An airport unmanned vehicle luggage stacking method and device and an airport unmanned vehicle

Abstract

This application discloses a baggage stacking method, apparatus, and airport unmanned vehicle (UAV) for airports, belonging to the field of airport logistics transfer technology. The baggage stacking method includes the following steps: acquiring pallet storage status and real-time baggage data; determining baggage stacking coordinates and target stacking pose based on pallet storage status and baggage data, and planning a baggage stacking route based on the baggage stacking coordinates; sending a first control command to the pallet unit according to the baggage stacking route, the first control command being used to cause the area controller to control the rotation of the drive wheel to move the baggage to the position of the baggage stacking coordinates. By decomposing the complete baggage stacking route to each area controller for control, the pallet unit where the baggage is located is accurately located, and the position and posture of the baggage are independently controlled by the pallet unit, ensuring the efficiency and accuracy of baggage stacking.

Description

Technical Field

[0001] This application relates to the field of airport logistics transfer technology, specifically to a baggage stacking method, device, and airport unmanned vehicle. Background Technology

[0002] Airport unmanned vehicles refer to autonomous vehicles that operate in specific areas of an airport (airside and landside), and are mainly used in scenarios such as baggage transportation, passenger transfer, cargo transfer, security patrol, and facility maintenance.

[0003] In existing technologies, airport unmanned vehicles typically adopt an integrated pallet structure, and baggage stacking relies on a global control system for unified scheduling. This lacks the ability to dynamically and accurately locate and independently control baggage, making it difficult to guarantee the efficiency and accuracy of baggage stacking. Summary of the Invention

[0004] This application provides a method, apparatus, and airport unmanned vehicle for stacking luggage, aiming to solve the problem in related technologies where airport unmanned vehicles lack the ability to dynamically and accurately locate and independently control luggage, making it difficult to guarantee the efficiency and accuracy of luggage stacking.

[0005] In a first aspect, embodiments of this application provide a baggage stacking method for an airport unmanned vehicle. The airport unmanned vehicle includes a pallet, which is composed of multiple pallet units arranged in an array. Each pallet unit includes an area controller and multiple drive wheels. The area controller is used to control the rotation of the drive wheels. The baggage stacking method includes the following steps: Obtain pallet storage status and real-time baggage data; Determine the baggage palletizing coordinates and target palletizing pose based on the pallet storage status and baggage data, and plan the baggage palletizing route based on the baggage palletizing coordinates; The first control command is sent to the pallet unit according to the baggage stacking route. The first control command is used to make the area controller control the drive wheel to rotate, so as to move the baggage to the location of the baggage stacking coordinates.

[0006] In some embodiments, the airport driverless vehicle also includes a vision sensor positioned above the pallet; Obtain real-time baggage data, including: Acquire 3D point cloud data of luggage collected by a visual sensor; The 3D point cloud data is processed using 3D vision algorithms to obtain baggage data, which includes baggage category, baggage size, baggage posture, and baggage location.

[0007] In some embodiments, determining baggage palletizing coordinates and target palletizing pose based on pallet storage status and baggage data, and planning baggage palletizing routes based on the baggage palletizing coordinates, includes: Calculate the appropriate luggage palletizing coordinates and target palletizing pose based on the pallet storage status and luggage size; Starting from the current location of the luggage and ending at the location of the luggage stacking coordinates, a luggage stacking route is planned based on the location boundaries of the luggage already stacked in the pallet storage status.

[0008] In some embodiments, the transmission wheels are evenly distributed around a center point. The first control command is sent to the pallet unit according to the baggage palletizing route, including: Determine the current pallet unit where the luggage is located based on its position. Based on the baggage palletizing route, determine the target area that the baggage needs to be moved to on the pallet unit, generate the first control command and send it to the area controller of the pallet unit; The area controller controls the rotation of the drive wheels, adjusts the posture of the luggage, and moves the luggage to the target area along the rotation direction of the drive wheels. All pallet units on the baggage palletizing route are controlled sequentially to move the baggage to the location of the baggage palletizing coordinates.

[0009] In some embodiments, controlling the rotation of the drive wheels via an area controller to adjust the posture of the luggage includes: The first control command is analyzed by the area controller to determine the target posture of the luggage; The corresponding drive wheels are driven to rotate by the area controller, so that multiple drive wheels work together. The differential rotation of the drive wheels drives the luggage to rotate, and the luggage posture is obtained in real time based on the luggage data. The area controller determines the attitude deviation based on the luggage's attitude and the target attitude; If the attitude deviation is greater than the preset attitude threshold, the area controller will re-drive the corresponding transmission wheel to rotate according to the attitude deviation. The luggage's posture is adjusted when the posture deviation is less than the preset posture threshold.

[0010] In some embodiments, controlling the rotation of the drive wheels via an area controller to move luggage along the rotation direction of the drive wheels to a target area includes: The area controller parses the first control command to determine the target location of the luggage; The corresponding drive wheels are driven to rotate by the area controller, so that multiple drive wheels work together to move the luggage horizontally or vertically, and the luggage position is obtained in real time based on the luggage data; The area controller determines the location deviation based on the luggage's location and the target location; If the position deviation is greater than the preset position threshold, the area controller will re-drive the corresponding transmission wheel to rotate according to the position deviation. If the position deviation is less than the preset position threshold, the luggage position adjustment is completed.

[0011] In some embodiments, the pallet unit further includes an array of photoelectric sensors; Luggage stacking methods also include: Obtain the scanning data periodically collected by the photoelectric sensor array in the pallet unit where the luggage stacking coordinates are located; Once the luggage arrives at the luggage palletizing coordinates, the scan data is fused with the luggage data to determine the luggage's occupancy status and palletizing pose at the luggage palletizing coordinates. The deviation value is determined based on the palletizing pose and the target palletizing pose. A second control command is generated based on the deviation value and sent to the pallet unit. The area controller parses the second control command and controls the transmission wheel to adjust the palletizing posture so that the deviation value is less than the preset deviation threshold. If the deviation value is less than the preset deviation threshold, the final pose of the luggage is obtained and the pallet storage status is updated according to the final pose to complete the luggage stacking.

