Intelligent parcel sorting method, system and equipment under multi-machine cooperation scene
By dynamically allocating and locking the no-load car, combining the real-time status information of the package and the car, the sorting path is optimized, and the problem of insufficient sorting efficiency and accuracy in the coordinated scenario of multiple machines is solved, and efficient and accurate package sorting is achieved.
Patent Information
- Application Number
- CN202411011035.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-26
- Publication Date
- 2024-12-06
AI Technical Summary
The existing intelligent parcel sorting system lacks unified central management and optimization decisions in multi-machine collaboration scenarios, resulting in low sorting efficiency and insufficient accuracy, and prone to path conflicts, package collisions and misalignment problems.
By receiving and processing the attribute information input by the package information identification device, dynamically allocating and locking available no-load trolleys, implementing package-up strategies based on the status information and package attributes of the associated trolleys, optimizing trolley allocation and path decisions, and reducing path conflicts and missessions risk.
It improves sorting efficiency and accuracy, reduces path conflicts and idle waiting time between cars, reduces labor costs and error rates, and realizes an efficient and accurate package sorting operation platform.
Smart Images

Figure CN119076389A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of package sorting, and in particular to an intelligent package sorting method, system and device in a multi-machine collaboration scenario. Background Art
[0002] With the rapid development of e-commerce and logistics industries, parcel sorting has become a key link. Traditional parcel sorting methods mainly rely on manual labor, which is not only inefficient but also prone to errors. In recent years, with the advancement of technology, intelligent and automated sorting systems have gradually become the mainstream of the industry. However, in the existing intelligent sorting systems, most still face problems such as low sorting efficiency, accuracy needs to be improved, and lack of coordination.
[0003] Especially in the scenario of multi-machine collaborative sorting, how to efficiently coordinate multiple intelligent sorting devices to ensure that parcels can be sorted to the designated location accurately and quickly is an urgent problem to be solved. Traditional sorting methods often lack centralized unified management and optimized decision-making, resulting in path conflicts, package collisions, and missorting between sorting carts, which seriously affects sorting efficiency and accuracy. Summary of the invention
[0004] 1. Technical issues to be resolved
[0005] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides an intelligent parcel sorting method, system and equipment in a multi-machine collaborative scenario, which solves the technical problems that the existing sorting system lacks central unified management and optimized decision-making, resulting in path conflicts, package collisions and missorting during the sorting process.
[0006] (II) Technical solution
[0007] In order to achieve the above object, the main technical solutions adopted by the present invention include:
[0008] In a first aspect, an embodiment of the present invention provides an intelligent package sorting method in a multi-machine collaboration scenario, comprising:
[0009] S1, receiving and processing attribute information of each package input from multiple package information identification devices, and determining the target package loading port of each package;
[0010] S2. Dynamically allocate and lock available empty carts based on the target unloading slot of the parcel and the status information of each intelligent sorting cart;
[0011] S3, obtaining the status information of the associated trolleys and the parcel information carried by the target unloading compartment or the adjacent unloading compartment set range on the predetermined path of each locked empty trolley;
[0012] S4. Based on the status information of the associated cart and the attributes of the parcels it carries, execute the loading strategy: if the target loading slot of the locked empty cart is inconsistent with the target loading slot of the associated cart, or if the target loading slot of the locked empty cart is consistent with the target loading slot of the associated cart, but there is at least one difference in the parcel attributes of the parcel that the locked empty cart is ready to load and the parcel carried by the associated cart, instruct the locked cart to receive the parcel and go to the target loading slot to drop the parcel;
[0013] S5. If the uploading decision logic does not trigger the uploading operation, the currently locked empty cart is abandoned, and other empty carts are dynamically reallocated and locked, and steps S3 and S4 are repeated.
[0014] Optionally, receiving and processing attribute information of each package input from a plurality of package information identification devices, and determining a target package drop-off slot for each package includes:
[0015] S11. synchronously or asynchronously receiving package attribute information transmitted from a plurality of package information identification devices through a preset data interface, the package attribute information including: key information including size, weight, shape, real-time appearance, content type of the package, and auxiliary information including place of shipment, destination, transportation priority, and transportation requirements;
[0016] S12, preprocessing the received package attribute information including data cleaning, formatting, standardization and verification;
[0017] S13, using the built-in algorithm model to perform in-depth analysis on key information in the preprocessed package attribute information, including package pattern recognition, grouping of packages with similar attributes, anomaly detection, and feature coding and description, and continuously self-optimize and adjust according to real-time feedback and historical data, and output a set of numerical or text information describing the package characteristics as a preliminary evaluation result;
[0018] S14, combining the preliminary evaluation results with the auxiliary information of the package to perform data alignment and data integration, and performing at least one operation of weighted scoring, association analysis, feature splicing, and t-distribution neighborhood embedding algorithm to form a package feature portrait;
[0019] S15. Based on the package feature portrait and historical sorting data, use the built-in algorithm model to predict the future load situation of the grid;
[0020] S16. Filter out a list of slots that can accommodate the package based on the feature profile of the package, and select the most suitable slot as the target slot for unloading the package from the filtered list of slots in combination with the future load conditions of the slots;
[0021] S17: The determined target packaging slot information is associated with the corresponding package attribute information and stored.
[0022] Optionally, according to the acquired target unloading slot of the parcel and the status information of each intelligent sorting vehicle, dynamically allocating and locking the available empty vehicles includes:
[0023] S21, receiving or retrieving target delivery slot information of each package;
[0024] S22, real-time acquisition of status information of all intelligent sorting vehicles in the sorting area;
[0025] S23. Filter out a list of candidate trolleys that are currently in an empty state and are not performing other tasks according to the status information of all intelligent sorting trolleys;
[0026] S24, based on the acquired status information, historical performance parameters and synergy effect parameters, perform availability evaluation on each vehicle in the candidate vehicle list;
[0027] S25, dynamically assigning the parcel to the optimal cart based on the target parcel drop-off slot and the evaluation results of the available carts;
[0028] S26. After the trolley is allocated and locked, the status information of the trolley is obtained in real time to track and manage the trolley and the package in real time.
