A Smart Cargo Inspection Method, System, Device and Storage Medium for Airport Departures

Abstract

The present application discloses an intelligent cargo inspection method, system, device and storage medium for air departure. The method includes: receiving cargo information, cargo corresponding flight information collected from the previous link of the air departure service, and container information collected from the departure link; according to the cargo information and container information corresponding to each flight, using a hybrid heuristic algorithm to obtain the optimal stacking plan for the cargo corresponding to each flight; loading the optimal stacking plan into the palletizing and film wrapping equipment to guide the robotic arm of the palletizing and film wrapping equipment to accurately stack the cargo and automatically complete the film wrapping process; adopting a dual verification mechanism of a grating and a profiling door to detect the container that has completed the film wrapping process and is loaded with cargo, and after the detection is error-free, performing a net covering process on the container that has completed the film wrapping process and is loaded with cargo, and simultaneously updating the data in the back-end database of the departure service. The present application can improve the automation level of air departure cargo inspection, reduce the need for manual intervention, and improve work efficiency.

Description

Technical Field

[0001] This application relates to the technical field of cargo handling in the outbound aviation business, and specifically relates to an intelligent cargo inspection method, system, device, and storage medium for outbound aviation. Background Art

[0002] Traditional aviation logistics operations highly rely on manual intervention. Although it can meet basic needs, problems such as low efficiency and easy occurrence of errors and omissions have gradually emerged. With the popularization of the application of emerging technologies such as the Internet of Things and artificial intelligence, aviation logistics has gradually entered the automation stage, which not only greatly improves the speed and accuracy of cargo handling, but also injects new vitality and development potential into the industry. In this context, intelligent cargo inspection for outbound aviation often covers multiple core links, including the cargo transportation link, the cargo assembly and stacking link, the cargo film wrapping link, and the cargo net covering and outbound link, etc. The collaborative optimization of each link is crucial for improving the overall efficiency.

[0003] Currently, in actual operations, in order to optimize the entire process of outbound aviation cargo inspection, a series of basic means are widely adopted in the industry. For example, bar code scanners are used for information verification during cargo transportation, or basic sensors are used to detect the status of the entire process of cargo outbound. In some scenarios, attempts are also made to combine electronic display boards to display key progress information to assist on-site staff in reasonably arranging the work sequence.

[0004] However, existing automated solutions mainly focus on the optimization of the detection itself in a single link, lacking optimization and improvement for multiple links in the entire cargo inspection chain, especially the detection preparation link, and lacking intelligent decision-making support, with a relatively low overall automation level. Therefore, how to comprehensively improve the automation level of cargo inspection and reduce the excessive dependence on manual labor has become an important direction for promoting the technological progress of intelligent cargo inspection for outbound aviation. Summary of the Invention

[0005] In order to improve the automation level of outbound aviation cargo inspection, reduce the need for manual intervention, and improve work efficiency, this application provides an intelligent cargo inspection method, system, device, and storage medium for outbound aviation.

[0006] In a first aspect, this application provides an intelligent cargo inspection method for outbound aviation, including:

[0007] Receiving cargo information and corresponding flight information of the cargo collected from the previous link of the outbound aviation business; the previous link includes cargo collection, security inspection, classification, and warehousing; receiving container information collected from the outbound link of the outbound aviation business;

[0008] According to the cargo information and container information corresponding to each flight, use a hybrid heuristic algorithm to obtain the optimal stowage plan for the cargo corresponding to each flight; the hybrid heuristic algorithm includes: using a greedy bin-packing algorithm to generate an initial stowage plan, using a genetic algorithm to globally optimize the generated initial stowage plan, and using a simulated annealing algorithm to locally optimize the generated initial stowage plan;

[0009] Load the optimal stowage plan into the palletizing and stretch-wrapping equipment to guide the robotic arm of the palletizing and stretch-wrapping equipment to stack the cargo and automatically complete the stretch-wrapping process;

[0010] Adopt a dual verification mechanism of a grating and a profiling door to detect the shape and integrity of the container that has completed the stretch-wrapping process and is loaded with cargo. After the detection is correct, perform net covering on the container that has completed the stretch-wrapping process and is loaded with cargo, and at the same time update the data of the cargo and container to be loaded in the outbound business background database.

[0011] By adopting the above solution, collect the cargo information in the previous outbound link and the container information in the outbound link, and use a hybrid heuristic algorithm to generate an optimal stowage plan to guide the robotic arm of the palletizing and stretch-wrapping equipment to accurately execute the stowage task, and automatically complete stowage and stretch-wrapping, providing a basis for quickly detecting the container using a dual verification mechanism of a grating and a profiling door, and directly performing net covering after meeting the qualified detection requirements, realizing a high degree of automation in multiple links of cargo outbound inspection, greatly shortening the cargo turnover time and reducing the labor cost.

[0012] Preferably, it further includes:

[0013] Receive flight change information, and adaptively adjust the priority of the cargo corresponding to the changed flight based on the flight change information, so that the cargo corresponding to the flight with a closer departure time has a higher priority;

[0014] Adjust the main and secondary objectives of cargo stowage according to the priority of the cargo corresponding to each flight, so that for the cargo with a priority higher than the preset priority, the main objectives of cargo stowage are to minimize the total stowage duration of cargo stowage and maximize the cargo stowage stability score, and the secondary objective is to maximize the container space utilization rate; for the cargo with a priority lower than the preset priority, the main objectives of cargo stowage are to maximize the container space utilization rate and maximize the cargo stowage stability score, and the secondary objective is to minimize the total stowage duration of cargo stowage; for the cargo with a priority equal to the preset priority, the main objectives of cargo stowage are to minimize the total stowage duration of cargo stowage, maximize the cargo stowage stability score, and maximize the container space utilization rate, and there is no secondary objective;

[0015] According to the primary and secondary objectives of the adjusted cargo stacking, the fitness function in the genetic algorithm is designed accordingly. For the first fitness function where the primary objective is to minimize the total stacking duration of the cargo stacking and maximize the cargo stacking stability score, the stacking duration parameter weight and the cargo stacking stability score parameter weight are designed to be greater than the space utilization rate parameter weight. For the second fitness function where the primary objective is to maximize the space utilization rate of the container and maximize the cargo stacking stability score, the space utilization rate parameter weight and the cargo stacking stability parameter weight are designed to be greater than the stacking duration parameter weight. For the third fitness function where the primary objective is to minimize the total stacking duration of the cargo stacking, maximize the cargo stacking stability score, and maximize the space utilization rate of the container, the space utilization rate parameter weight, the cargo stacking stability score parameter weight, and the stacking duration parameter weight are designed to be the same.