[0012] In some embodiments, the pallet has multiple layers; After sending a first control command to the area controller according to the baggage stacking route, the baggage stacking method further includes: Determine the pallet storage status. If the pallet storage status indicates that the current pallet is full, use the next pallet to receive the baggage.

[0013] Secondly, this application provides a baggage stacking device for an airport unmanned vehicle. The airport unmanned vehicle includes a pallet, which is composed of multiple pallet units arranged in an array. Each pallet unit includes an area controller and multiple drive wheels. The area controller is used to control the rotation of the drive wheels. Baggage stacking equipment includes: The data acquisition module is used to acquire the storage status of pallets and real-time baggage data; The route planning module is used to determine the luggage palletizing coordinates and target palletizing pose based on the pallet storage status and luggage data, and to plan the luggage palletizing route based on the luggage palletizing coordinates. The pose control module is used to send a first control command to the pallet unit according to the baggage stacking route. The first control command is used to make the area controller control the drive wheel to rotate, so as to move the baggage to the position where the baggage stacking coordinates are located.

[0014] Thirdly, this application provides an airport unmanned vehicle, including the baggage stacking device described in the second aspect.

[0015] This application breaks down the complete baggage palletizing route into individual area controllers for control, accurately locates the pallet unit where the baggage is located, and independently controls the position and orientation of the baggage through the pallet unit, ensuring the efficiency and accuracy of baggage stacking. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the pallet structure of an airport unmanned vehicle baggage stacking method provided by an exemplary embodiment of this disclosure; Figure 2 This is a schematic diagram of the pallet unit from one perspective of an exemplary embodiment of an airport unmanned vehicle baggage stacking method provided in this disclosure. Figure 3 This is a schematic diagram of the pallet unit from another perspective of an exemplary embodiment of an airport unmanned vehicle baggage stacking method provided in this disclosure; Figure 4 This is a flowchart illustrating an airport unmanned vehicle baggage stacking method provided by an exemplary embodiment of this disclosure; Figure 5 This is a schematic flowchart of S101 of an airport unmanned vehicle baggage stacking method provided by an exemplary embodiment of this disclosure; Figure 6 This is a three-dimensional point cloud data map of a baggage stacking method for an airport unmanned vehicle provided by an exemplary embodiment of this disclosure; Figure 7 This is a downsampled diagram of a baggage stacking method for an airport unmanned vehicle provided by an exemplary embodiment of this disclosure; Figure 8 This is a baggage posture diagram of a baggage stacking method for an airport unmanned vehicle provided by an exemplary embodiment of this disclosure; Figure 9 This is a schematic flowchart of S102 of an airport unmanned vehicle baggage stacking method provided by an exemplary embodiment of this disclosure; Figure 10 This is a schematic flowchart of S103 of an airport unmanned vehicle baggage stacking method provided by an exemplary embodiment of this disclosure; Figure 11 This is a schematic diagram of the attitude adjustment process in S403 of an airport unmanned vehicle baggage stacking method provided by an exemplary embodiment of the present disclosure. Figure 12This is a schematic diagram of the adjustment of position in S403 of a baggage stacking method for an airport unmanned vehicle provided by an exemplary embodiment of this disclosure; Figure 13 This is a flowchart illustrating another embodiment of an airport unmanned vehicle baggage stacking method provided by the exemplary embodiments of this disclosure; Figure 14 This is a schematic diagram of the structure of a multi-layer pallet for a baggage stacking method for an airport unmanned vehicle provided in an exemplary embodiment of this disclosure; Figure 15 This is a schematic diagram of the structure of a baggage stacking device for an airport unmanned vehicle provided in an exemplary embodiment of this disclosure; Figure 16 This is a schematic diagram of the structure of an airport unmanned vehicle provided by an exemplary embodiment of the present disclosure; Figure 17 This is a flowchart illustrating the usage of an airport unmanned vehicle according to an exemplary embodiment of this disclosure; Figure 18 This is a flowchart illustrating the process of acquiring 3D point cloud data by an airport unmanned vehicle according to an exemplary embodiment of this disclosure; Figure 19 This is a flowchart illustrating the attitude adjustment of an airport unmanned vehicle according to an exemplary embodiment of this disclosure; Figure 20 This is a flowchart illustrating the adjustment of the position of an airport unmanned vehicle according to an exemplary embodiment of this disclosure; Figure 21 This is a flowchart illustrating the fine-tuning pose of an airport unmanned vehicle according to an exemplary embodiment of this disclosure.

[0018] Explanation of icon numbers: 100. Pallet; 110. Pallet Unit; 111. Pallet Rack; 112. Drive Wheel; 113. Photoelectric Sensor Array; 114. Drive Ball Bearing; 115. Motor; 116. Area Controller; 200. Baggage Stacking Device; 201. Data Acquisition Module; 202. Route Planning Module; 203. Pose Control Module; 300. Central Control Layer; 301. Main Control Computer; 302. Communication Module; 303. Power Management Controller; 400. Cargo Box Real-time Control Layer; 401. Vision Sensor; 402. Safety Monitoring System; 403. Feeding Device Controller; 500. Perception and Navigation Layer; 501. Map Calculation Unit; 502. LiDAR; 503. Navigation Module; 504. Inertial Measurement Unit; 600. Cargo Box; 601. Feed Inlet Drive Device; 602. Lifting Motor. Detailed Implementation

[0019] The technical solutions of 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. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] In the description of this application, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are used 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 or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0021] "A and / or B" includes the following three combinations: A only, B only, and a combination of A and B.

[0022] The use of "applies to" or "configured to" in this application implies open and inclusive language, which does not exclude the applicability to or configuration to devices performing additional tasks or steps. Additionally, the use of "based on" implies openness and inclusivity, because processes, steps, calculations, or other actions "based on" one or more of the stated conditions or values ​​may in practice be based on additional conditions or values ​​beyond those stated.

[0023] In this application, the term "exemplary" is used to mean "used as an example, illustration, or description." Any embodiment described as "exemplary" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use this application. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that this application can be made without using these specific details. In other instances, well-known structures and processes are not described in detail to avoid obscuring the description of this application with unnecessary detail. Therefore, this application is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0024] In one aspect, embodiments of this application provide a method for stacking luggage for an airport unmanned vehicle.