[0029] Optionally, based on the acquired state information, historical performance parameters, and synergy effect parameters, the availability evaluation of each vehicle in the candidate vehicle list includes:
[0030] S241, obtaining the status information of the vehicle including the position and the driving speed, and performing fuzzy processing on the status information;
[0031] S242, performing fuzzy reasoning according to fuzzy rules set based on expert knowledge and fuzzy processed state information to obtain an initial state evaluation value;
[0032] S243, introducing a state decay function, dynamically adjusting the initial state evaluation value according to the continuous working time and task amount of the trolley, and obtaining an adjusted state evaluation value;
[0033] S244, obtaining historical performance parameters of the vehicle including travel distance, task completion time, fault records, and historical task completion rate;
[0034] S245, respectively calculating the standard deviation of the driving distance, task completion time, fault record and historical task completion rate, assigning weights, and then calculating the weighted average standard deviation;
[0035] S246, converting the weighted average standard deviation into a stability index to obtain a stable value of the historical performance parameter;
[0036] S247, obtaining the communication frequency, response time and collaborative task success rate of the car;
[0037] S248. Using graph theory and network analysis methods, each car is abstracted as a node, and the interaction relationship between each car is abstracted as an edge to form an interaction network graph;
[0038] S249. In the interactive network diagram, calculate the betweenness centrality value of each car:
[0039] S2410, according to the betweenness centrality value, combined with the communication frequency, response time and collaborative task success rate, using a complex network analysis algorithm to obtain a synergy effect coefficient by weighted summation;
[0040] S2411. Compare the adjusted state evaluation value, the stable value of the historical performance parameter and the synergy effect coefficient with the set threshold value to obtain the evaluation result of the available car.
[0041] Optionally,
[0042] The state decay function is:
[0043] α=e -λt-μw ;
[0044] Among them, α is the adjusted state evaluation value coefficient, ranging from 0 to 1, λ and μ are attenuation coefficients, t is the continuous working time of the car, and w is the amount of tasks completed by the car;
[0045] The adjusted status assessment value is:
[0046] S a =S o ×α;
[0047] Among them, S a is the adjusted state assessment value, S o is the initial state evaluation value;
[0048] The formula for calculating the standard deviation is:
[0049]
[0050] Where N is the number of samples of historical performance parameters, x i is the observed value of each historical performance parameter, and μ0 is the average observed value of the historical performance parameter;
[0051] The formula for calculating the weighted mean standard deviation is:
[0052]
[0053] Among them, w D is the weighted average standard deviation, M is the number of historical performance parameters, and w j is the weight of the jth historical performance parameter, and s Dj is the standard deviation of the jth historical performance parameter;
[0054] The stability index is:
[0055]
[0056] Among them, ε is a positive number less than a certain value;
[0057] The betweenness centrality value is:
[0058]
[0059] Where V is the set of nodes in the network, including all cars, s and t represent any two different nodes in the network, and s≠v≠t, which means that when calculating the betweenness centrality of node v, the node pair of node v itself is not included, σ st is the number of all shortest paths from node s to node t. The shortest path refers to the path with the least number of edges or the smallest weight required to connect two nodes. st (v) is the number of paths that pass through node v among all the shortest paths from node s to node t;
[0060] The synergy coefficient is:
[0061] SC=w1×C B (v) +w2×CE+w3×CSR;
[0062] Among them, w1, w2, w3 are the betweenness centrality values C B (v) The weights of communication efficiency CE and collaborative task success rate CSR.
[0063] Optionally, based on the evaluation results of the target unloading slot of the parcel and the available carts, dynamically allocating the parcel to the optimal cart includes:
[0064] S251, sorting the available cars according to the evaluation results from high to low, thereby forming a sorted car list;
[0065] S252: in the sorted list of carts, assign the package to be distributed to the optimal cart with the highest evaluation value;
[0066] S253. If there are multiple optimal cart groups with the same evaluation value in the sorted cart list, based on the target drop-off point for each package, a secondary screening is performed by calculating the estimated time and cost for each cart in the optimal cart group to reach the target drop-off point, so as to determine the optimal cart;
[0067] S254: After the allocation is completed, the status information of the trolley is updated, including the information of the allocated packages and the estimated arrival time of the trolley;
[0068] S255. Update the status of the package, mark it as allocated, and record the allocated vehicle number and estimated delivery time.
[0069] Optionally, based on the status information of the associated trolley and the attributes of the packages carried, a loading strategy is executed: if the target loading slot of the locked empty trolley is inconsistent with the target loading slot of the associated trolley, or if the target loading slot of the locked empty trolley is consistent with the target loading slot of the associated trolley, but there is at least one difference in the package attributes of the package that the locked empty trolley is ready to load and the package carried by the associated trolley, then instructing the locked trolley to receive the package and go to the target loading slot to drop the package includes:
[0070] S41, obtaining a target bag-dropping opening of the associated vehicle, and comparing the target bag-dropping opening of the associated vehicle with the target bag-dropping opening of the locked empty vehicle;
[0071] S42a, if the slots are not consistent, directly instruct the locked trolley to receive the package and go to its target slot to drop the package;
[0072] S42b, if the number of grids is consistent, further determine the weight information of the package to be loaded on the locked cart;
[0073] S43a, if the weight of the package to be loaded on the locked trolley is greater than the set weight threshold, directly instruct the locked trolley to receive the package and go to the target unloading compartment to drop the package;
[0074] S43a, if the weight of the package to be loaded on the locked trolley is not greater than the set weight threshold, obtain the weight information of the package carried by the associated trolley, and compare it according to the set first package threshold and second package threshold. When the weight of the package of the associated trolley is less than the first package threshold or greater than the second package threshold, directly instruct the locked trolley to receive the package and go to its target unloading compartment to drop the package;
[0075] S44, if the weight of the parcel of the associated cart is not less than the first parcel threshold or not greater than the second parcel threshold, continue to determine the height information of the parcel to be loaded on the locked cart;
[0076] S45a, if the height of the package to be loaded is greater than the set height threshold, directly instruct the locked trolley to receive the package and go to the target unloading compartment to drop the package;
[0077] S45b. If the height of the package to be loaded is not greater than the set height threshold, obtain the height information of the package carried by the associated trolley. If the height of the package of the associated trolley is less than the set height threshold, directly instruct the locked trolley to receive the package and go to its target unloading compartment to drop the package.
[0078] Optionally, if the uploading decision logic does not trigger the uploading operation, the currently locked empty cart is abandoned, and other empty carts are dynamically reallocated and locked, and after repeating step S3 and step S4, the following further includes:
[0079] S61: When a car receives a package and arrives at the target unloading slot, the status information of the car is updated, and the allocation and loading strategies of other cars are adjusted and optimized in real time according to the received status update information and historical data;
[0080] S62. If an abnormal situation occurs during the package distribution or loading process, it can automatically detect and make automatic adjustments including reallocating packages and calling spare carts.
[0081] In a second aspect, the present invention provides an intelligent parcel sorting system in a multi-machine collaborative scenario, comprising:
[0082] The allocation module is used to receive and process the attribute information of each package input from multiple package information identification devices, determine the target unloading compartment for each package, and dynamically allocate and lock the available empty carts according to the target unloading compartment for the package and the status information of each intelligent sorting cart;
[0083] An information acquisition module is used to obtain the status information of the associated trolleys and the package information carried by the target unloading compartment or the adjacent unloading compartment set range on the predetermined path of each locked empty trolley;
[0084] The package loading decision module is used to execute the package loading strategy based on the status information of the associated trolley and the attributes of the packages carried by the trolley: if the target package loading slot of the locked empty trolley is inconsistent with the target package loading slot of the associated trolley, or if the target package loading slot of the locked empty trolley is consistent with the target package loading slot of the associated trolley, but there is at least one difference in the package attributes of the package that the locked empty trolley is ready to load and the package carried by the associated trolley, then the locked trolley is instructed to receive the package and go to the target package loading slot to drop the package;
[0085] The iterative allocation module is used to abandon the currently locked empty cart when the packaging decision logic does not trigger the packaging operation, dynamically reallocate and lock other empty carts, and repeat the work of the information acquisition module and the packaging decision module.
