An international logistics task state management method, device and equipment based on edge AI technology and a medium

Abstract

The application relates to an international logistics task state management method and device based on edge AI technology, equipment and a medium, and belongs to the field of logistics management. The method comprises the following steps: uploading international logistics information by a shipping end and creating an international logistics task state; shipping by the shipping end and synchronizing the international logistics information and the international logistics task state to a scheduling end by improving edge AI technology, updating the international logistics task state by the scheduling end; adjusting the queue length and the incoming flow rate in the process of synchronizing the information to the scheduling end by a flow controller, realizing flow control; and synchronizing the international logistics information and the updated international logistics task state to a receiving end by the scheduling end. The application realizes effective transmission management of information and states in the international material conveying process.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of logistics management, and particularly relates to an international logistics task state management method, device and equipment based on edge AI technology and a medium. BACKGROUND

[0002] Transportation is an important part of the international logistics industry and is the center of various business activities in the international logistics process, playing a decisive role. With the continuous development of modern logistics, the current international logistics transportation industry has great development potential, but in fact there are still problems such as low degree of information management. At the same time, due to the rapid business expansion of international logistics enterprises, international logistics transportation business is increasing, and the original transportation system cannot support the expansion of business and cannot uniformly manage the international logistics task state.

[0003] In recent years, AI technology has developed rapidly, but the cloud computing mode requires uploading data to the cloud, which may have potential privacy problems, especially in the international logistics industry which requires more privacy. At the same time, the high delay of cloud computing transmission will also limit the performance of AI applications. Therefore, edge AI technology has gradually become a trend. The importance of edge computing in meeting real-time, privacy protection, and resource management continues to increase, especially in handling large-scale AI data streams, but it also has the disadvantages of delay and packet loss. Too large delay causes time lag experienced from the source end to the target end, and packet loss may cause incomplete information, further affecting real-time and integrity. SUMMARY

[0004] To solve the above problems in the prior art, the application provides an international logistics task state management method, device, equipment and medium based on edge AI technology.

[0005] The object of the application can be achieved by the following technical solutions:

[0006] An international logistics task state management method based on edge AI technology, the implementation of the international logistics task state management method includes the following steps:

[0007] The shipment end uploads international logistics information and creates an international logistics task state;

[0008] The shipment end ships and synchronizes the international logistics information and the international logistics task state to the dispatch end through improved edge AI technology, and the dispatch end updates the international logistics task state;

[0009] The queue length and incoming flow rate in the process of synchronizing information to the dispatch end are adjusted by the flow controller to realize flow control;

[0010] The dispatch end synchronizes the international logistics information and the updated international logistics task state to the consignee end.

[0011] Preferably, the international logistics information includes international logistics weight, international logistics article category, international logistics sending and receiving address information, and international logistics note information, and the international logistics task state includes task confirmation pending state, task confirmation state, task vehicle dispatching state, task consignment note binding state, task consignment note complete receiving state, task rejection state, task modification state, and task cancellation state.

[0012] Preferably, the improved edge AI technology includes a data plane layer, a control layer, and a knowledge layer.

[0013] Preferably, the data plane layer is composed of edge nodes, and the hosts communicate through MPI, the control layer includes a node monitoring module and a decision-making issuing module, and the knowledge layer contains a decision core module.

[0014] Preferably, the decision core module is implemented through an intelligent routing algorithm, and the intelligent routing algorithm includes: setting a weight parameter using a random initialization method, the weight parameter including a first weight parameter, a second weight parameter, and a third weight parameter, the first weight parameter and the second weight parameter having a value range of closed interval [-1, 0], and the third weight parameter having a value range of closed interval [0, 1], increasing the number of algorithm training times and calculating the algorithm state every i seconds, the algorithm state including link latency, packet loss rate, and link utilization rate, when the algorithm state is not in a termination state, determining a routing action according to the algorithm state and calculating an incentive value, obtaining the algorithm state at i+1 seconds, and storing the algorithm state, the routing action, and the incentive value into an experience pool; when the link latency and the link capacity meet the limit conditions, updating the intelligent routing algorithm.

