A method, device and working machine for detecting a position of a vehicle container

Abstract

The application provides a vehicle hopper pose detection method and device and a working machine. The vehicle hopper pose detection method obtains first point cloud data of a region where a target vehicle hopper is located, determines an initial detection box of the target vehicle hopper based on the first point cloud data, determines a current value of an environmental constraint height corresponding to the target vehicle hopper based on the initial detection box and the first point cloud data, and when it is determined that the current value of the environmental constraint height is valid, adjusts the initial detection box in height based on the current value of the environmental constraint height and a geometric constraint height value of the target vehicle hopper to obtain a current detection box of the target vehicle hopper, wherein the environmental constraint height is used to represent the height of a preset reference surface in the region where the target vehicle hopper is located, and the geometric constraint height value is used to represent the distance between a preset position of the target vehicle hopper and the preset reference surface, so that the current detection box of the target vehicle hopper can be effectively ensured to be close to the upper edge of the side plate of the target vehicle hopper, and the effective improvement of the vehicle hopper pose detection precision is realized.

Description

Technical Field

[0001] This application relates to the field of engineering machinery technology, specifically to the field of automatic sensing in engineering machinery technology, and more specifically to a method, device and working machinery for detecting the position and posture of a truck bed. Background Technology

[0002] Excavation and loading scenarios are one of the key application areas for unmanned and intelligent construction machinery. In this scenario, the ability to accurately detect the position and orientation of the truck bed is an essential prerequisite for achieving unmanned and intelligent excavation and loading.

[0003] During the excavation and loading operation, it is necessary to realize real-time obstacle avoidance and full load judgment of the truck bed. Therefore, in the process of position detection of the truck bed, it is necessary to ensure that the truck bed detection frame is close to the upper edge of the side plate of the truck bed, that is, the upper edge of the truck bed detection frame is flush with the upper edge of the side plate of the truck bed, which puts forward higher requirements for the position detection accuracy of the truck bed.

[0004] Therefore, it is necessary to provide a truck bed pose detection method so that the truck bed detection frame can be closely attached to the edge of the side plate of the truck bed. Summary of the Invention

[0005] To solve the above-mentioned technical problems, this application provides a truck bed position detection method, device and operating machinery to achieve the purpose of the truck bed detection frame being closely attached to the edge of the side plate of the truck bed.

[0006] To achieve the above technical objectives, the embodiments of this application provide the following technical solutions:

[0007] Firstly, the embodiments of this specification provide a method for detecting the position and orientation of a truck bed, including:

[0008] Acquire the first point cloud data of the area where the target truck bed is located, and determine the initial detection frame of the target truck bed based on the first point cloud data;

[0009] Based on the initial detection frame and the first point cloud data, the current value of the environmental constraint height corresponding to the target truck bed is determined; wherein, the environmental constraint height is used to characterize the height of a preset reference surface in the area where the target truck bed is located;

[0010] When the current value of the environmental constraint height is determined to be valid, the initial detection frame is height adjusted based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed, and the height-adjusted initial detection frame is used as the current detection frame of the target truck bed; wherein, the current detection frame is used to characterize the current pose of the target truck bed, and the geometric constraint height value is used to characterize the distance between the preset position of the target truck bed and the preset reference surface.

[0011] Optionally, determining the current value of the environmental constraint height corresponding to the target truck bed based on the initial detection box and the first point cloud data includes:

[0012] Based on the initial detection frame and the size information of the vehicle to which the target truck bed belongs, the current clipping frame corresponding to the preset reference surface is determined;

[0013] Based on the current cropping frame and the first point cloud data, the second point cloud data corresponding to the preset reference surface is obtained;

[0014] Based on the second point cloud data, the current value of the environmental constraint height is determined.

[0015] Optionally, determining the current clipping frame corresponding to the preset reference plane based on the initial detection frame and the size information of the vehicle to which the target truck bed belongs includes:

[0016] Based on the initial detection frame and the size information of the vehicle to which the target truck bed belongs, the position of the preset clipping frame is adjusted, and the adjusted preset clipping frame is used as the current clipping frame corresponding to the preset reference surface.

[0017] Optionally, determining the current value of the environmental constraint height based on the second point cloud data includes:

[0018] If the number of points in the second point cloud data is less than the preset number, then the initial value of the environmental constraint height or the historical value of the environmental constraint height at the previous moment shall be used as the current value of the environmental constraint height.

[0019] If the number of point clouds in the second point cloud data is greater than or equal to the preset number, then the current value of the environmental constraint height is determined based on the height information of each point cloud in the second point cloud data.

[0020] Optionally, determining whether the current value of the environmental constraint height is valid includes:

[0021] If the current value of the environmental constraint height is the initial value of the environmental constraint height, then the current value of the environmental constraint height is invalid.

[0022] If the current value of the environmental constraint height is not the initial value of the environmental constraint height, then the current value of the environmental constraint height is valid.

[0023] Optionally, adjusting the height of the initial detection box based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed includes:

[0024] Based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed, the target height of the target truck bed is determined;

[0025] Based on the target height, the height of the initial detection frame is adjusted.

[0026] Optionally, it also includes:

[0027] When the current value of the environmental constraint height is determined to be invalid, the initial detection box is used as the current detection box of the target truck bed.

[0028] Optionally, acquiring the first point cloud data of the area where the target truck bed is located includes:

[0029] The raw point cloud data collected by multiple lidar sensors are acquired respectively; wherein, the multiple lidar sensors are used to collect point cloud data of the area where the target truck bed is located from different angles, and each lidar sensor collects the raw point cloud data based on the control signal sent by the preset time synchronization device;

[0030] The original point cloud data collected by each of the lidars is spatially synchronized to obtain the first point cloud data.

[0031] Optionally, the step of spatially synchronizing the raw point cloud data collected by each of the lidars to obtain the first point cloud data includes:

[0032] Based on the preset point cloud transformation matrix and the rotation information of the target operating machinery, the original point cloud data collected by each of the lidars is spatially synchronized to obtain the first point cloud data.