[0086] In a third aspect, an embodiment of the present invention provides an intelligent parcel sorting device in a multi-machine collaboration scenario, including:
[0087] A package information identification device, used to identify the attribute information of the package;
[0088] Multiple intelligent sorting carts, each equipped with wireless communication equipment for receiving and sending data, as well as performing package transportation and loading tasks;
[0089] The central collaborative controller exchanges data with the package information identification device and the intelligent sorting vehicle through wireless communication equipment, and executes the intelligent package sorting method in the multi-machine collaborative scenario as described above to realize multi-machine collaborative intelligent package sorting.
[0090] (III) Beneficial effects
[0091] The beneficial effects of the present invention are:
[0092] First, the intelligent parcel sorting method in a multi-machine collaborative scenario proposed by the present invention can simultaneously and quickly identify the information of different parcels, including key attributes such as weight, size, and shape, through multiple parcel information recognition devices, and accurately determine the target grid where each parcel should be sorted. This parallel processing method significantly improves the initial efficiency and accuracy of sorting.
[0093] Secondly, the present invention manages multiple intelligent sorting carts in a unified manner. Each cart is equipped with a communication device that can exchange status information with the central collaborative controller in real time, such as whether it is empty, current location, driving speed, etc. This real-time information exchange mechanism ensures that the central collaborative controller can grasp the latest status of each cart at any time, providing accurate data support for the subsequent dynamic allocation and locking of empty carts.
[0094] Next, the present invention dynamically allocates and locks available empty carts according to the target slot of the parcel and the real-time status of each cart. This step not only optimizes the utilization efficiency of the carts, but also reduces the path conflicts and idle waiting time between carts, thereby improving the overall sorting efficiency.
[0095] More importantly, when executing the decision logic for loading packages, the status information of adjacent carts on the predetermined path of the locked empty cart or carts near the same grid will be considered. In this way, the present invention can effectively reduce or avoid the risk of package collision and missorting at the same grid, further improving the accuracy and efficiency of sorting.
[0096] Therefore, the multi-machine collaborative intelligent parcel sorting solution proposed in the present invention realizes an efficient, accurate and collaborative parcel sorting operation platform by integrating multiple parcel information identification devices, multiple intelligent sorting vehicles and a communication network for real-time data exchange. This systematic solution not only improves the sorting efficiency of the logistics industry, but also reduces labor costs and error rates, providing strong technical support for the sustainable development of the logistics industry. BRIEF DESCRIPTION OF THE DRAWINGS
[0097] Figure 1 A schematic flow chart of an intelligent parcel sorting method in a multi-machine collaboration scenario is provided for an embodiment of the present invention;
[0098] Figure 2 A specific flow chart of step S1 of an intelligent package sorting method in a multi-machine collaboration scenario is provided for an embodiment of the present invention;
[0099] Figure 3 A specific flow chart of step S2 of a smart package sorting method in a multi-machine collaboration scenario is provided for an embodiment of the present invention;
[0100] Figure 4 A specific flow chart of step S24 of a method for intelligent parcel sorting in a multi-machine collaboration scenario is provided for an embodiment of the present invention;
[0101] Figure 5 A specific flow chart of step S25 of a method for intelligent parcel sorting in a multi-machine collaboration scenario is provided for an embodiment of the present invention;
[0102] Figure 6 A specific flow chart of step S4 of an intelligent package sorting method in a multi-machine collaboration scenario is provided for an embodiment of the present invention. DETAILED DESCRIPTION
[0103] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.
[0104] like Figure 1As shown, an intelligent parcel sorting method in a multi-machine collaboration scenario proposed in an embodiment of the present invention includes: S1, receiving and processing the attribute information of each parcel input from multiple parcel information identification devices, and determining the target parcel unloading compartment for each parcel; S2, dynamically allocating and locking available empty carts according to the acquired target parcel unloading compartment and the status information of each intelligent sorting cart; S3, acquiring the status information of the associated carts within the set range of the target parcel unloading compartment or the adjacent parcel unloading compartment on the predetermined path of each locked empty cart and the information of the parcels carried; S4, based on the status information of the associated carts and the parcels carried Package attributes, execute the loading strategy: if the target loading slot of the locked empty cart is inconsistent with the target loading slot of the associated cart, or if the target loading slot of the locked empty cart is consistent with the target loading slot of the associated cart, but there is at least one difference in the package attributes between the package that the locked empty cart is ready to load and the package carried by the associated cart, then instruct the locked cart to receive the package and go to the target loading slot to drop the package; S5, if the loading decision logic does not trigger the loading operation, abandon the currently locked empty cart, dynamically reallocate and lock other empty carts, and repeat steps S3 and S4.
[0105] First, the intelligent parcel sorting method in a multi-machine collaborative scenario proposed by the present invention can simultaneously and quickly identify the information of different parcels, including key attributes such as weight, size, and shape, through multiple parcel information recognition devices, and accurately determine the target grid where each parcel should be sorted. This parallel processing method significantly improves the initial efficiency and accuracy of sorting.
[0106] Secondly, the present invention manages multiple intelligent sorting carts in a unified manner. Each cart is equipped with a communication device that can exchange status information with the central collaborative controller in real time, such as whether it is empty, current location, driving speed, etc. This real-time information exchange mechanism ensures that the central collaborative controller can grasp the latest status of each cart at any time, providing accurate data support for the subsequent dynamic allocation and locking of empty carts.
[0107] Next, the present invention dynamically allocates and locks available empty carts according to the target slot of the parcel and the real-time status of each cart. This step not only optimizes the utilization efficiency of the carts, but also reduces the path conflicts and idle waiting time between carts, thereby improving the overall sorting efficiency.
[0108] More importantly, when executing the decision logic for loading packages, the status information of adjacent carts on the predetermined path of the locked empty cart or carts near the same grid will be considered. In this way, the present invention can effectively reduce or avoid the risk of package collision and missorting at the same grid, further improving the accuracy and efficiency of sorting.
[0109] Therefore, the multi-machine collaborative intelligent parcel sorting solution proposed in the present invention realizes an efficient, accurate and collaborative parcel sorting operation platform by integrating multiple parcel information identification devices, multiple intelligent sorting vehicles and a communication network for real-time data exchange. This systematic solution not only improves the sorting efficiency of the logistics industry, but also reduces labor costs and error rates, providing strong technical support for the sustainable development of the logistics industry.