[0025] like Figure 1 As shown, the airport unmanned vehicle includes a pallet 100, which is composed of multiple pallet units 110 arranged in an array.

[0026] like Figure 2 As shown, the pallet unit 110 includes a pallet frame 111 and a plurality of drive wheels 112, which are mounted on the pallet frame 111.

[0027] The pallet 100 is the component of the airport's unmanned vehicle used to carry luggage. The pallet unit 110 is the smallest independent execution module of the pallet 100. They are evenly distributed in an array to form a complete pallet 100 plane and can move luggage independently or collaboratively. Each pallet unit 110 is equipped with an area controller 116 and multiple drive wheels 112. The area controller 116 controls the rotation of the drive wheels 112 to move the luggage horizontally or vertically.

[0028] like Figure 3 As shown, the pallet unit 110 also includes an area controller 116, a photoelectric sensor array 113, and a motor 115. The area controller 116 is used to control the rotation of the drive wheel 112, the photoelectric sensor array 113 is used to detect whether the luggage is above the pallet unit 110, and the motor 115 is used to control the rotation of the drive wheel 112.

[0029] In some embodiments, the pallet unit 110 further includes a drive ball 114, which assists luggage in moving on the pallet unit 110 and reduces resistance during luggage movement.

[0030] like Figure 4 As shown, the luggage stacking method includes the following steps: S101. Obtain the pallet storage status and real-time baggage data.

[0031] Specifically, the pallet storage status is used to characterize the space occupied by luggage on the current pallet, and the luggage data is used to identify the real-time location of luggage on the pallet, such as luggage type, size, volume, position, and posture.

[0032] By leveraging pallet storage status and real-time baggage data, we can provide a real-time and accurate data foundation for determining baggage palletizing coordinates and target palletizing poses, as well as for route planning in subsequent steps, thus ensuring the accuracy of calculations in these steps.

[0033] S102. Determine the luggage palletizing coordinates and target palletizing pose based on the pallet storage status and luggage data, and plan the luggage palletizing route based on the luggage palletizing coordinates.

[0034] Specifically, the baggage stacking coordinates are the spatial coordinates of the baggage's position on the pallet, and the target stacking pose is the angle and orientation of the baggage on the pallet. By using the baggage stacking coordinates and the target stacking pose, the utilization rate of the pallet space is maximized, and baggage overlap and exceeding the pallet boundary are avoided.

[0035] The baggage palletizing route is the path that baggage takes to move to the baggage palletizing coordinates. It is a collision-free and short-path baggage plan based on the storage status of the pallets, ensuring that the baggage movement is smooth and without interference.

[0036] Based on the storage status of pallets and baggage data, the entire baggage palletizing process is decided and route planned, and the basic data is transformed into executable control objectives so that the pallet units can execute them in subsequent steps.

[0037] S103. Send a first control command to the pallet unit according to the baggage stacking route. The first control command is used to make the area controller control the drive wheel to rotate so as to move the baggage to the location of the baggage stacking coordinates.

[0038] Specifically, the first control command is generated based on the baggage stacking route. The area controller parses the command to calculate the target speed, torque and direction of the drive wheels, and drives the baggage to move along the baggage stacking route to the baggage stacking coordinates.

[0039] This application breaks down the complete baggage palletizing route into individual area controllers for control, accurately locates the pallet unit where the baggage is located, and independently controls the position and orientation of the baggage through the pallet unit, ensuring the efficiency and accuracy of baggage stacking.

[0040] In some embodiments, the airport autonomous vehicle also includes a vision sensor positioned above the pallet. The vision sensor can be a 3D camera, such as a 3D speckle camera. The vision sensor is fixedly mounted directly above the pallet and the inlet, vertically oriented towards the pallet plane, and can acquire 3D point cloud and color images of the luggage.

[0041] like Figure 5 As shown, real-time baggage data is obtained, including: S201. Acquire the 3D point cloud data of the luggage collected by the visual sensor.

[0042] Specifically, 3D point cloud data is a collection of coordinate points acquired through visual sensors, used to represent the outline, shape, and spatial location of luggage. Acquiring 3D point cloud data via visual sensors avoids interference from luggage appearance and ambient lighting, ensuring accurate and stable data. The entire acquisition process is non-contact, preventing any contact with the luggage and avoiding scratches or damage.

[0043] S202. Process the 3D point cloud data based on 3D vision algorithms to obtain luggage data.

[0044] Specifically, baggage data includes baggage category, baggage size, baggage orientation, and baggage location. Baggage category refers to the identified baggage type, such as suitcase, travel bag, backpack, etc. Baggage size refers to the baggage's length, width, height, volume, and other geometric parameters. Baggage orientation refers to the baggage's rotation angle and spatial orientation on the pallet; baggage location refers to the baggage's position coordinates on the pallet.

[0045] In one example, the 3D vision algorithm could be the ADD-3D vision algorithm. like Figure 6 As shown, this is a 3D point cloud data of luggage directly above the conveyor belt at the inlet, collected by a visual sensor. From left to right, it shows the color image, depth image, and point cloud image of the luggage.

[0046] The acquisition frequency of the 3D point cloud data is linked to the running speed of the feed inlet conveyor belt; the acquisition frequency is set accordingly. satisfy: ; in The sampling frequency; The sampling frequency is expressed in frames per second (fps). The length of the minimum luggage is given in meters (m).

[0047] By controlling the acquisition frequency, at least three complete frames of point cloud data are collected for each piece of luggage to avoid data loss. The raw point cloud data obtained is represented as a set. ,in For the first The three-dimensional spatial coordinates of each point ( The coordinates are in the horizontal direction. The coordinates for the direction of conveyor belt movement. (Vertical height coordinates) This represents the total number of points in a single frame of the point cloud.