[0015] Preferably, the intelligent routing algorithm specifically includes:

[0016] A network topology is represented by an undirected graph G(V, E, C), where V is a data network node, E is a node connection path between the data network nodes, and C is a link capacity on the node connection path, node traffic is obtained, the node traffic including node sending traffic and node receiving traffic, node time is obtained, the node time including node sending time and node receiving time, and host packet processing time is obtained;

[0017] The link latency, the packet loss rate, and the link utilization rate are calculated according to the node traffic, the node time, and the host packet processing time, and the calculation formula of the link latency is where delay is the link latency, N is the number of data network nodes, Tr is the node sending time, Tr' is the node receiving time, and Tp is the host packet processing time. jNode receiving time of the jth data network node, Ts j Node sending time of the jth data network node, Tc j Host packet processing time of the jth data network node; the formula for calculating the packet loss rate is Wherein, loss is the packet loss rate, Dr j Node receiving traffic of the jth data network node, Ds j Node sending traffic of the jth data network node; the formula for calculating the link utilization is Wherein, utilization is the link utilization, C j Link capacity of the jth data network node;

[0018] The formula for calculating the incentive value is Wherein, R is the incentive value, a is the first weight parameter, β is the second weight parameter, and γ is the third weight parameter;

[0019] The constraint condition is Wherein, TD is the maximum tolerated delay threshold of the link, and TC is the maximum capacity of the link.

[0020] Preferably, the traffic controller generates a reference trajectory based on a given reference input, compares the actual system response with the reference trajectory using the reference trajectory method, adjusts the reference input to approach or reach the expected system behavior, performs rolling optimization on the reference trajectory, calculates the optimization result, adjusts the queue length and the incoming flow rate using the optimization result, adjusts the system parameters using the least square method, performs outgoing flow rate prediction on the adjusted system, compares the prediction result with the actual output, calculates the prediction error, and transmits the prediction error to feedback correction, which adjusts the traffic based on a cost function.

[0021] An international logistics task state management device based on edge AI technology, used to perform the international logistics task state management method described above, comprising a task creation module, a task synchronization module, and a traffic control module.

[0022] The task creation module is used to upload international logistics information by the shipping end and create an international logistics task state, wherein the international logistics information includes international logistics weight, international logistics item category, international logistics sending and receiving address information, and international logistics note information, and the international logistics task state includes a task to be confirmed state, a task confirmed state, a task has been assigned a car state, a task has been bound to a letter of entrustment state, a task letter of entrustment all received state, a task rejected state, a task modification state, and a task void state.

[0023] The task synchronization module is used for shipping end shipping and synchronizing the international logistics information and the international logistics task state to the scheduling end through improved edge AI technology, and the scheduling end updates the international logistics task state; the international logistics information and the updated international logistics task state are synchronized to the receiving end.

[0024] The flow control module is used for adjusting the queue length and the incoming flow rate in the process of information synchronization to the scheduling end through the flow controller, so as to realize flow control.

[0025] An electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the program to implement the international logistics task state management method described above.

[0026] A storage medium containing computer executable instructions for executing the international logistics task state management method described above when executed by a computer processor.

[0027] The beneficial effects of the present application are:

[0028] (1) The edge AI technology is used for international logistics task state management, data processing is performed locally instead of uploading to the cloud, potential data leakage risks are reduced, and edge computing can significantly reduce network delay and provide faster response speed;

[0029] (2) The improved edge AI technology is realized through the data plane layer, the control layer, and the knowledge layer, which is superior to traditional edge algorithms in terms of data packet loss rate and link delay;

[0030] (3) The real-time sensing capability of the control layer quickly identifies network congestion, transmission bottlenecks, and device failures, so that adjustment and optimization measures can be taken in time;

[0031] (4) The knowledge layer can replay the experience pool mechanism, continuously learn from historical experience, optimize flow table rules, improve data flow routing, improve network operation efficiency, realize intelligent adaptation to complex network environments, ensure load balancing of edge nodes, and make important contributions to flexibility and performance optimization of the overall architecture;

[0032] (5) MPI enables different nodes in the edge AI technology to work together to cope with complex network tasks, more quickly adapt to traffic changes, adjust routing, and optimize data transmission paths. At the same time, the MPI communication mechanism helps to avoid single point failure and improve system robustness and reliability;

[0033] (6) The flow controller dynamically adjusts the queue length and the incoming flow rate to ensure stable operation of the system in high concurrency scenarios and to guarantee network performance and service quality. Attached Figure Description

[0034] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0035] Figure 1 This is a flowchart illustrating the steps of an international logistics task status management method based on edge AI technology according to the present invention. Detailed Implementation