[0110] In order to better understand the above technical solution, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.
[0111] Specifically, an embodiment of the present invention provides an intelligent package sorting method in a multi-machine collaboration scenario, including:
[0112] S1. Receive and process attribute information of each package input from multiple package information identification devices, and determine the target unloading opening of each package.
[0113] Furthermore, if Figure 2 As shown, step S1 includes:
[0114] S11. Synchronously or asynchronously receiving package attribute information transmitted from multiple package information identification devices through a preset data interface.
[0115] The package attribute information includes: key information including the package size, weight, shape, real-time appearance, content type, and auxiliary information including the place of shipment, destination, transportation priority, and transportation requirements.
[0116] S12. Preprocessing the received package attribute information including data cleaning, formatting, standardization and verification.
[0117] First, the cleaned, formatted, verified and standardized parcel attribute information is obtained from the data preprocessing step. This data will be organized into a format suitable for processing by built-in algorithms and self-learning models to ensure the accuracy and completeness of the information.
[0118] S13. Use the built-in algorithm model to conduct in-depth analysis of key information in the pre-processed package attribute information, including package pattern recognition, grouping of packages with similar attributes, anomaly detection, and feature encoding and description. Continuously perform self-optimization and adjustment based on real-time feedback and historical data, and output a set of numerical or text information describing the package characteristics as a preliminary evaluation result.
[0119] First, the built-in advanced algorithm is used to deeply analyze the key information in the pre-processed package attribute information, such as size, weight, shape, etc. Through the self-learning model, it can continuously learn from historical sorting data and real-time feedback to optimize the accuracy of feature extraction and classification, thereby improving the accuracy of the preliminary evaluation results. The preliminary evaluation results include the classification label of the package (such as "large", "small", "fragile", etc.), abnormal identification (such as "abnormal size", "abnormal weight", etc.) or other attribute evaluations related to the sorting decision.
[0120] S14. Combine the preliminary evaluation results with the auxiliary information of the package to perform data alignment and data integration, and form a package feature portrait through at least one operation of weighted scoring, association analysis, feature splicing, and t-distribution neighborhood embedding algorithm.
[0121] The preliminary evaluation results and auxiliary information are aligned to ensure that the relevant information of each package can be accurately matched. Then, this information is integrated to form a unified data structure. After that, a specific fusion algorithm or technology is used to fuse the preliminary evaluation results with the auxiliary information. Through fusion processing, a more comprehensive and richer package feature portrait can be generated. This feature portrait not only reflects the physical properties of the package, but also integrates its transportation needs and business characteristics, providing a richer information basis for subsequent sorting decisions.
[0122] S15. Based on the package feature portrait and historical sorting data, use the built-in algorithm model to predict the future load situation of the grid.
[0123] Predict or obtain future load data in real time, including the number of packages to be processed at each grid, processing speed, remaining capacity, etc. Predict and analyze load trends in the future based on historical data and real-time data.
[0124] S16. Filter out a list of slots that can accommodate the package based on the feature profile of the package. From the filtered list of slots, select the most suitable slot as the target slot for unloading the package in combination with the future load conditions of the slots.
[0125] Based on the package's feature profile, such as size and weight, a list of slots that can accommodate the package is screened out. In the screened list of slots, the current and future loads of the slots, as well as the parcel's processing priority, are further considered to select the most suitable slot as the target slot for unloading the package.
[0126] S17: The determined target unloading slot information is associated with the corresponding package attribute information and stored for subsequent intelligent sorting vehicle allocation and package loading decision-making.
[0127] S2. Dynamically allocate and lock available empty carts based on the target unloading slot of the parcel and the status information of each intelligent sorting cart.
[0128] Furthermore, if Figure 3 As shown, step S2 includes:
[0129] S21, receiving or retrieving the target delivery slot information of each package.
[0130] S22. Acquire the status information of all intelligent sorting vehicles in the sorting area in real time.
[0131] S23. According to the status information of all intelligent sorting carts, a list of candidate carts that are currently in an empty state and are not performing other tasks is screened out.
[0132] S24: Based on the acquired status information, historical performance parameters, and synergy effect parameters, the availability of each car in the candidate car list is evaluated.
[0133] S25. Based on the evaluation results of the target unloading slot of the parcel and the available carts, the parcel is dynamically allocated to the optimal cart.
[0134] S26. After the trolley is allocated and locked, the status information of the trolley is obtained in real time to track and manage the trolley and the package in real time.
[0135] Once the optimal car is determined, an instruction is immediately sent to the car to notify it to go to the designated location to pick up the package. After the package is allocated, the transportation process of the car is monitored in real time to ensure that the task is carried out as planned. If an emergency occurs (such as a car failure, path congestion, etc.), the allocation strategy can be adjusted in real time to reassign the car to the affected package.
[0136] After each task is completed, feedback data on task execution is collected, including task completion time, path efficiency, etc. These feedback data are used to continuously optimize the allocation algorithm and the comprehensive evaluation model of the car to improve the efficiency and accuracy of future task allocation.
[0137] Through this dynamic allocation step, the present invention can efficiently and accurately allocate packages in a constantly changing environment, thereby ensuring smooth operation and efficient performance of the entire system.
[0138] Furthermore, if Figure 4 As shown, step S24 includes:
[0139] S241. Obtain the status information of the vehicle including the position and driving speed, and perform fuzzy processing on the status information.
[0140] In this step, a fuzzy logic algorithm is used to comprehensively consider the current position and speed of the car and other state information to comprehensively evaluate these parameters. First, the precise values such as position and driving speed are converted into membership values in fuzzy sets. For example, driving speed can be divided into fuzzy sets such as low-speed driving (Vs<30km / h), medium-speed driving (30km / h≤Vs<60km / h) and high-speed driving (Vs≥60km / h). Secondly, establish fuzzy rules: according to actual needs, establish fuzzy logic rules, such as "if the speed is low, the car state is poor". Then, fuzzy reasoning and decision-making: reasoning is performed based on fuzzy rules to obtain a fuzzy evaluation of the car state. Through fuzzy comprehensive evaluation methods, such as weighted average method, a comprehensive evaluation index of the car state is obtained.
[0141] S242, performing fuzzy reasoning according to the fuzzy rules set based on expert knowledge and the fuzzy processed state information to obtain an initial state evaluation value.
[0142] S243. Introduce a state decay function, dynamically adjust the initial state evaluation value according to the continuous working time and task amount of the vehicle, and obtain the adjusted state evaluation value.
[0143] The present invention innovatively introduces a state decay function to dynamically adjust the state evaluation value of the vehicle according to its working time and task volume. For example, as the working time increases, the power reduction or mechanical wear may lead to performance degradation, which will be reflected in the evaluation result through the state decay function.
