Online loading method for luggage in civil aviation airport

Abstract

The invention relates to the technical field of airport logistics, in particular to a civil aviation airport luggage online loading method which comprises the following steps: S1, before luggage is tipped over according to a specified sequence and slides into a chute, a soft package and a hard package in the luggage are distinguished through image analysis, and parameter information of the hard package luggage is acquired; s2, loading a loading optimization model based on maximum utilization of a loading space, and analyzing and calculating to obtain a stacking sorting scheme; s3, optimizing a stacking sorting scheme through a sparrow search algorithm in combination with an ROV strategy and a golden sine strategy; s4, the mechanical arm grabs the hard luggage sliding into the bottom of the sliding groove, and stacking is completed; the soft luggage sliding into the bottom of the sliding groove is placed in a reserved cache area; and S5, repeatedly executing the steps S2-S4 until the loading space of the trailer is filled with the hard package, and finally placing the soft package in the cache region at the uppermost part. According to the invention, sequential stacking of departure luggage and maximum utilization of a trailer loading space can be completed, the workload of front-line personnel is reduced, and the efficiency of a whole stacking system assembly line is improved.

Description

Technical Field

[0001] The invention relates to the technical field of airport logistics, and in particular to an online loading method for luggage at a civil aviation airport. Background Art

[0002] Faced with the increasing passenger throughput at airports, it is imperative to realize the automated loading and unloading of luggage. However, when the aircraft departs from the civil aviation airport, how to stack the checked luggage has always been a difficult problem. At present, through automation technology, the traditional manual handling of outbound luggage has been gradually transformed into intelligent ones, which effectively reduces the operating pressure of front-line positions, and many domestic airports have begun to actively explore intelligent luggage sorting and handling technologies. However, traditional research is mostly limited to the stacking of cartons with known quantities and known size information, and the luggage of passengers in the airport is highly random. Only the luggage information that has been completed and entered the tray sorting machine is known. Therefore, it is an important problem to be solved to efficiently use this known information to optimize the stacking of luggage to achieve high-density and high-efficiency stacking of checked luggage for subsequent transportation of luggage. Summary of the invention

[0003] The purpose of the present invention is to provide an online loading method for civil aviation airport luggage, which can enable the robot to automatically plan the position and posture trajectory from the grabbing point to the target point according to the provided luggage information, complete the disorderly stacking of outbound luggage and maximize the use of trailer loading space, reduce the workload of front-line personnel, and improve the efficiency of the entire stacking system assembly line.

[0004] In order to achieve the above object, a method for online loading of luggage at a civil aviation airport is provided, comprising the following steps:

[0005] S1. Before the luggage is tipped over in a specified order and slides into the chute, image analysis and recognition are performed on the luggage to distinguish between soft bags and hard bags in the luggage, and parameter information of the hard baggage is obtained, wherein the parameter information includes size information, quantity information and sequence information;

[0006] S2. Loading a loading optimization model based on maximizing the use of loading space; according to the stacking type information of the hard baggage stacked on the trailer, the parameter information of the hard baggage and the stacking constraint conditions, the loading optimization model is used to analyze and calculate the key points of the stacking of the hard baggage that has been identified by image analysis, and obtain the number of key points; then, according to the number information of the hard baggage and the number of key points, the number of stacked baggage is selected, and a stacking sorting scheme is obtained by analysis and calculation; the stacking constraint conditions include the volume constraint corresponding to the hard baggage size information and the trailer loading space, and the sequence constraint corresponding to the sequence information;

[0007] S3, optimize the stacking sorting scheme by using the sparrow search algorithm combined with the ROV strategy and the golden sine strategy;

[0008] S4, enabling the robotic arm to grab the hard baggage that slides into the bottom of the chute, and complete the palletizing according to the palletizing sorting plan; wherein the soft baggage that slides into the bottom of the chute is grabbed by the robotic arm and placed in the reserved buffer area;

[0009] S5. After completing the palletizing of all the hard baggage corresponding to the palletizing sorting plan, repeat steps S2-S4 until the hard baggage fills the trailer loading space, and finally the soft baggage in the buffer area is placed on the top.

