Stereoscopic warehouse checking method and system based on unmanned aerial vehicle

Abstract

The invention relates to a stereoscopic warehouse checking method and system based on an unmanned aerial vehicle, and the method comprises the steps: receiving a cargo checking task, and obtaining the layout information of a stereoscopic warehouse; determining the position of a to-be-checked goods location in the goods checking task in the stereoscopic warehouse; according to the layout information of the stereoscopic warehouse and the position of the to-be-checked goods location in the stereoscopic warehouse, determining a safe flight area and a hovering point of the unmanned aerial vehicle based on a path planning algorithm, and generating a flight task; issuing a flight task to the unmanned aerial vehicle, and calculating pose information of the unmanned aerial vehicle during execution of the flight task in real time through an inertial measurement unit; based on the observation image of the unmanned aerial vehicle visual system, correcting the accumulated error of the pose information through a Kalman filtering algorithm; rGB color images and depth images collected by the unmanned aerial vehicle at the to-be-checked goods locations are obtained in real time; determining whether the box stack is in an abnormal condition according to the depth image; and for the box stack which is not in the abnormal working condition, the number of goods in the box stack is determined according to the RGB color image.

Description

Technical Field

[0001] The present application relates to the technical field of warehouse inventory counting, and in particular to a method and system for inventory counting in a stereoscopic warehouse based on an unmanned aerial vehicle. Background Art

[0002] In warehouse management, especially for automated warehouse management, inventory work is an indispensable basic link to ensure efficient operation. Regular, daily and irregular inventory modes enable enterprises to monitor inventory status, assist in business decision-making, and optimize the overall warehouse management process. In order to achieve efficient inventory management, the inventory system should have the characteristics of high precision, non-continuous operation, and low time occupancy rate to ensure that the impact on regular business can be minimized even during the inventory, while maintaining a high degree of flexibility and system integration.

[0003] In the prior art, inventory counting in automated warehouses is generally achieved through manual visual counting and RFID (Radio Frequency Identification) counting. Manual visual counting refers to the warehouse manager directly observing and recording the status and quantity of the inventory in the warehouse to ensure the accuracy of the inventory data. RFID counting refers to automatically reading the RFID tag information attached to the items through radio frequency identification technology.

[0004] Manual visual inventory is inefficient and consumes a lot of time and human resources. It is prone to misreading, omissions or writing errors due to reliance on manual operations, which affects the accuracy of inventory data. At the same time, the inventory process is time-consuming and cannot provide real-time inventory status updates, which may lead to delayed management decisions. Frequent manual inventory counts will increase labor costs, and warehouse operations may need to be suspended during the inventory count, indirectly increasing operating costs.

[0005] Although RFID inventory has improved efficiency and accuracy, it also has some disadvantages. For example, the initial system construction and labeling costs are high, the reading of items made of special materials such as metals or liquids may be unstable, and signal interference may cause reading errors in dense storage environments. In addition, the effective operation of the RFID system depends on good technical support and maintenance. Once a technical failure occurs, it may affect the smooth progress of the entire inventory process. If the label is damaged or falls off, the product information cannot be correctly identified. Summary of the invention

[0006] In order to at least to some extent overcome the problems of high cost, high maintenance difficulty and low stability of the inventory system in the automated stereoscopic warehouse inventory method in the related art, the present application provides a stereoscopic warehouse inventory method and system based on drones.

[0007] The scheme of this application is as follows: According to a first aspect of an embodiment of the present application, a method for counting inventory in a high-rise warehouse based on a drone is provided, comprising: Receive cargo inventory tasks and obtain layout information of the three-dimensional warehouse; Determine the location of the cargo to be counted in the cargo inventory task in the three-dimensional warehouse; According to the layout information of the three-dimensional warehouse and the location of the goods to be counted in the three-dimensional warehouse, based on the path planning algorithm, the safe flight area and hovering point of the drone are determined, and the flight mission is generated; Sending the flight mission to the drone, and calculating the position and posture information of the drone in real time when performing the flight mission through an inertial measurement unit; Based on the observation image of the UAV vision system, the accumulated error of the posture information is corrected by a Kalman filter algorithm; Real-time acquisition of RGB color images and depth images collected by drones at each location to be counted; Determining whether the box stack is in an abnormal working condition according to the depth image; For a box stack that is not in an abnormal working condition, the quantity of goods in the box stack is determined based on the RGB color image.

[0008] Preferably, obtaining the layout information of the stereoscopic warehouse includes: Obtain a three-dimensional drawing of the stereoscopic warehouse, and determine the number of shelves in the stereoscopic warehouse, the volume parameters of each shelf, the position of the stacker track, the width of the shelf aisle, and the specific location of the entrance and exit based on the three-dimensional drawing of the stereoscopic warehouse; Obtain on-site investigation and verification results, and verify the three-dimensional drawings of the stereoscopic warehouse according to the on-site investigation and verification results.

[0009] Preferably, the method further comprises: Bind each shelf channel in the stereoscopic warehouse to each drone one by one; According to the layout information of the shelf aisle and the position of the goods to be counted in the shelf aisle, based on the path planning algorithm, the safe flight area and hovering point of the drone in its corresponding shelf aisle are determined, and the flight mission of the drone is generated.

[0010] Preferably, the method further comprises: When the UAV is performing a flight mission, the stacker in the shelf channel corresponding to the UAV is controlled to withdraw to the entrance and exit and suspend work.