[0036] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0037] Working principle and usage process of this invention:

[0038] Please see Figure 1 A method for managing the status of international logistics tasks based on edge AI technology, comprising:

[0039] S1: The shipper uploads international logistics information and creates an international logistics task status. The international logistics information includes international logistics weight, international logistics item category, international logistics sender and receiver address information, and international logistics remarks. The international logistics task status includes task pending confirmation, task confirmed, task dispatched, task bound to authorization letter, task authorization letter fully received, task rejected, task modified, and task cancelled. The task pending confirmation status means the logistics is waiting for carrier confirmation. The task confirmed status means the carrier has confirmed acceptance of the task. The task dispatched status means the logistics task has been dispatched. The task bound to authorization letter status means the logistics personnel have accepted the logistics task and created the corresponding authorization letter. The task authorization letter fully received status means the consignee has signed all authorization letters. The task rejected status means the logistics transportation request has been rejected and the goods have been returned to the shipper. The task modified status means the international logistics information is incorrect and is being corrected. The task cancelled status means the logistics task is cancelled and all operations are stopped.

[0040] S2: The shipping end ships the goods and synchronizes the international logistics information and the international logistics task status to the scheduling end through improved edge AI technology. The scheduling end updates the international logistics task status.

[0041] S3: Flow control is achieved by adjusting the queue length and incoming flow rate during the process of synchronizing information to the scheduling end through the flow controller;

[0042] S4: Synchronize the international logistics information and the updated international logistics task status to the receiving end.

[0043] In the embodiment, the improved edge AI technology includes a data plane layer, a control layer, and a knowledge layer, and is superior to traditional edge algorithms in terms of packet loss rate and link delay. Through the real-time sensing capability of the control layer, network congestion, transmission bottlenecks, and device failures and the like can be quickly identified, so that adjustment and optimization measures can be taken in a timely manner. Through the knowledge layer, the experience pool playback mechanism can be used to continuously learn historical experience, optimize flow table rules, improve data flow routing, improve the operating efficiency of the network, intelligently adapt to complex network environments, ensure load balancing of edge nodes, and make important contributions to flexibility and performance optimization of the overall architecture.

[0044] In the embodiment, the data plane layer is composed of edge nodes, and the hosts communicate through MPI. The control layer includes a node monitoring module and a decision-making issuing module. The knowledge layer includes a decision-making core module. MPI enables different nodes in the edge AI technology to work cooperatively to jointly cope with complex network tasks, more quickly adapt to traffic changes, adjust routing, and optimize data transmission paths. At the same time, the communication mechanism of MPI helps to avoid single-point failures and improve the robustness and reliability of the system.

[0045] The MPI includes an SDN controller and an SDN-MPI daemon. The SDN controller is deployed on a management node to intelligently control the entire cluster system. The SDN-MPI daemon runs in the background on all computing nodes and is responsible for initialization and information communication of the MPI.

[0046] The node monitoring module realizes real-time sensing of network conditions by continuously monitoring network resources and intensive AI data streams transmitted by terminal devices, identifies network congestion, transmission bottlenecks, and device failures and the like, so that adjustment and optimization measures can be taken in a timely manner. The decision-making issuing module is used to formulate issuing rules, optimize data flow routing, and ensure load balancing of edge nodes.

[0047] In the embodiment, the decision-making core module is implemented through an intelligent routing algorithm. The intelligent routing algorithm includes: using a random initialization method to set a weight parameter, the weight parameter including a first weight parameter, a second weight parameter, and a third weight parameter, the first weight parameter and the second weight parameter having a value range of closed interval [-1, 0], the third weight parameter having a value range of closed interval [0, 1], increasing the number of algorithm training times and calculating the algorithm state every i seconds, the algorithm state including link delay, packet loss rate, and link utilization rate, when the algorithm state is not in a termination state, determining a routing action according to the algorithm state and calculating an incentive value, obtaining the algorithm state at i+1 seconds, and storing the algorithm state, the routing action, and the incentive value into an experience pool; and when the link delay and the link capacity satisfy a limit condition, updating the intelligent routing algorithm.

