Security inspection machine ramp package conveying control method and system based on kinetic constraint

Abstract

The application discloses a kind of based on dynamics restraint's security inspection machine slope package control method and system, it is related to intelligent security inspection technical field.The method includes: real-time acquisition package depth image, and reconstructs the three-dimensional point cloud data of package;The three-dimensional point cloud data is handled, and the size of the oriented bounding box of package, centroid height and placement yaw angle are extracted;Based on the scanning progress state of previous package, and in combination with the oriented bounding box and placement yaw angle of current package, calculate anti-overlapping speed factor;According to the centroid height and equivalent dynamic friction coefficient of current package, calculate the anti-toppling critical acceleration of current package, and generate S-shaped jerk control curve;According to the anti-overlapping speed factor and the S-shaped jerk control curve, generate drive instruction, dynamically adjust the operating state of entry slope motor.The application solves the technical problems that package toppling, sliding and belt elastic oscillation caused by anti-overlapping emergency stop in the prior art.

Description

Technical Field

[0001] This invention relates to the field of intelligent security inspection technology, and in particular to a method and system for controlling the ramp-driven baggage delivery of a security inspection machine based on dynamic constraints. Background Technology

[0002] In public places such as airports, subways, and train stations, X-ray security scanners are core equipment for ensuring security. To improve security efficiency, a powered ramp conveyor belt is usually installed at the entrance of the scanner to automatically feed packages into the scanner channel. However, in actual operation, the entrance ramp package feeding system faces two major technical challenges: 1) Package Overlap Issue: If a package enters the security scanner before the previous package has fully entered (i.e., while the X-ray source is still scanning), an overlap will occur on the X-ray image, causing manual or automatic image interpretation to fail. Existing technologies typically use photoelectric switches for simple "start / stop" control or set a fixed anti-overlap distance threshold. This crude control method either results in extremely low package entry efficiency or physical overlap occurs when the spacing calculation is inaccurate.

[0003] 2) Package tipping and sliding issues: To prevent overlap, ramp motors often need to decelerate or stop abruptly. Due to the slope angle, the huge inertial force generated by sudden deceleration can easily cause packages with a high center of gravity to tip backward, or cause packages with a low coefficient of friction to slide downward and collide with packages behind them.

[0004] 3) Belt elastic oscillation problem: Existing anti-tipping control usually uses a fixed acceleration threshold for amplitude limiting. However, in high-frequency discrete control cycles, the anti-overlap emergency stop command will cause a sudden change in motor output. The mechanical elasticity of the inclined belt itself will produce deformation oscillation (sudden acceleration change), causing micro-slippage on the belt, which in turn causes visual ranging failure.

[0005] Currently, existing technologies often treat "anti-overlap" and "anti-tipping" as independent modules and simply connect them in series, failing to fundamentally resolve the coupling conflict between the anti-overlap emergency stop command and the mechanical dynamics of the belt. Summary of the Invention

[0006] This invention provides a method and system for controlling the ramp-fed package delivery of a security inspection machine based on dynamic constraints. This invention relates to the field of intelligent security inspection technology. The invention aims to solve the technical problems in existing technologies, such as package tipping, slippage, and belt elastic vibration caused by sudden stops to prevent overlapping.

[0007] In a first aspect, embodiments of the present invention provide a slope-based baggage delivery control method for a security inspection machine based on dynamic constraints, the method comprising: Real-time acquisition of depth images of packages in the ramp area at the entrance of the security inspection machine, and reconstruction of the three-dimensional point cloud data of the packages; The three-dimensional point cloud data is processed to extract the dimensions, centroid height, and placement yaw angle of the oriented bounding box on the slope plane. Based on the scanning progress status of the previous package in the security inspection machine channel, and combined with the oriented bounding box and placement yaw angle of the current package, the anti-overlap speed factor is calculated. Based on the current center of gravity height and equivalent dynamic friction coefficient of the package, calculate the critical acceleration against tipping of the package at the current slope angle, and generate the S-shaped acceleration control curve of the slope motor. Based on the anti-overlap speed factor and the S-shaped jerk control curve, drive commands are generated to dynamically adjust the operating status of the inlet ramp motor.

[0008] The technical solution provided in this application has at least the following beneficial effects: The spatial anti-overlap velocity factor is directly multiplied by the physical anti-tipping S-curve, and a dynamic slip risk index based on anti-tipping and anti-slip critical values ​​is introduced. When the anti-overlap logic requires emergency braking, the adaptive acceleration boundary automatically converges to a safer physical baseline, eliminating the engineering contradiction of overturning caused by emergency braking for anti-overlap purposes from the algorithm's underlying layer. For the case of a package placed at an angle on a slope, instead of using the package's own physical length, the extreme difference of the projection of the bottom point cloud along the belt movement direction is extracted to calculate the bottom support half-length, which truly reflects the effect of the angled package on the acceleration. The effective anti-tipping torque in the velocity direction avoids the failure of anti-tipping calculation; the jerk of the control command is extracted and scaled by the square of the time step and injected into the velocity command as a feedforward, which effectively offsets the lag of the system's mechanical inertia and elastic deformation, eliminates the micro-slippage of the package caused by belt vibration, and ensures the continuous effectiveness of 3D visual ranging; the anti-overlap spacing is no longer a fixed threshold, but dynamically constructs the retention polygon by combining the progress of the main belt inside the security inspection machine and the size of the previous package, which maximizes the throughput of the security inspection machine while ensuring absolute non-overlap.

[0009] In one optional implementation, depth images of packages in the ramp area at the entrance of the security checkpoint are acquired in real time, and the 3D point cloud data of the packages is reconstructed, including: A 3D structured light camera positioned directly above the ramp at the entrance of the security checkpoint is used to acquire depth images of packages in the ramp detection area in real time at a preset frame rate. Based on the pre-calibrated camera intrinsic parameter matrix of the 3D structured light camera, the depth image is back-projected onto the camera's three-dimensional coordinate system, and the three-dimensional coordinates of each pixel in the camera coordinate system are calculated to obtain the initial three-dimensional point cloud data. Using the random sampling consensus algorithm, plane fitting is performed on the initial 3D point cloud data to extract the plane with the most occupied points and that conforms to the physical characteristics of the slope, thus obtaining the general equation parameters of the slope background plane. Based on the general equation parameters of the slope background plane, the Euclidean distance from the initial 3D point cloud data to the slope background plane is calculated, and points whose Euclidean distance is less than or equal to the background removal threshold are identified as slope background points and removed, resulting in background-free 3D point cloud data containing only wrapping information. Voxel filtering is used to downsample and denoise the background-removed 3D point cloud data to obtain simplified and enclosed 3D point cloud data; Using hand-eye calibration, the rotation matrix and translation vector between the pre-acquired camera coordinate system and the belt coordinate system are determined. Based on the rotation matrix and translation vector, a homogeneous coordinate transformation is performed on each point in the simplified wrapped 3D point cloud data, and its new coordinates in the belt coordinate system are calculated to obtain the final 3D point cloud data wrapped in the belt coordinate system.