[0033] The preset point cloud transformation matrix is ​​determined based on the historical point cloud data collected by each of the lidars. Each lidar is installed on the upper body of the target working machine, which is used to load the target truck bed.

[0034] Secondly, embodiments of this specification provide a truck bed position detection device, comprising:

[0035] The first processing module is used to acquire the first point cloud data of the area where the target truck bed is located, and to determine the initial detection frame of the target truck bed based on the first point cloud data.

[0036] The second processing module is used to determine the current value of the environmental constraint height corresponding to the target truck bed based on the initial detection frame and the first point cloud data; wherein, the environmental constraint height is used to characterize the height of a preset reference surface in the area where the target truck bed is located;

[0037] The third processing module is used to adjust the height of the initial detection frame based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed when it is determined that the current value of the environmental constraint height is valid, and to use the height-adjusted initial detection frame as the current detection frame of the target truck bed; wherein, the current detection frame is used to characterize the current pose of the target truck bed, and the geometric constraint height value is used to characterize the distance between the preset position of the target truck bed and the preset reference surface.

[0038] Thirdly, embodiments of this specification provide a working machine, which includes the truck bed position detection device as described above, or employs the truck bed position detection method as described in any of the above embodiments.

[0039] Fourthly, embodiments of this specification provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the truck bed pose detection method as described in any of the preceding claims.

[0040] Fifthly, embodiments of this specification provide a computer program product or a computer program, the computer program product including a computer program stored in a computer-readable storage medium; the processor of the computer device reads the computer program from the computer-readable storage medium, and when the processor executes the computer program, it implements the truck bed pose detection method as described in any of the preceding claims.

[0041] As can be seen from the above technical solutions, the embodiments of this application provide a truck bed pose detection method, device, and operating machinery. The truck bed pose detection method acquires first point cloud data of the area where the target truck bed is located, determines an initial detection frame of the target truck bed based on the first point cloud data, and determines the current value of the environmental constraint height corresponding to the target truck bed based on the initial detection frame and the first point cloud data. When the current value of the environmental constraint height is determined to be valid, the initial detection frame is height-adjusted based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed to obtain the current detection frame of the target truck bed. The environmental constraint height characterizes the height of a preset reference surface in the area where the target truck bed is located, and the geometric constraint height value characterizes the distance between the preset position of the target truck bed and the preset reference surface. By comprehensively considering the environmental constraint height and the geometric constraint height of the target truck bed, the current detection frame of the target truck bed can be effectively ensured to be close to the edge of the side plate of the target truck bed, thereby effectively improving the accuracy of truck bed pose detection. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0043] Figure 1 A flowchart illustrating a truck bed pose detection method provided in one embodiment of this specification;

[0044] Figure 2 A schematic diagram of geometric constraint height values ​​provided for one embodiment of this specification;

[0045] Figure 3 A schematic diagram showing the distribution of multiple lidars provided for one embodiment of this specification;

[0046] Figure 4 A schematic diagram of the structure of a truck bed posture detection device provided for one embodiment of this specification;

[0047] Figure label:

[0048] 301: First lidar; 302: Second lidar; 303: Third lidar; 304: Fourth lidar; 305: Fifth lidar. Detailed Implementation

[0049] Unless otherwise defined, the technical or scientific terms used in the embodiments of this specification shall have the ordinary meaning understood by one of ordinary skill in the art to which this specification pertains. The terms "first," "second," and similar terms used in the embodiments of this specification do not indicate any order, quantity, or importance, but are merely used to avoid confusion of constituent elements.

[0050] Unless the context otherwise requires, throughout this specification, "a plurality of" means "at least two," and "including" is interpreted as open-ended or encompassing, that is, "including, but not limited to." In the description of this specification, terms such as "one embodiment," "some embodiments," "exemplary embodiment," "example," "specific example," or "some examples" are intended to indicate that a particular feature, structure, material, or characteristic associated with that embodiment or example is included in at least one embodiment or example of this specification. The illustrative representations of the above terms do not necessarily refer to the same embodiment or example.

[0051] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this specification.

[0052] Overview

[0053] As described in the background section, excavation and loading scenarios are a crucial application area for the unmanned and intelligent development of construction machinery. In this scenario, accurate detection of the truck bed's position is an essential prerequisite for achieving unmanned and intelligent excavation and loading. During excavation and loading operations, real-time obstacle avoidance and truck bed full-load determination are required. Therefore, higher requirements are placed on the accuracy of truck bed position detection. The detection frame must be flush with the upper edge of the truck bed's side panel, meaning the upper edge of the detection frame should be aligned with the upper edge of the side panel. Based on high-precision truck bed detection frame information, the real-time obstacle avoidance module can have greater obstacle avoidance planning space, significantly reducing the probability of collisions and making it easier to plan reasonable working device movement paths, thus reducing the excavator's fuel consumption. Meanwhile, the truck bed full-load determination module can accurately determine the fullness of the truck bed based on the degree to which the material point cloud in the truck bed rises above the detection frame flush with the upper edge of the side panel, enabling reasonable truck load-bearing and planning of excavation cycles.

[0054] Currently, object detection methods are mainly based on deep learning and are primarily applied in autonomous driving scenarios. In autonomous driving scenarios, the road surface is usually relatively flat, and the detection bounding box only needs to cover the target object. However, in the loading and unloading scenario, there are various terrain features such as platforms, ground, and slopes. The scenario is complex and the terrain is easily changeable, which interferes with the inference performance of deep learning-based object detection methods, making it impossible for the truck bed detection bounding box to fit snugly against the edge of the truck bed's side panel.

[0055] To address the issue that traditional target detection methods cannot guarantee that the truck bed detection frame is tightly attached to the edge of the truck bed's side panel, related technologies employ a method that involves acquiring first point cloud data of the target truck bed's location and determining an initial detection frame based on this data. Furthermore, based on the initial detection frame and the first point cloud data, the current value of the environmental constraint height corresponding to the target truck bed is determined. When the current environmental constraint height is deemed valid, the initial detection frame is height-adjusted based on both the current environmental constraint height and the geometric constraint height of the target truck bed to obtain the current detection frame. Here, the environmental constraint height characterizes the height of a preset reference surface in the target truck bed's location, while the geometric constraint height characterizes the distance between the target truck bed's preset position and the preset reference surface. By comprehensively considering both the environmental constraint height and the geometric constraint height of the target truck bed, the current detection frame can be effectively ensured to be tightly attached to the edge of the target truck bed's side panel, thereby significantly improving the accuracy of truck bed pose detection.