[0016] By adopting the above solution, the flight cargo priority is adjusted in real-time in response to flight schedule changes, and the stacking optimization algorithm is adapted to adjust the cargo stacking strategy, realizing intelligent decision-making support for outbound cargo in the cargo assembly and stacking process, facilitating subsequent timely and rapid container inspection, and improving the efficiency of outbound cargo inspection.

[0017] Preferably, the automatic completion of the film wrapping process further includes:

[0018] Calculating the center of gravity offset of the container under the optimal stacking plan;

[0019] Inputting the container information and the calculated center of gravity offset into the film wrapping process acquisition network to obtain the film wrapping tension, material, and wrapping method in the film wrapping process, so that the center of gravity offset of the container is reduced to a preset threshold after film wrapping; the film wrapping process acquisition network selects a convolutional neural network and is trained and generated using historical container information, the center of gravity offset of the container, and the film wrapping tension, material, and wrapping method in the film wrapping process.

[0020] Wrapping the container according to the obtained film wrapping tension, material, and wrapping method in the film wrapping process.

[0021] By adopting the above solution, the specific parameters of the film wrapping process are intelligently selected according to the center of gravity offset under the optimal stacking plan, ensuring stable film wrapping of the stacked cargo, facilitating subsequent rapid completion of cargo film wrapping inspection, and improving the efficiency of outbound cargo inspection.

[0022] Preferably, it further includes:

[0023] Integrating a film wrapping tension sensor array and a machine vision device on the stacking and film wrapping equipment, and using the integrated film wrapping tension sensor array and machine vision device to real-time monitor the film wrapping tension at each angle of the container and the image of the container being wrapped in the film wrapping process.

[0024] Analyze the shape of the container wrapped with film detected in real time during the film wrapping process, compare the shape of the container wrapped with film with the shape of the target container wrapped with film in the film wrapping process or compare the film wrapping tension at each angle of the container with the target film wrapping tension in the film wrapping process. If the comparison result shows that the error is greater than the preset error value, trigger the film wrapping self-correction program and re-perform the film wrapping.

[0025] By adopting the above solution, use sensors and machine vision to detect the film wrapping process from multiple angles, and perform film wrapping correction in a timely manner according to the detection results, ensuring the accuracy of film wrapping and improving the automation level of the detection of outbound goods.

[0026] Preferably, it further includes:

[0027] Add multiple verification devices at key positions of the conveyor belt for transporting goods in the outbound link. The multiple verification devices include an intelligent camera array and an RFID reader / writer;

[0028] Use the multiple verification devices to detect the goods and perform image analysis to determine whether the current goods information is consistent with the recorded information and whether the transportation path conforms to the preset goods transportation path of the corresponding flight. Once it is determined that there is any situation where the goods information is inconsistent with the recorded information or does not conform to the preset goods transportation path of the corresponding flight, trigger an alarm and perform real-time tracking on the current goods for display on the display device.

[0029] By adopting the above solution, use multiple verification methods such as an intelligent camera array and an RFID reader / writer to improve the accuracy and integrity of outbound goods during the transmission stage, and improve the automation level of the detection of outbound goods.

[0030] Preferably, it further includes:

[0031] Obtain the first feedback data of the user on the outbound goods and the second feedback data of the airline staff on the outbound goods. The first feedback data includes: the goods damage rate; the second feedback data includes: the usage cost of the container;

[0032] Once the goods damage rate of the outbound goods reaches the preset damage rate or the proportion of the usage cost of the container being greater than the preset cost within the preset time period is greater than the preset proportion, select the optimized hybrid heuristic algorithm.

[0033] By adopting the above solution, use the feedback mechanism to optimize the stacking algorithm of outbound goods, further ensuring the correct stacking of outbound goods and facilitating subsequent rapid passing of the stacking detection.

[0034] Preferably, it further includes:

[0035] Determine the loading position requirements for each piece of cargo based on the cargo information. The loading position requirements include: adjacent loading positions for associated cargo, and the loading position of the cargo with a high extraction priority is close to the entrance position of the container.

[0036] Take whether the cargo in the stacking plan meets the loading position requirements as a penalty term parameter and incorporate it into the fitness function in the genetic algorithm.

[0037] By adopting the above solution, determine the specific loading position requirements for each piece of cargo according to the cargo information, and convert these requirements into penalty term parameters in the genetic algorithm, optimizing the rationality of the spatial layout of the cargo stacking while enhancing the user experience, improving user satisfaction, reducing the time-consuming of the optimization algorithm, ensuring the correct stacking of the outbound cargo, and facilitating the subsequent rapid passing of the stacking inspection.

[0038] In a second aspect, the present application provides an intelligent cargo inspection system for air outbound, including:

[0039] An outbound cargo data collection module, used to receive the cargo information and the corresponding flight information of the cargo collected from the previous links of the air outbound business; the previous links include cargo receipt, security inspection, classification, and warehousing; receive the container information collected from the outbound link of the air outbound business.

[0040] An outbound cargo stacking plan acquisition module, used to obtain the optimal stacking plan for the cargo corresponding to each flight according to the cargo information and container information corresponding to each flight, using a hybrid heuristic algorithm; the hybrid heuristic algorithm includes: generating an initial stacking plan using the greedy bin-packing algorithm, globally optimizing the generated initial stacking plan using the genetic algorithm, and locally optimizing the generated initial stacking plan using the simulated annealing algorithm.

