CN-121660577-B - Unmanned vehicle cluster collaborative distribution method and system for intelligent logistics park

CN121660577BCN 121660577 BCN121660577 BCN 121660577BCN-121660577-B

Abstract

The invention discloses an unmanned vehicle cluster collaborative distribution method and system for an intelligent logistics park, and relates to the technical field of logistics distribution collaborative control, wherein the method comprises the steps of pre-judging and acquiring a first predicted emergency by utilizing an emergency prediction model based on real-time logistics monitoring data of the intelligent logistics park; and introducing flexible countermeasure indexes to evaluate the initial cooperative distribution strategy group, optimizing the difference value between the flexible countermeasure indexes and the preset flexible countermeasure indexes, and outputting a first cooperative distribution strategy group. The method solves the technical problems of poor adaptability and slow response of the path planning surface to sudden and unpredictable events, which lead to the reduction of the overall distribution efficiency of the unmanned vehicle cluster and the insufficient operation reliability in the prior art, and achieves the technical effects of improving the overall distribution efficiency and the operation reliability by dynamically enhancing the anti-interference capability and the self-adaptability of the unmanned vehicle cluster collaborative distribution strategy through predictive evaluation and flexible optimization.

Inventors

ZHANG LINGLING
CHEN YUMEI

Assignees

南通理工学院

Dates

Publication Date: 20260508
Application Date: 20260206

Claims (7)