[0048] Preprocessing is required for 3D point cloud data. The raw point cloud data contains environmental noise (such as light reflection interference and conveyor belt surface texture interference) and invalid points (such as points outside the measurement range). Preprocessing steps are needed to filter out interference and optimize point cloud quality. Specifically, this includes two sub-steps: denoising and downsampling. Gaussian Denoising: A Gaussian filtering algorithm is used to remove high-frequency noise for each point in the point cloud. Its filtered coordinates The calculation formula is: ; in: For point The set of neighborhood points, The number of neighboring points; The Gaussian function is expressed as follows: ; The standard deviation is Gaussian, used to adjust the smoothness of the filter. For point With neighboring points The Euclidean distance is as follows: .

[0049] Downsampling: A voxel grid downsampling algorithm is used to reduce the amount of point cloud data, improve the efficiency of subsequent algorithms, and preserve the core geometric features of the luggage. The voxel grid size is set, and the point cloud space is divided into several voxels. The average coordinates of all points within each voxel are taken as the representative point of that voxel, resulting in a preprocessed point cloud set. ,in The total number of points in the point cloud after downsampling ( ).

[0050] Downsampling pairs, for example Figure 7 As shown.

[0051] By combining the feature extraction and model registration modules of the Point Cloud Library (PCL), baggage category recognition, 3D shape reconstruction, and pose analysis are completed. The specific implementation is as follows: Category Recognition: An improved PointNet deep learning model is used, with a preprocessed point cloud set P as input. The pcl::Feature module of the point cloud library extracts geometric features such as normal vectors and curvature, as well as texture features, which are then fed into the model to complete the recognition of luggage categories (e.g., suitcases, travel bags, backpacks, etc.). The core output of the model is the category probability vector, calculated using the following formula: ; in: The transformation network is used to spatially align the input point cloud, eliminating the influence of luggage placement posture on the recognition results, and outputting an aligned point cloud set. During spatial alignment, the PCL library function pcl::transformPointCloud is called to transform the point cloud coordinates. It is a multilayer perceptron used to extract high-order geometric and texture features of point clouds. The input features include normal vector features extracted from the PCL library (calculated through the pcl::NormalEstimation class), curvature features, and RGB texture features. The activation function maps the feature vector to the category probability vector y. The category corresponding to the element with the highest probability in y is the identified luggage category, denoted as C (C∈{suitcase, travel bag, backpack,...}).

[0052] Shape Analysis: Based on the preprocessed point cloud set P, the convex hull of the luggage point cloud is constructed using the pcl::ConvexHull point cloud library. The complete 3D shape of the luggage is obtained through a 3D reconstruction algorithm, and shape parameters (such as volume V, maximum circumscribed dimension, etc.) are extracted. Minimum external size The volume V is calculated using the convex hull integration method, relying on the convex hull vertex data from the PCL library. The formula is as follows: ; in, , , , The coordinates of the tetrahedron vertices on the convex hull surface of the luggage are obtained through the PCL library function pcl::ConvexHull::getVertices(), where K is the total number of tetrahedrons obtained from the convex hull decomposition.

[0053] Pose analysis: The PCL library's pcl::IterativeClosestPoint class is called, employing the Iterative Closest Point (ICP) algorithm to compare the identified baggage point cloud P with the standard point cloud model for that category of baggage. (Pre-stored in the system in pcl::PointXYZRGB format) Registration is performed to calculate the real-time spatial attitude parameters of the luggage (rotation matrix R, translation vector t). The core of the ICP algorithm is to minimize the Euclidean distance error between point clouds; the objective function is: ; in, For standard point cloud models The i-th point in the equation is given by R, a 3×3 rotation matrix (describing the spatial rotational attitude of the luggage), and t, a 3×1 translation vector (describing the spatial positional offset of the luggage). The equations are obtained iteratively through the ICP algorithm module of the PCL library.

[0054] The minimum value of the objective function E(R,t) is obtained by iteratively solving for it using the gradient descent method, resulting in the optimal rotation matrix R and translation vector t. The specific iterative steps (implemented using the ICP parameter configuration of the PCL library) are as follows: a1. Initialize the rotation matrix (Identity matrix), translation vector (Zero vector), set the initial transformation matrix of ICP using PCL library functions; a2. For the current iteration number k, call the corresponding function in the PCL library's ICP module to calculate the point cloud P and the transformed standard point cloud. The nearest matching pair; a3. Based on the matched pairs, find the solution that minimizes the error E(R,t). and This is automatically completed by the internal algorithm of the PCL library; a4. Calculate the error difference ,like ( The convergence threshold is set to 10. -6 If the values ​​are set using PCL library functions, the iteration terminates and the optimal R and t are output; otherwise, steps ②-③ are repeated.

[0055] Finally, the real-time attitude of the luggage in the world coordinate system can be determined using the rotation matrix R and the translation vector t. The attitude resolution error is ≤0.5°. The attitude of the luggage is as follows: Figure 8 As shown.

[0056] In some embodiments, such as Figure 9 As shown, the baggage palletizing coordinates and target palletizing pose are determined based on the pallet storage status and baggage data, and the baggage palletizing route is planned based on the baggage palletizing coordinates, including: S301. Calculate the pallet stacking coordinates and target stacking pose of the suitcase based on the pallet storage status and luggage size.

[0057] Specifically, based on the storage status of the pallets, the available areas on the pallets are determined, the available space that can accommodate the size of the luggage is selected, and the luggage stacking coordinates and target stacking positions are determined.

[0058] By calculating the luggage stacking coordinates and target stacking pose, spatial matching of luggage is achieved to meet space utilization requirements and increase luggage loading capacity.

[0059] S302. Starting from the current location of the luggage and ending at the location of the luggage stacking coordinates, plan the luggage stacking route based on the location boundaries of the luggage already stacked in the pallet storage status.

[0060] Specifically, based on the current location of the luggage and the location of the luggage stacking coordinates as the endpoint, and based on the position boundaries of the luggage already stacked in the pallet storage state, a collision-free and interference-free luggage stacking route is planned to ensure smooth luggage movement and avoid collisions and squeezing during the luggage movement process.