[0056] Meanwhile, in order to improve the stability and reliability of the determination result of the current value of the environmental constraint height, the truck bed pose detection method provided in the embodiments of this specification also limits the method for determining the current value of the environmental constraint height corresponding to the target truck bed.

[0057] Furthermore, in order to further improve the recognition accuracy of the current detection box of the target truck bed, the truck bed pose detection method provided in this specification also limits the method for adjusting the height of the initial detection box and the method for obtaining the first point cloud data of the area where the target truck bed is located.

[0058] Based on the above inventive concept, the truck bed position detection method provided in the embodiments of this specification will be described exemplarily below.

[0059] Exemplary methods

[0060] This specification provides a method for detecting the position and orientation of a truck bed, such as... Figure 1 As shown, it includes:

[0061] S101. Obtain the first point cloud data of the area where the target truck bed is located, and determine the initial detection frame of the target truck bed based on the first point cloud data.

[0062] Specifically, the target truck bed is the truck bed to be subjected to pose detection. The area where the target truck bed is located may include the parking area of ​​the vehicle to which the target truck bed belongs, as well as the surrounding area of ​​the parking area of ​​the vehicle to which the target truck bed belongs.

[0063] The first point cloud data of the area where the target truck bed is located can be three-dimensional lidar point cloud data, which can be acquired by lidar. Compared with traditional cameras, lidar has a wider field of view and can return more accurate distance measurements. In addition, lidar is not easily affected by changes in the environment such as lighting, thus effectively improving the reliability of perception results in the perception of complex, changeable and harsh excavation and loading scenarios.

[0064] In implementation, multiple lidar units can be deployed. These units can collect point cloud data of the target truck bed area from different angles, thus avoiding the influence of lidar blind spots on the target truck bed's pose detection results. The lidar units can be mounted on the excavating machinery, enabling the machinery to perform unmanned operation based on the point cloud data collected by each lidar unit and the resulting pose detection data.

[0065] Understandably, the excavating machinery can collect point cloud data of the area where the target truck bed is located in real time during the operation, and can perform pose detection of the target truck bed based on the point cloud data collected at any given moment.

[0066] In determining the initial detection bounding box of the target truck bed based on the first point cloud data, the first point cloud data can be input into a preset target detection model to obtain the initial detection bounding box of the target truck bed. The target detection model can use point cloud target detection methods such as PointPillars, PointNet, PointNet++, VoxelNet, and CenterPoint to determine the initial detection bounding box of the target truck bed; no specific limitation is made here.

[0067] In one optional implementation, the target detection model can map the three-dimensional first point cloud data to a two-dimensional image space, and then, based on the features of the two-dimensional image space, determine the size and position of the initial detection box of the target truck bed in the X and Y axis directions using deep learning methods. Simultaneously, the target detection model can also determine the size and position of the initial detection box of the target truck bed in the Z-axis direction based on the features of the first point cloud data on the Z-axis, where the Z-axis can be the depth direction of the target truck bed. The process of mapping the three-dimensional first point cloud data to the two-dimensional image space may include the following steps:

[0068] 1. Divide the first point cloud data into multiple volumetric grids according to the X and Y axes. That is, for any volumetric grid among the multiple volumetric grids, it is composed of the point cloud data distributed to that volumetric grid from the first point cloud data.

[0069] For any point cloud, a vector (x, y, z, r, x) with dimension D = 9 can be used. cy c , z c x p y p Let (x, y, z, r) represent the three-dimensional coordinates and reflection intensity of the point cloud, respectively. c y c , z c (x) represents the geometric center of the cylindrical mesh containing the point cloud; p y p ) is (xx c yy c This reflects the relative position of the point cloud to the geometric center of the cylindrical mesh it is located in.

[0070] 2. Tensor representation of the first point cloud data: If the first point cloud data includes P non-empty volumetric grids, and the maximum number of points in each volumetric grid is N, then the first point cloud data can be represented as a tensor (D, P, N).

[0071] Third, the tensor representation of the first point cloud data is subjected to dimensionality reduction and feature extraction, so that the vector dimension of each point cloud changes from D to C. Thus, the tensor representation of the first point cloud data changes from (D, P, N) to (C, P, N). For any volumetric cylindrical mesh, max pooling can be performed on each point cloud in the volumetric cylindrical mesh to obtain the feature map of the first point cloud data in the (C, P) dimension.

[0072] Fourth, based on the position of each volumetric grid, the feature map of the first point cloud data in the (C, P) dimension is expanded into pseudo-image features; where, for any volumetric grid, P can be converted into the position (H, W) of the volumetric grid on the X and Y axes, thus obtaining the feature representation in the form of (C, H, W).

[0073] By mapping the three-dimensional first point cloud data to the two-dimensional image space through the above steps, the representation of the first point cloud data can be made more compact and dense, avoiding the redundant computation caused by the sparsity and irregularity of point cloud data in methods such as voxel method and point-by-point method, and improving the recognition efficiency of the initial detection box. At the same time, based on the features of the two-dimensional image, mature deep learning methods can be used to identify the initial detection box of the target truck bed, improving the reliability of the recognition results of the initial detection box in the X and Y axis directions, and facilitating the deployment of the target detection model in the actual working environment.

[0074] S102. Based on the initial detection frame and the first point cloud data, determine the current value of the environmental constraint height corresponding to the target truck bed; wherein, the environmental constraint height is used to characterize the height of the preset reference surface in the area where the target truck bed is located.