[0011] Preferably, based on the observation image of the UAV vision system, the accumulated error of the posture information is corrected by a Kalman filter algorithm, including: Constructing a state covariance matrix, and performing error covariance prediction on the posture information through a Kalman filter algorithm; Determine whether the current observation image of the UAV vision system is a key frame; If it is a key frame, the state covariance matrix is ​​expanded, the feature points in the key frame are tracked and removed, and new feature points are extracted; Determine whether the tracking of feature points is completed; If the tracking of the feature point has been completed, compare the length of the tracking frame with the minimum tracking threshold; If the length of the tracking frame is greater than the minimum tracking threshold, the three-dimensional coordinates of the feature points are calculated, the reprojection errors of the feature points are calculated according to the three-dimensional coordinates of the feature points to obtain observation information, and the accumulated errors of the posture information are corrected according to the observation information; If the tracking of the feature point is not finished, compare the length of the tracking frame with the maximum tracking threshold; If the length of the tracking frame is greater than the maximum tracking threshold, the three-dimensional coordinates of the feature points are calculated, and the tracking frames are screened. In the screened tracking frames, the reprojection errors of the feature points are calculated according to the three-dimensional coordinates of the feature points to obtain observation information, and the accumulated error of the posture information is corrected according to the observation information.

[0012] Preferably, the method further comprises: Construct the state update estimation equation from time k-1 to time k: ; in, represents the system state estimate at time k; represents the system state estimate at time k-1; represents the gain matrix at time k, represents the measurement value or residual at time k; Represents the measurement model matrix at time k; The state vector Decomposed into the inertial measurement unit vector part and the drone vision system vector part, the state vector matrix is ​​expressed as: ; in, Represents the state vector of the entire system; represents the vector part of the inertial measurement unit; Represents the vector part of the drone vision system; The inertial measurement unit vector is expressed as: ; in; Represents the carrier attitude quaternion; represents the velocity vector; represents the position vector; Indicates the gyroscope zero bias; Indicates the accelerometer zero bias; The quaternion representing the installation deviation angle between the drone vision system and the inertial measurement unit; Represents the UAV vision system and inertial measurement unit arm vector; The error state vector of the inertial measurement unit is expressed as: ; in; Represents the rotation angle error of the inertial measurement unit relative to the world coordinate system; Represents the angular velocity error of the inertial measurement unit relative to the world coordinate system; Represents the translation error from the world coordinate system to the inertial measurement unit; Represents the gyro bias of the inertial measurement unit; Represents the accelerometer bias of the inertial measurement unit; Represents the rotation angle error of the UAV vision system relative to the inertial measurement unit; represents the translation error from the inertial measurement unit to the UAV vision system; The angular velocity error of the inertial measurement unit relative to the world coordinate system is expressed as: ; in, and Represents the rotation matrix between the world coordinate system W and the inertial measurement unit; represents the velocity vector; The state prediction model of Kalman filter is expressed as: ; in; Increment representing the rate of change of the inertial measurement unit state; represents the state transfer matrix; Represents the increment of the state variable; represents the control input matrix; represents process noise; State transition matrix It is expressed as: ; in; represents the angular velocity of the Earth's rotation; Represents the direction cosine matrix from the carrier coordinate system to the world coordinate system; Represents the gravitational acceleration in the world coordinate system; The conversion formula of gravity acceleration is expressed as: ; Direction cosine matrix from earth coordinate system to world coordinate system It is expressed as: ; in; represents the gravitational acceleration in the earth coordinate system; Indicates the latitude of the world coordinate system relative to the earth coordinate system; Represents the longitude of the world coordinate system relative to the earth coordinate system; Inertial Measurement Unit Noise Transfer Matrix It is expressed as: ; White noise vector It is expressed as: ; in; represents the white noise of the gyroscope; represents the white noise of the accelerometer; White noise representing gyroscope bias; White noise representing the accelerometer bias; White noise representing attitude angle; White noise representing position; The UAV vision system vector is represented as: ; in, Indicates the quaternion rotation attitude from the world coordinate system to the drone vision system coordinate system in the first frame; Indicates the position of the drone vision system in the world coordinate system at the first frame; Indicates the quaternion rotation attitude from the world coordinate system to the drone vision system coordinate system at the Nth frame; Indicates the position of the drone vision system in the world coordinate system at the Nth frame; The error state vector related to the UAV vision system is expressed as: ; in, and Represents the attitude error and position error of the UAV vision system in the first frame; Indicates the rotation error from the world coordinate system to the drone vision system coordinate system at the Nth frame; Indicates the position error of the drone vision system in the world coordinate system at the Nth frame; The state error vector of the N+1th frame is: ; ; in, Indicates the rotation error from the world coordinate system to the drone vision system coordinate system at the N+1th frame; Indicates the position error of the UAV vision system in the world coordinate system at the N+1th frame; represents the 3×3 identity matrix; represents the 3×3 identity matrix; Represents the rotation matrix from the inertial measurement unit coordinate system to the world coordinate system; represents the position of the drone vision system in the inertial measurement unit coordinate system, Represents the operation of converting a vector into an antisymmetric matrix; The augmented state covariance matrix It is expressed as: ; ; ; in, represents the state covariance matrix after augmentation; represents the identity matrix of order 6N+15; represents the Jacobian matrix, which is used to describe the relationship between state error vectors; represents the original state covariance matrix; represents the covariance between the UAV vision system state error and the conversion error from the inertial measurement unit coordinate system to the UAV vision system coordinate system in the N+1th frame; Represents the covariance of the UAV visual system state error itself in the N+1th frame; Calculate the estimated 3D coordinates of the jth feature point in the i-th frame image in the drone vision system coordinate system: ; in, represents the estimated 3D coordinate value of the j-th feature point in the coordinate system of the UAV vision system in the i-th frame image; Indicates the X-axis coordinate of the feature point in the UAV vision system coordinate system; Indicates the Y-axis coordinate of the feature point in the UAV vision system coordinate system; Indicates the Z-axis coordinate of the feature point in the UAV vision system coordinate system; Represents the rotation matrix from the world coordinate system to the coordinate system of the drone vision system in the i-th frame; Represents the three-dimensional coordinates of the jth feature point in the world coordinate system; Indicates the position of the UAV vision system in the world coordinate system at the i-th frame; Calculate the estimated two-dimensional coordinates of the j-th feature point in the coordinate system of the drone vision system in the i-th frame image: ; in, represents the estimated two-dimensional coordinate value of the j-th feature point in the coordinate system of the drone vision system in the i-th frame image; Indicates the depth of the jth feature point in the i-th frame UAV vision system coordinate system; represents the X-axis coordinate of the j-th feature point in the i-th frame UAV vision system coordinate system; Represents the Y-axis coordinate of the j-th feature point in the i-th frame UAV vision system coordinate system; The observation model is linearly represented by the system error state and the feature point position error: ; in, represents the residual of the i-th observation in the j-th frame; represents the influence of the system error state in the jth frame on the i-th observation value; represents the system error state vector; Represents the influence of the feature point position error in the jth frame on the i-th observation value; Represents the position error vector of the feature point in the world coordinate system, represents the noise term of the i-th observation in the j-th frame; in, It is expressed as: ; ; Transform the observation model into: .