[0061] In some embodiments, the drive wheels are evenly distributed around a center point, and the directions of the drive wheels include at least lateral and longitudinal directions. In some examples, there are four drive wheels, evenly distributed around the center point. Two of the drive wheels rotate laterally, and the other two rotate longitudinally, with the lateral drive wheels adjacent to the longitudinal drive wheels. This embodiment uses the differential rotation of the different drive wheels to drive the luggage above to move laterally, longitudinally, or rotate.

[0062] like Figure 10 As shown, a first control command is sent to the pallet unit according to the baggage palletizing route, including: S401. Determine the pallet unit where the luggage is currently located based on its location.

[0063] Specifically, the location of the luggage is obtained through visual sensors to determine the current pallet unit where the luggage is located, thereby identifying the pallet unit that needs to be controlled.

[0064] S402. Determine the target area that the luggage needs to be moved to on the pallet unit according to the luggage stacking route, and generate the first control command to send to the area controller of the pallet unit.

[0065] Specifically, the baggage palletizing route is broken down into the unit movement route for each pallet unit along the baggage palletizing route, and the first control command for each pallet unit is determined based on the unit movement route. The target area is the target position for the unit movement route.

[0066] S403. The area controller controls the rotation of the drive wheels to adjust the posture of the luggage and move the luggage to the target area along the rotation direction of the drive wheels.

[0067] Specifically, the area controller parses the first control command and controls the drive wheels to rotate. The simultaneous rotation of the four drive wheels can drive the luggage to rotate. The simultaneous rotation of two drive wheels in the same direction can drive the luggage to move laterally or longitudinally, moving the luggage to the target area along the unit movement route.

[0068] S404: Sequentially control all pallet units on the baggage palletizing route to move the baggage to the location of the baggage palletizing coordinates.

[0069] Specifically, after the luggage moves from one pallet unit to the target area of ​​that pallet unit, it enters the starting position of the next pallet unit on the luggage palletizing route, and returns to step S402 to generate the first control command for the next pallet unit. All pallet units on the luggage palletizing route are controlled sequentially, thereby moving the luggage to the location specified by the luggage palletizing coordinates.

[0070] In some embodiments, such as Figure 11As shown, the rotation of the drive wheels is controlled by the area controller to adjust the posture of the luggage, including: S501: The first control command is parsed by the area controller to determine the target posture of the luggage.

[0071] Specifically, the target posture refers to the angle and orientation of the luggage after it has moved within the current pallet unit. The area controller parses the first control command and identifies the target posture of the luggage, which is then used to drive the rotation of the drive wheels in subsequent steps.

[0072] S502: Drive the corresponding transmission wheel to rotate through the area controller so that multiple transmission wheels work together. Drive the luggage to rotate through the differential rotation of the transmission wheels, and obtain the luggage posture in real time based on the luggage data obtained by the vision controller.

[0073] Specifically, the angle and direction of the luggage's rotation are determined based on the luggage's posture and the target posture. The luggage is rotated by four transmission wheels at different speeds or torques, and the speed difference is used to achieve the luggage's in-situ turning. The current posture of the luggage is obtained in real time based on the luggage data.

[0074] S503, the area controller determines the attitude deviation based on the luggage attitude and the target attitude.

[0075] Specifically, the area controller calculates the attitude deviation between the current luggage posture and the target posture to determine whether the luggage posture meets the requirements.

[0076] S504. When the attitude deviation is greater than the preset attitude threshold, the area controller re-drives the corresponding transmission wheel to rotate according to the attitude deviation.

[0077] Specifically, the preset attitude threshold is the critical value of the attitude deviation. When the attitude deviation exceeds the preset attitude threshold, the drive wheel needs to be re-driven to rotate, causing the luggage to rotate and reduce the attitude deviation. The preset attitude threshold can be 0.5°.

[0078] S505. When the posture deviation is less than the preset posture threshold, the posture adjustment of the luggage is completed.

[0079] Specifically, the luggage's posture is adjusted when the posture deviation is less than a preset posture threshold, and the posture adjustment of the luggage is achieved through a separate area controller.

[0080] In some embodiments, such as Figure 12 As shown, the area controller controls the rotation of the drive wheels to move the luggage to the target area along the direction of the drive wheel rotation, including: S601. The first control command is parsed by the area controller to determine the target location of the luggage.

[0081] Specifically, the target location is the coordinate position of the luggage after it has moved within the current pallet unit. The area controller parses the first control command, identifies the target location of the luggage, and uses this information to drive the movement of the drive wheels in subsequent steps.

[0082] S602: Drive the corresponding drive wheels to rotate through the area controller so that multiple drive wheels work together to move the luggage horizontally or vertically, and obtain the luggage position in real time based on the luggage data.

[0083] Specifically, the path that the luggage needs to move is determined based on the luggage location and the target location. The luggage is moved laterally or longitudinally by rotating two wheels that rotate in the same direction, and the luggage location is obtained in real time based on the luggage data obtained from the vision controller.

[0084] S603, the area controller determines the positional deviation based on the luggage's location and the target location.

[0085] Specifically, the area controller calculates the positional deviation between the current luggage position and the target position to be reached on the current plane to determine whether the luggage's posture meets the requirements.

[0086] S604. If the position deviation is greater than the preset position threshold, the area controller will re-drive the corresponding transmission wheel to rotate according to the position deviation.

[0087] Specifically, the preset position threshold is the critical value of the position deviation. When the position deviation is greater than the preset position threshold, the drive wheel needs to be re-driven to rotate, so that the luggage can be moved laterally or longitudinally to reduce the position deviation.

[0088] S605. If the position deviation is less than the preset position threshold, the position of the luggage is adjusted.

[0089] Specifically, the luggage position is adjusted when the position deviation is less than a preset position threshold, and the luggage position adjustment is achieved through a separate area controller.

[0090] In some embodiments, such as Figure 13 As shown, the pallet unit also includes a photoelectric sensor array, which is used to detect whether there is luggage above the current pallet unit.

[0091] Luggage stacking methods also include: S701. Obtain the scanning data periodically collected by the photoelectric sensor array in the pallet unit where the luggage stacking coordinates are located.