[0075] Specifically, the environmental constraint height corresponding to the target truck bed is used to characterize the height of the preset reference surface in the area where the target truck bed is located. The preset reference surface can be a reference surface with a fixed positional relationship to the target truck bed, such as the roof of the vehicle to which the target truck bed belongs, or the ground near the area where the target truck bed is parked. Since the vehicle to which the target truck bed belongs needs to keep the truck bed as close as possible to the high platform working surface when performing excavation and loading operations, the ground terrain here is complex and has large undulations, making it unsuitable as an environmental constraint. Therefore, the ground near the front of the vehicle to which the target truck bed belongs can be selected as the preset reference surface to perform environmental constraints during the target truck bed pose detection process.

[0076] The current value of the environmental constraint height corresponding to the target truck bed is the value determined based on the first point cloud data collected at the current moment.

[0077] S103. When it is determined that the current value of the environmental constraint height is valid, the height of the initial detection frame is adjusted based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed, and the height-adjusted initial detection frame is used as the current detection frame of the target truck bed; wherein, the current detection frame is used to characterize the current pose of the target truck bed, and the geometric constraint height value is used to characterize the distance between the preset position of the target truck bed and the preset reference surface.

[0078] Specifically, the preset position can be the centerline of the target truck bed along the length direction of the vehicle (the direction connecting the front and rear of the vehicle), etc. The geometric constraint height value of the target truck bed is used to characterize the distance between the preset position of the target truck bed and the preset reference surface, and can be a pre-set constant. In implementation, the geometric constraint height value can be as follows: Figure 2 As shown, Figure 2 In the first geometric constraint height value, the distance from the centerline of the target truck bed in the length direction to the roof of the vehicle to which the target truck bed belongs is the distance from the centerline of the target truck bed in the length direction to the point where the ground of the area where the target truck bed is parked is tangent to the tire of the target truck bed.

[0079] After determining the current value of the environmental constraint height, it can be determined whether the current value of the environmental constraint height is valid. For example, the validity of the current value of the environmental constraint height can be determined based on the comparison between the current value of the environmental constraint height and the initial value of the environmental constraint height.

[0080] If the current value of the environmental constraint height is valid, the initial detection frame of the target truck bed can be adjusted based on the current value of the environmental constraint height and the geometric constraint height of the target truck bed. For example, the target height of the target truck bed can be determined based on the current value of the environmental constraint height and the geometric constraint height of the target truck bed, and the initial detection frame of the target truck bed can be adjusted according to the target height to obtain the current detection frame of the target truck bed. In this way, the environmental constraint height and the geometric constraint height of the target truck bed can be comprehensively considered in the process of determining the current detection frame of the target truck bed, so that the current detection frame of the target truck bed can be closely attached to the edge of the side plate of the target truck bed.

[0081] The current bounding box of the target truck bed is the target position of the bounding box of the target truck bed determined based on the first point cloud data collected at the current moment. The current bounding box of the target truck bed can be used to represent the current pose of the target truck bed, that is, the pose of the target truck bed at the current moment.

[0082] Understandably, if the current value of the environmental constraint height is invalid, the initial detection box of the target truck bed can be directly used as the current detection box of the target truck bed to ensure the validity of the pose detection results.

[0083] Therefore, the truck bed pose detection method provided in this specification acquires the first point cloud data of the area where the target truck bed is located, determines the initial detection frame of the target truck bed based on the first point cloud data, and determines the current value of the environmental constraint height corresponding to the target truck bed based on the initial detection frame and the first point cloud data. When the current value of the environmental constraint height is determined to be valid, the height of the initial detection frame is adjusted based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed to obtain the current detection frame of the target truck bed. The environmental constraint height is used to characterize the height of the preset reference surface in the area where the target truck bed is located, and the geometric constraint height value is used to characterize the distance between the preset position of the target truck bed and the preset reference surface. By comprehensively considering the environmental constraint height and the geometric constraint height of the target truck bed, the current detection frame of the target truck bed can be effectively ensured to be close to the edge of the side plate of the target truck bed, thereby effectively improving the accuracy of truck bed pose detection.

[0084] To improve the stability and reliability of the determination of the current value of the environmental constraint height, in one embodiment of this specification, determining the current value of the environmental constraint height corresponding to the target truck bed based on the initial detection frame and the first point cloud data includes:

[0085] Based on the initial detection frame and the size information of the vehicle to which the target truck bed belongs, the current clipping frame corresponding to the preset reference surface is determined;

[0086] Based on the current cropping frame and the first point cloud data, the second point cloud data corresponding to the preset reference surface is obtained;

[0087] Based on the second point cloud data, the current value of the environmental constraint height is determined.

[0088] Specifically, the current clipping frame corresponding to the preset reference surface can be a clipping frame that includes the preset reference surface, for example, it can be a three-dimensional clipping frame.

[0089] In implementation, the current clipping frame corresponding to the preset reference surface can be determined based on the initial detection frame of the target truck bed and the size information of the vehicle to which the target truck bed belongs. For example, the position of the target truck bed can be determined based on the position of the initial detection frame of the target truck bed, and the position of the current clipping frame corresponding to the preset reference surface can be determined based on the position of the target truck bed and the size information of the vehicle to which the target truck bed belongs. The size of the current clipping frame can be a preset size. Considering the size information of the vehicle to which the target truck bed belongs and the issue of point cloud occlusion, the size of the current clipping frame can be set large enough to encompass a sufficient amount of point cloud data on the preset reference surface.

[0090] After determining the current clipping frame corresponding to the preset reference plane, candidate point cloud data can be obtained from the first point cloud data based on the current clipping frame. The candidate point cloud data is the point cloud data enclosed by the current clipping frame in the first point cloud data. Specifically, based on the type of the preset reference plane, second point cloud data corresponding to the preset reference plane can be obtained from the candidate point cloud data. For example, if the preset reference plane is the roof of the vehicle to which the target truck bed belongs, the topmost point cloud data in the candidate point cloud data can be used as the second point cloud data corresponding to the preset reference plane. If the preset reference plane is the ground near the area where the target truck bed is parked, the bottommost point cloud data in the candidate point cloud data can be used as the second point cloud data corresponding to the preset reference plane.