[0010] Furthermore, the steps of selecting the number of stacked luggage in step S2 are as follows:

[0011] S201, when performing analysis and calculation of key points for stacking hard baggage that has been identified through image analysis using the loading optimization model, analyze the key points of the remaining loading space on the trailer that can be used to place the hard baggage, set the number of key points to M, and obtain the coordinates of the key points; set the number of hard baggage that has been identified through image analysis and not yet stacked in step S1 to N;

[0012] S202, judging the relationship between M and N. If M is greater than N, only the information of the first N pieces of hard baggage is used for the current palletizing plan, and a palletizing sorting scheme is obtained according to the coordinates of the N key points. If M is less than N, only the information of the first M pieces of hard baggage is used for the current palletizing plan, and a palletizing sorting scheme is obtained according to the coordinates of the M key points.

[0013] Furthermore, the specific process of loading the optimization model to perform key point analysis and calculation in step S2 includes the following steps:

[0014] S2011. Establish a two-dimensional coordinate system corresponding to the stack type of the stacked hard baggage according to the stack type information, find the point in the two-dimensional coordinate system that changes from horizontal to vertical, and obtain the two-dimensional key point; then find the hard baggage containing the key point, and generate an envelope; the stack type information is the position information of the stacked hard baggage and the maximum volume of the loading space in the trailer; the size information is the length, width and height of the baggage, and the maximum volume is the maximum allowable length, width and depth of the loading space in the trailer;

[0015] S2012, reconstructing the three-dimensional coordinate system, searching for corner points on the envelope line that change from vertical to horizontal, until all corner points in the three-dimensional coordinate system are traversed, obtaining all corner points as key points, and recording the coordinates of the key points;

[0016] The initial placement point is the point (0, 0, 0) in the spatial coordinate system; the coordinates of the key point are the vertex coordinates of the intersection of the left view, the rear view, and the bottom view after the hard baggage is placed;

[0017] S2013. After each stacking of M or N hard baggage, the trailer's stacking information is acquired again, and new key points are generated according to the above steps, and the coordinates of the new key points are acquired, and then output as the number and coordinates of the key points that can be placed in the stacking.

[0018] Further, the specific process of optimizing the stacking sorting scheme in step S3 is as follows:

[0019] S301, optimizing the stacking sorting scheme by using a sparrow search algorithm;

[0020] S302, discretizing the sparrow population through the ROV strategy and converting it into a discrete stacking sorting scheme;

[0021] S303, calculating the fitness of the individuals of the sparrow population according to the fitness function; obtaining the best position and its fitness, and the worst position and its fitness;

[0022] S304, updating the finder position information in the sparrow search algorithm according to the golden sine strategy;

[0023] S305, updating the joiner location information and the alerter location information in the sparrow search algorithm;

[0024] S306, judging whether the iterative termination condition of the sparrow search algorithm is met, if so, terminating the algorithm and outputting the optimal stacking sorting scheme; if not, calculating a new fitness function, re-executing step S302, and iterating.

[0025] Furthermore, the stacking constraints include:

[0026] Size constraint: means that the size of each piece of luggage must be smaller than the trailer box;

[0027]

[0028] Volume constraint: After the baggage is stacked, the size of the stack must be smaller than the trailer stacking space;

[0029]

[0030] Orthogonal constraint: means that the luggage can only be placed horizontally or orthogonally to the trailer;

[0031]

[0032] Embedding constraint: means that luggage cannot be embedded in each other;

[0033]

[0034] Posture constraint: Luggage loading posture is only allowed to be placed upside down, not vertically;

[0035] u≠v

[0036] in,

[0037] Sequence constraint: The baggage stacking order is consistent with the sorter conveying order;

[0038]

[0039] Among them, l i , w i ,h i are the length, width and height of the i-th piece of luggage to be loaded; L, W and H are the length, width and height of the interior of the trailer; x li ,y li , z li are the vertex coordinates of the intersection of the left view, rear view, and bottom view of the i-th piece of luggage that has been loaded, respectively. ri ,y ri , z ri S are the vertex coordinates of the intersection of the right view, front view, and top view of the i-th completed baggage; i is the baggage sequence on the tray sorter, P i is the actual baggage stacking order, and N is the number of baggage to be stacked.