[0048] In the embodiment, the intelligent routing algorithm can be implemented by the following steps:

[0049] The network topology is represented by an undirected graph G(V, E, C), where V is the data network nodes, E is the node connection path between the data network nodes, and C is the link capacity on the node connection path. The node traffic is obtained, including the node sending traffic and the node receiving traffic. The node sending traffic is the traffic sent by each data network node, and the node receiving traffic is the traffic received by each data network node. The node time is obtained, including the node sending time and the node receiving time. The node sending time is the time when the data network node starts to send traffic, and the node receiving time is the time when the data network node ends to receive traffic. The host packet processing time is obtained.

[0050] The link delay, the packet loss rate, and the link utilization are calculated. The calculation formula of the link delay is where delay is the link delay, N is the number of data network nodes, Tr j is the node receiving time of the jth data network node, Ts j is the node sending time of the jth data network node, and Tc j is the host packet processing time of the jth data network node. The calculation formula of the packet loss rate is where loss is the packet loss rate, Dr j is the node receiving traffic of the jth data network node, and Ds j is the node sending traffic of the jth data network node. The calculation formula of the link utilization is where utilization is the link utilization, C j is the link capacity of the jth data network node.

[0051] The calculation formula of the incentive value is where R is the incentive value, α is the first weight parameter, β is the second weight parameter, and γ is the third weight parameter.

[0052] The constraint condition is where TD is the maximum tolerable delay threshold of the link, and TC is the maximum capacity of the link.

[0053] In the embodiment, the flow controller generates a reference trajectory based on a given reference input, compares an actual system response with the reference trajectory using a reference trajectory method, adjusts the reference input to approach or reach a desired system behavior, performs rolling optimization on the reference trajectory, calculates an optimization result, adjusts the queue length and the incoming flow rate using the optimization result, adjusts system parameters using a least square method, performs outgoing flow rate prediction on the adjusted system, compares the prediction result with an actual output, calculates a prediction error, and transmits the prediction error to feedback correction, which adjusts the flow based on a cost function to ensure stable operation of the system in a high concurrency scenario and guarantee network performance and service quality.

[0054] An international logistics task state management device based on edge AI technology, comprising a task creation module, a task synchronization module, and a flow control module.

[0055] The task creation module is configured to upload international logistics information by a shipping end and create an international logistics task state, wherein the international logistics information comprises international logistics weight, international logistics item category, international logistics sender and receiver address information, and international logistics remark information, and the international logistics task state comprises a task to be confirmed state, a task confirmed state, a task has been assigned a vehicle state, a task has been bound to a letter of entrustment state, a task letter of entrustment all received state, a task rejected state, a task modified state, and a task canceled state.

[0056] The task synchronization module is configured to ship by the shipping end and synchronize the international logistics information and the international logistics task state to a dispatching end through improved edge AI technology, and update the international logistics task state by the dispatching end, and synchronize the international logistics information and the updated international logistics task state to a receiving end.

[0057] The flow control module is configured to adjust a queue length and an incoming flow rate in the process of synchronizing information to the dispatching end by a flow controller, and realize flow control.

[0058] The computer storage medium of the embodiments of the present application can adopt any combination of one or more computer readable media. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination thereof. More specific examples (non-exhaustive list) of the computer readable storage medium include: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus or device.

[0059] The computer readable signal medium can include a data signal propagated in baseband or propagated as a carrier wave, in which computer readable program code is embodied. Such propagated data signals can take a wide variety of forms, including but not limited to electro-magnetic signals, optical signals, or any suitable combination thereof. Computer readable signal medium can also be any computer readable medium that is not a storage medium, that is capable of storing the program for use by or in connection with the instruction execution system, apparatus, or device.

[0060] The program code embodied on the computer readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the above. The computer program code for carrying out operations of the present application can be written in any suitable programming language, including object oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0061] The above merely describes preferred embodiments of the present application, and is not intended to limit the present application in any form. Although the present application has been disclosed with the preferred embodiments as above, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above disclosed technical content to obtain equivalent embodiments with equivalent changes, as long as the changes or modifications do not deviate from the technical solution of the present application. Any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application still falls within the scope of the technical solution of the present application.