[0144] The state decay function is:
[0145] α=e -λt-μw ;
[0146] Among them, α is the adjusted state evaluation value coefficient, ranging from 0 to 1, λ and μ are attenuation coefficients, t is the continuous working time of the car, and w is the amount of tasks completed by the car;
[0147] The adjusted status assessment value is:
[0148] S a =S o ×α;
[0149] Among them, S a is the adjusted state assessment value, S o is the initial state evaluation value.
[0150] S244. Obtain the historical performance parameters of the vehicle including travel distance, task completion time, fault records, and historical task completion rate.
[0151] S245, respectively calculating the standard deviation of the driving distance, task completion time, fault record and historical task completion rate, assigning weights, and then calculating the weighted average standard deviation.
[0152] S246. Convert the weighted average standard deviation into a stability index to obtain a stable value of the historical performance parameter.
[0153] in,
[0154] The formula for calculating the standard deviation is:
[0155]
[0156] Where N is the number of samples of historical performance parameters, x i is the observed value of each historical performance parameter, and μ0 is the average observed value of the historical performance parameter;
[0157] In order to obtain an overall performance stability index, the standard deviation of each performance parameter can be combined by weighted average or other appropriate methods. For example, if there are three performance parameters (driving distance, task completion time, and historical task completion rate), they can be assigned weights (such as 0.4, 0.3, 0.3) and then the weighted average standard deviation can be calculated. The formula for calculating the weighted average standard deviation is:
[0158]
[0159] Among them, w D is the weighted average standard deviation, M is the number of historical performance parameters, and w j is the weight of the jth historical performance parameter, and s Dj is the standard deviation of the jth historical performance parameter;
[0160] Since the standard deviation measures the fluctuation of the data, the smaller the standard deviation, the more stable the data. In order to convert the standard deviation into a positive indicator (i.e., stability index), its reciprocal or other appropriate mathematical transformations can be used. The stability index is:
[0161]
[0162] Here, ε is a positive number less than a certain value.
[0163] Therefore, the present invention innovatively proposes a "performance stability index". This index is obtained by calculating the standard deviation of the historical performance data of the car. The smaller the standard deviation, the more stable the performance of the car.
[0164] S247. Obtain the communication frequency, response time and collaborative task success rate of the car.
[0165] S248. Using graph theory and network analysis methods, each car is abstracted as a node, and the interaction relationship between each car is abstracted as an edge to form an interaction network graph.
[0166] The present invention uses graph theory to abstract each component (cart, sensor, central collaborative controller, etc.) as a node, and the interaction relationship between components (such as communication, collaboration, etc.) as an edge. In this way, the entire sorting system can be represented as an interactive network graph.
[0167] S249. In the interaction network diagram, calculate the betweenness centrality value of each car.
[0168] S2410. Based on the betweenness centrality value, combined with the communication frequency, response time and collaborative task success rate, a complex network analysis algorithm is used to obtain a synergy coefficient by weighted summation.
[0169] S2411. Compare the adjusted state evaluation value, the stable value of the historical performance parameter, and the synergy coefficient with the set thresholds to obtain the evaluation results of the available cars. Only when all the performance indicators of the car meet or exceed the set thresholds will the car be considered available. Based on the comparison results, a list of available cars will be generated, which meet the preset standards in terms of state, historical performance, and synergy. This list can be used as the basis for task allocation and scheduling.
[0170] in,
[0171] The betweenness centrality value is:
[0172]
[0173] Where V is the set of nodes in the network, including all cars, s and t represent any two different nodes in the network, and s≠v≠t, which means that when calculating the betweenness centrality of node v, the node pair of node v itself is not included, σ st is the number of all shortest paths from node s to node t. The shortest path refers to the path with the least number of edges or the smallest weight required to connect two nodes. st (v) is the number of paths that pass through node v among all the shortest paths from node s to node t;
[0174] The synergy coefficient is:
[0175] SC=w1×C B (v) +w2×CE+w3×CSR;
[0176] Among them, w1, w2, w3 are the betweenness centrality values C B(v) Weights of communication efficiency CE and collaborative task success rate CSR. These weights can be adjusted according to actual conditions to reflect the importance of different parameters. Communication efficiency can be a measure of communication speed, accuracy or reliability; collaborative task success rate is the proportion of cars that successfully complete collaborative tasks.
[0177] Therefore, the present invention innovatively introduces the concept of "betweenness centrality" in social network analysis to measure the collaborative importance of each car in the entire sorting system. Betweenness centrality is determined by calculating the number of times the car serves as a bridge for the shortest path, thereby reflecting the key role of the car in promoting information transfer and collaboration between components. In addition, a "synergy coefficient" is innovatively proposed. This coefficient is calculated by weighted summation based on multiple key parameters such as the betweenness centrality of the car, communication efficiency, and success rate of collaborative tasks. This coefficient can comprehensively reflect the synergy and overall contribution between the cars.
[0178] Furthermore, if Figure 5 As shown, step S25 includes:
[0179] S251. Sort the available cars according to the evaluation results from high to low, so as to form a sorted car list.
[0180] S252: In the sorted list of carts, the package to be distributed is assigned to the optimal cart with the highest evaluation value. This ensures that the package can be delivered to the destination quickly and efficiently.
[0181] S253. If there are multiple optimal cart groups with the same evaluation value in the sorted cart list, based on the target drop-off point for each package, a secondary screening is performed by calculating the estimated time and cost for each cart in the optimal cart group to reach the target drop-off point to determine the optimal cart. Such a screening process can more accurately match carts and packages and improve logistics efficiency.
[0182] S254. After the allocation is completed, the status information of the trolley is updated, including the information of the allocated packages and the estimated arrival time of the trolley.
[0183] After the parcel is assigned, the status information of the cart is updated. This includes information about the assigned parcel, such as the weight, size, destination, etc. of the parcel, as well as the estimated arrival time of the cart. The update of this information helps to track the operation of the cart and the delivery progress of the parcel in real time.
[0184] S255. Update the status of the package, mark it as allocated, and record the allocated vehicle number and estimated delivery time.
[0185] In this step, the status of the package is updated, marked as allocated, and the allocated vehicle number and estimated delivery time are recorded. This makes it easier to track and manage the package in the future, ensuring that the package can be delivered to the customer on time and accurately. At the same time, these records can also serve as a basis for optimization and improvement.
[0186] S3. Obtain the status information of the associated trolleys and the package information carried by the target unloading compartment or the adjacent unloading compartment on the predetermined path of each locked empty trolley.
[0187] S4. Execute the loading strategy based on the status information of the associated cart and the attributes of the packages it carries: if the target loading opening of the locked empty cart is inconsistent with the target loading opening of the associated cart, or if the target loading opening of the locked empty cart is consistent with the target loading opening of the associated cart, but there is at least one difference in the package attributes of the package that the locked empty cart is ready to load and the package carried by the associated cart, then instruct the locked cart to receive the package and go to the target loading opening to drop the package.