[0013] Preferably, the method further comprises: Obtain historical storage location images, annotate the front face, top face, and front face of the pallet of the entire stack in the historical storage location images with target detection frames, and use the annotated data as training data; A target detection model is trained based on the training data.

[0014] Preferably, the method further comprises: Based on the target detection model, confidence screening is performed on the front face of the box stack, the upper top face and the front face of the pallet in the RGB color image; Based on the pre-set pallet interest area, the effectiveness of the pallet in this location is screened; The working conditions of the screened effective pallets are divided into: no pallet and no goods, pallet but no goods, and pallet with goods; The quantity of goods directly output from effective pallets for the working conditions of no pallet and no goods or pallet and no goods is zero; Based on the pre-set cargo area of ​​interest and the depth image of the valid pallet with pallets and goods, the abnormal working condition of the box stack is identified; The front end face and upper top face of the abnormal box stack are removed, and the detection frames of the side stack and rear stack of the abnormal box stack are removed.

[0015] Preferably, determining the quantity of goods in the box stack according to the RGB color image comprises: The whole stack is layered according to the detection frame on the front end of the box stack; Determine whether each layer is full and calculate the quantity of goods on each layer except the top layer of the pallet; Determine the effective front face of the box stack; According to the effective front end surface of the box stack, the effective upper top surface of the box stack is determined, and the quantity of goods in the effective upper top surface of the box stack is identified; Calculate the sum of the quantity of goods in each layer of the stack except the top layer and the quantity of goods in the effective upper top surface of the stack to obtain the quantity of goods in the stack.

[0016] According to a second aspect of an embodiment of the present application, a drone-based three-dimensional warehouse inventory system is provided, comprising: Processor and memory; The processor and the memory are connected via a communication bus: Wherein, the processor is used to call and execute the program stored in the memory; The memory is used to store a program, and the program is at least used to execute a drone-based three-dimensional warehouse inventory method as described in any one of the above items.

[0017] The technical solution provided by this application may have the following beneficial effects: Compared with manual visual inventory, this technical solution can quickly reach the designated location without the need for personnel to enter the warehouse for inventory, which greatly reduces the inventory time. At the same time, the RGBD visual equipment carried by the drone combines deep learning target detection and visual counting technology to achieve accurate inventory counting. This reduces the safety risks of operators, Compared with RFID inventory counting, this technical solution does not need to equip each item with an RFID tag, saving the tag cost and the time for tag installation and maintenance. The image information provided by the RGBD visual device is not limited to simple presence detection, but can also provide more information about abnormal status of storage locations. RFID technology may be affected by the environment, resulting in reading failure. The visual-based inventory method is not limited by these factors and can work stably.

[0018] In summary, this technical solution is superior to traditional manual visual inventory and RFID inventory in terms of efficiency, accuracy, safety and cost-effectiveness, and is particularly suitable for large and complex warehousing environments.

[0019] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0021] Figure 1 This is a flowchart of a drone-based three-dimensional warehouse inventory method provided by an embodiment of the present application; Figure 2 This is a flow chart of correcting the accumulated error of posture information by using a Kalman filter algorithm based on an observation image of a drone vision system provided by an embodiment of the present application; Figure 3 It is a schematic diagram of a process for determining whether a box stack is in an abnormal working condition according to a depth image provided by an embodiment of the present application; Figure 4 It is a schematic diagram of a process for determining the quantity of goods in a box stack according to an RGB color image provided by an embodiment of the present application; Figure 5 It is a structural schematic diagram of a drone-based three-dimensional warehouse inventory system provided by an embodiment of the present application.

[0022] Reference numerals: processor-51; memory-52. DETAILED DESCRIPTION

[0023] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0024] Embodiment 1 Figure 1 This is a flowchart of a method for inventorying a stereoscopic warehouse based on a drone provided by an embodiment of the present application, referring to Figure 1 , a drone-based three-dimensional warehouse inventory method, comprising: S11: Receive the task of taking inventory of goods and obtain the layout information of the three-dimensional warehouse; The operator selects the cargo location to be counted through the WMS (Warehouse Management System). The WMS automatically generates a list of tasks to be counted and sends it to the WCS (Warehouse Control System). The warehouse control system (also known as this system) receives the cargo counting task and then processes it further.

[0025] It should be noted that obtaining the layout information of the stereoscopic warehouse includes: Obtain a three-dimensional drawing of the stereoscopic warehouse, and determine the number of shelves in the stereoscopic warehouse, the volume parameters of each shelf, the position of the stacker track, the width of the shelf aisle, and the specific location of the entrance and exit based on the three-dimensional drawing of the stereoscopic warehouse; Obtain the on-site survey and verification results, and verify the three-dimensional drawings of the high-rise warehouse based on the on-site survey and verification results.

[0026] In order to ensure that the drone can perform inventory tasks efficiently and accurately at designated cargo locations, it is first necessary to obtain detailed layout information of the automated warehouse. This includes accurate three-dimensional drawings, covering the number of shelves in the warehouse, the volume parameters of each shelf, the location of the stacker track, the width of the shelf aisle, and the specific location of the entrance and exit. Then it is necessary to verify the accuracy of the existing drawings through on-site inspection and timely correct any discrepancies with the actual environment.