[0040] Furthermore, the fitness function is defined as the minimum difference between the height and bottom area of ​​the luggage to be placed and the luggage on the left, rear and bottom sides. The fitness function calculation formula is as follows:

[0041]

[0042] Among them, l i , b i are the height differences between the luggage to be stacked and the luggage on the left and back of the placement point, s j 、s i They are the support area and the bottom area for the luggage to be placed.

[0043] Furthermore, in step S304, the update formula for improving the finder position information in the sparrow search algorithm by the golden sine strategy is as follows:

[0044]

[0045] x1=-π+(1-τ)×2π, x2=-π+τ×2π

[0046]

[0047] Among them, r1∈[0,2π], r2∈[0,π], which respectively determine the next individual's moving distance and moving direction;

[0048] R2∈[0,1] represents the warning value, and ST∈[0.5,1] represents the alert value;

[0049] is the current optimal position, Q is a random number that obeys the normal distribution; L represents a 1×d all-1 matrix, d is the dimension of the solution space, and x1 and x2 are coefficients obtained according to the golden section coefficient τ.

[0050] Principles and advantages:

[0051] This scheme studies the online loading problem of airline baggage, takes the trailer space utilization rate as the optimization target under the constraints of volume and loading sequence, and establishes a loading optimization model based on maximizing loading space. The sparrow search algorithm is improved, and the online loading of baggage based on ROV improved sparrow search fusion key point analysis is carried out. According to the baggage information provided by the decision system in the early stage, the robot's mechanical arm automatically plans the position and posture trajectory planning from the grasping point to the target point, completes the disordered stacking of outbound baggage and maximizes the utilization of trailer loading space, thereby reducing the workload of front-line personnel and improving the efficiency of the entire stacking system assembly line. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a flowchart of an online loading method for luggage at a civil aviation airport according to an embodiment of the present invention;

[0053] Figure 2 It is a schematic diagram of the structure of the pallet sorter, chute, robot arm and trailer;

[0054] Figure 3 Schematic diagram of standard luggage model

[0055] Figure 4 Schematic diagram for key point analysis;

[0056] Figure 5 It is a schematic diagram of discretization processing based on ROV strategy;

[0057] Figure 6 This is a diagram showing the luggage placement. DETAILED DESCRIPTION

[0058] The following is further described in detail through specific implementation methods:

[0059] Example

[0060] An online loading method for civil aviation airport luggage, basically as follows Figure 1 As shown, the following steps are included:

[0061] S1. Before the luggage is tipped over by the tray sorter in a specified order and slides into the chute, a depth camera is used to perform image analysis and recognition on the material, posture and size of the luggage on the sorter to distinguish between soft bags and hard bags in the luggage, and obtain parameter information of the hard baggage; the parameter information includes size information, quantity information N and sequence information; Regarding posture, during the luggage transportation process, since there is a bag tipping and bag splitting mechanism before the luggage enters the sorter, and the robot arm will not change the luggage from a lying posture to a standing posture, the posture of the luggage to be stacked is one and only the following two, such as Figure 6 shown.