[0092] Specifically, photoelectric sensors in the pallet unit where the baggage is located periodically collect scanning data to determine whether the baggage has moved to the pallet unit where the baggage is located. The photoelectric sensors can scan synchronously once every 300ms to acquire scanning data at a fixed frequency.

[0093] S702. When the luggage arrives at the luggage stacking coordinates, the scan data is fused with the luggage data to determine the luggage's occupancy status and stacking position at the luggage stacking coordinates.

[0094] Specifically, the scanning data obtained by photoelectric sensors and the luggage data obtained by visual sensors are fused to calculate and determine the luggage size, posture and position. The fusion of multi-sensor data eliminates the error of single perception, accurately determines the luggage stacking status, and provides a reliable basis for subsequent deviation calculation.

[0095] S703. Determine the deviation value based on the palletizing pose and the target palletizing pose, generate a second control command based on the deviation value, and send the second control command to the pallet unit.

[0096] Specifically, the palletizing pose refers to the position and orientation of the baggage after it arrives at the baggage palletizing coordinates. After the baggage arrives at the baggage palletizing coordinates, its position and orientation may deviate from the target position and orientation. At this time, the deviation value is determined based on the current palletizing pose and the target palletizing pose. If the deviation value is less than the preset position threshold and the preset orientation threshold, no further adjustment is needed, and the baggage is stacked. If the deviation value is greater than the preset position threshold and the preset orientation threshold, the baggage's orientation needs to be further fine-tuned. At this time, a second control command is generated based on the deviation value and sent to the pallet unit where the baggage is located at the palletizing coordinates.

[0097] S704. The area controller parses the second control command and controls the transmission wheel to adjust the palletizing posture so that the deviation value is less than the preset deviation threshold.

[0098] Specifically, the area controller parses the second control command and uses four transmission wheels to rotate at different speeds or with different torques. By using the speed difference, the luggage can be turned in place to adjust its posture, or two rotating wheels rotating in the same direction can be used to move the luggage laterally or longitudinally to adjust its position.

[0099] S705. If the deviation value is less than the preset deviation threshold, obtain the final pose of the luggage and update the pallet storage status according to the final pose to complete the luggage stacking.

[0100] If the deviation value is less than the preset position threshold and preset attitude threshold, no further adjustment is needed, and the baggage stacking is complete. At this point, the current baggage position and size are updated in the pallet storage status, ready for the next baggage to be stored.

[0101] In some embodiments, the pallet 100 in the airport driverless vehicle has multiple layers, such as in some examples, like Figure 14 As shown, pallet 100 has three layers, and each layer of pallet 100 can perform the above-described baggage stacking method.

[0102] After sending a first control command to the area controller according to the baggage stacking route, the baggage stacking method further includes: Determine the pallet storage status. If the pallet storage status indicates that the current pallet is full, use the next pallet to receive the baggage.

[0103] Specifically, after each piece of luggage is stacked, it is determined whether there is still space in the pallet to store new luggage. If the pallet is full, it is indicated that the pallet is full, and luggage is received through the next pallet, improving the flexibility and storage capacity of luggage stacking.

[0104] Secondly, this application provides a baggage stacking device for an airport unmanned vehicle. The airport unmanned vehicle includes a pallet 100, which is composed of multiple arrayed pallet units 110. Each pallet unit 110 includes an area controller 116 and multiple drive wheels 112. The area controller 116 is used to control the rotation of the drive wheels 112.

[0105] like Figure 15 As shown, the luggage stacking device 200 includes a data acquisition module 201, a route planning module 202, and a pose control module 203.

[0106] The data acquisition module 201 is used to acquire the storage status of pallets 100 and real-time baggage data. This data provides a real-time and accurate foundation for subsequent steps, including determining baggage palletizing coordinates and target palletizing poses, as well as route planning, ensuring the accuracy of calculations in subsequent steps.

[0107] The route planning module 202 is used to determine the baggage palletizing coordinates and target palletizing pose based on the storage status of the pallet 100 and baggage data, and to plan the baggage palletizing route based on the baggage palletizing coordinates. It completes the decision-making and path planning for the entire baggage palletizing process based on the storage status of the pallet 100 and baggage data, and converts the basic data into executable control objectives so that the pallet unit 110 can execute them in subsequent steps.

[0108] The pose control module 203 sends a first control command to the pallet unit 110 according to the baggage stacking route. The first control command causes the area controller 116 to control the drive wheel 112 to rotate, thereby moving the baggage to the position where the baggage is located at the stacking coordinates. The complete baggage stacking route is decomposed and controlled by each area controller 116, accurately locating the pallet unit 110 where the baggage is located, and independently controlling the position and posture of the baggage through the pallet unit 110, ensuring the efficiency and accuracy of baggage stacking.

[0109] Thirdly, this application provides an airport unmanned vehicle, such as... Figure 16 As shown, the airport unmanned vehicle includes a central control layer 300, a cargo compartment real-time control layer 400, a perception and navigation layer 500, and a cargo compartment 600.

[0110] The central control layer 300 includes a main control computer 301, a communication module 302, and a power management controller 303.

[0111] The main control computer 301 coordinates all subsystems, analyzes task allocation resources, processes sensing data, and makes decisions. The communication module 302 ensures real-time data exchange between all subsystems. The power management controller 303 controls the overall power distribution.

[0112] The cargo compartment real-time control layer 400 includes a vision sensor 401, a photoelectric sensor array 113, a safety monitoring system 402, an area controller 116, and a feeding device controller 403.

[0113] A vision sensor 401 is used to acquire 3D point cloud data of the luggage. A photoelectric sensor array 113 is used to sense whether luggage is present above the pallet unit 110 where the luggage palletizing coordinates are located. A safety monitoring system 402 is used to identify fault codes fed back by various systems, and can report system faults in real time, generate fault logs, and ensure operational safety. An area controller 116 is installed in each pallet unit 110 to control the rotation of the drive wheel 112. A feeding device controller 403 is used to control the feeding port drive device 601 and the lifting motor 602.

[0114] The perception and navigation layer 500 includes a map calculation unit 501, a lidar 502, a navigation module 503, and an inertial measurement unit 504.