[0027] S12: Determine the location of the cargo to be counted in the cargo inventory task in the three-dimensional warehouse; S13: According to the layout information of the three-dimensional warehouse and the location of the goods to be counted in the three-dimensional warehouse, based on the path planning algorithm, the safe flight area and hovering point of the UAV are determined, and the flight mission is generated; It should be noted that the method also includes: Bind each shelf channel in the stereoscopic warehouse to each drone one by one; According to the layout information of the shelf aisle and the position of the goods to be counted in the shelf aisle, based on the path planning algorithm, the safe flight area and hovering point of the drone in its corresponding shelf aisle are determined, and the flight mission of the drone is generated.

[0028] It should be noted that the present embodiment adopts a single lane configuration of drones. Each lane is equipped with a dedicated drone, and each drone is responsible for the aerial photography and inventory task of a single lane, which can reduce the difficulty of obstacle avoidance and improve work efficiency.

[0029] It should be noted that the method also includes: When the UAV is performing a flight mission, the stacker in the shelf channel corresponding to the UAV is controlled to withdraw to the entrance and exit and suspend work.

[0030] While the drone is performing inventory counting tasks, the stacker in the aisle should stop working and retreat to the entry and exit end to avoid mutual interference and ensure safe operation.

[0031] In practice, based on the specific inventory tasks of each lane, the path planning algorithm is used in combination with the three-dimensional information of the vertical warehouse to calculate the flight path of the drone. This process takes into account factors such as the location of the goods to be counted, the location of obstacles on shelves and other fixed structures. The overall process of path planning: starting from the fixed starting point of the drone hangar, go to each inventory location in turn. (For example, inventory location 1 → inventory location 2 →... → inventory location N), and finally return to the fixed starting point of the drone.

[0032] S14: Sending a flight mission to the UAV, and using the inertial measurement unit to calculate the position and posture information of the UAV in real time when performing the flight mission; The inertial measurement unit is used to calculate the attitude, position, and speed information of the UAV in real time when performing a flight mission as the UAV's position information.

[0033] S15: Based on the observation image of the UAV vision system, the accumulated error of the posture information is corrected through the Kalman filter algorithm; S16: Real-time acquisition of RGB color images and depth images collected by the drone at each cargo location to be counted; S17: Determine whether the box stack is in an abnormal working condition according to the depth image; S18: For the box stack that is not in an abnormal working condition, determine the quantity of goods in the box stack according to the RGB color image.

[0034] It should be noted that drones usually rely on GPS (Global Positioning System) for navigation, but this reliance is challenged in the automated warehouse environment. The automated warehouse is equipped with tightly arranged and high-height shelves and stores high-density goods. These factors will not only block or reflect GPS signals, causing signal attenuation, but also produce multipath effects, seriously affecting positioning accuracy. Therefore, when performing flight operations in such an environment, drones may not be able to rely on GPS for accurate navigation and positioning.

[0035] Therefore, the technical problems to be solved by this technical solution are mainly divided into two parts: one is the precise positioning of the drone, and the other is the inventory in the warehouse.

[0036] In this technical solution, precise positioning of the UAV is achieved through inertial navigation plus visual positioning, in which the inertial measurement unit (accelerometer and gyroscope) is used as the main positioning sensor to calculate the UAV's posture information in real time, and the UAV vision system is used as the observation quantity to correct the accumulated error of inertial navigation through the Kalman filter algorithm.

[0037] Inventory counting can achieve efficient counting of whole pallets of goods in designated storage locations through advanced visual recognition technology. This process combines RGB color images and depth images, and uses computer vision algorithms to complete automatic counting of pallets. Specifically, RGB color images are mainly used for visual counting; while depth images use three-dimensional spatial data to confirm whether the pallet is in an abnormal condition.

[0038] Compared with manual visual inventory, this technical solution can quickly reach the designated location without the need for personnel to enter the warehouse for inventory, which greatly reduces the inventory time. At the same time, the RGBD visual equipment carried by the drone combines deep learning target detection and visual counting technology to achieve accurate inventory counting. This reduces the safety risks of operators, Compared with RFID inventory counting, this technical solution does not need to equip each item with an RFID tag, saving the tag cost and the time for tag installation and maintenance. The image information provided by the RGBD visual device is not limited to simple presence detection, but can also provide more information about abnormal status of storage locations. RFID technology may be affected by the environment, resulting in reading failure. The visual-based inventory method is not limited by these factors and can work stably.

[0039] In summary, this technical solution is superior to traditional manual visual inventory and RFID inventory in terms of efficiency, accuracy, safety and cost-effectiveness, and is particularly suitable for large and complex warehousing environments.

[0040] Embodiment 2 This embodiment describes how to achieve accurate positioning of a drone.

[0041] Reference Figure 2 Based on the observation image of the UAV vision system, the accumulated error of the posture information is corrected through the Kalman filter algorithm, including: S21: Construct a state covariance matrix and use the Kalman filter algorithm to predict the error covariance of the posture information; S22: Determine whether the current observation image of the UAV visual system is a key frame; S23: If it is a key frame, the state covariance matrix is ​​expanded, the feature points in the key frame are tracked and removed, and new feature points are extracted; S24: Determine whether the tracking of the feature points is completed; S25: If the tracking of the feature point has been completed, compare the length of the tracking frame with the minimum tracking threshold; S26: if the length of the tracking frame is greater than the minimum tracking threshold, the three-dimensional coordinates of the feature points are calculated, the reprojection errors of the feature points are calculated according to the three-dimensional coordinates of the feature points to obtain observation information, and the accumulated errors of the posture information are corrected according to the observation information; S27: If the tracking of the feature point is not finished, compare the length of the tracking frame with the maximum tracking threshold; S28: If the length of the tracking frame is greater than the maximum tracking threshold, the three-dimensional coordinates of the feature points are calculated, and the tracking frames are screened. In the screened tracking frames, the reprojection errors of the feature points are calculated according to the three-dimensional coordinates of the feature points to obtain observation information, and the accumulated errors of the posture information are corrected according to the observation information.