[0041] An outbound cargo film wrapping execution module, used to load the optimal stacking plan onto the palletizing and film wrapping equipment to guide the robotic arm of the palletizing and film wrapping equipment to stack the cargo and automatically complete the film wrapping process.

[0042] An outbound cargo verification and data update module, used to perform shape and integrity detection on the container loaded with cargo after the film wrapping process is completed using a dual verification mechanism of a grating and a profiling door, perform net covering on the container loaded with cargo after the film wrapping process is completed and the detection is correct, and simultaneously update the data of the cargo and the container to be loaded in the outbound business background database.

[0043] By adopting the above solution, receive the cargo and related information from multiple links, comprehensively use multiple algorithms to generate the optimal stacking plan, guide the robotic arm of the palletizing and film wrapping equipment to accurately stack the cargo and complete the film wrapping process, and at the same time ensure the loading quality and data consistency of the cargo with the help of the dual verification mechanism, directly or indirectly promoting the efficient and accurate outbound cargo inspection and realizing a highly automated intelligent cargo inspection system for air outbound.

[0044] In a third aspect, the present application provides a computer-readable storage medium, which includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the method as described above.

[0045] In a fourth aspect, the present application provides a computer device, which includes a memory, a processor, and a program stored and executable on the memory. When the program is executed by the processor, it implements the steps of the method as described above.

[0046] In summary, the present application has the following beneficial effects:

[0047] 1. By adopting a hybrid heuristic algorithm to generate an optimal stacking scheme, the stacking efficiency of goods and the space utilization rate of containers are improved, indirectly ensuring a quick pass through the inspection of the goods stacking link; introducing a dual verification mechanism of a grating and a profiling door to ensure the accuracy of the detection in the goods stacking link and improve the automation degree of the detection in the goods stacking link for outbound goods;

[0048] 2. Combining with the dynamic flight change information, adjusting the priority of goods and then matching an appropriate stacking optimization algorithm to respond to emergencies of outbound goods and improve the stacking efficiency of outbound goods, indirectly ensuring a quick pass through the inspection of the goods stacking link;

[0049] 3. Introducing an adaptive goods stretch wrapping mechanism to complete stable stretch wrapping of goods for adaptation, so as to quickly pass through the detection in the goods stretch wrapping stage; combining multiple stretch wrapping detection techniques to timely query and self-correct incorrect stretch wrapping of outbound goods, and improving the accuracy and automation degree of the detection in the goods stretch wrapping link for outbound goods. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a flowchart of the intelligent cargo inspection method for air outbound in a specific embodiment;

[0051] Figure 2 It is a full-process diagram of the previous link of the outbound business and the outbound link of the outbound business in the intelligent cargo inspection method for air outbound in a specific embodiment;

[0052] Figure 3 It is a full-link diagram of the outbound business in the intelligent cargo inspection method for air outbound in a specific embodiment;

[0053] Figure 4 It is a structural schematic diagram of the intelligent cargo inspection system for air outbound in a specific embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] To make the objectives, technical solutions, and advantages of this application more clear and understandable, the following further details this application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0055] The intelligent cargo inspection for outbound air transportation covers the entire process of cargo outbound, including various links such as cargo extraction from the warehouse, cargo transportation, cargo stacking, cargo film wrapping, cargo net covering, and cargo out of the warehouse (i.e., cargo outbound). Considering that the existing outbound cargo detection technology has a low degree of automation and is difficult to support intelligent decision-making, resulting in low efficiency and prone to errors in the intelligent cargo inspection for outbound air transportation, and the labor cost is relatively high. To improve the automation level of the intelligent cargo inspection for outbound air transportation, reduce the need for manual intervention, and improve work efficiency; as Figure 1 shown, the embodiments of this application disclose an intelligent cargo inspection method for outbound air transportation, and the specific steps include:

[0056] S1. Receive the cargo information, the cargo corresponding flight information collected from the previous link of the outbound air transportation business, and the container information collected from the outbound link.

[0057] Specifically, as Figure 2 shown, the previous link refers to the previous link of the outbound air transportation business, including: cargo acceptance, cargo security inspection, cargo classification, and cargo warehousing, etc. Through the information collection devices in the previous link, the information of the subsequent cargo to be outbound is collected, including: the size, type, packaging type, vulnerability, handling priority, and the flight number to be loaded, etc., of the cargo, as well as the cargo corresponding flight information of the subsequent cargo to be outbound, including: flight number, flight operation, and the status of delay or cancellation. All the information collected above is stored in the outbound business background database.

[0058] To complete the cargo outbound, it is necessary to uniformly pack the cargo to be outbound according to the corresponding flight number and load it into the container. Therefore, the information of the used and unused containers in the cargo outbound business is generally monitored in real time and stored in the outbound business background database. Correspondingly, receive the container information collected from the outbound link of the outbound air transportation business, including: information such as the size, maximum load-bearing capacity, available quantity, and current loading status of the container (such as: PMC, PAG board).

[0059] S2. According to the cargo information and container information corresponding to each flight, use the hybrid heuristic algorithm to obtain the optimal stacking plan for the cargo corresponding to each flight.

[0060] As Figure 3As shown in the figure, to improve the intelligent cargo inspection automation level for outbound flights, it is not only necessary to optimize the cargo inspection itself in each link, but also to optimize the basis of cargo inspection in advance to facilitate subsequent rapid passing of the inspection. From this perspective, by optimizing the stacking plan in the outbound cargo stacking link, it can not only improve the efficiency of cargo stacking and then quickly enter the inspection stage, but also ensure the correctness of cargo stacking, quickly pass the stacking inspection, and avoid being repeatedly stuck in the stacking inspection stage due to cargo stacking problems.