[0115] The map computing unit 501 is used for localization using the SLAM algorithm, the lidar 502 is used to build an environmental map to achieve real-time localization and obstacle avoidance, the navigation module 503 is used for path planning of the airport unmanned vehicle, and the inertial measurement unit 504 is used for load localization to provide accurate attitude and heading.

[0116] The cargo box 600 includes multi-layer pallets 100, a feed inlet transmission device 601, and lifting electrodes.

[0117] The pallet 100 consists of multiple pallet units 110 arranged in an array. Each pallet unit 110 includes an area controller 116 and multiple drive wheels 112. The area controller 116 is used to control the rotation of the drive wheels 112.

[0118] The pallet 100 is the component of the airport's unmanned vehicle used to carry luggage. The pallet unit 110 is the smallest independent execution module of the pallet 100. They are evenly distributed in an array to form a complete pallet 100 plane and can move luggage independently or collaboratively. Each pallet unit 110 is equipped with an area controller 116 and multiple drive wheels 112. The area controller 116 controls the rotation of the drive wheels 112 to move the luggage horizontally or vertically.

[0119] The drive wheels 112 are evenly distributed around a center point. The directions of the drive wheels 112 include at least the horizontal and the vertical. In some examples, there are four drive wheels 112 evenly distributed around the center point. The directions of two adjacent drive wheels 112 are perpendicular to each other. That is, the two drive wheels 112 adjacent to one of the drive wheels 112 with the horizontal rotation direction are both in the vertical direction. There are two drive wheels 112 with the horizontal direction and two drive wheels 112 with the vertical direction. By the differential rotation of the different drive wheels 112, the luggage above can be driven to move horizontally, vertically or rotate.

[0120] Each drive wheel 112 corresponds to a motor 115. The area controller 116 is electrically connected to the motor 115 to control the rotation of each drive wheel 112.

[0121] In addition, airport unmanned vehicles also include systems such as chassis drive systems and support systems. For example, a thermal management system heats and insulates the battery and key components, while a protection system protects the airport unmanned vehicle. These are all existing technologies and will not be elaborated upon here.

[0122] like Figure 17 As shown, the main control computer 301 is used to perform the following steps: The feeding device controller 403 controls the lifting motor 602 to align the feeding port transmission device 601 with the uppermost pallet 100 and the airport conveyor belt. A vision sensor 401 monitors the baggage. When baggage arrives at the designated handling area, the feeding port transmission device 601 is triggered to transport the baggage at a matching speed. The vision sensor 401 monitors whether the baggage has arrived at the feeding port. After arrival, the baggage is stacked. Once stacking is complete, the system determines whether the task is finished. If not, it checks whether the current pallet 100 is fully stacked. If not, it waits and processes the next baggage. If the current pallet 100 is fully stacked, the feeding device controller 403 controls the lifting motor 602 to align the feeding port transmission device 601 with the next pallet 100.

[0123] like Figure 18 As shown, the main control computer 301 is also used to perform the following steps: The baggage detection mode is initiated by starting the baggage recognition model via the visual sensor 401. When the visual sensor 401 at the baggage inlet detects baggage, it acquires depth maps and point cloud data, performs point cloud processing, classifies baggage based on a deep learning model, and calculates the baggage edge volume, size, and pose information to obtain structured 3D point cloud data. Simultaneously, the visual sensor 401 above the pallet 100 is activated. The visual sensor 401 above the pallet 100 acquires depth maps and point cloud data, performs point cloud processing, classifies baggage based on a deep learning model, and calculates the baggage edge volume, size, and pose information to obtain structured 3D point cloud data. This structured point cloud data is used to plan the baggage palletizing path based on the 3D point cloud data from the visual sensor 401 at the baggage inlet and the visual sensor 401 above the pallet 100, and then sends the planned palletizing path to the area controller 116.

[0124] like Figure 19 As shown, the area controller 116 is used to perform the following steps: The area controller 116 awaits adjustment commands. Upon receiving a luggage posture adjustment command from the main control computer 301, it parses the command and locks onto the area of ​​the pallet unit 110 where the target is located. It generates control commands and assigns them to the motors 115 of the corresponding drive wheels 112, causing each drive wheel 112 to work in tandem, achieving in-situ turning, rotation, or translation of the luggage through differential rotation. It reads encoder feedback in real time for closed-loop position control. If an abnormality is detected in motor 115, it stops motor 115 and reports the abnormality. If no abnormality is detected in motor 115, it acquires the real-time luggage posture from the vision sensor 401, calculates the deviation between the luggage posture and the target posture, adjusts the luggage posture to ensure the deviation is less than the allowable error range, and stops motor 115 from driving the drive wheels 112.

[0125] The area controller 116 controls multiple omnidirectional wheel motors 115 to adjust the posture of luggage, enabling luggage to turn 360° in place and multiple pieces of luggage to move simultaneously.

[0126] like Figure 20 As shown, the main control computer 301 is also used to perform the following steps: A control command is sent to the area controller 116, which parses the command to obtain the baggage identification and target baggage stacking position. It then reads the storage status of the pallet 100 to obtain a real-time occupancy map and plans a collision-free stacking route for each baggage. The area controller 116 decomposes the baggage stacking route into longitudinal and lateral drive sequences, assigning them to the corresponding drive wheels 112 to move the baggage laterally or longitudinally along the path. Closed-loop trajectory tracking is performed using encoder feedback. If path blockage or severe deviation is detected, the drive wheels 112 stop moving and a fault is reported. After the baggage moves to the baggage stacking coordinates, it is fine-tuned and aligned using the pallet unit 110 where the stacking coordinates are located. This completes the stacking operation, updates the storage status of the pallet 100, and continues until all baggage is stacked.

[0127] like Figure 21 As shown, the main control computer 301 is also used to perform the following steps: The photoelectric sensor array 113 on the pallet unit 110 periodically scans to obtain scan data, performs data fusion calculation to determine the real-time occupancy status of the palletizing area, calculates the positional and orientation deviations of the luggage, and determines whether the positional and orientation deviations exceed the threshold. If the positional and orientation deviations exceed the threshold, the luggage position needs to be fine-tuned. Based on the deviation, the adjustment value is calculated, and a fine-tuning command is sent to the area controller 116 so that the area controller 116 drives the transmission wheel 112 to rotate, thereby adjusting the position or orientation of the luggage to meet the threshold range.