[0042] It should be noted that the method also includes: Construct the state update estimation equation from k-1 to k: ; in, represents the system state estimate at time k; represents the system state estimate at time k-1; represents the gain matrix at time k, represents the measurement value or residual at time k; Represents the measurement model matrix at time k; The state vector Decomposed into the inertial measurement unit vector part and the drone vision system vector part, the state vector matrix is ​​expressed as: ; in, Represents the state vector of the entire system; represents the vector part of the inertial measurement unit; Represents the vector part of the drone vision system; The inertial measurement unit vector is expressed as: ; in; Represents the carrier attitude quaternion; represents the velocity vector; represents the position vector; Indicates the gyroscope zero bias; Indicates the accelerometer zero bias; The quaternion representing the installation deviation angle between the drone vision system and the inertial measurement unit; Represents the UAV vision system and inertial measurement unit arm vector; The error state vector of the inertial measurement unit is expressed as: ; in; Represents the rotation angle error of the inertial measurement unit relative to the world coordinate system; Represents the angular velocity error of the inertial measurement unit relative to the world coordinate system; Represents the translation error from the world coordinate system to the inertial measurement unit; Represents the gyro bias of the inertial measurement unit; Represents the accelerometer bias of the inertial measurement unit; Represents the rotation angle error of the UAV vision system relative to the inertial measurement unit; represents the translation error from the inertial measurement unit to the UAV vision system; The angular velocity error of the inertial measurement unit relative to the world coordinate system is expressed as: ; in, and Represents the rotation matrix between the world coordinate system W and the inertial measurement unit; represents the velocity vector; The state prediction model of Kalman filter is expressed as: ; in; Increment representing the rate of change of the inertial measurement unit state; represents the state transfer matrix; Represents the increment of the state variable; represents the control input matrix; represents process noise; State transition matrix It is expressed as: ; in; represents the angular velocity of the Earth's rotation; Represents the direction cosine matrix from the carrier coordinate system to the world coordinate system; Represents the gravitational acceleration in the world coordinate system; The conversion formula of gravity acceleration is expressed as: ; Direction cosine matrix from earth coordinate system to world coordinate system It is expressed as: ; in; represents the gravitational acceleration in the earth coordinate system; Indicates the latitude of the world coordinate system relative to the earth coordinate system; Represents the longitude of the world coordinate system relative to the earth coordinate system; Inertial Measurement Unit Noise Transfer Matrix It is expressed as: ; White noise vector It is expressed as: ; in; represents the white noise of the gyroscope; represents the white noise of the accelerometer; White noise representing the gyroscope bias; White noise representing the accelerometer bias; White noise representing attitude angle; White noise representing position; The UAV vision system vector is represented as: ; in, Indicates the quaternion rotation attitude from the world coordinate system to the drone vision system coordinate system in the first frame; Indicates the position of the drone vision system in the world coordinate system at the first frame; Indicates the quaternion rotation attitude from the world coordinate system to the drone vision system coordinate system at the Nth frame; Indicates the position of the drone vision system in the world coordinate system at the Nth frame; The error state vector related to the UAV vision system is expressed as: ; in, and Represents the attitude error and position error of the UAV vision system in the first frame; Indicates the rotation error from the world coordinate system to the drone vision system coordinate system at the Nth frame; Indicates the position error of the drone vision system in the world coordinate system at the Nth frame; It should be noted that every time the UAV vision system calculates the key frame to obtain the pose measurement value, the new key frame camera pose state needs to be added to the original state vector and the state covariance matrix of the Kalman filter needs to be expanded.

[0043] The state error vector of the N+1th frame is: ; ; in, Indicates the rotation error from the world coordinate system to the drone vision system coordinate system at the N+1th frame; Indicates the position error of the UAV vision system in the world coordinate system at the N+1th frame; represents the 3×3 identity matrix; represents the 3×3 identity matrix; Represents the rotation matrix from the inertial measurement unit coordinate system to the world coordinate system; represents the position of the drone vision system in the inertial measurement unit coordinate system, Represents the operation of converting a vector into an antisymmetric matrix; The augmented state covariance matrix It is expressed as: ; ; ; in, represents the state covariance matrix after augmentation; represents the identity matrix of order 6N+15; represents the Jacobian matrix, which is used to describe the relationship between state error vectors; represents the original state covariance matrix; represents the covariance between the UAV vision system state error and the conversion error from the inertial measurement unit coordinate system to the UAV vision system coordinate system in the N+1th frame; Represents the covariance of the UAV visual system state error itself in the N+1th frame; In this integrated navigation, the UAV vision system is used as an observation sensor, and its observation model It is defined as the reprojection error of the key frame feature points of the UAV vision system, that is, the error obtained by comparing the pixel coordinates (key point projection position) with the position obtained by projecting the 3D point according to the current estimated pose.

[0044] Calculate the estimated 3D coordinates of the jth feature point in the i-th frame image in the drone vision system coordinate system: ; in, represents the estimated 3D coordinate value of the j-th feature point in the coordinate system of the UAV vision system in the i-th frame image; Indicates the X-axis coordinate of the feature point in the UAV vision system coordinate system; Indicates the Y-axis coordinate of the feature point in the UAV vision system coordinate system; Indicates the Z-axis coordinate of the feature point in the UAV vision system coordinate system; Represents the rotation matrix from the world coordinate system to the coordinate system of the drone vision system in the i-th frame; Represents the three-dimensional coordinates of the jth feature point in the world coordinate system; Indicates the position of the UAV vision system in the world coordinate system at the i-th frame; Calculate the estimated two-dimensional coordinates of the j-th feature point in the coordinate system of the drone vision system in the i-th frame image: ; in, represents the estimated two-dimensional coordinate value of the j-th feature point in the coordinate system of the drone vision system in the i-th frame image; Indicates the depth of the jth feature point in the i-th frame UAV vision system coordinate system; represents the X-axis coordinate of the j-th feature point in the i-th frame UAV vision system coordinate system; Represents the Y-axis coordinate of the j-th feature point in the i-th frame UAV vision system coordinate system; The observation model is linearly represented by the system error state and the feature point position error: ; in, represents the residual of the i-th observation in the j-th frame; represents the influence of the system error state in the jth frame on the i-th observation value; represents the system error state vector; Represents the influence of the feature point position error in the jth frame on the i-th observation value; Represents the position error vector of the feature point in the world coordinate system, represents the noise term of the i-th observation in the j-th frame; in, It is expressed as: ; ; Transform the observation model into: .