[0061] Specifically, according to the cargo information and container information corresponding to each flight, a hybrid heuristic algorithm is used to obtain the optimal stacking plan for the cargo corresponding to each flight; among them, the hybrid heuristic algorithm includes:

[0062] First, use the greedy bin-packing algorithm to generate an initial stacking plan. The specific steps include:

[0063] Arrange all the goods to be loaded corresponding to the same flight in descending order of volume;

[0064] Determine the flight priority order according to the flight operation time. For each piece of cargo corresponding to the flight with the highest priority in the priority order, select the container that best matches it in the current remaining space, where the matching rule is determined by comparing the size of the cargo and the remaining space of the container;

[0065] Try to place the cargo in the selected container and check whether all physical and weight constraints are met. If they are met, update the status of the container; if not, try the next best-matching container;

[0066] Repeat the above steps of matching containers and placing cargo until all cargo is placed or no more cargo can be placed in the current container, and obtain the initial stacking plan for all the goods to be loaded corresponding to the same flight.

[0067] Secondly, use the genetic algorithm to globally optimize the generated initial stacking plan. The specific steps include:

[0068] Encode the loading order and placement position of all the goods to be loaded corresponding to the same flight into chromosomes. For example: use integer encoding, each integer represents the encoding of a piece of cargo, and the order of the chromosomes represents the loading order of the cargo;

[0069] Based on the initial stacking plan, randomly generate multiple initial solutions as the initial population, and each solution represents a possible way of loading cargo;

[0070] The fitness function of the design synthesis evaluates the quality of each solution. Since the cargo stacking scheme is a multi-objective optimization problem, the comprehensive fitness function balances the conflicts between cargo loadings. The cargo stacking objectives include: maximizing the space utilization rate of the container, maximizing the stability score of the cargo stacking, and minimizing the total stacking duration of the cargo stacking. And the objective calculation is often based on partial constraints. For example, the stability of the cargo stacking can be comprehensively determined according to factors such as whether the compressive strength of each cargo at the corresponding stacking layer exceeds the preset compressive capacity of the corresponding cargo and the center of gravity position of the overall cargo. Furthermore, the designed fitness function parameter terms include: space utilization rate, handling time, and stability score. The final comprehensive fitness function is obtained through weighted calculation;

[0071] Select excellent individuals for reproduction according to the fitness function, and generate new solutions through crossover operations (such as exchanging part of the genes of two chromosomes) and mutation operations (such as changing the position or rotation direction of a certain cargo);

[0072] Repeat the selection, crossover, and mutation operations until the termination condition is met, such as reaching the maximum number of iterations or the fitness improvement is less than a certain threshold, and then obtain the global optimization solution accordingly.

[0073] Finally, use the simulated annealing algorithm to perform local optimization on the generated initial stacking scheme, specifically including:

[0074] Based on the global optimization solution obtained by performing global optimization on the initial stacking scheme, perturb the global optimization solution, such as: exchanging the positions of two cargos, or adjusting the rotation direction or position of the cargo;

[0075] Judge whether the perturbed solution is better than the global optimization solution (i.e., has a higher fitness). If so, accept this solution as the new current solution; otherwise, accept the inferior solution with a certain preset probability to avoid falling into the local optimum;

[0076] Repeat the perturbation and acceptance operations until the termination condition is met (such as: reaching the maximum number of iterations), and obtain the optimal solution, that is, the optimal stacking scheme.

[0077] S3. Load the optimal stacking scheme into the palletizing and stretch-wrapping equipment to use the optimal stacking scheme to guide the robotic arm to accurately stack the cargo and automatically complete the stretch-wrapping process.

[0078] Specifically, in order to improve the automation degree of the whole process of outbound cargo inspection, load the optimal stacking scheme into the palletizing and stretch-wrapping equipment to guide the robotic arm of the palletizing and stretch-wrapping equipment to accurately stack the cargo.

[0079] For better transportation, it is often necessary to further wrap and shape the container where the goods are stacked. After the goods are accurately stacked, the wrapping process is completed using the robotic arm of the palletizing and wrapping equipment. While completing the wrapping process, the monitoring equipment installed on the palletizing and wrapping equipment is used to monitor the wrapping process in real time to prevent errors in the wrapping operation that may lead to subsequent verification failures.

[0080] S4 uses a dual verification mechanism of light barriers and contour doors to inspect the shape and integrity of containers that have completed the wrapping process and are loaded with goods.

[0081] Specifically, in addition to stacking optimization during the outbound cargo stacking stage to speed up the stacking process, inspections are carried out on the carrying containers to prevent deformation of the containers after stacking and wrapping and before leaving the warehouse, which may lead to failure to leave the warehouse normally. In order to ensure the accuracy of the inspection and improve the degree of automation, a dual verification mechanism of gratings and contour doors is selected to perform shape and integrity inspections on the containers that have completed the wrapping process and are loaded with goods. Among them, the grating is used in advance to perform preliminary inspections on the shape and size of the containers that have completed the wrapping process and are loaded with goods. After the inspection is passed, the contour door is used to continue the verification. Once it passes smoothly, it indicates that the verification is successful.

[0082] S5. After the inspection is correct, the container that has completed the film wrapping process and is loaded with goods is covered with a net, and the data of the goods and containers to be loaded in the outbound business background database are updated at the same time.

[0083] Specifically, when the container that has completed the film wrapping process and is loaded with goods is detected to be correct, the container is pushed to the area to be shipped out and the container that has completed the film wrapping process and is loaded with goods is automatically covered with a net cover by a covering device, and the data of the goods to be loaded and the container in the outbound business background database are updated at the same time, such as deleting the loaded goods and the used containers when updating the outbound business background database.

[0084] In addition, when a container that has completed the wrapping process and is loaded with goods fails to be inspected, the container is pushed to the manual review area so that the user can choose to push it to the palletizing wrapping area or the grating and contour door double verification area to re-perform palletizing wrapping or verification.

[0085] By adopting the above methods, the degree of automation of the entire process of cargo outbound inspection can be comprehensively improved, reducing manual intervention and improving work efficiency.