[0010] In one optional implementation, the 3D point cloud data is processed to extract the dimensions, centroid height, and placement yaw angle of the bounding box on the slope plane, including: The final 3D point cloud data wrapped in the belt coordinate system is projected onto the slope plane to obtain a 2D projection point set. Then, the points in the final 3D point cloud data whose Z-axis component is in the minimum value range are extracted and projected onto the slope plane to obtain the bottom 2D point set. Based on the two-dimensional point set on the bottom surface, calculate the principal axis direction wrapped on the slope plane and extract the corresponding placement yaw angle; Based on the placement yaw angle, construct a two-dimensional rotation matrix, and based on the two-dimensional rotation matrix, rotate the two-dimensional projection point set to an aligned coordinate system parallel to the main axis of the package, and obtain the rotated two-dimensional projection point set; In the aligned coordinate system, the extreme values ​​are obtained along the principal axis and the vertical direction to calculate the length and width of the package and obtain the dimensions of the oriented bounding box; The centroid height is obtained by taking the weighted average of the Z-axis components of all points in the final 3D point cloud data wrapped in the belt coordinate system.

[0011] In one alternative implementation, the principal axis direction wrapped on the slope plane is calculated based on the two-dimensional point set on the bottom surface, thereby extracting the placement yaw angle, including: Calculate the covariance matrix of the two-dimensional point set on the bottom surface; Eigenvalue decomposition is performed on the covariance matrix to obtain the first eigenvalue, the second eigenvalue, and the corresponding eigenvectors. Using the eigenvector corresponding to the largest first eigenvalue as the principal axis direction of the wrapping bottom surface, the angle between the principal axis direction and the belt movement direction is calculated to obtain the placement yaw angle.

[0012] In one optional implementation, based on the scanning progress status of the previous package within the security inspection machine channel, and in conjunction with the oriented bounding box and placement yaw angle of the current package, an anti-overlap velocity factor is calculated, including: By interacting with the main control system of the security inspection machine, the start time and current time of the previous package entering the security inspection machine channel are obtained, and the displacement progress of the previous package in the security inspection machine channel is calculated. Extract the scanning progress status when the previous package enters, the scanning progress status includes the corresponding historical oriented bounding box size, historical placement yaw angle and two-dimensional coordinate value of historical geometric center; If the difference between the length of the historical bounding box of the previous package and the displacement progress is greater than 0, it is determined that there is still a part of the tail of the previous package stuck in the entrance slope area. A two-dimensional polygon vertex set of the stuck part of the previous package in the slope belt coordinate system is constructed and the next step is carried out. Otherwise, the anti-overlap speed factor of 1 is directly generated. Define a local coordinate system with the geometric center of the current package as the origin, and define an initial set of standard vertex coordinates consisting of the four standard vertex coordinates of the current package when it is not rotated, in combination with the size of the directed bounding box of the current package. Based on the placement yaw angle, construct a two-dimensional rotation matrix, and based on the two-dimensional rotation matrix, rotate and translate the initial standard vertex coordinate set to the belt global coordinate system, and calculate the actual standard vertex coordinate set composed of the four actual two-dimensional polygon vertex sets of the currently wrapped directed bounding box on the slope plane. Taking the opposite direction of the belt movement as the positive X-axis, extract the minimum value of the vertex X-coordinate of all actual standard vertices from the actual standard vertex coordinate set, and use it as the X-coordinate of the front end of the current package; Extract the maximum value of the vertex X coordinates of all actual standard vertices from the vertex set of the two-dimensional polygon, and use it as the tail X coordinate of the previous wrapper. Calculate the difference between the X coordinates of the front end and the X coordinates of the rear end to obtain the minimum parallel projection distance between the current package and the previous package in the direction of belt movement. If the minimum parallel projection distance is less than 0, it is determined that the current package and the previous package have overlapped in the X-axis projection, and proceed to the next step. Based on the current real-time operating speed of the inclined conveyor belt, a dynamic safety threshold for anti-overlap determination is generated; Based on the minimum parallel projection spacing and dynamic security threshold, calculate the anti-overlap security margin required for the current package to enter the security inspection machine channel, and generate the corresponding anti-overlap speed factor.

[0013] In one optional implementation, the method for generating the anti-overlap speed factor includes: If the anti-overlap safety margin is greater than or equal to 0, the spacing is deemed sufficient, and operation is allowed at the maximum speed of the anti-tipping plan, generating an anti-overlap speed factor of 1. If the anti-overlap safety margin is less than 0 and greater than or equal to the negative value of the fixed mechanical margin, it is determined that the spacing is insufficient but there is no collision. It is necessary to slow down proportionally and creep to wait for the channel to be emptied. The anti-overlap speed factor is the ratio of the minimum parallel projection spacing to the dynamic safety threshold. If the anti-overlap safety margin is less than the negative value of the fixed mechanical margin, it is determined that physical overlap or serious interference has occurred, and a forced stop must be initiated, generating an anti-overlap speed factor of 0.

[0014] In one optional implementation, based on the current package's center of gravity height and equivalent kinetic friction coefficient, the critical anti-tipping acceleration of the current package at the current slope angle is calculated, and an S-shaped acceleration control curve for the slope motor is generated, including: Extract the extreme value difference of the two-dimensional point set on the bottom surface along the direction of belt movement, calculate the length of the bottom surface along the belt direction, and calculate the half length of the bottom support required to resist the overturning moment. Based on torque balance, the first critical acceleration at which the package tilts over the bottom rear support edge is calculated. Based on d'Alembert's principle, the second critical acceleration at which the package does not slide downwards is calculated. An anti-shake safety factor is introduced to calculate the current package's allowable anti-tipping critical acceleration. Based on the critical anti-tipping acceleration, the system's preset maximum acceleration, and the target constant speed required for the security inspection machine to enter the bag, calculate the key time nodes of the S-shaped trajectory; Based on key time points, generate the S-shaped acceleration control curve of the ramp motor.