[0062] S2. Loading a loading optimization model based on maximizing the use of loading space; according to the stacking type information of the hard baggage stacked on the trailer, the parameter information of the hard baggage and the stacking constraint conditions, the loading optimization model is used to analyze and calculate the key points of the stacking of the hard baggage that has been identified by image analysis, and obtain the number of key points; then, according to the number information of the hard baggage and the number of key points, the number of stacked baggage is selected, and a stacking sorting scheme is obtained by analysis and calculation; the stacking constraint conditions include the volume constraint corresponding to the hard baggage size information and the trailer loading space, and the sequence constraint corresponding to the sequence information;

[0063] In luggage stacking, except for soft bags, the luggage only has different sizes, but the shapes are almost the same, which are rectangular. Conventional luggage has certain rigidity and load-bearing capacity, so it can be stacked without considering the deformation of the luggage due to squeezing. The specific model is as follows Figure 3 As shown in the figure, all non-soft baggage in the algorithm model is considered as a cuboid. While considering the volume constraint and order constraint, it is also necessary to consider the baggage volume constraint, size constraint, embedding constraint, etc.

[0064] The stacking constraints are as follows:

[0065] Size constraint: means that the size of each piece of luggage must be smaller than the trailer box;

[0066]

[0067] Volume constraint: After the baggage is stacked, the size of the stack must be smaller than the trailer stacking space;

[0068]

[0069] Orthogonal constraint: means that the luggage can only be placed horizontally or orthogonally to the trailer;

[0070]

[0071] Embedding constraint: means that luggage cannot be embedded in each other;

[0072]

[0073] Posture constraint: Luggage loading posture is only allowed to be placed upside down, not vertically;

[0074] u≠v

[0075] in,

[0076] Sequence constraint: The baggage stacking order is consistent with the sorter conveying order;

[0077]

[0078] Among them, l i , w i ,h i are the length, width and height of the i-th piece of luggage to be loaded; L, W and H are the length, width and height of the interior of the trailer; x li ,y li , z li are the vertex coordinates of the intersection of the left view, rear view, and bottom view of the i-th piece of luggage that has been loaded, respectively. ri ,y ri , z ri S are the vertex coordinates of the intersection of the right view, front view, and top view of the i-th completed baggage; i is the baggage sequence on the tray sorter, P i is the actual baggage stacking order, and N is the number of baggage to be stacked.

[0079] The steps of selecting the number of palletized luggage in step S2 are as follows:

[0080] S201, when performing analysis and calculation of key points for stacking hard baggage that has been identified through image analysis using the loading optimization model, analyze the key points of the remaining loading space on the trailer that can be used to place the hard baggage, set the number of key points to M, and obtain the coordinates of the key points; set the number of hard baggage that has been identified through image analysis and not yet stacked in step S1 to N;

[0081] S202, judging the relationship between M and N. If M is greater than N, only the information of the first N pieces of hard baggage is used for the current palletizing plan, and a palletizing sorting scheme is obtained according to the coordinates of the N key points. If M is less than N, only the information of the first M pieces of hard baggage is used for the current palletizing plan, and a palletizing sorting scheme is obtained according to the coordinates of the M key points.

[0082] The specific process of loading the optimization model to perform key point analysis and calculation in step S2 includes the following steps:

[0083] S2011. Establish a two-dimensional coordinate system corresponding to the stack type of the stacked hard baggage according to the stack type information, find the point that changes from horizontal to vertical in the two-dimensional coordinate system, and obtain the two-dimensional key point, such as Figure 4 As shown by the black dots; then look for hard baggage containing key points, such as Figure 4 baggage 2, baggage 3, and baggage 4 in the trailer, and generate an envelope; the stacking type information is the position information of the stacked hard baggage and the maximum volume of the loading space in the trailer; the size information is the length, width, and height of the baggage, and the maximum volume is the maximum allowable length, width, and depth of the loading space in the trailer;

[0084] S2012, reconstructing the three-dimensional coordinate system, searching for corner points on the envelope line that change from vertical to horizontal, until all corner points in the three-dimensional coordinate system are traversed, obtaining all corner points as key points, and recording the coordinates of the key points;

[0085] The initial placement point is the point (0, 0, 0) in the spatial coordinate system; the coordinates of the key point are the vertex coordinates of the intersection of the left view, the rear view, and the bottom view after the hard baggage is placed;

[0086] S2013. After each stacking of M or N hard baggage, the trailer's stacking information is acquired again, and new key points are generated according to the above steps, and the coordinates of the new key points are acquired, and then output as the number and coordinates of the key points that can be placed in the stacking.