[0086] In a specific embodiment, considering that the actual cargo departure process often leads to an advance or delay in the departure time of cargo due to changes in flights, in order to respond to the changes in a timely manner, it is necessary to dynamically adjust the stacking strategy to ensure efficient operation under complex circumstances; the method also includes:

[0087] Receive flight change information, where the change information includes flight delay and delay time, flight advance and advance time, flight cancellation and adjustment information.

[0088] Based on the flight change information, adaptively adjust the priority of the goods corresponding to the changed flight so that the goods corresponding to the flight with a closer departure time have a higher priority; specifically, determine the operating time of the changed flight according to the flight information and the flight change information, and set the corresponding goods priority as the first priority according to whether the distance from the current time is the first time period, set the corresponding goods priority as the second priority according to whether the distance from the current time is the second time period, and set the corresponding goods priority as the third priority according to whether the distance from the current time is the third time period, where the durations of the first time period, the second time period, and the third time period gradually increase; the priority levels of the first priority, the second priority, and the third priority gradually decrease.

[0089] Considering that the stacking of goods corresponding to flights with closer operating times focuses more on efficiency and can sacrifice space utilization as much as possible. On the contrary, the stacking of goods corresponding to flights with farther operating times focuses more on space utilization to save costs. The overall stability of the goods stacking needs to be satisfied as much as possible in any case; therefore, adjust the main and secondary objectives of the goods stacking according to the priority of the goods corresponding to each flight, so that the main objective of the goods stacking with a priority higher than the preset priority is to minimize the total stacking duration of the goods stacking and maximize the stacking stability score, and the secondary objective is to maximize the space utilization rate of the container. The main objective of the goods stacking with a priority lower than the preset priority is to maximize the space utilization rate of the container and maximize the stacking stability score, and the secondary objective is to minimize the total stacking duration of the goods stacking. The main objective of the goods stacking with a priority equal to the preset priority is to minimize the total stacking duration of the goods stacking, maximize the stacking stability score, and maximize the space utilization rate of the container, without secondary objectives; in this embodiment, the preset priority is the second priority.

[0090] According to the main and secondary objectives of the adjusted stacking of goods, the fitness function in the genetic algorithm is designed accordingly, so that the main objective is to minimize the total stacking duration of goods stacking and maximize the stacking stability score of goods stacking. The first fitness function corresponding to the designed stacking duration parameter weight and the stacking stability score parameter weight of goods stacking is greater than the space utilization rate parameter weight. For example, the weight ratio of the stacking duration parameter: the stacking stability score parameter of goods stacking: the space utilization rate parameter is: 4:4:2; the main objective is to maximize the space utilization rate of the container and maximize the stacking stability score of goods stacking. The second fitness function corresponding to the designed space utilization rate parameter weight and the parameter weight of the stacking stability of goods stacking is greater than the stacking duration parameter weight. For example, the weight ratio of the stacking duration parameter: the stacking stability score parameter of goods stacking: the space utilization rate parameter is: 2:4:4; the main objective is to minimize the total stacking duration of goods stacking, maximize the stacking stability score of goods stacking, and maximize the space utilization rate of the container. The third fitness function corresponding to the designed space utilization rate parameter weight, the stacking stability score parameter weight, and the stacking duration parameter weight is the same or the difference is less than the preset weight difference.

[0091] In a specific embodiment, in order to further improve the automation degree of the whole process of outbound goods inspection, the wrapping process with good wrapping effect is adaptively selected to complete the wrapping, so as to quickly enter the next link and successfully complete the wrapping inspection. The automatic wrapping process in the method further includes:

[0092] Calculate the centroid offset of the container under the optimal stacking scheme;

[0093] Input the container information and the calculated centroid offset into the wrapping process acquisition network to obtain the wrapping tension, material, and wrapping method in the wrapping process. Among them, the wrapping process can be divided into the initial stage, the middle stage, and the later stage. The wrapping methods include: the number of wrapping layers, the wrapping shape (each part of the container). The wrapping process acquisition network selects a convolutional neural network and is trained and generated using historical container information, the centroid offset of the container, and the wrapping tension, material, and wrapping method in the wrapping process. Among them, the wrapping tension, material, and wrapping method in the historical wrapping process can be determined according to the methods proposed by experienced wrapping experts to reduce the centroid offset to a preset threshold after the wrapping process is executed.

[0094] Finally, wrap the container according to the wrapping tension, material, and wrapping method in the obtained wrapping process, so that the centroid offset of the container is reduced to a preset threshold after wrapping.

[0095] In addition, in addition to performing adaptive wrapping during the wrapping stage of outbound goods to select a wrapping process with good wrapping effect, multi-level detection and timely correction of the wrapping inspection can also be performed to ensure a good wrapping effect, successfully pass the subsequent inspection, and improve the inspection efficiency. The method further includes:

[0096] Integrate a film-wrapping tension sensor array and a machine vision device on the stack-coding and film-wrapping equipment, and use the integrated film-wrapping tension sensor array and machine vision device to monitor the film-wrapping tension at each angle of the container and the image of the container wrapped with film in the film-wrapping process in real time; among them, each angle can be set corresponding to the size of the container, such as: angles facing the front, back, left, right of the container, etc.

[0097] Analyze the shape of the container wrapped with film in the film-wrapping process detected in real time, compare the shape of the container wrapped with film with the shape of the target container wrapped with film in the film-wrapping process or compare the film-wrapping tension at each angle of the container with the target film-wrapping tension in the film-wrapping process. If the comparison result shows that the error between the two is greater than the preset error value, trigger the film-wrapping self-correction program and re-perform film-wrapping.

[0098] Specifically, if the film-wrapping self-correction program is triggered at any stage of the specific early, middle, or late stage of the determined film-wrapping process, then unwind the film in the reverse direction and return to the starting point of this stage to re-perform film-wrapping.