[0188] Furthermore, if Figure 6 As shown, step S4 includes:
[0189] S41, obtaining the target bag unloading opening of the associated trolley, and comparing the target bag unloading opening of the associated trolley with the target bag unloading opening of the locked empty trolley.
[0190] S42a, if the slots are not consistent, directly instruct the locked trolley to receive the package and go to its target slot to drop the package.
[0191] S42b. If the number of grid openings is consistent, further determine the weight information of the package that is ready to be loaded on the locked cart.
[0192] S43a: If the weight of the package to be loaded on the locked trolley is greater than the set weight threshold, the locked trolley is directly instructed to receive the package and go to the target unloading compartment to drop the package.
[0193] S43b, if the weight of the package to be loaded on the locked trolley is not greater than the set weight threshold, obtain the weight information of the package carried by the associated trolley, and compare it according to the set first package threshold and second package threshold. When the weight of the package of the associated trolley is less than the first package threshold or greater than the second package threshold, directly instruct the locked trolley to receive the package and go to its target package unloading compartment to drop the package. The first package threshold < the second package threshold.
[0194] S44: If the weight of the package of the associated cart is not less than the first package threshold or not greater than the second package threshold, continue to determine the height information of the package to be loaded on the locked cart.
[0195] S45a: If the height of the package to be loaded is greater than the set height threshold, the locked trolley is directly instructed to receive the package and go to the target unloading compartment to drop the package.
[0196] S45b. If the height of the package to be loaded is not greater than the set height threshold, obtain the height information of the package carried by the associated trolley. If the height of the package of the associated trolley is less than the set height threshold, directly instruct the locked trolley to receive the package and go to its target unloading compartment to drop the package.
[0197] The present invention dynamically manages the intelligent sorting trolley and optimizes the decision logic for loading packages, aiming to improve sorting efficiency and reduce errors, thereby avoiding the situation where two or more trolleys in the front and rear need to go to the same grid, causing the packages to collide at the grid and rebound, resulting in missorting.
[0198] S5. If the uploading decision logic does not trigger the uploading operation, the currently locked empty cart is abandoned, and other empty carts are dynamically reallocated and locked, and steps S3 and S4 are repeated.
[0199] If the loading operation is not triggered after executing the loading decision logic, the currently locked empty cart will be abandoned. Subsequently, dynamic allocation is re-performed to find and lock other empty carts to replace the previous carts. After completing the allocation and locking of the new cart, steps S3 and S4 will be repeated, that is, the matching evaluation of the package and the cart and the final decision before loading will be re-performed to ensure that each package can be effectively allocated and loaded onto the appropriate cart. This step is designed to improve flexibility and efficiency, ensure that the strategy can be quickly adjusted when the initial allocation is unsuccessful, reduce resource waste, and optimize the overall logistics distribution process.
[0200] Furthermore, after step S5, the method further includes:
[0201] S61. When a car receives a package and arrives at the target unloading compartment, the status information of the car is updated, and the allocation and loading strategies of other cars are adjusted and optimized in real time based on the received status update information and historical data. This dynamic optimization aims to improve the overall logistics efficiency and ensure that the packages can be delivered faster and more accurately.
[0202] S62. If an abnormal situation occurs during the package distribution or loading process, it can automatically detect and make automatic adjustments including reallocating packages and calling spare carts.
[0203] During the distribution or loading of packages, the present invention has the ability to automatically detect abnormal situations. Once abnormal situations such as vehicle failure, package information errors, and transportation route obstructions are detected, an immediate response is made. Depending on the specific circumstances of the abnormality, the package is reallocated to other available vehicles, or a spare vehicle is called to ensure the smooth progress of the logistics process. This automated exception handling mechanism significantly improves stability and reliability, reduces the need for manual intervention, and ensures that the package is delivered on time and accurately.
[0204] In a specific implementation, the overall implementation method of the above steps is:
[0205] (1) Identify the package information, obtain the weight and other information, and obtain the number of slots for the package.
[0206] (2) Lock the sorting trolley and obtain information on whether the trolley is empty. If so, lock the trolley. Otherwise, determine whether the trolley is empty in the next step until an empty trolley A is found and lock the trolley number A.
[0207] (3) Obtain the number of slots for unloading the parcel carried by A-1 (the first trolley, or the first two trolleys). If the number of slots is inconsistent with that of A, load the parcel directly.
[0208] (4) If the number of slots is the same, determine the weight of the package to be loaded. If the weight is greater than the set threshold, load it directly. (In this case, you can also sacrifice efficiency and choose not to load it, and proceed to the next step).
[0209] (5) Otherwise, obtain the information of package A-1. If its weight is less than the set first package threshold, directly load it into the package. (In this case, you can also sacrifice efficiency and choose not to load it into the package and proceed to the next step).
[0210] (6) Otherwise, obtain the information of package A-1. If its weight is greater than the set second package threshold, directly load it into the package. (In this case, you can also sacrifice efficiency and choose not to load it into the package and proceed to the next step).
[0211] (7) Determine the height information of the package to be loaded. If the height is greater than the set threshold, load the package directly. (In this case, you can also sacrifice efficiency and choose not to load the package and proceed to the next step).
[0212] (8) Otherwise, obtain the information of package A-1. If the height is less than the set threshold, directly load the package.
[0213] (9) Otherwise, give up locking the trolley and re-lock the next trolley, repeating the above (2)-(8).
[0214] If any of steps (3)-(8) triggers a packet upload, step (1) is executed again to start the cycle.
[0215] In addition, an embodiment of the present invention provides an intelligent parcel sorting system in a multi-machine collaboration scenario, including:
[0216] The allocation module is used to receive and process the attribute information of each package input from multiple package information identification devices, determine the target unloading grid for each package, and dynamically allocate and lock the available empty carts based on the target unloading grid for the package and the status information of each intelligent sorting cart.
[0217] The information acquisition module is used to obtain the status information of the associated trolleys and the package information carried by the target unloading compartment or the adjacent unloading compartment set range on the predetermined path of each locked empty trolley.
[0218] The package loading decision module is used to execute the package loading strategy based on the status information of the associated cart and the attributes of the packages it carries: if the target package loading opening of the locked empty cart is inconsistent with the target package loading opening of the associated cart, or if the target package loading opening of the locked empty cart is consistent with the target package loading opening of the associated cart, but there is at least one difference in the package attributes of the package that the locked empty cart is ready to load and the package carried by the associated cart, then the locked cart is instructed to receive the package and go to the target package loading opening to drop the package.
[0219] The iterative allocation module is used to abandon the currently locked empty cart when the packaging decision logic does not trigger the packaging operation, dynamically reallocate and lock other empty carts, and repeat the work of the information acquisition module and the packaging decision module.