[0087] S3, optimize the stacking sorting scheme by using the sparrow search algorithm combined with the ROV strategy and the golden sine strategy;

[0088] The specific process of optimizing the stacking sorting scheme in step S3 is as follows:

[0089] S301, optimize the stacking sorting scheme by using the sparrow search algorithm; the basic process is as follows:

[0090] Assuming that the initial population size of sparrows is n, in step S202, if M is greater than N, only the information of the first N pieces of hard baggage is used for the current stacking planning, and the stacking sorting scheme is obtained according to the coordinates of the N key points. Therefore If it is less than N, only the information of the first M pieces of hard baggage is used for the current palletizing plan, and the palletizing sorting scheme is obtained according to the coordinates of the M key points, so n = M! The population individual is a palletizing sorting scheme with a sequential relationship.

[0091] The initial population can be expressed as follows:

[0092]

[0093] In Equation (1), d represents the dimension of the solution space, n represents the number of sparrows, and each row represents a sparrow individual, that is, a palletizing sorting scheme. The population fitness function can be expressed as shown in Equation (2):

[0094]

[0095] where: f is the individual fitness function.

[0096] The fitness function f is defined as the minimum difference between the height and the bottom area of the luggage to be placed and the luggage on the left, rear, and lower sides. The calculation formula of the fitness function is as follows:

[0097]

[0098] where, l i , b i are the height differences between the luggage to be palletized and the luggage on the left and rear sides of the placement point respectively, and s j , s i are the support area and the bottom area of the luggage to be placed respectively.

[0099] While setting the initial population size of the sparrows, set the basic parameters such as the upper and lower limits of the values of the sparrow search algorithm SSA and the number of iterations.

[0100] In the sparrow search algorithm SSA, the responsibility of the discoverer is to help the entire sparrow population find food and provide the foraging direction for the joiners. Therefore, the foraging search range of the discoverer is larger than that of the joiners. In each iteration process, the position update of the discoverer is described as shown in Equation (3):

[0101]

[0102] where t is the current iteration number, iter max is the maximum iteration number, represents the position information of the i-th sparrow in the j-th dimension at time t. R2 ∈ [0, 1] represents the warning value, and ST ∈ [0.5, 1] represents the warning threshold.

[0103] Specifically, when R2 ≥ ST, it means that some sparrow populations are in danger, and the sparrow population needs to find a new safe area to forage again; on the contrary, when R2 < ST, at this time the sparrow group is in a relatively safe environment and can expand the foraging range to obtain more food.

[0104] The joiners in the foraging state have two different states. At In the case of , it is in a state of food shortage and needs to go to other areas to forage. At the same time, some joiners will compete with the discoverer for food. If the joiner wins, it will get the food of the discoverer. The joiner position update formula is shown in (4):

[0105]

[0106] In formula (4), Indicates the optimal position occupied by the current finder in the t+1 iteration.

[0107] It is usually assumed that individuals who perceive danger are randomly generated from the population. As a vigilant, their position update is shown in formula (5):

[0108]

[0109] Where X best is the optimal position. β is the step size control parameter, and K is a random number in [-1,1]. i represents the individual fitness value, f g is the optimal fitness value, f w is the worst fitness value. ε is a constant to avoid the denominator being 0.