[0091] After obtaining the second point cloud data corresponding to the preset reference surface, the current value of the environmental constraint height can be determined based on the second point cloud data. For example, the current value of the environmental constraint height can be determined directly based on the height information of each point cloud in the second point cloud data; alternatively, it can be determined whether the second point cloud data is valid. If the second point cloud data is valid, the current value of the environmental constraint height can be determined based on the height information of each point cloud in the second point cloud data. If the second point cloud data is invalid, the initial value of the environmental constraint height or the historical value of the environmental constraint height at the previous moment can be used as the current value of the environmental constraint height. This effectively ensures the stability and reliability of the determination result of the current value of the environmental constraint height. Furthermore, in the process of determining the current detection box of the target truck bed based on the current value of the environmental constraint height, the recognition accuracy of the current detection box can be further improved, so that the current detection box of the target truck bed can be closely attached to the edge of the side plate of the target truck bed.

[0092] In one feasible implementation, determining the current clipping frame corresponding to the preset reference plane based on the initial detection frame and the size information of the vehicle to which the target truck bed belongs includes:

[0093] Based on the initial detection frame and the size information of the vehicle to which the target truck bed belongs, the position of the preset clipping frame is adjusted, and the adjusted preset clipping frame is used as the current clipping frame corresponding to the preset reference surface.

[0094] Specifically, the preset clipping frame can be pre-set based on the first dimension information of the vehicle to which the target truck bed belongs, as well as point cloud occlusion, etc. The preset clipping frame is a three-dimensional clipping frame used to determine the size of the current clipping frame; that is, the size of the current clipping frame is the same as the size of the preset clipping frame. The first dimension information of the vehicle to which the target truck bed belongs can be the distance from the centerline of the target truck bed along its length to the ground, the distance from the centerline of the target truck bed along its length to the roof of the truck bed, the size of the roof of the truck bed, etc.

[0095] In determining the current clipping frame corresponding to the preset reference surface based on the initial detection frame and the size information of the vehicle to which the target truck bed belongs, the preset clipping frame can be adjusted in position, such as by translation, based on the position of the initial detection frame of the target truck bed and the second size information of the vehicle to which the target truck bed belongs. The adjusted preset clipping frame is then used as the current clipping frame. The second size information of the vehicle to which the target truck bed belongs can be the distance from the center point of the target truck bed to the roof of the vehicle along its length, or the distance from the center point of the target truck bed to the front end of the vehicle.

[0096] In practice, preset points (such as vertices, center points, etc.) in the preset clipping frame can be translated to the center point of the initial detection frame of the target truck bed. Then, the preset clipping frame is translated according to the yaw angle of the vehicle to which the target truck bed belongs and the second dimension information of the vehicle to which the target truck bed belongs. Among them, the yaw angle of the vehicle to which the target truck bed belongs can be the angle between the center line of the initial detection frame in the length direction and the X-axis. Taking the ground near the front of the vehicle to which the target truck bed belongs as the preset reference surface as an example, the translation result of the preset point during the translation of the preset clipping frame according to the yaw angle of the vehicle to which the target truck bed belongs and the second dimension information of the vehicle to which the target truck bed belongs can be as shown in Equation (1):

[0097]

[0098] In the formula, (x m y m , z m) represents the coordinates of the preset point in the preset clipping box after translation, d represents the distance from the center point of the target truck bed to the front of the vehicle to which it belongs, (box_x, box_y, box_z) represents the coordinates of the center point of the initial detection box, and box_rt represents the yaw angle of the vehicle to which the target truck bed belongs.

[0099] Understandably, z m You can also adjust the height of box_z, specifically based on the height of the preset clipping box in the Z-axis direction, ensuring that the preset clipping box after position adjustment covers the point cloud corresponding to the preset reference plane to the greatest extent.

[0100] Since the initial detection bounding box of the target truck bed has high accuracy in the XY plane, adjusting the position of the preset clipping box based on the initial detection bounding box and the size information of the vehicle to which the target truck bed belongs ensures that the adjusted preset clipping box maximally encompasses the point cloud corresponding to the preset reference surface, thus improving the accuracy of determining the current value of the environmental constraint height. Furthermore, by considering the yaw angle of the vehicle to which the target truck bed belongs during the translation of the preset clipping box, the reliability of the current clipping box determination can be effectively guaranteed, further improving the accuracy of determining the current value of the environmental constraint height.

[0101] In one feasible implementation, determining the current value of the environmental constraint height based on the second point cloud data includes:

[0102] If the number of points in the second point cloud data is less than the preset number, then the initial value of the environmental constraint height or the historical value of the environmental constraint height at the previous moment shall be used as the current value of the environmental constraint height.

[0103] If the number of point clouds in the second point cloud data is greater than or equal to the preset number, then the current value of the environmental constraint height is determined based on the height information of each point cloud in the second point cloud data.

[0104] Specifically, the preset quantity can be set according to actual needs. If the number of point clouds in the second point cloud data is less than the preset quantity, it means that the reliability of the second point cloud data is poor, that is, the second point cloud data is invalid. In this case, the environmental constraint height can be left unupdated. That is, the initial value of the environmental constraint height or the historical value of the environmental constraint height at the previous moment can be used as the current value of the environmental constraint height.

[0105] For example, if the current time is the initial time, the initial value of the environmental constraint height can be used as the current value of the environmental constraint height. This initial value can be a preset value, such as 0. If the current time is not the initial time, the historical value of the environmental constraint height from the previous time can be used as the current value of the environmental constraint height. This historical value is the value of the environmental constraint height determined based on the first point cloud data collected at the previous time. It can be understood that the historical value of the environmental constraint height from the previous time can be either the initial value or not.

[0106] If the number of points in the second point cloud data is greater than or equal to a preset number, it indicates that the second point cloud data has high reliability, meaning it is valid. The current value of the environmental constraint height can be determined based on the height information of each point in the second point cloud data. For example, the mean or median height of each point in the second point cloud data can be used as the current value of the environmental constraint height. This effectively ensures the stability and reliability of the update results of the current environmental constraint height. Furthermore, in determining the current detection frame of the target truck bed based on the current environmental constraint height, it effectively ensures that the current detection frame of the target truck bed is closely attached to the edge of the side panel of the target truck bed.