[0128] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0129] The above provides a detailed description of the baggage stacking method, apparatus, and airport unmanned vehicle provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for stacking luggage in an airport unmanned vehicle, characterized in that, The airport unmanned vehicle includes a pallet, which is composed of multiple pallet units arranged in an array. Each pallet unit includes an area controller and multiple drive wheels. The area controller is used to control the rotation of the drive wheels. The luggage stacking method includes the following steps: Obtain pallet storage status and real-time baggage data; The luggage palletizing coordinates and target palletizing pose are determined based on the pallet storage status and the luggage data, and the luggage palletizing route is planned based on the luggage palletizing coordinates. A first control command is sent to the pallet unit according to the baggage stacking route. The first control command is used to cause the area controller to control the rotation of the drive wheel to move the baggage to the location of the baggage stacking coordinates.

2. The luggage stacking method according to claim 1, characterized in that, The airport unmanned vehicle also includes a vision sensor, which is positioned above the pallet. Obtain real-time baggage data, including: Acquire the three-dimensional point cloud data of the luggage collected by the visual sensor; The three-dimensional point cloud data is processed based on a 3D vision algorithm to obtain the luggage data, which includes luggage category, luggage size, luggage posture, and luggage location.

3. The luggage stacking method according to claim 2, characterized in that, Based on the pallet storage status and the baggage data, baggage palletizing coordinates and target palletizing pose are determined, and a baggage palletizing route is planned based on the baggage palletizing coordinates, including: The pallet storage status and luggage size are used to calculate the luggage palletizing coordinates and target palletizing pose to fit the luggage. Starting from the current location of the luggage and ending at the location of the luggage stacking coordinates, a luggage stacking route is planned based on the position boundaries of the luggage already stacked in the pallet storage state.

4. The luggage stacking method according to claim 3, characterized in that, The transmission wheels are evenly distributed around a central point. Send a first control command to the pallet unit according to the baggage palletizing route, including: The pallet unit where the luggage is currently located is determined based on the luggage's location; Based on the baggage stacking route, determine the target area where the baggage needs to be moved on the pallet unit, generate the first control command and send it to the area controller of the pallet unit; The area controller controls the rotation of the drive wheel to adjust the posture of the luggage and move the luggage to the target area along the rotation direction of the drive wheel; All pallet units on the baggage palletizing route are controlled sequentially to move the baggage to the location of the baggage palletizing coordinates.

5. The luggage stacking method according to claim 4, characterized in that, Controlling the rotation of the drive wheels via the area controller to adjust the posture of the luggage includes: The target posture of the luggage is determined by parsing the first control command through the area controller; The area controller drives the corresponding transmission wheel to rotate, so that multiple transmission wheels work together. The differential rotation of the transmission wheels drives the luggage to rotate, and the luggage posture is obtained in real time based on the luggage data. The area controller determines the attitude deviation based on the luggage attitude and the target attitude; If the attitude deviation is greater than a preset attitude threshold, the area controller will re-drive the corresponding transmission wheel to rotate based on the attitude deviation. When the posture deviation is less than the preset posture threshold, the posture adjustment of the luggage is completed.

6. The luggage stacking method according to claim 4, characterized in that, Controlling the rotation of the drive wheels via the area controller to move the luggage along the rotation direction of the drive wheels to the target area includes: The area controller parses the first control command to determine the target location of the luggage; The area controller drives the corresponding transmission wheel to rotate, so that multiple transmission wheels work together to move the luggage horizontally or vertically, and the luggage position is obtained in real time based on the luggage data; The area controller determines the positional deviation based on the luggage's location and the target location; If the position deviation is greater than a preset position threshold, the area controller will re-drive the corresponding transmission wheel to rotate based on the position deviation. If the position deviation is less than the preset position threshold, the position adjustment of the luggage is completed.

7. The luggage stacking method according to claim 4, characterized in that, The pallet unit also includes an array of photoelectric sensors; The luggage stacking method also includes: Obtain the scanning data periodically collected by the photoelectric sensor array in the pallet unit where the luggage stacking coordinates are located; When the luggage arrives at the luggage stacking coordinates, the scan data and the luggage data are fused to determine the luggage's occupancy status and stacking pose at the luggage stacking coordinates; A deviation value is determined based on the palletizing pose and the target palletizing pose, and a second control command is generated based on the deviation value and sent to the pallet unit. The area controller parses the second control command and controls the transmission wheel to adjust the palletizing posture so that the deviation value is less than a preset deviation threshold. If the deviation value is less than a preset deviation threshold, the final pose of the luggage is obtained and the pallet storage status is updated according to the final pose to complete the stacking of the luggage.

8. The baggage stacking method according to claim 2, characterized in that, The pallet has multiple layers; After sending a first control command to the area controller according to the baggage stacking route, wherein the first control command is used to cause the area controller to control the rotation of the drive wheel to move the baggage to the location of the baggage stacking coordinates, the baggage stacking method further includes: Determine the storage status of the pallet, and if the storage status indicates that the current pallet is full, use the next layer of pallets to receive the luggage.

9. A baggage stacking device for an airport unmanned vehicle, characterized in that, The airport unmanned vehicle includes a pallet, which is composed of multiple pallet units arranged in an array. Each pallet unit includes an area controller and multiple drive wheels. The area controller is used to control the rotation of the drive wheels. The baggage stacking device includes: The data acquisition module is used to acquire the storage status of pallets and real-time baggage data; The route planning module is used to determine the luggage palletizing coordinates and target palletizing pose based on the pallet storage status and luggage data, and to plan the luggage palletizing route based on the luggage palletizing coordinates. The pose control module is used to send a first control command to the pallet unit according to the baggage stacking route. The first control command is used to cause the area controller to control the rotation of the drive wheel to move the baggage to the position where the baggage stacking coordinates are located.

10. An airport unmanned vehicle, characterized in that, Includes the luggage stacking device as described in claim 9.

Images