[0110] S302, discretizing the sparrow population through the ROV strategy and converting it into a discrete stacking sorting scheme; the steps of discretizing the sparrow population through the ROV strategy in step S302 are as follows:

[0111] S3021, sorting all position elements of the sparrow individuals in ascending order; the position elements are the coordinates of the key points;

[0112] S3022. Assign a unique ROV value to each position element in ascending order, and a sequence number corresponding to the ascending order;

[0113] S3023. Rearrange the ROV values ​​according to the original arrangement order of the element positions and construct a palletizing sorting scheme.

[0114] like Figure 5 As shown, assume that the position elements are all random numbers in [-σ,σ], which constitute the individual sparrows. First, all the position elements of the sparrow individuals in the first row are arranged in ascending order from small to large, and the ascending order of the second row is obtained. The third row is the element number of the position element in the initial state. The fourth row is the stacking scheme of the position element according to the element number in the initial state. The fifth row assigns a unique ROV value to each element in natural order, and then rearranges the ROV value according to the original element position to construct the stacking scheme (stacking sorting scheme).

[0115] S303, calculating the fitness of the individuals of the sparrow population according to the fitness function; obtaining the best position and its fitness, and the worst position and its fitness;

[0116] S304, updating the finder position information in the sparrow search algorithm according to the golden sine strategy; introducing the golden section coefficient in the process of updating the finder position to improve the algorithm's optimization ability.

[0117] As shown in formula (6), assuming that there are s population individuals in the d-dimensional solution space, the position update of individual i (i = 1, 2, 3, ..., s) in the tth iteration can be expressed as:

[0118]

[0119] Where: r1∈[0,2π], r2∈[0,π], which respectively determine the next individual's moving distance and moving direction. is the global optimal position of the t-th iteration, x1 = -π + (1-τ) × 2π, x2 = -π + τ × 2π are the coefficients obtained according to the golden section coefficient, and the golden section coefficient is as follows:

[0120]

[0121] These coefficients can be used to move from the boundary of each dimension to the middle until the optimal value of each dimension is found.

[0122] The golden sine strategy is used to improve the update formula of the discoverer position in the sparrow search algorithm to improve the global search ability of the algorithm. The improved new position update method is as follows:

[0123]

[0124] in, is the current optimal position, Q is a random number that obeys the normal distribution; L represents a 1×d all-1 matrix, and d is the dimension of the solution space.

[0125] According to the above formula, in the improved position update formula, the discoverer will communicate with the optimal individual every time the position is updated, which improves the defect of lack of communication between individuals in the original algorithm and makes more full use of the information of the current optimal solution; on the other hand, the improved algorithm can avoid falling into the local optimal situation to a certain extent, effectively improving the problem of premature convergence of the original algorithm. At the same time, the search space is gradually narrowed by the golden section coefficient, and the search distance and direction of the sparrow are controlled by r1 and r2, which optimizes the optimization method of the sparrow search algorithm and further coordinates the global exploration and local development capabilities of the algorithm.

[0126] S305, updating the joiner location information and the alerter location information in the sparrow search algorithm;

[0127] S306, judging whether the iterative termination condition of the sparrow search algorithm is met, if so, terminating the algorithm and outputting the optimal stacking sorting scheme; if not, calculating a new fitness function, re-executing step S302, and iterating.

[0128] S4. During the palletizing process, the robot arm automatically plans a route to grab the hard baggage that slides into the bottom of the chute, and completes the palletizing according to the palletizing sorting plan; the robot arm automatically plans the route based on the existing image recognition algorithm and route planning algorithm. Among them, the soft baggage that slides into the bottom of the chute is grabbed by the robot arm and placed in the reserved buffer area.

[0129] S5. After completing the palletizing of all the baggage corresponding to the palletizing sorting scheme, repeat steps S2-S4 until the trailer loading space is filled with hard bags. As the packing progresses, until the program deems that the packing space can no longer be used, the soft bags in the buffer area are processed and placed on top of the stacked hard bags by the robotic arm. In the actual airport, the trailer is fixed with a net bag before being towed away by the tractor, so there is no risk of falling, thereby achieving a complete palletizing process.