[0045] With the state model, observation model and covariance matrix, the Kalman gain can be solved and the state of the UAV posture can be updated.

[0046] It should be noted that the above-mentioned visual navigation feature point detection pairing algorithm can apply the FAST (Features from Accelerated Segment Test) corner point detection algorithm; the sequence frame image feature point tracking can use the KLT (KanadeLucas Tomasi) algorithm; for the corner points of moving objects in the image, outliers need to be eliminated, which can be achieved using the two-point RANSAC algorithm assisted by an inertial measurement unit. The above three algorithms are all common algorithms and will not be described here.

[0047] Embodiment 3 This embodiment describes how to implement inventory counting in a warehouse.

[0048] The object of the stereoscopic warehouse inventory in this technical solution is mainly focused on the box-type pallet stereoscopic warehouse. At the same time, before visual counting of the box-type pallet, it is necessary to obtain the full-stack type information of the warehouse location, including the full-stack quantity and the number of each layer of the full-stack. It should be noted that the method also includes: Obtain historical storage location images, annotate the front face, top face, and front face of the pallet of the entire stack in the historical storage location images with target detection frames, and use the annotated data as training data; Train an object detection model based on the training data.

[0049] In the preparation stage, the object detection model needs to be trained through historical storage location images.

[0050] Preferably, the target detection model uses the YOLO target detection model.

[0051] After the target detection model is trained, refer to Figure 3 , the method further comprises: S31: Confidence screening of the front face of the box stack, the top face and the front face of the pallet in the RGB color image based on the target detection model; First, confidence screening is performed on the front face of the box stack, the top face, and the front face of the pallet to remove detection frames with poor results.

[0052] S32: screening the effectiveness of the pallets in the current cargo location based on the pre-set pallet interest area; Since the storage location images taken by the drone in the early stage are at a specified height, the pallet will appear in a fixed position, so the pallet ROI (region of interest) is set to ensure that the rear pallet will not enter the recognition process.

[0053] S33: classifying the working conditions of the screened valid pallets; the pallet working conditions include: no pallet and no goods, pallet but no goods, and pallet with goods; According to the number of valid pallets obtained, the working conditions of the screened valid pallets are divided into three working conditions: no pallet and no goods, pallet but no goods, and pallet with goods (normal pallet counting).

[0054] S34: The quantity of goods directly output from effective pallets for the working conditions of no pallet and no goods or pallet and no goods is zero; There is no need to count valid pallets with no goods or pallets with no goods, and the results can be returned directly.

[0055] S35: Based on the preset cargo location interest area and the depth image of the valid pallet with pallets and goods, identify the abnormal working condition of the box stack; In view of the abnormal situations that whole pallets of goods may be scattered or skewed in actual engineering situations, the acquired depth map is used to generate a point cloud and set the cargo location ROI. If it exceeds the range, it is judged as an abnormal state.

[0056] S36: removing the front end face and the upper top face of the abnormal box stack, and removing the detection frames of the side stack and the rear stack of the abnormal box stack.

[0057] The detection frames of the side and rear stacks are removed. The removal principle depends on the spatial coordinates of the pallet and the spatial relationship between the front end and top surface of the box stack.

[0058] It should be noted that, refer to Figure 4 , determine the quantity of goods in the box stack based on the RGB color image, including: S41: Layering the entire stack according to the front end detection frame of the box stack; Whole stack stratification: The whole stack is stratified according to the front face detection frame of the box stack. The conditions for judging the same layer are: ; in, Indicates the layer distance threshold for judging the front face as the same layer. and It is expressed as the minimum and maximum values ​​of the front face detection frame in the y direction. and Expressed as the minimum and maximum values ​​of the front face detection frame in the x direction.

[0059] S42: Determine whether each layer is full, and calculate the quantity of goods in each layer of the box stack except the top layer; Due to parallax in shooting angles, it is necessary to identify whether each layer is full or not.

[0060] The specific judgment conditions are: Layer width + perspective compensation value > tray width Layer bottom border > all top boxes .

[0061] The perspective compensation value can be set by yourself.

[0062] All top frames Indicates the maximum value of all top surface boxes in the y direction.

[0063] S43: Determine the effective front end face of the box stack; The effective front end face is further screened out based on the front end face involved in the non-full layer. The effective front end face is the basis for selecting the upper top face.

[0064] The screening criteria for valid front faces are: ; The y-threshold and x-threshold can be set by yourself. and Indicates the distance of the front face detection frame in the y and x directions.

[0065] Front face frame Indicates the minimum value of the front face frame in the y direction; Front face frame Indicates the minimum value of the front face frame in the x direction; Front face frame Indicates the maximum value of the front face frame in the x direction; Upper top frame Indicates the maximum value of the top surface box in the y direction; Upper top frame Indicates the minimum value of the upper top surface box in the x direction; Upper top frame Indicates the maximum value of the top surface box in the x direction.

[0066] S44: determining the effective upper top surface of the box stack according to the effective front end surface of the box stack, and identifying the quantity of goods in the effective upper top surface of the box stack; In the process of counting the top surface frames, the top surface frames exposed in the lower layer will also be detected, and now the effective top surface detection frames are counted.

[0067] The specific judgment conditions are: ; Top frame Indicates the center value of the top surface box in the y direction; Effective front face frame Indicates the minimum value of the effective front face frame in the y direction; Effective front face frame Indicates the minimum value of the effective front face frame in the x direction; Top frame Indicates the center value of the top surface box in the x direction; Effective front face frame Indicates the maximum value of the effective front face box in the x direction.

[0068] S45: Calculate the sum of the quantity of goods in each layer of the box stack except the top layer and the quantity of goods in the effective upper top surface of the box stack to obtain the quantity of goods in the box stack.