[0220] Furthermore, an embodiment of the present invention provides an intelligent parcel sorting device in a multi-machine collaborative scenario, comprising: a parcel information identification device for identifying the attribute information of a parcel; a plurality of intelligent sorting carts, each of which is equipped with a wireless communication device for receiving and sending data, and performing the transport and loading tasks of the parcel; a central collaborative controller for performing data exchange with the parcel information identification device and the intelligent sorting cart through the wireless communication device, and executing the method as described above, so as to realize the intelligent parcel sorting of multi-machine collaborative
[0221] The intelligent parcel sorting equipment identifies parcel information through the parcel information identification device. The central collaborative controller dynamically assigns tasks to the smart sorting cart based on the parcel information and the real-time status of the cart, and sends instructions through the wireless communication device. After receiving the instructions, the cart performs the corresponding transportation and delivery tasks. During the whole process, the central collaborative controller monitors and adjusts the task allocation and packaging strategies of each cart in real time to achieve efficient and accurate parcel sorting.
[0222] Since the system / device described in the above embodiments of the present invention is a system / device used to implement the method of the above embodiments of the present invention, a person skilled in the art can understand the specific structure and deformation of the system / device based on the method described in the above embodiments of the present invention, and thus will not be described in detail here. All systems / devices used in the method of the above embodiments of the present invention belong to the scope of protection of the present invention.
[0223] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0224] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions.
[0225] It should be noted that in the claims, any reference numerals placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In the claims enumerating several means, several of these means may be embodied by the same hardware. The use of the words first, second, third, etc., is for convenience of expression only and does not indicate any order. These words may be understood as part of the component name.
[0226] In addition, it should be noted that, in the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.
[0227] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments after knowing the basic creative concept. Therefore, the claims should be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.
[0228] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include these modifications and variations.
Claims
1. An intelligent parcel sorting method in a multi-machine collaborative scenario, characterized in that: include: S1, receiving and processing attribute information of each package input from multiple package information identification devices, and determining the target package loading port of each package; S2. Dynamically allocate and lock available empty carts based on the target unloading slot of the parcel and the status information of each intelligent sorting cart; S3, obtaining the status information of the associated trolleys and the parcel information carried by the target unloading compartment or the adjacent unloading compartment set range on the predetermined path of each locked empty trolley; S4. Based on the status information of the associated cart and the attributes of the parcels it carries, execute the loading strategy: if the target loading slot of the locked empty cart is inconsistent with the target loading slot of the associated cart, or if the target loading slot of the locked empty cart is consistent with the target loading slot of the associated cart, but there is at least one difference in the parcel attributes of the parcel that the locked empty cart is ready to load and the parcel carried by the associated cart, instruct the locked cart to receive the parcel and go to the target loading slot to drop the parcel; S5. If the uploading decision logic does not trigger the uploading operation, the currently locked empty cart is abandoned, and other empty carts are dynamically reallocated and locked, and steps S3 and S4 are repeated.
2. The intelligent package sorting method in a multi-machine collaborative scenario as claimed in claim 1, characterized in that: Receiving and processing the attribute information of each package inputted from a plurality of package information identification devices, and determining the target unloading slot of each package comprises: S11. synchronously or asynchronously receiving package attribute information transmitted from a plurality of package information identification devices through a preset data interface, the package attribute information including: key information including size, weight, shape, real-time appearance, content type of the package, and auxiliary information including place of shipment, destination, transportation priority, and transportation requirements; S12, preprocessing the received package attribute information including data cleaning, formatting, standardization and verification; S13, using the built-in algorithm model to perform in-depth analysis on key information in the preprocessed package attribute information, including package pattern recognition, grouping of packages with similar attributes, anomaly detection, and feature coding and description, and continuously self-optimize and adjust according to real-time feedback and historical data, and output a set of numerical or text information describing the package characteristics as a preliminary evaluation result; S14, combining the preliminary evaluation results with the auxiliary information of the package to perform data alignment and data integration, and performing at least one operation of weighted scoring, association analysis, feature splicing, and t-distribution neighborhood embedding algorithm to form a package feature portrait; S15. Based on the package feature portrait and historical sorting data, use the built-in algorithm model to predict the future load situation of the grid; S16. Filter out a list of slots that can accommodate the package based on the feature profile of the package, and select the most suitable slot as the target slot for unloading the package from the filtered list of slots in combination with the future load conditions of the slots; S17: The determined target packaging slot information is associated with the corresponding package attribute information and stored.
3. The intelligent package sorting method in a multi-machine collaborative scenario as claimed in claim 1, characterized in that: According to the target unloading slot of the parcel and the status information of each intelligent sorting vehicle, the available empty vehicles are dynamically allocated and locked, including: S21, receiving or retrieving target delivery slot information of each package; S22, real-time acquisition of status information of all intelligent sorting vehicles in the sorting area; S23. Filter out a list of candidate trolleys that are currently in an empty state and are not performing other tasks according to the status information of all intelligent sorting trolleys; S24, based on the acquired status information, historical performance parameters and synergy effect parameters, perform availability evaluation on each vehicle in the candidate vehicle list; S25, dynamically assigning the parcel to the optimal cart based on the target parcel drop-off slot and the evaluation results of the available carts; S26. After the trolley is allocated and locked, the status information of the trolley is obtained in real time to track and manage the trolley and the package in real time.
4. The intelligent package sorting method in a multi-machine collaborative scenario as claimed in claim 3, characterized in that: Based on the acquired status information, historical performance parameters and synergy effect parameters, the availability evaluation of each car in the candidate car list includes: S241, obtaining the status information of the vehicle including the position and the driving speed, and performing fuzzy processing on the status information; S242, performing fuzzy reasoning according to fuzzy rules set based on expert knowledge and fuzzy processed state information to obtain an initial state evaluation value; S243, introducing a state decay function, dynamically adjusting the initial state evaluation value according to the continuous working time and task amount of the trolley, and obtaining an adjusted state evaluation value; S244, obtaining historical performance parameters of the vehicle including travel distance, task completion time, fault records, and historical task completion rate; S245, respectively calculating the standard deviation of the driving distance, task completion time, fault record and historical task completion rate, assigning weights, and then calculating the weighted average standard deviation; S246, converting the weighted average standard deviation into a stability index to obtain a stable value of the historical performance parameter; S247, obtaining the communication frequency, response time and collaborative task success rate of the car; S248. Using graph theory and network analysis methods, each car is abstracted as a node, and the interaction relationship between each car is abstracted as an edge to form an interaction network graph; S249. In the interactive network diagram, calculate the betweenness centrality value of each car: S2410, according to the betweenness centrality value, combined with the communication frequency, response time and collaborative task success rate, using a complex network analysis algorithm to obtain a synergy effect coefficient by weighted summation; S2411. Compare the adjusted state evaluation value, the stable value of the historical performance parameter and the synergy effect coefficient with the set threshold value to obtain the evaluation result of the available car.