[0130] During the stacking process, after obtaining the parameter information of the hard baggage that has been checked in, the robotic arm will grab one hard baggage as it falls down from the chute, without having to wait until all the hard baggage has fallen down before grabbing it.

[0131] The above is only an embodiment of the present invention. The common sense such as the known specific structure and characteristics in the scheme is not described in detail here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field of the invention before the application date or priority date, can obtain all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The protection scope required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. An online loading method for luggage at a civil aviation airport, characterized in that: The following steps are involved: S1. Before the luggage is tipped over in a specified order and slides into the chute, image analysis and recognition are performed on the luggage to distinguish between soft bags and hard bags in the luggage, and parameter information of the hard baggage is obtained, wherein the parameter information includes size information, quantity information and sequence information; S2. Loading a loading optimization model based on maximizing the use of loading space; according to the stacking type information of the hard baggage stacked on the trailer, the parameter information of the hard baggage and the stacking constraint conditions, the loading optimization model is used to analyze and calculate the key points of the stacking of the hard baggage that has been identified by image analysis, and obtain the number of key points; then, according to the number information of the hard baggage and the number of key points, the number of stacked baggage is selected, and a stacking sorting scheme is obtained by analysis and calculation; the stacking constraint conditions include the volume constraint corresponding to the hard baggage size information and the trailer loading space, and the sequence constraint corresponding to the sequence information; S3, optimize the stacking sorting scheme by using the sparrow search algorithm combined with the ROV strategy and the golden sine strategy; S4, enabling the robotic arm to grab the hard baggage that slides into the bottom of the chute, and complete the palletizing according to the palletizing sorting plan; wherein the soft baggage that slides into the bottom of the chute is grabbed by the robotic arm and placed in the reserved buffer area; S5. After completing the palletizing of all the hard baggage corresponding to the palletizing sorting plan, repeat steps S2-S4 until the hard baggage fills the trailer loading space, and finally the soft baggage in the buffer area is placed on the top.

2. The online loading method for luggage at a civil aviation airport according to claim 1, characterized in that: The steps of selecting the number of stacked luggage in step S2 are as follows: S201, when performing analysis and calculation of key points for stacking hard baggage that has been identified through image analysis using the loading optimization model, analyze the key points of the remaining loading space on the trailer that can be used to place the hard baggage, set the number of key points to M, and obtain the coordinates of the key points; set the number of hard baggage that has been identified through image analysis and not yet stacked in step S1 to N; S202, judging the relationship between M and N. If M is greater than N, only the information of the first N pieces of hard baggage is used for the current palletizing plan, and a palletizing sorting scheme is obtained according to the coordinates of the N key points. If M is less than N, only the information of the first M pieces of hard baggage is used for the current palletizing plan, and a palletizing sorting scheme is obtained according to the coordinates of the M key points.

3. The method for online loading of luggage at a civil aviation airport according to claim 2, characterized in that: The specific process of loading the optimization model to perform key point analysis and calculation in step S2 includes the following steps: S2011. Establish a two-dimensional coordinate system corresponding to the stack type of the stacked hard baggage according to the stack type information, find the point in the two-dimensional coordinate system that changes from horizontal to vertical, and obtain the two-dimensional key point; then find the hard baggage containing the key point, and generate an envelope; the stack type information is the position information of the stacked hard baggage and the maximum volume of the loading space in the trailer; the size information is the length, width and height of the baggage, and the maximum volume is the maximum allowable length, width and depth of the loading space in the trailer; S2012, reconstructing the three-dimensional coordinate system, searching for corner points on the envelope line that change from vertical to horizontal, until all corner points in the three-dimensional coordinate system are traversed, obtaining all corner points as key points, and recording the coordinates of the key points; The initial placement point is the point (0, 0, 0) in the spatial coordinate system; the coordinates of the key point are the vertex coordinates of the intersection of the left view, the rear view, and the bottom view after the hard baggage is placed; S2013. After each stacking of M or N hard baggage, the trailer's stacking information is acquired again, and new key points are generated according to the above steps, and the coordinates of the new key points are acquired, and then output as the number and coordinates of the key points that can be placed in the stacking.

