehicle speed of 120 km / h, the measured depth was 12.7 mm, the measured value was 12.4 mm, and the error was 2.4%, verifying the dynamic adaptability.

[0055] The implementation principle of this utility model is as follows: through the synergistic effect of four principles—multi-view spatial layout optimization (tilt angle + horizontal spacing + installation height), adaptive adjustment of three-dimensional slide rail (lateral + longitudinal + vertical), three-dimensional reconstruction of multi-view images, and stability guarantee of slide rail device 12 (composite fixing + slide rail locking)—high precision, high efficiency, and high compatibility (covering more than 95% of vehicle models) of rut detection are achieved, solving the core pain points of existing technologies in terms of efficiency, cost, and dynamic adaptability.

[0056] The embodiments described herein are preferred embodiments of this utility model and are not intended to limit the scope of protection of this utility model. Therefore, all equivalent changes made to the structure, shape, and principle of this utility model should be included within the scope of protection of this utility model.

Claims

1. A vehicle-mounted adaptive multi-view vision road rut detection device, characterized in that, Includes a base (1) and a detection device (2) mounted on the base (1). The detection device (2) includes a first camera (4), a second camera (5) and a third camera (6) arranged at equal intervals on the gimbal frame (3) along the transverse direction of the vehicle body. The first camera (4), the second camera (5) and the third camera (6) achieve multi-angle three-dimensional coverage of the same rut area by combining different downward angles through the corresponding gimbal (13). The base (1) includes a fixed seat (11) and a slide rail device (12). The fixed seat (11) is used to detachably fix the base (1) to the vehicle body. The slide rail device (12) includes a transverse slide rail (121), a longitudinal slide rail (122) and a vertical slide rail (123). The slide rail device (12) is installed on the fixed seat (11) and is used to adjust the spatial position of the detection device (2). The slide rail device (12) is provided with a connecting seat (7) for detachably fixing the detection device (2).

2. The vehicle-mounted adaptive multi-view vision road rut detection device according to claim 1, characterized in that, The first camera (4) is positioned at a 90° downward angle in the middle of the gimbal frame (3), the second camera (5) is positioned at a 40° downward angle on the left side of the gimbal frame (3), and the third camera (6) is positioned at a 45° downward angle on the right side of the gimbal frame (3).

3. The vehicle-mounted adaptive multi-view vision road rut detection device according to claim 1, characterized in that, The horizontal distance between the first camera (4) and the second camera (5), and the horizontal distance between the first camera (4) and the third camera (6) are both 500mm. The installation height of the first camera (4), the second camera (5) and the third camera (6) on the gimbal frame (3) is 1000mm.

4. The vehicle-mounted adaptive multi-view vision road rut detection device according to claim 1, characterized in that, The transverse slide rail (121) is composed of an inner slide rail (1211), a middle slide rail (1212), and an outer slide rail (1213) nested in sequence. The outer slide rail (1213) is fixedly connected to the fixed base (11). The top of the inner slide rail (1211) is provided with a first slide rail (8). The transverse slide rail (121) is used to adjust the left and right position of the detection device (2) along the width direction of the vehicle. The longitudinal slide rail (122) is slidably connected to the first slide rail (8). The longitudinal slide rail (122) is provided with an L-shaped connecting plate (9). The longitudinal slide rail (122) is used to adjust the front and rear position of the detection device (2) along the vehicle driving direction. The vertical slide rail (123) is fixedly connected to the L-shaped connecting plate (9), and the vertical slide rail (123) is used to adjust the vertical position of the detection device (2) along the vehicle height direction.

5. The vehicle-mounted adaptive multi-view vision road rut detection device according to claim 4, characterized in that, The vertical slide rail (123) has an adjustment range of 300-800mm.

6. The vehicle-mounted adaptive multi-view vision road rut detection device according to claim 4, characterized in that... The adjustment range of the transverse slide rail (121) is 800-2500mm.

7. The vehicle-mounted adaptive multi-view vision road rut detection device according to claim 4, characterized in that... The transverse slide rail (121) and the longitudinal slide rail (122) are locked to the positioning hole by spring pins, and the vertical slide rail (123) is equipped with an anti-reverse pin.

8. A vehicle-mounted adaptive multi-view vision road rut detection device according to claim 6, characterized in that... The vertical slide rail (123) is provided with a second slide rail (10) that is slidably connected to it, and the second slide rail (10) is fixedly connected to the connecting seat (7).

9. The vehicle-mounted adaptive multi-view vision road rut detection device according to claim 1, characterized in that, The fixed base (11) includes an electromagnetic adsorption module and a vacuum negative pressure adsorption module; The electromagnetic adsorption module uses the magnetic force generated by energizing to adsorb the metal surface of the vehicle body. The vacuum negative pressure adsorption module creates negative pressure by drawing a vacuum to adsorb non-metallic surfaces of the vehicle body.

Images