[0015] In one alternative implementation, the first stage is an acceleration stage, the second stage is a uniform acceleration stage, and the third stage is a deceleration stage.

[0016] In one optional implementation, a drive command is generated based on the anti-overlap speed factor and the S-shaped jerk control curve to dynamically adjust the operating state of the inlet ramp motor, including: Multiply the anti-tipping base speed of the current control cycle in the S-shaped acceleration control curve with the corresponding anti-overlap speed factor to generate the dynamic target speed of the current control cycle. Based on the dynamic target velocity of the current control cycle, the actual command velocity issued in the previous control cycle, the first critical acceleration, and the second critical acceleration, a nonlinear adaptive acceleration boundary for the current control cycle is constructed. Based on the adaptive acceleration boundary, the maximum allowable velocity change within a single control cycle is calculated. The expected velocity change in the current control cycle is then limited to obtain the actual allowable change. Finally, the shaped actual command velocity is calculated. Based on the adaptive acceleration boundaries of the current control cycle and the previous control cycle, the jerk of the current control cycle is extracted and injected into the velocity command as a feedforward quantity to obtain the final physical target velocity. The final physical target speed determined by the physical speed calibration is converted into a duty cycle control quantity that can be recognized by the motor driver, resulting in a quantized motor PWM duty cycle control signal. The motor PWM duty cycle control signal is used as a drive command, encapsulated into a control data frame of a specific industrial bus, and sent to the servo driver or frequency converter of the ramp motor through the communication interface to dynamically adjust the operating status of the inlet ramp motor.

[0017] Secondly, embodiments of the present invention provide a security inspection machine ramp baggage feeding control system based on dynamic constraints, used to implement a security inspection machine ramp baggage feeding control method, the system comprising: The 3D vision acquisition module is used to acquire depth images of packages in the ramp area at the entrance of the security inspection machine in real time and reconstruct the 3D point cloud data of the packages. The feature extraction module is used to process the three-dimensional point cloud data and extract the size, centroid height, and placement yaw angle of the oriented bounding box on the slope plane. The anti-overlap decision module is used to calculate the anti-overlap speed factor based on the scanning progress status of the previous package in the security inspection machine channel, combined with the oriented bounding box and placement yaw angle of the current package. The anti-tipping dynamics module is used to calculate the critical anti-tipping acceleration of the current package under the current slope angle based on the current center of gravity height and equivalent dynamic friction coefficient, and generate the S-shaped acceleration control curve of the slope motor. The composite adaptive drive module is used to generate drive commands based on the anti-overlap speed factor and the S-shaped jerk control curve, and dynamically adjust the operating state of the inlet ramp motor.

[0018] A third aspect of this invention provides an electronic device, which includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by at least one processor, such that the at least one processor can perform the method proposed in the first aspect of the present invention.

[0019] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in the first aspect of the present invention. Attached Figure Description

[0020] Figure 1This is a schematic diagram of the electronic device structure of the hardware operating environment involved in the embodiments of the present invention; Figure 2 This is a flowchart illustrating the steps of a slope-based baggage delivery control method for a security inspection machine based on dynamic constraints, provided in an embodiment of the present invention. Figure 3 This is a functional module diagram of a slope-based bag delivery control system for a security inspection machine based on dynamic constraints, provided in an embodiment of the present invention. Detailed Implementation

[0021] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0022] The present invention will be further described below with reference to the accompanying drawings.

[0023] Reference Figure 1 , Figure 1 This is a schematic diagram of the electronic device structure of the hardware operating environment involved in the embodiments of the present invention.

[0024] like Figure 1 As shown, the electronic device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be a high-speed random access memory (RAM) or a stable non-volatile memory (NVM), such as a disk drive. The memory 1005 may also optionally be a storage device independent of the aforementioned processor 1001.

[0025] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0026] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a data storage module, a network communication module, a user interface module, and an electronic program for a security inspection machine ramp package delivery control system based on dynamic constraints.

[0027] exist Figure 1 In the electronic device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the electronic device of the present invention can be set in the electronic device. The electronic device calls the electronic program of the security inspection machine ramp package feeding control system based on dynamic constraints stored in the memory 1005 through the processor 1001, and executes the security inspection machine ramp package feeding control method based on dynamic constraints provided in the embodiment of the present invention.

[0028] Reference Figure 2 The present invention provides a method for controlling the ramp-driven baggage delivery of a security inspection machine based on dynamic constraints, the method comprising: S201: Real-time acquisition of depth images of packages in the ramp area at the entrance of the security inspection machine, and reconstruction of the three-dimensional point cloud data of the packages; S202: Process the three-dimensional point cloud data to extract the dimensions, centroid height, and placement yaw angle of the oriented bounding box on the slope plane; S203: Based on the scanning progress status of the previous package in the security inspection machine channel, and combined with the oriented bounding box and placement yaw angle of the current package, calculate the anti-overlap speed factor. S204: Based on the current center of gravity height and equivalent dynamic friction coefficient of the package, calculate the critical acceleration against tipping of the current package at the current slope angle, and generate the S-shaped acceleration control curve of the slope motor. S205: Based on the anti-overlap speed factor and the S-shaped acceleration control curve, generate drive commands to dynamically adjust the operating status of the inlet ramp motor.

[0029] The technical solution provided in this application has at least the following beneficial effects: The spatial anti-overlap velocity factor is directly multiplied by the physical anti-tipping S-curve, and a dynamic slip risk index based on anti-tipping and anti-slip critical values ​​is introduced. When the anti-overlap logic requires emergency braking, the adaptive acceleration boundary automatically converges to a safer physical baseline, eliminating the engineering contradiction of overturning caused by emergency braking for anti-overlap purposes from the algorithm's underlying layer. For the case of a package placed at an angle on a slope, instead of using the package's own physical length, the extreme difference of the projection of the bottom point cloud along the belt movement direction is extracted to calculate the bottom support half-length, which truly reflects the effect of the angled package on the acceleration. The effective anti-tipping torque in the velocity direction avoids the failure of anti-tipping calculation; the jerk of the control command is extracted and scaled by the square of the time step and injected into the velocity command as a feedforward, which effectively offsets the lag of the system's mechanical inertia and elastic deformation, eliminates the micro-slippage of the package caused by belt vibration, and ensures the continuous effectiveness of 3D visual ranging; the anti-overlap spacing is no longer a fixed threshold, but dynamically constructs the retention polygon by combining the progress of the main belt inside the security inspection machine and the size of the previous package, which maximizes the throughput of the security inspection machine while ensuring absolute non-overlap.