[0107] In one feasible implementation, determining whether the current value of the environmental constraint height is valid includes:

[0108] If the current value of the environmental constraint height is the initial value of the environmental constraint height, then the current value of the environmental constraint height is invalid.

[0109] If the current value of the environmental constraint height is not the initial value of the environmental constraint height, then the current value of the environmental constraint height is valid.

[0110] Specifically, if the current value of the environmental constraint height is the initial value of the environmental constraint height, it means that the environmental constraint height has not been updated at the current time and at all times before the current time. That is, at the current time and at all times before the current time, the second point cloud data corresponding to the preset reference surface are all invalid point cloud data. In this case, the current value of the environmental constraint height can be considered invalid.

[0111] If the current value of the environmental constraint height is not the initial value, it indicates that at least one time, including the current time and all previous times, the environmental constraint height has been updated. That is, at least one time, including the current time and all previous times, the second point cloud data corresponding to the preset reference surface is valid point cloud data. In this case, the current value of the environmental constraint height can be considered valid, thus enabling a quick and reliable determination of its validity. Simultaneously, when the current value of the environmental constraint height is valid, the initial detection frame of the target truck bed is height-adjusted based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed, ensuring that the height-adjusted initial detection frame is flush against the edge of the target truck bed's side panel.

[0112] To further improve the recognition accuracy of the current detection frame of the target truck bed, in one embodiment of this specification, adjusting the height of the initial detection frame based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed includes:

[0113] Based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed, the target height of the target truck bed is determined;

[0114] Based on the target height, the height of the initial detection frame is adjusted.

[0115] Specifically, the current value of the environmental constraint height and the geometric constraint height value of the target truck bed can be added or subtracted to obtain the target height of the target truck bed.

[0116] For example, when the preset reference plane is the roof of the vehicle to which the target truck bed belongs, the geometric constraint height value of the target truck bed can be adopted. Figure 2 The target height of the target truck bed is determined based on the first geometric constraint height value in the environmental constraint height calculation and the difference between the current environmental constraint height value and the first geometric constraint height value. When the preset reference surface is the ground near the area where the target truck bed is parked, the geometric constraint height value of the target truck bed can be determined using... Figure 2 The target height of the target truck bed is determined by the sum of the current value of the environmental constraint height and the second geometric constraint height value.

[0117] After determining the target height of the target truck bed, the initial detection frame of the target truck bed can be adjusted based on the target height. For example, after adjusting the height of the initial detection frame, the height of the center point of the initial detection frame is the target height, so that the current detection frame of the target truck bed can be close to the edge of the side plate of the target truck bed.

[0118] To ensure the validity of the target truck bed pose detection results, one embodiment of this specification further includes:

[0119] When the current value of the environmental constraint height is determined to be invalid, the initial detection box is used as the current detection box of the target truck bed.

[0120] Specifically, when the current value of the environmental constraint height is determined to be invalid, if the height of the initial detection box is adjusted based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed, it cannot be guaranteed that the initial detection box after height adjustment can be closely attached to the edge of the side plate of the target truck bed. In this case, the initial detection box of the target truck bed can be directly used as the current detection box of the target truck bed to ensure the validity of the pose detection result of the target truck bed.

[0121] To improve the comprehensiveness and effectiveness of the first point cloud data of the area where the target truck bed is located, in one embodiment of this specification, acquiring the first point cloud data of the area where the target truck bed is located includes:

[0122] The raw point cloud data collected by multiple lidar sensors are acquired respectively; wherein, the multiple lidar sensors are used to collect point cloud data of the area where the target truck bed is located from different angles, and each lidar sensor collects the raw point cloud data based on the control signal sent by the preset time synchronization device;

[0123] The original point cloud data collected by each of the lidars is spatially synchronized to obtain the first point cloud data.

[0124] Specifically, multiple lidars can be set up, each used to collect point cloud data of the area where the target truck bed is located from different angles, in order to obtain the maximum field of view and achieve comprehensive and effective collection of point cloud data of the area where the target truck bed is located.

[0125] The preset time synchronization device can use PTP (Precision Time Protocol) to synchronize the time of each lidar, ensuring that point cloud data is acquired at the same time. Specifically, the preset time synchronization device can send pulse signals from the same pulse source as control signals to each lidar to trigger point cloud acquisition, ensuring the consistency of the acquisition time for each lidar. Simultaneously, the preset time synchronization device can also add timestamps to the raw point cloud data acquired by each lidar based on a preset system time source.

[0126] During the excavation operation, point cloud data of the target truck bed area can be collected in real time using various lidar sensors. For any given point cloud collection moment, the raw point cloud data collected by each lidar sensor at that moment can be spatially synchronized to fuse the raw point cloud data and obtain the first point cloud data corresponding to that moment. This effectively ensures the comprehensiveness and validity of the first point cloud data of the target truck bed area.

[0127] In one feasible implementation, the step of spatially synchronizing the raw point cloud data collected by each of the lidars to obtain the first point cloud data includes:

[0128] Based on the preset point cloud transformation matrix and the rotation information of the target operating machinery, the original point cloud data collected by each of the lidars is spatially synchronized to obtain the first point cloud data.

[0129] The preset point cloud transformation matrix is ​​determined based on the historical point cloud data collected by each of the lidars. Each lidar is installed on the upper body of the target working machine, which is used to load the target truck bed.

[0130] Specifically, the target working machinery is used to load the target truck bed. That is, the target working machinery can be working machinery that performs excavation operations, such as excavators or loaders. The target working machinery can include an upper body and a lower body. To facilitate excavation operations, the upper body can rotate during operation. Multiple lidar sensors can be rigidly connected and installed on the upper body of the target working machinery to effectively collect the first point cloud data of the area where the target truck bed is located during the excavation operation.