1. The unmanned vehicle cluster collaborative distribution method for the intelligent logistics park is characterized by comprising the following steps of: Acquiring a real-time logistics monitoring data set of a target intelligent logistics park; Inputting the real-time logistics monitoring data set into an emergency prediction model for prediction to obtain a first predicted emergency, wherein the emergency prediction model is obtained by training a historical emergency set of the target intelligent logistics park; Carrying out unmanned vehicle cluster collaborative distribution analysis according to target distribution order information and unmanned vehicle cluster state information to obtain an initial collaborative distribution strategy group, wherein the initial collaborative distribution strategy group comprises distribution planning paths obtained by optimizing all unmanned vehicles; Evaluating the flexible countermeasure indexes of the initial cooperative distribution strategy group according to the first predicted emergency, optimizing the initial cooperative distribution strategy group according to the index difference value of the flexible countermeasure indexes and the preset flexible countermeasure indexes, and outputting a first cooperative distribution strategy group; wherein, the flexible countermeasure index of the initial collaborative distribution strategy group is evaluated according to the first predicted emergency, and the method comprises the following steps: Identifying an event type of the first predicted incident; Determining a sudden disturbance vector according to the event type, wherein the sudden disturbance vector comprises at least one of a path disturbance range, a delay disturbance duration and a vehicle state disturbance; evaluating a plurality of flexible countermeasure factors of each delivery strategy in the initial collaborative delivery strategy group according to the sudden disturbance vector, wherein the flexible countermeasure factors comprise path reconfigurability, delivery task mobility, vehicle state redundancy and affected path proportion; weighting and summing the flexible countermeasure factors to obtain a flexible countermeasure index; calculating an index difference value between the flexible countermeasure index and a preset flexible countermeasure index; When the index difference value is larger than a preset threshold value, triggering an optimization instruction, wherein the optimization instruction comprises at least one of a path reconstruction optimization instruction, a task migration optimization instruction, a state compensation optimization instruction and a collaborative reconstruction optimization instruction; the initial collaborative distribution strategy group is optimized according to the triggered optimization instruction, and the method comprises the following steps: Generating task optimization constraints according to the triggered optimization instructions; and re-planning a group collaborative distribution path of the unmanned vehicle cluster according to the task optimization constraint and the order-vehicle matching profit matrix, and outputting a first collaborative distribution strategy group.
2. The unmanned vehicle cluster collaborative distribution method for a smart logistics park according to claim 1, wherein the incident prediction model is obtained by training a set of historical incidents for the target smart logistics park, the method comprising: extracting a historical logistics monitoring data set sample corresponding to the historical emergency set; performing feature extraction on the historical logistics monitoring data set sample to obtain a logistics monitoring time sequence feature sample, a logistics monitoring flow feature sample and a logistics monitoring environment feature sample; Marking training samples according to the logistics monitoring time sequence characteristic samples, the logistics monitoring flow characteristic samples and the logistics monitoring environment characteristic samples, and obtaining training sample data based on a historical time window; And carrying out time sequence prediction network training according to the training sample data to obtain an emergency prediction model.
3. The unmanned vehicle cluster co-distribution method of the intelligent logistics park of claim 2, wherein training sample data based on the historical time window is obtained; the training sample data comprises event type labels of each historical emergency event and two classification labels representing occurrence of the event, and is based on a logistics monitoring time sequence characteristic alignment sample, a logistics monitoring flow characteristic alignment sample and a logistics monitoring environment characteristic alignment sample of each historical emergency event in a corresponding historical time window.
4. The unmanned vehicle cluster collaborative distribution method of the intelligent logistics park according to claim 1, wherein the unmanned vehicle cluster collaborative distribution analysis is performed according to the target distribution order information and the unmanned vehicle cluster state information to obtain an initial collaborative distribution strategy group, the method comprising: analyzing the order distribution demand vector of the target distribution order information; Analyzing the vehicle distribution capacity vector of the unmanned vehicle cluster state information; Constructing an order-vehicle matching profit matrix according to the order distribution demand vector and the vehicle distribution capacity vector, wherein a profit scoring item of the order-vehicle matching profit matrix comprises a path reachability score, an energy consumption cost score and a time window conformity score; and planning a group collaborative distribution path of the unmanned vehicle cluster according to the order-vehicle matching profit matrix, and outputting an initial collaborative distribution strategy group.
5. The unmanned vehicle cluster collaborative distribution method of an intelligent logistics park according to claim 1, wherein the real-time logistics monitoring dataset is input into an emergency prediction model for prediction, and after the first predicted emergency is obtained, the method further comprises: Inputting the real-time logistics monitoring data set into an emergency prediction model for prediction, and obtaining the first N predicted emergency events with the occurrence probability larger than the expected occurrence probability; Evaluating N flexible countermeasure indexes of the initial collaborative distribution strategy group according to the N predicted incidents; calculating N index difference values of the N flexible countermeasure indexes and a preset flexible countermeasure index; and carrying out cooperative optimization on the initial cooperative distribution strategy group according to the N index difference values, and outputting a first cooperative distribution strategy group.
6. The unmanned vehicle cluster collaborative distribution method for an intelligent logistics park of claim 5, wherein the initial collaborative distribution strategy group is collaborative optimized according to the N index differences, and a first collaborative distribution strategy group is output, the method comprising: Identifying N optimization instructions corresponding to the N index difference values; and extracting superposition optimization instructions of the N optimization instructions, performing group cooperative distribution path cooperative optimization on the unmanned vehicle cluster according to the superposition optimization instructions, and outputting a first cooperative distribution strategy group.
7. An unmanned vehicle cluster collaborative distribution system of an intelligent logistics park, wherein the system is used for implementing the unmanned vehicle cluster collaborative distribution method of the intelligent logistics park according to any one of claims 1 to 6, the system comprising: the logistics monitoring data acquisition module is used for acquiring a real-time logistics monitoring data set of the target intelligent logistics park; The predicted emergency acquisition module is used for inputting the real-time logistics monitoring data set into an emergency prediction model to predict so as to acquire a first predicted emergency, and the emergency prediction model is used for training and acquiring the historical emergency set of the target intelligent logistics park; the collaborative distribution analysis module is used for carrying out collaborative distribution analysis of the unmanned vehicle cluster according to the target distribution order information and the unmanned vehicle cluster state information to obtain an initial collaborative distribution strategy group, wherein the initial collaborative distribution strategy group comprises distribution planning paths obtained by optimizing all unmanned vehicles; and the collaborative distribution strategy optimization module is used for evaluating the flexible countermeasure indexes of the initial collaborative distribution strategy group according to the first predicted emergency, optimizing the initial collaborative distribution strategy group according to the index difference value of the flexible countermeasure indexes and the preset flexible countermeasure indexes, and outputting a first collaborative distribution strategy group.