[0069] The final quantity of goods in the stack is the sum of the quantity of goods in each layer of the stack except the top layer and the quantity of goods in the effective upper top surface of the stack.

[0070] Embodiment 4 A drone-based three-dimensional warehouse inventory system, referring to Figure 5 ,include: Processor 51 and memory 52; The processor 51 and the memory 52 are connected via a communication bus: The processor 51 is used to call and execute the program stored in the memory 52; The memory 52 is used to store a program, and the program is used to execute at least one of the above embodiments of the inventory counting method for a high-bay warehouse based on a drone.

[0071] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0072] It should be noted that, in the description of this application, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of this application, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0073] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0074] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0075] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0076] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0077] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0078] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0079] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A drone-based three-dimensional warehouse inventory method, characterized in that: include: Receive cargo inventory tasks and obtain layout information of the three-dimensional warehouse; Determine the location of the cargo to be counted in the cargo inventory task in the three-dimensional warehouse; According to the layout information of the three-dimensional warehouse and the location of the goods to be counted in the three-dimensional warehouse, based on the path planning algorithm, the safe flight area and hovering point of the drone are determined, and the flight mission is generated; Sending the flight mission to the drone, and calculating the position and posture information of the drone in real time when performing the flight mission through an inertial measurement unit; Based on the observation image of the UAV vision system, the accumulated error of the posture information is corrected by a Kalman filter algorithm; Real-time acquisition of RGB color images and depth images collected by drones at each location to be counted; Determining whether the box stack is in an abnormal working condition according to the depth image; For a box stack that is not in an abnormal working condition, the quantity of goods in the box stack is determined based on the RGB color image.

2. The method according to claim 1, characterized in that Get the layout information of the high-bay warehouse, including: Obtain a three-dimensional drawing of the stereoscopic warehouse, and determine the number of shelves in the stereoscopic warehouse, the volume parameters of each shelf, the position of the stacker track, the width of the shelf aisle, and the specific location of the entrance and exit based on the three-dimensional drawing of the stereoscopic warehouse; Obtain on-site investigation and verification results, and verify the three-dimensional drawings of the stereoscopic warehouse according to the on-site investigation and verification results.

3. The method according to claim 1, characterized in that The method further comprises: Bind each shelf channel in the stereoscopic warehouse to each drone one by one; According to the layout information of the shelf aisle and the position of the goods to be counted in the shelf aisle, based on the path planning algorithm, the safe flight area and hovering point of the drone in its corresponding shelf aisle are determined, and the flight mission of the drone is generated.

4. The method according to claim 1, characterized in that: The method further comprises: When the UAV is performing a flight mission, the stacker in the shelf channel corresponding to the UAV is controlled to withdraw to the entrance and exit and suspend work.

5. The method according to claim 1, characterized in that: Based on the observation image of the UAV vision system, the accumulated error of the posture information is corrected by the Kalman filter algorithm, including: Constructing a state covariance matrix, and performing error covariance prediction on the posture information through a Kalman filter algorithm; Determine whether the current observation image of the UAV vision system is a key frame; If it is a key frame, the state covariance matrix is ​​expanded, the feature points in the key frame are tracked and removed, and new feature points are extracted; Determine whether the tracking of feature points is completed; If the tracking of the feature point has been completed, compare the length of the tracking frame with the minimum tracking threshold; If the length of the tracking frame is greater than the minimum tracking threshold, the three-dimensional coordinates of the feature points are calculated, the reprojection errors of the feature points are calculated according to the three-dimensional coordinates of the feature points to obtain observation information, and the accumulated errors of the posture information are corrected according to the observation information; If the tracking of the feature point is not finished, compare the length of the tracking frame with the maximum tracking threshold; If the length of the tracking frame is greater than the maximum tracking threshold, the three-dimensional coordinates of the feature points are calculated, and the tracking frames are screened. In the screened tracking frames, the reprojection errors of the feature points are calculated according to the three-dimensional coordinates of the feature points to obtain observation information, and the accumulated error of the posture information is corrected according to the observation information.