[0131] In a preferred embodiment, five lidars can be configured, namely, a first lidar 301, a second lidar 302, a third lidar 303, a fourth lidar 304, and a fifth lidar 305. For example... Figure 3 As shown, the first lidar 301 can be installed on the top of the cab of the target machinery to ensure that the first lidar 301 can obtain the maximum working field of view and is not easily affected by the vibration of the working device in the target machinery; the second lidar 302 can be installed at the root of the boom of the target machinery, that is, at the ladder of the vehicle body, to ensure that when the target machinery is performing a large turning range, the area where the target truck bed is located is still within the field of view of the second lidar 302; the third lidar 303 and the fifth lidar 305 are respectively installed on both sides of the vehicle body of the target machinery, and the fourth lidar 304 is installed at the rear of the vehicle body of the target machinery to compensate for the blind spots of the first lidar 301 and the second lidar 302.

[0132] A preset point cloud transformation matrix is ​​used to characterize the coordinate transformation relationship between the coordinate system of each lidar and the coordinate system of the target machinery. This preset point cloud transformation matrix can be determined based on historical point cloud data collected by each lidar. For example, coarse registration can be performed on the historical point cloud data collected by each lidar based on their position information, and fine registration can be performed on the coarsely registered historical point cloud data using point cloud registration methods such as GICP (Generalized Iterative Closest Point) and NDT (Normal Distribution Transform) to obtain the preset point cloud transformation matrix.

[0133] During the spatial synchronization processing of the raw point cloud data collected by each lidar, a first coordinate transformation can be performed on the raw point cloud data collected by each lidar based on a preset point cloud transformation matrix to synchronize the raw point cloud data collected by each lidar to the upper body coordinate system of the target working machinery. Then, based on the rotation information of the target working machinery, a second coordinate transformation can be performed on the raw point cloud data after the first coordinate transformation to synchronize the raw point cloud data collected by each lidar to the lower body coordinate system of the target working machinery. Thus, through the first and second coordinate transformation processes, the spatial synchronization of the raw point cloud data collected by each lidar can be completed, and the fusion of the raw point cloud data collected by each lidar is achieved through spatial synchronization, thereby obtaining the first point cloud data of the area where the target truck bed is located. Since the target truck bed is parked on the ground, synchronizing the raw point cloud data collected by each lidar to the lower body coordinate system of the target working machinery further improves the effectiveness of the first point cloud data.

[0134] Exemplary device

[0135] refer to Figure 4 This specification also provides a truck bed position detection device, comprising:

[0136] The first processing module 401 is used to acquire the first point cloud data of the area where the target truck bed is located, and to determine the initial detection frame of the target truck bed based on the first point cloud data.

[0137] The second processing module 402 is used to determine the current value of the environmental constraint height corresponding to the target truck bed based on the initial detection frame and the first point cloud data; wherein, the environmental constraint height is used to characterize the height of a preset reference surface in the area where the target truck bed is located;

[0138] The third processing module 403 is used to adjust the height of the initial detection frame based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed when it is determined that the current value of the environmental constraint height is valid, and to use the height-adjusted initial detection frame as the current detection frame of the target truck bed; wherein, the current detection frame is used to characterize the current pose of the target truck bed, and the geometric constraint height value is used to characterize the distance between the preset position of the target truck bed and the preset reference surface.

[0139] For example, the second processing module 402 is specifically used for:

[0140] Based on the initial detection frame and the size information of the vehicle to which the target truck bed belongs, the current clipping frame corresponding to the preset reference surface is determined;

[0141] Based on the current cropping frame and the first point cloud data, the second point cloud data corresponding to the preset reference surface is obtained;

[0142] Based on the second point cloud data, the current value of the environmental constraint height is determined.

[0143] For example, the second processing module 402 is specifically used for:

[0144] Based on the initial detection frame and the size information of the vehicle to which the target truck bed belongs, the position of the preset clipping frame is adjusted, and the adjusted preset clipping frame is used as the current clipping frame corresponding to the preset reference surface.

[0145] For example, the second processing module 402 is specifically used for:

[0146] If the number of points in the second point cloud data is less than the preset number, then the initial value of the environmental constraint height or the historical value of the environmental constraint height at the previous moment shall be used as the current value of the environmental constraint height.

[0147] If the number of point clouds in the second point cloud data is greater than or equal to the preset number, then the current value of the environmental constraint height is determined based on the height information of each point cloud in the second point cloud data.

[0148] For example, the third processing module 403 is specifically used for:

[0149] If the current value of the environmental constraint height is the initial value of the environmental constraint height, then the current value of the environmental constraint height is invalid.

[0150] If the current value of the environmental constraint height is not the initial value of the environmental constraint height, then the current value of the environmental constraint height is valid.

[0151] For example, the third processing module 403 is specifically used for:

[0152] Based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed, the target height of the target truck bed is determined;

[0153] Based on the target height, the height of the initial detection frame is adjusted.

[0154] For example, the third processing module 403 is further configured to:

[0155] When the current value of the environmental constraint height is determined to be invalid, the initial detection box is used as the current detection box of the target truck bed.

[0156] For example, the first processing module 401 is specifically used for:

[0157] The raw point cloud data collected by multiple lidar sensors are acquired respectively; wherein, the multiple lidar sensors are used to collect point cloud data of the area where the target truck bed is located from different angles, and each lidar sensor collects the raw point cloud data based on the control signal sent by the preset time synchronization device;

[0158] The original point cloud data collected by each of the lidars is spatially synchronized to obtain the first point cloud data.

[0159] For example, the first processing module 401 is specifically used for:

[0160] Based on the preset point cloud transformation matrix and the rotation information of the target operating machinery, the original point cloud data collected by each of the lidars is spatially synchronized to obtain the first point cloud data.

[0161] The preset point cloud transformation matrix is ​​determined based on the historical point cloud data collected by each of the lidars. Each lidar is installed on the upper body of the target working machine, which is used to load the target truck bed.

[0162] The apparatus provided in this embodiment belongs to the same concept as the method provided in the above embodiments of this application, and can execute the method provided in any of the above embodiments of this application, possessing the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the specific processing content of the method provided in the above embodiments of this application, and will not be repeated here.