[0030] In one optional implementation, depth images of packages in the ramp area at the entrance of the security checkpoint are acquired in real time, and the 3D point cloud data of the packages is reconstructed, including: S2011: Using a 3D structured light camera positioned directly above the ramp at the entrance of the security checkpoint, the depth image of the package in the ramp detection area is acquired in real time at a preset frame rate. In this embodiment, a 3D structured light camera is used and installed 0.8 meters directly above the slope, looking down at the slope detection area. The camera's preset frame rate is set to 20 fps, and the camera outputs a grayscale image with depth information in real time, where each pixel value represents the physical distance from the point to the camera. S2012: Based on the camera intrinsic parameter matrix pre-calibrated by the 3D structured light camera, the depth image is back-projected onto the camera's three-dimensional coordinate system, and the three-dimensional coordinates of each pixel in the camera coordinate system are calculated to obtain the initial three-dimensional point cloud data. S2013: Using the random sampling consensus algorithm, the initial 3D point cloud data is fitted to a plane, and the plane with the most points and conforming to the physical characteristics of the slope is extracted to obtain the general equation parameters of the slope background plane. In this embodiment, since the initial point cloud includes the surface and wrapping of the sloping conveyor belt, the belt background needs to be removed. A random sampling consensus algorithm is used: three points are randomly selected each time to fit the plane equation. Count the number of other points whose distance to the plane is less than 3 mm. After 100 iterations, select the plane parameter with the most interior points. General equation parameters for the background plane of the slope; S2014: Based on the general equation parameters of the slope background plane, calculate the Euclidean distance from the initial 3D point cloud data to the slope background plane, and remove points whose Euclidean distance is less than or equal to the background removal threshold as slope background points, thus obtaining background-free 3D point cloud data containing only wrapping information. The formula is: In the formula, For the initial 3D point cloud data, the first i The Euclidean distance from each point to the background plane of the slope; i Point index; These are parameters for the general equation; The general equation for the background plane of the slope; For the initial 3D point cloud data, the first i The spatial coordinates of a point in the 3D camera coordinate system; like If the size is ≤3mm, it is considered a background point and is removed. S2015: Voxel filtering is used to downsample and denoise the background-removed 3D point cloud data to obtain simplified and wrapped 3D point cloud data; In this embodiment, the remaining background-removed point cloud is downsampled using a voxel grid with a side length of 5 mm. The centroid of all points in the grid is used as the new representative point of the grid, thereby greatly reducing the amount of data and obtaining simplified and wrapped 3D point cloud data. S2016: Use hand-eye calibration to determine the rotation matrix and translation vector between the pre-acquired camera coordinate system and the belt coordinate system. Based on the rotation matrix and translation vector, perform homogeneous coordinate transformation on each point in the simplified wrapped 3D point cloud data, calculate its new coordinates in the belt coordinate system, and obtain the final 3D point cloud data wrapped in the belt coordinate system.

[0031] In one optional implementation, the 3D point cloud data is processed to extract the dimensions, centroid height, and placement yaw angle of the bounding box on the slope plane, including: S2021: Project the final 3D point cloud data wrapped in the belt coordinate system onto the slope plane to obtain a 2D projection point set, and extract the points in the final 3D point cloud data whose Z-axis component is in the minimum value range, and project them onto the slope plane to obtain the bottom 2D point set. In this embodiment, points with Z-axis components in the range of 0 mm to 3 mm are extracted in the belt coordinate system and projected onto the XY plane to obtain a two-dimensional point set on the bottom surface (used to calculate the principal axis direction); at the same time, all point clouds are projected onto the XY plane to obtain a fully enclosed two-dimensional projection point set. S2022: Based on the two-dimensional point set on the bottom surface, calculate the principal axis direction wrapped on the slope plane and extract the corresponding placement yaw angle; S2023: Based on the placement yaw angle, construct a two-dimensional rotation matrix, and based on the two-dimensional rotation matrix, rotate the two-dimensional projection point set to an aligned coordinate system parallel to the main axis of the package, obtaining the rotated two-dimensional projection point set. The formula is: In the formula, For the 2D projection point set after rotation i The two-dimensional coordinates of a point in the aligned coordinate system; It is a two-dimensional rotation matrix; To place the yaw angle; For the first two-dimensional projection point set i Two-dimensional coordinates of each point; i Point index; S2024: In the aligned coordinate system, find the extreme values ​​along the principal axis and the vertical direction, calculate the length and width of the bounding box, and obtain the dimensions of the directed bounding box. The formula is: In the formula, For the 2D projection point set after rotation i The components of the two-dimensional coordinates of a point in the aligned coordinate system on the X-axis; For the 2D projection point set after rotation i The components of the two-dimensional coordinates of a point in the aligned coordinate system on the Y-axis; These are functions for finding the maximum and minimum values; The length and width of the package constitute the dimensions of the oriented bounding box; S2025: Calculate the weighted average of the Z-axis components of all points in the final 3D point cloud data wrapped in the belt coordinate system to obtain the centroid height. The formula is: In the formula, The height of the center of mass; This represents the number of points in the final 3D point cloud data. For the final 3D point cloud data, the first i The spatial coordinates of a point in the belt coordinate system are represented by the Z-axis component. i For point indexes.

[0032] In one alternative implementation, the principal axis direction wrapped on the slope plane is calculated based on the two-dimensional point set on the bottom surface, thereby extracting the placement yaw angle, including: S20221: Calculate the covariance matrix of the two-dimensional point set on the bottom surface, using the following formula: In the formula, It is the covariance matrix; This represents the number of points in the two-dimensional point set on the bottom surface. The coordinates of the geometric center of the two-dimensional point set on the bottom surface; For the first point in the two-dimensional point set on the bottom surface j Two-dimensional coordinates of each point; j Point index; S20222: Perform eigenvalue decomposition on the covariance matrix to obtain the first eigenvalue, the second eigenvalue, and the corresponding eigenvectors; S20223: Using the eigenvector corresponding to the largest first eigenvalue as the principal axis direction of the wrapping bottom surface, calculate the angle between the principal axis direction and the belt movement direction to obtain the placement yaw angle, the formula is: In the formula, To place the yaw angle; It is the arctangent function; The components of the eigenvector corresponding to the largest first eigenvalue on the X and Y axes.