Claims

1. A method for managing the status of international logistics tasks based on edge AI technology, characterized in that, The implementation of the international logistics task status management method includes the following steps: The shipping end uploads international logistics information and creates an international logistics task status. The international logistics information includes international logistics weight, international logistics item category, international logistics delivery and receipt address information, and international logistics remarks information. The international logistics task status includes task pending confirmation, task confirmed, task dispatched, task bound to authorization letter, task authorization letter fully received, task rejected, task modified, and task cancelled. The shipping end dispatches goods and synchronizes the international logistics information and the international logistics task status to the scheduling end via improved edge AI technology. The scheduling end updates the international logistics task status. The improved edge AI technology includes a data plane layer, a control layer, and a knowledge layer. The data plane layer consists of edge nodes that communicate with each other via MPI. The control layer includes a node monitoring module and a decision-making module. The knowledge layer contains a core decision-making module, which is implemented using an intelligent routing algorithm. The intelligent routing algorithm includes setting weight parameters using a random initialization method. The weight parameters include a first weight parameter, a second weight parameter, and a third weight parameter. The first and second weight parameters have values ​​within a closed interval. The value range of the third weight parameter is [-1,0], which is a closed interval [0,1]. The number of algorithm training iterations is increased, and the algorithm state is calculated every i seconds. The algorithm state includes link latency, packet loss rate, and link utilization. When the algorithm state is not in a terminated state, the routing action is determined based on the algorithm state, and the incentive value is calculated. The algorithm state at i+1 seconds is obtained, and the algorithm state, the routing action, and the incentive value are stored in the experience pool. When the link latency and link capacity meet the constraints, the intelligent routing algorithm is updated. The intelligent routing algorithm specifically includes: using an undirected graph G(V,E,C) to represent the network topology, where V is a data network node, E is the node connection path between the data network nodes, and C is the link capacity on the node connection path; obtaining node traffic, which includes node sending traffic and node receiving traffic; obtaining node time, which includes node sending time and node receiving time; and obtaining host data packet processing time. The link latency, packet loss rate, and link utilization are calculated based on the node traffic, node time, and host data packet processing time. The formula for calculating the link latency is as follows: Where delay is the link delay, and N is the number of data network nodes. Let be the node reception time of the j-th data network node. Let j be the node transmission time of the j-th data network node. Let be the host data packet processing time of the j-th data network node; the packet loss rate is calculated using the formula: loss Where loss is the packet loss rate. For the j-th data network node to receive traffic, Send traffic to the j-th data network node; The formula for calculating the link utilization is as follows: Where utilization is the link utilization rate, and Cj is the link capacity of the j-th data network node; The formula for calculating the excitation value Where R is the incentive value, α is the first weighting parameter, β is the second weighting parameter, and γ is the third weighting parameter; The limiting conditions are: Where TD is the maximum tolerable delay threshold of the link, and TC is the maximum capacity of the link; Flow control is achieved by adjusting the queue length and incoming flow rate during the information synchronization process to the scheduling end using a flow controller. The flow controller generates a reference trajectory based on a given reference input, compares the actual system response with the reference trajectory using a reference trajectory method, and adjusts the reference input to approximate or achieve the desired system behavior. Rolling optimization is performed on the reference trajectory, and the optimization result is calculated. The queue length and incoming flow rate are adjusted using the optimization result, and the system parameters are adjusted using the least squares method. The outgoing flow rate of the adjusted system is predicted, and the prediction result is compared with the actual output to calculate the prediction error. The prediction error is then passed to feedback correction, which performs the flow control based on a cost function. The dispatcher synchronizes the international logistics information and the updated international logistics task status to the receiving end.

2. An international logistics task status management device based on edge AI technology, characterized in that, The device is applied to the international logistics task status management method as described in claim 1, including a task creation module, a task synchronization module, and a flow control module; The task creation module is used by the shipping end to upload international logistics information and create international logistics task status. The international logistics information includes international logistics weight, international logistics item category, international logistics delivery and receipt address information, and international logistics remarks information. The international logistics task status includes task pending confirmation status, task confirmed status, task dispatched status, task bound to authorization letter status, task authorization letter fully received status, task rejected status, task modified status, and task cancelled status. The task synchronization module is used by the shipping end to ship goods and synchronize the international logistics information and the international logistics task status to the scheduling end through improved edge AI technology. The scheduling end updates the international logistics task status; and synchronizes the international logistics information and the updated international logistics task status to the receiving end. The flow control module is used to adjust the queue length and incoming flow rate during the process of synchronizing information to the scheduling end through the flow controller, thereby achieving flow control.

3. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the international logistics task status management method as described in claim 1.

4. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform the international logistics task status management method as described in claim 1.

Images