[0163] Exemplary operating machinery

[0164] This specification also provides a working machine, which includes a truck bed position detection device as described in any of the above embodiments, or a truck bed position detection method as described in any of the above embodiments.

[0165] The operating machinery mentioned includes construction machinery such as excavators and loaders.

[0166] Exemplary computer program products and storage media

[0167] In addition to the methods and devices described above, the truck bed pose detection method provided in the embodiments of this specification can also be a computer program product, which includes computer program instructions that, when executed by a processor, cause the processor to perform the steps in the truck bed pose detection method according to various embodiments of this specification as described in the "Exemplary Methods" section above.

[0168] The computer program product described herein can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments described herein. These programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0169] Furthermore, embodiments of this specification also provide a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor of the steps in the truck bed pose detection method according to various embodiments of this specification as described in the "Exemplary Methods" section above.

[0170] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this specification can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0171] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0172] The embodiments described above are merely illustrative of several implementation methods outlined in this specification. While the descriptions are specific and detailed, they should not be construed as limiting the scope of the solutions provided in this specification. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this specification, and these all fall within the scope of protection of this specification. Therefore, the scope of protection for this patent should be determined by the appended claims.

Claims

1. A method for detecting the position and orientation of a truck bed, characterized in that, include: Acquire the first point cloud data of the area where the target truck bed is located, and determine the initial detection frame of the target truck bed based on the first point cloud data; Based on the initial detection frame and the first point cloud data, the current value of the environmental constraint height corresponding to the target truck bed is determined; wherein, the environmental constraint height is used to characterize the height of a preset reference surface in the area where the target truck bed is located; When the current value of the environmental constraint height is determined to be valid, the initial detection frame is height adjusted based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed, and the height-adjusted initial detection frame is used as the current detection frame of the target truck bed; wherein, the current detection frame is used to characterize the current pose of the target truck bed, and the geometric constraint height value is used to characterize the distance between the preset position of the target truck bed and the preset reference surface.

2. The method according to claim 1, characterized in that, The step of determining the current value of the environmental constraint height corresponding to the target truck bed based on the initial detection frame and the first point cloud data includes: Based on the initial detection frame and the size information of the vehicle to which the target truck bed belongs, the current clipping frame corresponding to the preset reference surface is determined; Based on the current cropping frame and the first point cloud data, the second point cloud data corresponding to the preset reference surface is obtained; Based on the second point cloud data, the current value of the environmental constraint height is determined.

3. The method according to claim 2, characterized in that, The step of determining the current clipping frame corresponding to the preset reference plane based on the initial detection frame and the size information of the vehicle to which the target truck bed belongs includes: Based on the initial detection frame and the size information of the vehicle to which the target truck bed belongs, the position of the preset clipping frame is adjusted, and the adjusted preset clipping frame is used as the current clipping frame corresponding to the preset reference surface.

4. The method according to claim 2, characterized in that, Determining the current value of the environmental constraint height based on the second point cloud data includes: If the number of points in the second point cloud data is less than the preset number, then the initial value of the environmental constraint height or the historical value of the environmental constraint height at the previous moment shall be used as the current value of the environmental constraint height. If the number of point clouds in the second point cloud data is greater than or equal to the preset number, then the current value of the environmental constraint height is determined based on the height information of each point cloud in the second point cloud data.

5. The method according to claim 4, characterized in that, Determining whether the current value of the environmental constraint height is valid includes: If the current value of the environmental constraint height is the initial value of the environmental constraint height, then the current value of the environmental constraint height is invalid. If the current value of the environmental constraint height is not the initial value of the environmental constraint height, then the current value of the environmental constraint height is valid.

6. The method according to claim 1, characterized in that, The step of adjusting the height of the initial detection frame based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed includes: Based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed, the target height of the target truck bed is determined; Based on the target height, the height of the initial detection frame is adjusted.

7. The method according to claim 1, characterized in that, Also includes: When the current value of the environmental constraint height is determined to be invalid, the initial detection box is used as the current detection box of the target truck bed.

8. The method according to any one of claims 1 to 7, characterized in that, The acquisition of the first point cloud data of the area where the target truck bed is located includes: The raw point cloud data collected by multiple lidar sensors are acquired respectively; wherein, the multiple lidar sensors are used to collect point cloud data of the area where the target truck bed is located from different angles, and each lidar sensor collects the raw point cloud data based on the control signal sent by the preset time synchronization device; The original point cloud data collected by each of the lidars is spatially synchronized to obtain the first point cloud data.

9. The method according to claim 8, characterized in that, The process of spatially synchronizing the raw point cloud data collected by each of the lidars to obtain the first point cloud data includes: Based on the preset point cloud transformation matrix and the rotation information of the target operating machinery, the original point cloud data collected by each of the lidars is spatially synchronized to obtain the first point cloud data. The preset point cloud transformation matrix is ​​determined based on the historical point cloud data collected by each of the lidars. Each lidar is installed on the upper body of the target working machine, which is used to load the target truck bed.

10. A vehicle bed position and posture detection device, characterized in that, include: The first processing module is used to acquire the first point cloud data of the area where the target truck bed is located, and to determine the initial detection frame of the target truck bed based on the first point cloud data. The second processing module is used to determine the current value of the environmental constraint height corresponding to the target truck bed based on the initial detection frame and the first point cloud data; wherein, the environmental constraint height is used to characterize the height of a preset reference surface in the area where the target truck bed is located; The third processing module is used to adjust the height of the initial detection frame based on the current value of the environmental constraint height and the geometric constraint height value of the target truck bed when it is determined that the current value of the environmental constraint height is valid, and to use the height-adjusted initial detection frame as the current detection frame of the target truck bed; wherein, the current detection frame is used to characterize the current pose of the target truck bed, and the geometric constraint height value is used to characterize the distance between the preset position of the target truck bed and the preset reference surface.

11. A type of operating machinery, characterized in that, The operating machinery includes the truck bed position detection device as described in claim 10.

Images