[0033] In one optional implementation, based on the scanning progress status of the previous package within the security inspection machine channel, and in conjunction with the oriented bounding box and placement yaw angle of the current package, an anti-overlap velocity factor is calculated, including: S2031: By interacting with the main control system of the security inspection machine, obtain the start time and current time of the previous package entering the security inspection machine channel, and calculate the displacement progress of the previous package within the security inspection machine channel. The formula is as follows: In the formula, This represents the displacement progress of the previous package within the security inspection machine channel. This refers to the current and starting times of the previous package entering the security screening channel; The main belt speed is set to 0.2 m / s; S2032: Extract the scanning progress status when the previous package enters, the scanning progress status includes the corresponding historical oriented bounding box size, historical placement yaw angle and two-dimensional coordinate value of historical geometric center; S2033: If the difference between the length of the historical bounding box of the previous package and the displacement progress is greater than 0, it is determined that the tail of the previous package is still stuck in the entrance ramp area. Construct a two-dimensional polygon vertex set of the stuck part of the previous package in the ramp belt coordinate system and proceed to the next step. Otherwise, directly generate an anti-overlap speed factor of 1. With the geometric center of the current package as the origin, define a local coordinate system. Based on the dimensions of the current package's directed bounding box, define an initial set of standard vertex coordinates consisting of the four standard vertex coordinates of the current package when it is not rotated. The formula is: In the formula, This is the initial set of standard vertex coordinates; The length and width of the package constitute the dimensions of the oriented bounding box; These are the coordinates of the four standard vertices of the current package when it is not rotated; The first vertex in the initial standard vertex coordinate set k Initial standard vertex coordinates; k Standard vertex index; S2034: Based on the placement yaw angle, construct a two-dimensional rotation matrix. Then, based on this matrix, rotate and translate the initial standard vertex coordinate set to the belt's global coordinate system. Calculate the actual standard vertex coordinate set formed by the four actual two-dimensional polygon vertex sets of the currently bounding box on the slope plane. The formula is: In the formula, This is the set of actual standard vertex coordinates; The first vertex in the actual standard vertex coordinate set k One actual standard vertex coordinate; The first vertex in the initial standard vertex coordinate set k Initial standard vertex coordinates; k Standard vertex index; It is a two-dimensional rotation matrix; To place the yaw angle; These are the two-dimensional coordinates of the geometric center of the current package; S2035: Taking the opposite direction of belt movement as the positive X-axis, extract the minimum value of the vertex X-coordinate of all actual standard vertices from the actual standard vertex coordinate set, and use it as the X-coordinate of the front end of the current package; S2036: Extract the maximum value of the vertex X coordinates of all actual standard vertices from the vertex set of the two-dimensional polygon, and use it as the tail X coordinate of the previous wrapper. S2037: Calculate the difference between the X coordinate of the front end and the X coordinate of the rear end to obtain the minimum parallel projection distance between the current package and the previous package in the direction of belt movement. If the minimum parallel projection distance is less than 0, it is determined that the current package and the previous package have overlapped in the X-axis projection and proceed to the next step. S2038: Based on the current real-time operating speed of the inclined conveyor belt, a dynamic safety threshold for anti-overlap determination is generated, using the following formula: In the formula, Dynamic security threshold; The current real-time operating speed of the inclined conveyor belt is assumed to be 0.3 m / s. The minimum timing margin required for the security inspection machine to complete the image exposure and data transmission of a single package (take 100 ms); The image stabilization safety factor (values ​​range from 1.1 to 1.3). Fixed mechanical allowance (e.g., 50mm) to prevent corner collisions; S2039: Based on the minimum parallel projection spacing and dynamic security threshold, calculate the anti-overlap security margin required for the current package to enter the security inspection machine channel, and generate the corresponding anti-overlap speed factor. The formula is as follows: In the formula, To prevent overlap, a safety margin is provided; Dynamic security threshold; This represents the minimum parallel projection spacing.

[0034] In one optional implementation, the method for generating the anti-overlap speed factor includes: If the anti-overlap safety margin (e.g., a spacing of 150mm) is greater than or equal to 0, then the spacing is deemed sufficient, and operation at the maximum speed of the anti-tipping plan is permitted, generating an anti-overlap speed factor of 1. If the anti-overlap safety margin is less than 0 and greater than or equal to the negative value of the fixed mechanical margin (e.g., if the spacing is 60mm and the margin is -26mm), it is determined that the spacing is insufficient but there is no collision. It is necessary to reduce the speed proportionally and creep to wait for the channel to be emptied. The anti-overlap speed factor is the ratio of the minimum parallel projection spacing to the dynamic safety threshold. If the anti-overlap safety margin is less than the negative value of the fixed mechanical margin (the spacing is less than the mechanical margin), then physical overlap or serious interference is determined to have occurred, and a forced stop must be initiated, generating an anti-overlap speed factor of 0.