[0099] In a specific embodiment, in order to further improve the automation level of the whole stage of outbound cargo inspection, multiple detection devices are integrated in the outbound cargo transportation stage to complete automatic detection. The method further includes:

[0100] Add multiple verification devices at each key position of the conveyor belt for transporting goods in the outbound link. The multiple verification devices include an intelligent camera array and an RFID reader / writer;

[0101] Use the multiple verification devices to detect the goods and perform image analysis and judgment to determine whether the current goods are consistent with the recorded information (such as: size, weight, etc.) and whether the transportation path conforms to the preset cargo transportation path corresponding to the flight (such as: the preset transportation path for the goods corresponding to Flight A is the f-section conveyor belt path). Once it is judged that there is any situation where the goods information is inconsistent with the recorded information or does not conform to the preset cargo transportation path corresponding to the flight, trigger an alarm and perform real-time tracking on the current goods to be displayed on the display device.

[0102] In a specific embodiment, considering using a feedback mechanism to continuously optimize the stacking layout, so as to improve the stacking efficiency or increase the probability of passing the inspection in sequence. The method further includes:

[0103] From the perspectives of the owners (users) of the outbound goods and the handlers (airline employees) of the outbound goods, obtain the first feedback data of the users on the outbound goods and the second feedback data of the airline employees on the outbound goods respectively. The first feedback data includes: the damage rate of the goods, the satisfaction of the goods extraction; the second feedback data includes: the usage cost of the container, the time taken for the outbound goods to complete the warehouse exit.

[0104] Once the data in the first feedback data of the outbound goods reaches the preset value of this data, such as:

[0105] The damaged rate of the goods reaches the preset damaged rate or there is data in the second feedback data that reaches the preset value ratio within the preset time period and is greater than the preset ratio. For example, the ratio of the usage cost of the container within the preset time period being greater than the preset cost is greater than the preset ratio, which indicates that the entire process of outbound inspection needs to be further improved to further improve efficiency and user satisfaction. Therefore, the hybrid heuristic algorithm is selected for optimization.

[0106] In addition, in order to further improve satisfaction and avoid the extra time consumption caused by optimizing the stacking algorithm, the method further includes:

[0107] Determine the loading position requirements of each piece of goods according to the goods information. The loading position requirements include: adjacent loading positions for associated goods, and the loading positions of goods with high extraction priorities being close to the container entrance position, etc.; among them, the extraction priority of the goods can be set manually.

[0108] Use whether the goods in the stacking plan meet the loading position requirements as a penalty term parameter and incorporate it into the fitness function in the genetic algorithm, that is, calculate the penalty term score corresponding to the ratio of the goods in the stacking plan that do not meet the loading position requirements and add it to the calculation of the comprehensive fitness function.

[0109] As Figure 4 shown, an embodiment of the present application provides an intelligent cargo inspection system for air outbound, including:

[0110] An outbound goods data collection module 101, configured to receive the goods information and the goods corresponding flight information collected in the previous link of the air outbound business; the previous link includes goods receipt, security inspection, classification, and warehousing; receive the container information collected in the outbound link of the air outbound business;

[0111] An outbound goods stacking plan acquisition module 102, configured to obtain the optimal stacking plan for the goods corresponding to each flight according to the goods information and container information corresponding to each flight, using the hybrid heuristic algorithm; the hybrid heuristic algorithm includes: using the greedy bin packing algorithm to generate an initial stacking plan, using the genetic algorithm to globally optimize the generated initial stacking plan, and using the simulated annealing algorithm to locally optimize the generated initial stacking plan;

[0112] An outbound goods film wrapping execution module 103, configured to load the optimal stacking plan onto the palletizing and film wrapping equipment to guide the robotic arm of the palletizing and film wrapping equipment to stack the goods and automatically complete the film wrapping process;

[0113] The outbound goods verification and data update module 104 is used to detect the shape and integrity of the container loaded with goods after the film wrapping process by using a dual verification mechanism of grating and profiling door. After the detection is correct, the container loaded with goods after the film wrapping process is covered with a net, and at the same time, the data of the goods and containers to be loaded in the outbound business background database is updated.

[0114] A specific embodiment further includes:

[0115] The outbound goods stacking plan optimization module 105 is used to receive flight change information, adaptively adjust the priority of the goods corresponding to the changed flight based on the flight change information, so that the goods corresponding to the flight with a closer departure time have a higher priority; adjust the main and secondary objectives of the goods stacking according to the priority of the goods corresponding to each flight, so that the main objective of the goods stacking with a priority higher than the preset priority is to minimize the total stacking duration of the goods stacking and maximize the stacking stability score of the goods stacking, and the secondary objective is to maximize the space utilization rate of the container. The main objective of the goods stacking with a priority lower than the preset priority is to maximize the space utilization rate of the container and maximize the stacking stability score of the goods stacking, and the secondary objective is to minimize the total stacking duration of the goods stacking. The main objective of the goods stacking with a priority equal to the preset priority is to minimize the total stacking duration of the goods stacking, maximize the stacking stability score of the goods stacking and maximize the space utilization rate of the container, and there is no secondary objective; according to the adjusted main and secondary objectives of the goods stacking, design the fitness function in the genetic algorithm accordingly, so that the main objective is to minimize the total stacking duration of the goods stacking and maximize the stacking stability score of the goods stacking, and the corresponding designed stacking duration parameter weight and goods stacking stability score parameter weight are greater than the space utilization rate parameter weight in the first fitness function. The main objective is to maximize the space utilization rate of the container and maximize the stacking stability score of the goods stacking, and the corresponding designed space utilization rate parameter weight and goods stacking stability parameter weight are greater than the stacking duration parameter weight in the second fitness function. The main objective is to minimize the total stacking duration of the goods stacking, maximize the stacking stability score of the goods stacking and maximize the space utilization rate of the container, and the corresponding designed space utilization rate parameter weight, goods stacking stability score parameter weight, and stacking duration parameter weight are the same in the third fitness function.

[0116] The outbound goods stacking plan optimization module 105 is also used to determine the loading position requirements of each goods according to the goods information. The loading position requirements include: the loading positions of associated goods are adjacent, and the goods with a high extraction priority are loaded near the entrance position of the container; use whether the goods in the stacking plan meet the loading position requirements as a penalty item parameter and incorporate it into the fitness function in the genetic algorithm.