5. The intelligent package sorting method in a multi-machine collaborative scenario as claimed in claim 4, characterized in that: The state decay function is: α=e -λt-μw ; Among them, α is the adjusted state evaluation value coefficient, ranging from 0 to 1, λ and μ are attenuation coefficients, t is the continuous working time of the car, and w is the amount of tasks completed by the car; The adjusted status assessment value is: S a =S o ×α; Among them, S a is the adjusted state assessment value, S o is the initial state evaluation value; The formula for calculating the standard deviation is: Where N is the number of samples of historical performance parameters, x i is the observed value of each historical performance parameter, and μ0 is the average observed value of the historical performance parameter; The formula for calculating the weighted mean standard deviation is: Among them, w D is the weighted average standard deviation, M is the number of historical performance parameters, and w j is the weight of the jth historical performance parameter, and is the standard deviation of the jth historical performance parameter; The stability index is: Among them, ε is a positive number less than a certain value; The betweenness centrality value is: Where V is the set of nodes in the network, including all cars, s and t represent any two different nodes in the network, and s≠v≠t, which means that when calculating the betweenness centrality of node v, the node pair of node v itself is not included, σ st is the number of all shortest paths from node s to node t. The shortest path refers to the path with the least number of edges or the smallest weight required to connect two nodes. st (v) is the number of paths that pass through node v among all the shortest paths from node s to node t; The synergy coefficient is: SC=w1×C B (v)+w2×CE+w3×CSR; Among them, w1, w2, w3 are the betweenness centrality values C B (v) The weights of communication efficiency CE and collaborative task success rate CSR.
6. The intelligent package sorting method in a multi-machine collaborative scenario as claimed in claim 3, characterized in that: Based on the evaluation results of the target drop-off slot of the package and the available carts, the package is dynamically assigned to the optimal cart including: S251, sorting the available cars according to the evaluation results from high to low, thereby forming a sorted car list; S252: in the sorted list of carts, assign the package to be distributed to the optimal cart with the highest evaluation value; S253. If there are multiple optimal cart groups with the same evaluation value in the sorted cart list, based on the target drop-off point for each package, a secondary screening is performed by calculating the estimated time and cost for each cart in the optimal cart group to reach the target drop-off point, so as to determine the optimal cart; S254: After the allocation is completed, the status information of the trolley is updated, including the information of the allocated packages and the estimated arrival time of the trolley; S255. Update the status of the package, mark it as allocated, and record the allocated vehicle number and estimated delivery time.
7. The intelligent package sorting method in a multi-machine collaborative scenario according to any one of claims 1 to 6, characterized in that: Based on the status information of the associated cart and the attributes of the parcels it carries, execute the loading strategy: if the target loading slot of the locked empty cart is inconsistent with the target loading slot of the associated cart, or if the target loading slot of the locked empty cart is consistent with the target loading slot of the associated cart, but there is at least one difference in the parcel attributes of the parcel that the locked empty cart is ready to load and the parcel carried by the associated cart, then instruct the locked cart to receive the parcel and go to the target loading slot to drop the parcel, including: S41, obtaining a target bag-dropping opening of the associated vehicle, and comparing the target bag-dropping opening of the associated vehicle with the target bag-dropping opening of the locked empty vehicle; S42a, if the slots are not consistent, directly instruct the locked trolley to receive the package and go to its target slot to drop the package; S42b, if the number of grids is consistent, further determine the weight information of the package to be loaded on the locked cart; S43a, if the weight of the package to be loaded on the locked trolley is greater than the set weight threshold, directly instruct the locked trolley to receive the package and go to the target unloading compartment to drop the package; S43a, if the weight of the package to be loaded on the locked trolley is not greater than the set weight threshold, obtain the weight information of the package carried by the associated trolley, and compare it according to the set first package threshold and second package threshold. When the weight of the package of the associated trolley is less than the first package threshold or greater than the second package threshold, directly instruct the locked trolley to receive the package and go to its target unloading compartment to drop the package; S44, if the weight of the parcel of the associated cart is not less than the first parcel threshold or not greater than the second parcel threshold, continue to determine the height information of the parcel to be loaded on the locked cart; S45a, if the height of the package to be loaded is greater than the set height threshold, directly instruct the locked trolley to receive the package and go to the target unloading compartment to drop the package; S45b. If the height of the package to be loaded is not greater than the set height threshold, obtain the height information of the package carried by the associated trolley. If the height of the package of the associated trolley is less than the set height threshold, directly instruct the locked trolley to receive the package and go to its target unloading compartment to drop the package.
8. The intelligent package sorting method in a multi-machine collaborative scenario according to any one of claims 1 to 6, characterized in that: If the uploading decision logic does not trigger the uploading operation, the currently locked empty cart is abandoned, and other empty carts are dynamically allocated and locked again. After repeating step S3 and step S4, the following steps are further included: S61: When a car receives a package and arrives at the target unloading slot, the status information of the car is updated, and the allocation and loading strategies of other cars are adjusted and optimized in real time according to the received status update information and historical data; S62. If an abnormal situation occurs during the package distribution or loading process, it can automatically detect and make automatic adjustments including reallocating packages and calling spare carts.
9. An intelligent parcel sorting system in a multi-machine collaborative scenario, characterized in that: include: The allocation module is used to receive and process the attribute information of each package input from multiple package information identification devices, determine the target unloading compartment for each package, and dynamically allocate and lock the available empty carts according to the target unloading compartment for the package and the status information of each intelligent sorting cart; An information acquisition module is used to obtain the status information of the associated trolleys and the package information carried by the target unloading compartment or the adjacent unloading compartment set range on the predetermined path of each locked empty trolley; The package loading decision module is used to execute the package loading strategy based on the status information of the associated trolley and the attributes of the packages carried by the trolley: if the target package loading slot of the locked empty trolley is inconsistent with the target package loading slot of the associated trolley, or if the target package loading slot of the locked empty trolley is consistent with the target package loading slot of the associated trolley, but there is at least one difference in the package attributes of the package that the locked empty trolley is ready to load and the package carried by the associated trolley, then the locked trolley is instructed to receive the package and go to the target package loading slot to drop the package; The iterative allocation module is used to abandon the currently locked empty cart when the packaging decision logic does not trigger the packaging operation, dynamically reallocate and lock other empty carts, and repeat the work of the information acquisition module and the packaging decision module.
10. An intelligent parcel sorting device in a multi-machine collaborative scenario, characterized in that: include: A package information identification device, used to identify the attribute information of the package; Multiple intelligent sorting carts, each equipped with wireless communication equipment for receiving and sending data, as well as performing package transportation and loading tasks; The central collaborative controller exchanges data with the package information identification device and the intelligent sorting vehicle through wireless communication equipment, and executes the intelligent package sorting method in the multi-machine collaborative scenario as described in any one of claims 1 to 8 to realize multi-machine collaborative intelligent package sorting.
Citation Information
Cited By
Electromechanical equipment cooperative control method and system based on Internet of Things
CN120848437A
Mechanical and electrical equipment collaborative control method and system based on internet of things
CN120848437B