[0035] In one optional implementation, based on the current package's center of gravity height and equivalent kinetic friction coefficient, the critical anti-tipping acceleration of the current package at the current slope angle is calculated, and an S-shaped acceleration control curve for the slope motor is generated, including: S2041: Extract the extreme value difference from the two-dimensional point set on the bottom surface along the direction of belt movement, calculate the length of the bottom surface along the belt direction, and calculate the half-length of the bottom support required to resist the overturning moment. The formula is as follows: In the formula, The bottom support is half the length; The length of the base; For the first point in the two-dimensional point set on the bottom surface jThe components of the two-dimensional coordinates of a point on the X-axis; j Point index; These are functions for finding the maximum and minimum values; S2042: Based on torque balance, calculate the first critical acceleration at which the package will tip over around the rear support edge of the bottom. Based on d'Alembert's principle, calculate the second critical acceleration at which the package will not slide downwards. Introducing a stabilization safety factor, calculate the current package's allowable critical acceleration against tipping. The formula is: In the formula, This is the first critical acceleration; It is the acceleration due to gravity; The slope angle is taken as 20°. The bottom support is half the length; The height of the center of mass; The equivalent dynamic friction coefficient between the current bottom surface of the wrapping and the inclined belt is obtained from the belt surface material and wrapping bottom material library (e.g., 0.4 for cardboard boxes and rubber belts). This is the second critical acceleration; The critical acceleration to resist tipping was calculated to be 1.5 m / s². The function is for finding the minimum value; The image stabilization safety factor (valued at 0.8 here); S2043: Calculate the key time points of the S-shaped trajectory based on the anti-tipping critical acceleration, the system's preset maximum acceleration, and the target constant speed required for the security inspection machine to enter the bag; The first stage is the acceleration stage, the second stage is the uniform acceleration stage, and the third stage is the deceleration stage. In other words, a standard S-shaped acceleration curve includes three stages: acceleration, uniform acceleration, and deceleration. The duration of the first stage (acceleration stage) and the local velocity increment reached are calculated using the following formula: In the formula, The duration of the first stage is calculated to be 0.15 seconds. The critical acceleration for resisting tipping; The maximum jerk preset by the system; The local velocity increment for the first stage is calculated to be 0.1125 m / s; Determine whether the maximum acceleration can be reached (i.e.) ,in, Let the target constant velocity required for the security scanner to enter the bag be 0.3 m / s. If this is satisfied, then calculate the duration of the second stage (uniform acceleration stage) using the following formula: In the formula, The duration of the second phase is calculated to be 0.05 s; The duration of the third phase (deceleration phase) is equal to the duration of the first phase; S2044: Based on key time nodes, generate the S-shaped acceleration control curve of the ramp motor, using the following formula: In the formula, For the first t The anti-tipping base velocity trajectory function at any given time is used to generate the anti-tipping base velocity, forming an S-shaped acceleration control curve that varies with time. t For time indexing; The critical acceleration for resisting tipping; The maximum jerk preset for the system is assumed to be 10 m / s³. The duration of the first phase; The duration of the second phase; This refers to the duration of the third phase.

[0036] In one optional implementation, a drive command is generated based on the anti-overlap speed factor and the S-shaped jerk control curve to dynamically adjust the operating state of the inlet ramp motor, including: S2051: Multiply the anti-tipping base speed of the current control cycle in the S-shaped acceleration control curve by the corresponding anti-overlap speed factor to generate the dynamic target speed of the current control cycle, as shown in the formula: In the formula, For the current control cycle Dynamic target speed; The current control period in the S-shaped acceleration control curve The anti-tipping base speed; For the current control cycle Anti-overlap speed factor; S2052: Based on the dynamic target velocity of the current control cycle, the actual command velocity issued in the previous control cycle, the first critical acceleration, and the second critical acceleration, construct the nonlinear adaptive acceleration boundary for the current control cycle, as shown in the formula: In the formula, The desired acceleration for the current control cycle; For the previous control cycle The actual speed at which instructions are issued; The system control cycle (e.g., 1ms or 10ms); This is a dynamic slip risk index, with a value range of (0,1). The closer it is to 1, the higher the slip risk. It is a natural constant; The risk sensitivity coefficient is set to 3. The first critical acceleration With the second critical acceleration The median safety margin in the middle; To adapt to the acceleration boundary, a rapid response is allowed when the deceleration is far from the danger zone, and when it approaches the sliding boundary, the boundary automatically contracts according to a cosine curve to smoothly suppress inertial forces. For the current control cycle Dynamic target speed; The first critical acceleration With the second critical acceleration The smaller value in; The first critical acceleration With the second critical acceleration The larger value in; S2056: Based on the adaptive acceleration boundary, calculate the maximum allowable velocity change within a single control cycle, limit the expected velocity change for the current control cycle to obtain the actual allowable change, and calculate the shaped actual command velocity using the following formula: In the formula, This represents the actual allowable variation. The system control cycle (e.g., 1ms or 10ms); The function is for finding the minimum value; For a sign function, when When >0, =1 (needs acceleration), when When <0, = 1 (Needs to slow down), when When =0, =0; For the current control cycle The actual command speed after shaping; For the previous control cycle The actual speed at which instructions are issued; The desired acceleration for the current control cycle; For the current control cycle Dynamic target speed; For adaptive acceleration boundaries; S2057: Based on the adaptive acceleration boundaries of the current control cycle and the previous control cycle, extract the jerk of the current control cycle and inject it as a feedforward quantity into the velocity command to obtain the final physical target velocity. The formula is as follows: In the formula, For the current control cycle acceleration; For the current control cycle and the previous control cycle Adaptive acceleration boundary; For the current control cycle The final physical target velocity; This is the belt elastic feedforward gain coefficient, used to compensate for the hysteresis of the system's mechanical inertia and elastic deformation, and its value is 0.05. For the current control cycle The feedforward compensation speed; Achieve dimensional matching (converting m / s) 3 (Converted to the equivalent m / s compensation amount). S2058: Converts the final physical target speed determined by the physical speed calibration into a duty cycle control quantity recognizable by the motor driver, obtaining the quantized motor pulse-width modulation (PWM) duty cycle control signal, as shown in the formula: In the formula, For the current control cycle The PWM duty cycle control signal for the motor; For the current control cycle The final physical target velocity; The rated maximum operating speed of the ramp motor is 1.0 m / s. The PWM resolution supported by the driver (e.g., 10000). It is a saturation constraint function; S2059: The motor PWM duty cycle control signal is used as a drive command, encapsulated into a control data frame of a specific industrial bus (such as EtherCAT or RS485 / Modbus), and sent to the servo driver or frequency converter of the ramp motor through the communication interface to dynamically adjust the operating status of the inlet ramp motor.

[0037] This invention also provides a slope-based baggage delivery control system 300 for a security inspection machine based on dynamic constraints, as described in the following embodiments. Figure 3 The system may include the following modules: The 3D vision acquisition module 301 is used to acquire depth images of packages in the ramp area at the entrance of the security inspection machine in real time and reconstruct the 3D point cloud data of the packages. Feature extraction module 302 is used to process the three-dimensional point cloud data and extract the size, centroid height and placement yaw angle of the oriented bounding box on the slope plane. The anti-overlap decision module 303 is used to calculate the anti-overlap speed factor based on the scanning progress status of the previous package in the security inspection machine channel, and in combination with the oriented bounding box and placement yaw angle of the current package. The anti-tipping dynamics module 304 is used to calculate the critical anti-tipping acceleration of the current package under the current slope angle based on the current center of gravity height and equivalent dynamic friction coefficient, and to generate the S-shaped acceleration control curve of the slope motor. The composite adaptive drive module 305 is used to generate drive commands based on the anti-overlap speed factor and the S-shaped jerk control curve, and dynamically adjust the operating state of the inlet ramp motor.