6. The method according to claim 5, characterized in that The method further comprises: Construct the state update estimation equation from k-1 to k: ； in, represents the system state estimate at time k; represents the system state estimate at time k-1; represents the gain matrix at time k, represents the measurement value or residual at time k; Represents the measurement model matrix at time k; The state vector Decomposed into the inertial measurement unit vector part and the drone vision system vector part, the state vector matrix is ​​expressed as: ； in, Represents the state vector of the entire system; represents the vector part of the inertial measurement unit; Represents the vector part of the drone vision system; The inertial measurement unit vector is expressed as: ； in; Represents the carrier attitude quaternion; represents the velocity vector; represents the position vector; Indicates the gyroscope zero bias; Indicates the accelerometer zero bias; The quaternion representing the installation deviation angle between the drone vision system and the inertial measurement unit; Represents the UAV vision system and inertial measurement unit arm vector; The error state vector of the inertial measurement unit is expressed as: ； in; Represents the rotation angle error of the inertial measurement unit relative to the world coordinate system; Represents the angular velocity error of the inertial measurement unit relative to the world coordinate system; Represents the translation error from the world coordinate system to the inertial measurement unit; Represents the gyro bias of the inertial measurement unit; Represents the accelerometer bias of the inertial measurement unit; Represents the rotation angle error of the UAV vision system relative to the inertial measurement unit; represents the translation error from the inertial measurement unit to the UAV vision system; The angular velocity error of the inertial measurement unit relative to the world coordinate system is expressed as: ； in, and Represents the rotation matrix between the world coordinate system W and the inertial measurement unit; represents the velocity vector; The state prediction model of Kalman filter is expressed as: ； in; Increment representing the rate of change of the inertial measurement unit state; represents the state transfer matrix; Represents the increment of the state variable; represents the control input matrix; represents process noise; State transition matrix It is expressed as: ； in; represents the angular velocity of the Earth's rotation; Represents the direction cosine matrix from the carrier coordinate system to the world coordinate system; Represents the gravitational acceleration in the world coordinate system; The conversion formula of gravity acceleration is expressed as: ； Direction cosine matrix from earth coordinate system to world coordinate system It is expressed as: ； in; represents the gravitational acceleration in the earth coordinate system; Indicates the latitude of the world coordinate system relative to the earth coordinate system; Represents the longitude of the world coordinate system relative to the earth coordinate system; Inertial Measurement Unit Noise Transfer Matrix It is expressed as: ； White noise vector It is expressed as: ； in; represents the white noise of the gyroscope; represents the white noise of the accelerometer; White noise representing the gyroscope bias; White noise representing the accelerometer bias; White noise representing attitude angle; White noise representing position; The UAV vision system vector is represented as: ； in, Indicates the quaternion rotation attitude from the world coordinate system to the drone vision system coordinate system in the first frame; Indicates the position of the drone vision system in the world coordinate system at the first frame; Indicates the quaternion rotation attitude from the world coordinate system to the drone vision system coordinate system at the Nth frame; Indicates the position of the drone vision system in the world coordinate system at the Nth frame; The error state vector related to the UAV vision system is expressed as: ； in, and Represents the attitude error and position error of the UAV vision system in the first frame; Indicates the rotation error from the world coordinate system to the drone vision system coordinate system at the Nth frame; Indicates the position error of the drone vision system in the world coordinate system at the Nth frame; The state error vector of the N+1th frame is: ； ； in, Indicates the rotation error from the world coordinate system to the drone vision system coordinate system at the N+1th frame; Indicates the position error of the UAV vision system in the world coordinate system at the N+1th frame; represents the 3×3 identity matrix; represents the 3×3 identity matrix; Represents the rotation matrix from the inertial measurement unit coordinate system to the world coordinate system; represents the position of the drone vision system in the inertial measurement unit coordinate system, Represents the operation of converting a vector into an antisymmetric matrix; The augmented state covariance matrix It is expressed as: ； ； ； in, represents the state covariance matrix after augmentation; represents the identity matrix of order 6N+15; represents the Jacobian matrix, which is used to describe the relationship between state error vectors; represents the original state covariance matrix; represents the covariance between the UAV vision system state error and the conversion error from the inertial measurement unit coordinate system to the UAV vision system coordinate system in the N+1th frame; Represents the covariance of the UAV visual system state error itself in the N+1th frame; Calculate the estimated 3D coordinates of the jth feature point in the i-th frame image in the drone vision system coordinate system: ； in, represents the estimated 3D coordinate value of the j-th feature point in the coordinate system of the UAV vision system in the i-th frame image; Indicates the X-axis coordinate of the feature point in the UAV vision system coordinate system; Indicates the Y-axis coordinate of the feature point in the UAV vision system coordinate system; Indicates the Z-axis coordinate of the feature point in the UAV vision system coordinate system; Represents the rotation matrix from the world coordinate system to the coordinate system of the drone vision system in the i-th frame; Represents the three-dimensional coordinates of the jth feature point in the world coordinate system; Indicates the position of the UAV vision system in the world coordinate system at the i-th frame; Calculate the estimated two-dimensional coordinates of the j-th feature point in the coordinate system of the drone vision system in the i-th frame image: ； in, represents the estimated two-dimensional coordinate value of the j-th feature point in the coordinate system of the drone vision system in the i-th frame image; Indicates the depth of the jth feature point in the i-th frame UAV vision system coordinate system; represents the X-axis coordinate of the j-th feature point in the i-th frame UAV vision system coordinate system; Represents the Y-axis coordinate of the j-th feature point in the i-th frame UAV vision system coordinate system; The observation model is linearly represented by the system error state and the feature point position error: ； in, represents the residual of the i-th observation in the j-th frame; represents the influence of the system error state in the jth frame on the i-th observation value; represents the system error state vector; Represents the influence of the feature point position error in the jth frame on the i-th observation value; Represents the position error vector of the feature point in the world coordinate system, represents the noise term of the i-th observation in the j-th frame; in, It is expressed as: ； ； Transform the observation model into: 。 7. The method according to claim 1, characterized in that The method further comprises: Obtain historical storage location images, annotate the front face, top face, and front face of the pallet of the entire stack in the historical storage location images with target detection frames, and use the annotated data as training data; A target detection model is trained based on the training data.

8. The method according to claim 7, characterized in that The method further comprises: Based on the target detection model, confidence screening is performed on the front face of the box stack, the upper top face and the front face of the pallet in the RGB color image; Based on the pre-set pallet interest area, the effectiveness of the pallet in this location is screened; The working conditions of the screened effective pallets are divided into: no pallet and no goods, pallet but no goods, and pallet with goods; The quantity of goods directly output from effective pallets for the working conditions of no pallet and no goods or pallet and no goods is zero; Based on the pre-set cargo area of ​​interest and the depth image of the valid pallet with pallets and goods, the abnormal working condition of the box stack is identified; The front end face and upper top face of the abnormal box stack are removed, and the detection frames of the side stack and rear stack of the abnormal box stack are removed.

9. The method according to claim 8, characterized in that Determining the quantity of goods in the box stack according to the RGB color image includes: The whole stack is layered according to the detection frame on the front end of the box stack; Determine whether each layer is full and calculate the quantity of goods on each layer except the top layer of the pallet; Determine the effective front face of the box stack; According to the effective front end surface of the box stack, the effective upper top surface of the box stack is determined, and the quantity of goods in the effective upper top surface of the box stack is identified; Calculate the sum of the quantity of goods in each layer of the stack except the top layer and the quantity of goods in the effective upper top surface of the stack to obtain the quantity of goods in the stack.

10. A drone-based three-dimensional warehouse inventory system, characterized in that: include: Processor and memory; The processor and the memory are connected via a communication bus: Wherein, the processor is used to call and execute the program stored in the memory; The memory is used to store a program, and the program is at least used to execute the drone-based three-dimensional warehouse inventory method described in any one of claims 1-9.

Images