[0117] A specific embodiment further includes:

[0118] The outbound goods film wrapping execution optimization module 106 is used to calculate the center of gravity offset of the container under the optimal stacking scheme; input the container information and the calculated center of gravity offset into the film wrapping process acquisition network to obtain the film wrapping tension, material, and winding method in the film wrapping process; the film wrapping process acquisition network selects a convolutional neural network and is trained and generated using historical container information, the center of gravity offset of the container, and the film wrapping tension, material, and winding method in the film wrapping process; perform film wrapping on the container according to the obtained film wrapping tension, material, and winding method in the film wrapping process; it is also used to integrate a film wrapping tension sensor array and a machine vision device on the stacking and film wrapping equipment, and use the integrated film wrapping tension sensor array and machine vision device to monitor the film wrapping tension at each angle of the container and the image of the container being film wrapped in real time during the film wrapping process; analyze the shape of the container being film wrapped in the film wrapping process in real time, compare the shape of the container being film wrapped with the shape of the target film wrapped container in the film wrapping process or compare the film wrapping tension at each angle of the container with the target film wrapping tension in the film wrapping process. If the comparison result shows that the error is greater than the preset error value, trigger the film wrapping self-correction program and re-perform film wrapping.

[0119] A specific embodiment further includes:

[0120] The outbound goods transportation detection module 107 is used to add multiple verification devices at key positions of the conveyor belt for transporting goods in the outbound link. The multiple verification devices include an intelligent camera array and an RFID reader / writer;

[0121] Use the multiple verification devices to detect the goods and perform image analysis to determine whether the current goods information is consistent with the recorded information and whether the transportation path conforms to the preset goods transportation path of the corresponding flight. Once it is determined that there is any situation where the goods information is inconsistent with the recorded information or does not conform to the preset goods transportation path of the corresponding flight, trigger an alarm and perform real-time tracking on the current goods for display on the display device.

[0122] A specific embodiment further includes:

[0123] The outbound goods data feedback module 108 is used to obtain the first feedback data of users on outbound goods and the second feedback data of airline staff on outbound goods. The first feedback data includes: the goods damage rate; the second feedback data includes: the usage cost of the container. Once the goods damage rate of the outbound goods reaches the preset damage rate or the proportion of the usage cost of the container being greater than the preset cost within the preset time period is greater than the preset proportion, select the optimized hybrid heuristic algorithm.

[0124] The embodiments of the present application also disclose a computer-readable storage medium.

[0125] Specifically, the computer-readable storage medium stores a computer program that can be loaded and executed by a processor, such as the intelligent cargo inspection method for outbound flights described above. The computer-readable storage medium includes various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs.

[0126] An embodiment of the present application also discloses a computer device.

[0127] Specifically, the computer device includes a memory and a processor. The memory stores a computer program that can be loaded and executed by the processor, such as the intelligent cargo inspection method for outbound flights described above.

[0128] The above are all preferred embodiments of the present application. The protection scope of the present application is not limited thereby. Any feature disclosed in this specification (including the abstract and drawings), unless specifically described, can be replaced by other equivalent or similar-purpose alternative features. That is, unless specifically described, each feature is only an example of a series of equivalent or similar features.

Claims

1. A smart cargo inspection method for outbound flights, characterized in that, Including: Receiving cargo information and cargo corresponding flight information collected in the previous links of the air departure operation; the previous links include cargo collection, security inspection, classification, and warehousing; Receiving container information collected in the departure link of the air departure operation; According to the cargo information and container information corresponding to each flight, using a hybrid heuristic algorithm to obtain the optimal stowage plan for the cargo corresponding to each flight; the hybrid heuristic algorithm includes: using a greedy bin-packing algorithm to generate an initial stowage plan, using a genetic algorithm to globally optimize the generated initial stowage plan, and using a simulated annealing algorithm to locally optimize the generated initial stowage plan; Loading the optimal stowage plan onto the palletizing and stretch-wrapping equipment to guide the robotic arm of the palletizing and stretch-wrapping equipment to stack the cargo and automatically complete the stretch-wrapping process; Adopting a dual verification mechanism of a grating and a profiling door to detect the shape and integrity of the container that has completed the stretch-wrapping process and is loaded with cargo. After the detection is correct, performing net covering on the container that has completed the stretch-wrapping process and is loaded with cargo, and simultaneously updating the data of the cargo and container to be loaded in the back-end database of the departure operation; The automatically completing the stretch-wrapping process further includes: Calculating the center-of-gravity offset of the container under the optimal stowage plan; Inputting the container information and the calculated center-of-gravity offset into the stretch-wrapping process acquisition network to obtain the stretch-wrapping tension, material, and winding method in the stretch-wrapping process so that the center-of-gravity offset of the container is reduced to a preset threshold after stretch-wrapping; the stretch-wrapping process acquisition network selects a convolutional neural network and is trained and generated using historical container information, the center-of-gravity offset of the container, and the stretch-wrapping tension, material, and winding method in the stretch-wrapping process; Stretch-wrapping the container according to the obtained stretch-wrapping tension, material, and winding method in the stretch-wrapping process; Also including: Integrating a stretch-wrapping tension sensor array and a machine vision device on the palletizing and stretch-wrapping equipment, and using the integrated stretch-wrapping tension sensor array and machine vision device to real-time monitor the stretch-wrapping tension at each angle of the container and the image of the container being stretch-wrapped in the stretch-wrapping process; Analyzing the shape of the container being stretch-wrapped in the stretch-wrapping process detected in real time, comparing the shape of the container being stretch-wrapped with the shape of the target stretch-wrapped container in the stretch-wrapping process or comparing the stretch-wrapping tension at each angle of the container with the target stretch-wrapping tension in the stretch-wrapping process. If the comparison result shows that the error is greater than the preset error value, triggering the stretch-wrapping self-correction program and re-performing the stretch-wrapping; 2. The intelligent cargo inspection method for airport departure according to claim 1, wherein, Also including: Receiving flight change information, and adaptively adjusting the priority of the cargo corresponding to the changed flight based on the flight change information so that the cargo corresponding to the flight with a closer departure time has a higher priority; Adjust the primary and secondary objectives of cargo stacking according to the cargo priority corresponding to each flight, so that the primary objective of stacking cargo with a priority higher than the preset priority is to minimize the total stacking duration of cargo stacking and maximize the cargo stacking stability score, and the secondary objective is to maximize the utilization rate of the container space. The primary objective of stacking cargo with a priority lower than the preset priority is to maximize the utilization rate of the container space and maximize the cargo stacking stability score, and the secondary objective is to minimize the total stacking duration of cargo stacking. The primary objective of stacking cargo with a priority equal to the preset priority is to minimize the total stacking duration of cargo stacking, maximize the cargo stacking stability score, and maximize the utilization rate of the container space, without secondary objectives; According to the adjusted primary and secondary objectives of cargo stacking, design the fitness function in the genetic algorithm accordingly, so that the primary objective is to minimize the total stacking duration of cargo stacking and maximize the cargo stacking stability score, corresponding to the first fitness function where the parameter weight of the stacking duration and the parameter weight of the cargo stacking stability score are greater than the parameter weight of the space utilization rate. The primary objective is to maximize the utilization rate of the container space and maximize the cargo stacking stability score, corresponding to the second fitness function where the parameter weight of the space utilization rate and the parameter weight of the cargo stacking stability are greater than the parameter weight of the stacking duration. The primary objective is to minimize the total stacking duration of cargo stacking, maximize the cargo stacking stability score, and maximize the utilization rate of the container space, corresponding to the third fitness function where the parameter weights of the space utilization rate, the cargo stacking stability score, and the stacking duration are the same.