[0038] Based on the same inventive concept, another embodiment of the present invention provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus. Memory, used to store computer programs; When the processor executes the program stored in the memory, it implements the slope package feeding control method of the security inspection machine based on dynamic constraints of the present invention.

[0039] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used in the diagram, but this does not indicate that there is only one bus or one type of bus. The communication interface is used for communication between the aforementioned terminal and other devices. The memory can include Random Access Memory (RAM) or Non-Volatile Memory (NVM), such as at least one disk storage device. Optionally, the memory can also be at least one storage device located remotely from the aforementioned processor.

[0040] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0041] Furthermore, to achieve the above objectives, embodiments of the present invention also propose a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the slope-based baggage delivery control method for security inspection machines based on dynamic constraints, as described in the embodiments of the present invention.

[0042] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, embodiments of the present invention can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of the present invention can take the form of computer program products implemented on one or more computer-usable hardware devices (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0043] The embodiments of the present invention are described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (apparatus), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0044] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0045] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0046] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. "And / or" indicates that either one or both can be chosen. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes the element.

[0047] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for controlling the ramp-driven baggage delivery of a security inspection machine based on dynamic constraints, characterized in that, The method includes: Real-time acquisition of depth images of packages in the ramp area at the entrance of the security inspection machine, and reconstruction of the three-dimensional point cloud data of the packages; The three-dimensional point cloud data is processed to extract the dimensions, centroid height, and placement yaw angle of the oriented bounding box on the slope plane. Based on the scanning progress status of the previous package in the security inspection machine channel, and combined with the oriented bounding box and placement yaw angle of the current package, the anti-overlap speed factor is calculated. Based on the current center of gravity height and equivalent dynamic friction coefficient of the package, calculate the critical acceleration against tipping of the package at the current slope angle, and generate the S-shaped acceleration control curve of the slope motor. Based on the anti-overlap speed factor and the S-shaped jerk control curve, drive commands are generated to dynamically adjust the operating status of the inlet ramp motor.

2. The slope-based baggage feeding control method for security inspection machines based on dynamic constraints according to claim 1, characterized in that, Real-time acquisition of depth images of packages in the ramp area at the entrance of the security checkpoint, and reconstruction of the package's 3D point cloud data, including: A 3D structured light camera positioned directly above the ramp at the entrance of the security checkpoint is used to acquire depth images of packages in the ramp detection area in real time at a preset frame rate. Based on the pre-calibrated camera intrinsic parameter matrix of the 3D structured light camera, the depth image is back-projected onto the camera's three-dimensional coordinate system, and the three-dimensional coordinates of each pixel in the camera coordinate system are calculated to obtain the initial three-dimensional point cloud data. Using the random sampling consensus algorithm, plane fitting is performed on the initial 3D point cloud data to extract the plane with the most occupied points and that conforms to the physical characteristics of the slope, thus obtaining the general equation parameters of the slope background plane. Based on the general equation parameters of the slope background plane, the Euclidean distance from the initial 3D point cloud data to the slope background plane is calculated, and points whose Euclidean distance is less than or equal to the background removal threshold are identified as slope background points and removed, resulting in background-free 3D point cloud data containing only wrapping information. Voxel filtering is used to downsample and denoise the background-removed 3D point cloud data to obtain simplified and enclosed 3D point cloud data; Using hand-eye calibration, the rotation matrix and translation vector between the pre-acquired camera coordinate system and the belt coordinate system are determined. Based on the rotation matrix and translation vector, a homogeneous coordinate transformation is performed on each point in the simplified wrapped 3D point cloud data, and its new coordinates in the belt coordinate system are calculated to obtain the final 3D point cloud data wrapped in the belt coordinate system.

3. The slope-based baggage feeding control method for security inspection machines based on dynamic constraints according to claim 2, characterized in that, The 3D point cloud data is processed to extract the dimensions, centroid height, and placement yaw angle of the bounding box on the slope plane, including: The final 3D point cloud data wrapped in the belt coordinate system is projected onto the slope plane to obtain a 2D projection point set. Then, the points in the final 3D point cloud data whose Z-axis component is in the minimum value range are extracted and projected onto the slope plane to obtain the bottom 2D point set. Based on the two-dimensional point set on the bottom surface, calculate the principal axis direction wrapped on the slope plane and extract the corresponding placement yaw angle; Based on the placement yaw angle, construct a two-dimensional rotation matrix, and based on the two-dimensional rotation matrix, rotate the two-dimensional projection point set to an aligned coordinate system parallel to the main axis of the package, and obtain the rotated two-dimensional projection point set; In the aligned coordinate system, the extreme values ​​are obtained along the principal axis and the vertical direction to calculate the length and width of the package and obtain the dimensions of the oriented bounding box; The centroid height is obtained by taking the weighted average of the Z-axis components of all points in the final 3D point cloud data wrapped in the belt coordinate system.

4. The slope-based baggage feeding control method for security inspection machines based on dynamic constraints according to claim 3, characterized in that, Based on the two-dimensional point set on the bottom surface, the principal axis direction wrapped on the slope plane is calculated to extract the placement yaw angle, including: Calculate the covariance matrix of the two-dimensional point set on the bottom surface; Eigenvalue decomposition is performed on the covariance matrix to obtain the first eigenvalue, the second eigenvalue, and the corresponding eigenvectors. Using the eigenvector corresponding to the largest first eigenvalue as the principal axis direction of the wrapping bottom surface, the angle between the principal axis direction and the belt movement direction is calculated to obtain the placement yaw angle.