4. The method for online loading of luggage at a civil aviation airport according to claim 3, characterized in that: The specific process of optimizing the stacking sorting scheme in step S3 is as follows: S301, optimizing the stacking sorting scheme by using a sparrow search algorithm; S302, discretizing the sparrow population through the ROV strategy and converting it into a discrete stacking sorting scheme; S303, calculating the fitness of the individuals of the sparrow population according to the fitness function; obtaining the best position and its fitness, and the worst position and its fitness; S304, updating the finder position information in the sparrow search algorithm according to the golden sine strategy; S305, updating the joiner location information and the alerter location information in the sparrow search algorithm; S306, judging whether the iterative termination condition of the sparrow search algorithm is met, if so, terminating the algorithm and outputting the optimal stacking sorting scheme; if not, calculating a new fitness function, re-executing step S302, and iterating.

5. The method for online loading of luggage at a civil aviation airport according to claim 4, characterized in that: The steps of discretizing the sparrow population by the ROV strategy in step S302 are as follows: S3021, sorting all position elements of the sparrow individuals in ascending order; the position elements are the coordinates of the key points; S3022. Assign a unique ROV value to each position element in ascending order, and a sequence number corresponding to the ascending order; S3023. Rearrange the ROV values ​​according to the original arrangement order of the element positions and construct a palletizing sorting scheme.

6. The method for online loading of luggage at a civil aviation airport according to claim 5, characterized in that: The stacking constraints include: Size constraint: means that the size of each piece of luggage must be smaller than the trailer box; Volume constraint: After the baggage is stacked, the size of the stack must be smaller than the trailer stacking space; Orthogonal constraint: means that the luggage can only be placed horizontally or orthogonally to the trailer; Embedding constraint: means that luggage cannot be embedded in each other; Posture constraint: Luggage loading posture is only allowed to be placed upside down, not vertically; u≠v in, Sequence constraint: The baggage stacking order is consistent with the sorter conveying order; Among them, l i , w i ,h i are the length, width and height of the i-th piece of luggage to be loaded; L, W and H are the length, width and height of the interior of the trailer; x li ,y li , z li are the vertex coordinates of the intersection of the left view, rear view, and bottom view of the i-th piece of luggage that has been loaded, respectively. ri ,y ri , z ri S are the vertex coordinates of the intersection of the right view, front view, and top view of the i-th completed baggage; i is the baggage sequence on the tray sorter, P i is the actual baggage stacking order, and N is the number of baggage to be stacked.

7. The method for online loading of luggage at a civil aviation airport according to claim 6, characterized in that: The fitness function is defined as the minimum difference between the height and bottom area of ​​the luggage to be placed and the luggage on the left, rear and bottom sides. The fitness function calculation formula is as follows: Among them, l i 、b i are the height differences between the luggage to be stacked and the luggage on the left and rear sides of the placement point, s j 、s i They are the support area and the bottom area for the luggage to be placed.

8. The method for online loading of luggage at a civil aviation airport according to claim 7, characterized in that: In step S304, the update formula for improving the finder position information in the sparrow search algorithm by the golden sine strategy is as follows: x1=-π+(1-τ)×2π, x2=-π+τ×2π Among them, r1∈[0,2π], r2∈[0,π], which respectively determine the next individual's moving distance and moving direction; R2∈[0,1] represents the warning value, and ST∈[0.5,1] represents the alert value; is the current optimal position, Q is a random number that obeys the normal distribution; L represents a 1×d all-1 matrix, d is the dimension of the solution space, and x1 and x2 are coefficients obtained according to the golden section coefficient τ.

Images