3. The intelligent cargo inspection method for outbound aviation according to claim 1, wherein, It also includes: Install multiple verification devices at each key position of the conveyor belt for transporting cargo during the outbound process. The multiple verification devices include an intelligent camera array and an RFID reader / writer; Use the multiple verification devices to detect the cargo and perform image analysis to determine whether the current cargo information is consistent with the recorded information and whether the transportation path conforms to the preset cargo transportation path of the corresponding flight. Once it is determined that there is any situation where the cargo information is inconsistent with the recorded information or does not conform to the preset cargo transportation path of the corresponding flight, an alarm is triggered and real-time tracking of the current cargo is performed for display on the display device.

4. The intelligent cargo inspection method for outbound aviation according to claim 1, wherein It also includes: Obtain the first feedback data of the user for the outbound cargo and the second feedback data of the airline staff for the outbound cargo. The first feedback data includes: the cargo damage rate; the second feedback data includes: the usage cost of the container; Once the cargo damage rate of the outbound cargo reaches the preset damage rate or the proportion of the usage cost of the container being greater than the preset cost within the preset time period is greater than the preset proportion, select the optimized hybrid heuristic algorithm.

5. The intelligent cargo inspection method for outbound aviation according to claim 4, wherein It also includes: Determine the loading position requirements for each cargo according to the cargo information. The loading position requirements include: the loading positions of associated cargo are adjacent, and the loading positions of goods with a high extraction priority are close to the entrance position of the container; Use whether the cargo in the stacking plan meets the loading position requirements as a penalty term parameter and incorporate it into the fitness function in the genetic algorithm.

6. An intelligent cargo inspection system for outbound flights, characterized in that, It includes: The outbound cargo data collection module is used to receive the cargo information and the corresponding flight information of the cargo collected in the previous links of the air outbound business; the previous links include cargo collection, security inspection, classification, and warehousing; it receives the container information collected in the outbound link of the air outbound business. The outbound cargo stacking plan acquisition module is used to obtain the optimal stacking plan for the cargo corresponding to each flight according to the cargo information and container information corresponding to each flight, using a hybrid heuristic algorithm; the hybrid heuristic algorithm includes: generating an initial stacking plan using the greedy bin-packing algorithm, globally optimizing the generated initial stacking plan using the genetic algorithm, and locally optimizing the generated initial stacking plan using the simulated annealing algorithm. The outbound cargo film wrapping execution module is used to load the optimal stacking plan into the palletizing and film wrapping equipment to guide the robotic arm of the palletizing and film wrapping equipment to stack the cargo and automatically complete the film wrapping process. The outbound cargo verification and data update module is used to perform shape and integrity detection on the container that has completed the film wrapping process and is loaded with cargo using a dual verification mechanism of a grating and a profiling gate. After the detection is correct, it performs net covering on the container that has completed the film wrapping process and is loaded with cargo, and at the same time updates the data of the cargo and container to be loaded in the outbound business background database. The outbound cargo film wrapping execution optimization module is used to calculate the center of gravity offset of the container under the optimal stacking plan; input the container information and the calculated center of gravity offset into the film wrapping process acquisition network to obtain the film wrapping tension, material, and winding method in the film wrapping process so that the center of gravity offset of the container is reduced to a preset threshold after film wrapping; the film wrapping process acquisition network selects a convolutional neural network and is trained and generated using historical container information, the center of gravity offset of the container, and the film wrapping tension, material, and winding method in the film wrapping process; perform film wrapping on the container according to the obtained film wrapping tension, material, and winding method in the film wrapping process. It is also used to integrate a film wrapping tension sensor array and a machine vision device on the palletizing and film wrapping equipment, and use the integrated film wrapping tension sensor array and machine vision device to real-time monitor the film wrapping tension at each angle of the container and the image of the container being wrapped in the film wrapping process; analyze the shape of the container being wrapped in the film wrapping process detected in real time, compare the shape of the container being wrapped with the shape of the target container to be wrapped in the film wrapping process or compare the film wrapping tension at each angle of the container with the target film wrapping tension in the film wrapping process. If the comparison result shows that the error is greater than the preset error value, trigger the film wrapping self-correction program and perform film wrapping again.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the method according to any one of claims 1 to 5.

8. A computer device, characterized in that, The computer device includes a memory, a processor, and a program stored and executable on the memory. When the program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 5.

Images