5. The slope baggage feeding control method for security inspection machines based on dynamic constraints according to claim 4, characterized in that, Based on the scanning progress status of the previous package in the security inspection machine channel, and combined with the oriented bounding box and placement yaw angle of the current package, the anti-overlap velocity factor is calculated, including: By interacting with the main control system of the security inspection machine, the start time and current time of the previous package entering the security inspection machine channel are obtained, and the displacement progress of the previous package in the security inspection machine channel is calculated. Extract the scanning progress status when the previous package enters, the scanning progress status includes the corresponding historical oriented bounding box size, historical placement yaw angle and two-dimensional coordinate value of historical geometric center; If the difference between the length of the historical bounding box of the previous package and the displacement progress is greater than 0, it is determined that there is still a part of the tail of the previous package stuck in the entrance slope area. A two-dimensional polygon vertex set of the stuck part of the previous package in the slope belt coordinate system is constructed and the next step is carried out. Otherwise, the anti-overlap speed factor of 1 is directly generated. Define a local coordinate system with the geometric center of the current package as the origin, and define an initial set of standard vertex coordinates consisting of the four standard vertex coordinates of the current package when it is not rotated, in combination with the size of the directed bounding box of the current package. Based on the placement yaw angle, construct a two-dimensional rotation matrix, and based on the two-dimensional rotation matrix, rotate and translate the initial standard vertex coordinate set to the belt global coordinate system, and calculate the actual standard vertex coordinate set composed of the four actual two-dimensional polygon vertex sets of the currently wrapped directed bounding box on the slope plane. Taking the opposite direction of the belt movement as the positive X-axis, extract the minimum value of the vertex X-coordinate of all actual standard vertices from the actual standard vertex coordinate set, and use it as the X-coordinate of the front end of the current package; Extract the maximum value of the vertex X coordinates of all actual standard vertices from the vertex set of the two-dimensional polygon, and use it as the tail X coordinate of the previous wrapper. Calculate the difference between the X coordinates of the front end and the X coordinates of the rear end to obtain the minimum parallel projection distance between the current package and the previous package in the direction of belt movement. If the minimum parallel projection distance is less than 0, it is determined that the current package and the previous package have overlapped in the X-axis projection, and proceed to the next step. Based on the current real-time operating speed of the inclined conveyor belt, a dynamic safety threshold for anti-overlap determination is generated; Based on the minimum parallel projection spacing and dynamic security threshold, calculate the anti-overlap security margin required for the current package to enter the security inspection machine channel, and generate the corresponding anti-overlap speed factor.

6. The slope-based baggage delivery control method for security inspection machines based on dynamic constraints according to claim 5, characterized in that, The method for generating the anti-overlap speed factor includes: If the anti-overlap safety margin is greater than or equal to 0, the spacing is deemed sufficient, and operation is allowed at the maximum speed of the anti-tipping plan, generating an anti-overlap speed factor of 1. If the anti-overlap safety margin is less than 0 and greater than or equal to the negative value of the fixed mechanical margin, it is determined that the spacing is insufficient but there is no collision. It is necessary to slow down proportionally and creep to wait for the channel to be emptied. The anti-overlap speed factor is the ratio of the minimum parallel projection spacing to the dynamic safety threshold. If the anti-overlap safety margin is less than the negative value of the fixed mechanical margin, it is determined that physical overlap or serious interference has occurred, and a forced stop must be initiated, generating an anti-overlap speed factor of 0.

7. The slope-based baggage feeding control method for security inspection machines based on dynamic constraints according to claim 6, characterized in that, Based on the current center of gravity height and equivalent kinetic friction coefficient of the package, calculate the critical anti-tipping acceleration of the package at the current slope angle, and generate the S-shaped acceleration control curve of the slope motor, including: Extract the extreme value difference of the two-dimensional point set on the bottom surface along the direction of belt movement, calculate the length of the bottom surface along the belt direction, and calculate the half length of the bottom support required to resist the overturning moment. Based on torque balance, the first critical acceleration at which the package tilts over the bottom rear support edge is calculated. Based on d'Alembert's principle, the second critical acceleration at which the package does not slide downwards is calculated. An anti-shake safety factor is introduced to calculate the current package's allowable anti-tipping critical acceleration. Based on the critical anti-tipping acceleration, the system's preset maximum acceleration, and the target constant speed required for the security inspection machine to enter the bag, calculate the key time nodes of the S-shaped trajectory; Based on key time points, generate the S-shaped acceleration control curve of the ramp motor.

8. The slope-based baggage feeding control method for security inspection machines based on dynamic constraints according to claim 7, characterized in that, The first stage is the acceleration stage, the second stage is the uniform acceleration stage, and the third stage is the deceleration stage.

9. The slope baggage feeding control method for security inspection machines based on dynamic constraints according to claim 8, characterized in that, Based on the anti-overlap speed factor and the S-shaped jerk control curve, drive commands are generated to dynamically adjust the operating state of the inlet ramp motor, including: Multiply the anti-tipping base speed of the current control cycle in the S-shaped acceleration control curve with the corresponding anti-overlap speed factor to generate the dynamic target speed of the current control cycle. Based on the dynamic target velocity of the current control cycle, the actual command velocity issued in the previous control cycle, the first critical acceleration, and the second critical acceleration, a nonlinear adaptive acceleration boundary for the current control cycle is constructed. Based on the adaptive acceleration boundary, the maximum allowable velocity change within a single control cycle is calculated. The expected velocity change in the current control cycle is then limited to obtain the actual allowable change. Finally, the shaped actual command velocity is calculated. Based on the adaptive acceleration boundaries of the current control cycle and the previous control cycle, the jerk of the current control cycle is extracted and injected into the velocity command as a feedforward quantity to obtain the final physical target velocity. The final physical target speed determined by the physical speed calibration is converted into a duty cycle control quantity that can be recognized by the motor driver, resulting in a quantized motor PWM duty cycle control signal. The motor PWM duty cycle control signal is used as a drive command, encapsulated into a control data frame of a specific industrial bus, and sent to the servo driver or frequency converter of the ramp motor through the communication interface to dynamically adjust the operating status of the inlet ramp motor.

10. A security inspection machine ramp baggage feeding control system based on dynamic constraints, used to implement the security inspection machine ramp baggage feeding control method as described in any one of claims 1-9, characterized in that, The system includes: The 3D vision acquisition module is used to acquire depth images of packages in the ramp area at the entrance of the security inspection machine in real time and reconstruct the 3D point cloud data of the packages. The feature extraction module is used to process the three-dimensional point cloud data and extract the size, centroid height, and placement yaw angle of the oriented bounding box on the slope plane. The anti-overlap decision module is used to calculate the anti-overlap speed factor based on the scanning progress status of the previous package in the security inspection machine channel, combined with the oriented bounding box and placement yaw angle of the current package. The anti-tipping dynamics module is used to calculate the critical anti-tipping acceleration of the current package under the current slope angle based on the current center of gravity height and equivalent dynamic friction coefficient, and generate the S-shaped acceleration control curve of the slope motor. The composite adaptive drive module is used to generate drive commands based on the anti-overlap speed factor and the S-shaped jerk control curve, and dynamically adjust the operating state of the inlet ramp motor.

Images