Description

Unmanned vehicle cluster collaborative distribution method and system for intelligent logistics park Technical Field The invention relates to the technical field of logistics distribution cooperative control, in particular to an unmanned vehicle cluster cooperative distribution method and system for an intelligent logistics park. Background The intelligent logistics park is used as a core node of a modern logistics system, the operation efficiency and the service quality of the intelligent logistics park directly influence the efficiency of the whole logistics chain, the traditional logistics park relies on manual scheduling and fixed path planning, and the intelligent logistics park always presents problems of slow response, low resource utilization rate, poor anti-interference capability and the like when facing increasing order quantity and complex and changeable operation environments, and particularly under the scene of sudden abnormal events such as high concurrent orders, seasonal peak or equipment faults, traffic jams, weather changes and the like, dynamic optimization and real-time adjustment are difficult to realize, so that distribution delay, cost rise and even service interruption are caused. In recent years, unmanned vehicle technology is widely applied to logistics distribution links, however, the dynamic property and uncertainty of an intelligent logistics park put higher demands on collaborative scheduling of unmanned vehicle clusters, for example, unpredictability of emergency events may disturb a preset distribution plan, and how to realize dynamic re-planning of paths on the premise of not interrupting overall operation becomes a key challenge in the intelligent logistics field. Therefore, in the related technology at the present stage, the path planning has the technical problems of poor adaptability to sudden and unpredictable events and slow response, so that the overall distribution efficiency of the unmanned vehicle cluster is reduced, and the operation reliability is insufficient. Disclosure of Invention The unmanned vehicle cluster collaborative distribution method and system solve the technical problems of low overall distribution efficiency and insufficient operation reliability of the unmanned vehicle cluster caused by poor adaptability and slow response of path planning facing sudden and unpredictable events in the prior art, and achieve the technical effects of dynamically enhancing the anti-interference capability and the self-adaptability of the unmanned vehicle cluster collaborative distribution strategy through predictive evaluation and flexible optimization and improving the overall distribution efficiency and the operation reliability. The application provides an unmanned vehicle cluster collaborative distribution method of an intelligent logistics park, which comprises the steps of obtaining a real-time logistics monitoring data set of the intelligent logistics park, inputting the real-time logistics monitoring data set into an emergency prediction model for prediction to obtain a first predicted emergency, training and obtaining a historical emergency set of the intelligent logistics park by the emergency prediction model, carrying out unmanned vehicle cluster collaborative distribution analysis according to target distribution order information and unmanned vehicle cluster state information to obtain an initial collaborative distribution strategy cluster, wherein the initial collaborative distribution strategy cluster comprises distribution planning paths obtained by optimizing all unmanned vehicles, evaluating flexible countermeasure indexes of the initial collaborative distribution strategy cluster according to the first predicted emergency, optimizing the initial collaborative distribution strategy cluster according to index difference values of the flexible countermeasure indexes and preset flexible countermeasure indexes, and outputting the first collaborative distribution strategy cluster. In a possible implementation manner, the unmanned vehicle cluster collaborative distribution method of the intelligent logistics park further performs the following processing of extracting historical logistics monitoring data set samples corresponding to the historical emergency set, performing feature extraction on the historical logistics monitoring data set samples to obtain logistics monitoring time sequence feature samples, logistics monitoring flow feature samples and logistics monitoring environment feature samples, performing training sample labeling according to the logistics monitoring time sequence feature samples, the logistics monitoring flow feature samples and the logistics monitoring environment feature samples to obtain training sample data based on a historical time window, and performing time sequence prediction network training according to the training sample data to obtain an emergency prediction model. In a possible implementation manner, the unmanned