Multi-path-oriented vehicle path information generation method and device, equipment and medium

Abstract

The embodiment of the invention discloses a multi-path-oriented vehicle path information generation method and device, equipment and a medium. A specific embodiment of the method comprises the steps of preprocessing initial data based on a preset first punctuality rate to obtain a preprocessing result; according to a preprocessing result, inserting each distribution node in a distribution node list which is not inserted into each virtual vehicle to obtain virtual vehicle data, and determining the inserted vehicle path information and the virtual vehicle data as initial vehicle path information; taking the initial vehicle path information as current vehicle path information, and based on a preset second punctuality rate and the current vehicle path information, executing the following steps: performing single-wheel optimization processing on the current vehicle path information to obtain a single-wheel optimization processing result; and determining a single-wheel optimization processing result as vehicle path information. According to the embodiment, through multiple processing, the efficiency of generating the vehicle path information is improved, and efficient planning and optimization of the vehicle path are realized.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology and include a method, apparatus, device, and medium for generating vehicle path information for multiple paths. Background Art

[0002] With the rapid development of smart logistics, vehicle routing optimization technology is gradually shifting from traditional single-trip optimization to more complex and efficient multi-path optimization. Currently, a common approach to multi-path optimization in vehicle routing problems (VRPs) involves adding one virtual vehicle at a time and then iterating the solution. If the number of vehicles added reaches a certain value and the solution cannot be found, it indicates that the last number of vehicles added is the minimum number required.

[0003] However, when using this method for multi-path optimization, the following technical issues often arise: Because the parameters of the iterative solution must be set for a certain number of rounds, and this number of rounds must be repeated to obtain the final minimum number of trips, each time a virtual vehicle is added and an iterative solution is performed, the same number of rounds must be repeated. If a result is obtained from adding a virtual vehicle at a certain time, this is considered the minimum number of virtual vehicles required; if no result is obtained, the solution must continue to add vehicles and iterate. This method is cumbersome and requires a large amount of repeated calculations, resulting in a waste of computing resources.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the Invention

[0005] The content of this disclosure is used to briefly introduce concepts that will be described in detail in the detailed description section below. The content of this disclosure is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] Some embodiments of the present disclosure propose an occlusion removal method, apparatus, electronic device, and computer-readable medium based on a neural radiation field to explain one or more of the technical problems mentioned in the above background technology section.

[0007] In a first aspect, some embodiments of the present disclosure propose a method for generating vehicle path information for multiple paths, the method comprising: preprocessing initial data based on a preset first punctuality rate to obtain a preprocessing result, wherein the initial data comprises initial vehicle configuration data and delivery node data, and the preprocessing result comprises inserted vehicle path information and a list of uninserted delivery nodes; in response to the uninserted delivery node list being not empty, inserting each delivery node in the uninserted delivery node list into each virtual vehicle according to the uninserted delivery node list and virtual vehicle parameter data to obtain virtual vehicle data, and inserting the inserted delivery node into each virtual vehicle according to the uninserted delivery node list and virtual vehicle parameter data to obtain virtual vehicle data; The input vehicle path information and the above-mentioned virtual vehicle data are determined as initial vehicle path information, wherein each delivery node in the above-mentioned uninserted delivery node list corresponds one-to-one to each virtual vehicle corresponding to the above-mentioned virtual vehicle data; the determined initial vehicle path information is used as current vehicle path information, and based on a preset second punctuality rate and the current vehicle path information, the following optimization steps are performed: a single-round optimization process is performed on the current vehicle path information to obtain a single-round optimization process result; in response to the single-round optimization process result not meeting the preset vehicle optimization condition or the single-round optimization process meeting the preset resource constraint condition, the single-round optimization process result is determined as the vehicle path information.

[0008] In a second aspect, some embodiments of the present disclosure provide a multi-path oriented vehicle path information generation device, the device comprising: a preprocessing unit configured to preprocess initial data based on a preset first punctuality rate to obtain a preprocessing result, wherein the initial data includes initial vehicle configuration data and delivery node data, and the preprocessing result includes inserted vehicle path information and a list of uninserted delivery nodes; an insertion unit configured to, in response to the uninserted delivery node list being not empty, insert each delivery node in the uninserted delivery node list into each virtual vehicle according to the uninserted delivery node list and virtual vehicle parameter data to obtain virtual vehicle data; and The inserted vehicle path information and the virtual vehicle data are determined as initial vehicle path information, wherein each delivery node in the uninserted delivery node list corresponds one-to-one to each virtual vehicle corresponding to the virtual vehicle data; the optimization unit is configured to use the determined initial vehicle path information as current vehicle path information, and based on a preset second punctuality rate and the current vehicle path information, perform the following optimization steps: perform a single-round optimization process on the current vehicle path information to obtain a single-round optimization process result; in response to the single-round optimization process result not meeting a preset vehicle optimization condition or the single-round optimization process meeting a preset resource constraint condition, determine the single-round optimization process result as the vehicle path information.

[0009] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by one or more processors, the one or more processors implement the method described in any implementation of the first aspect above.

[0010] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation of the first aspect is implemented.

[0011] In a fifth aspect, some embodiments of the present disclosure provide a computer program product, including a computer program, which implements the method described in any implementation of the first aspect when executed by a processor.

[0012] The above-mentioned embodiments of the present disclosure have the following beneficial effects: Through the multi-path vehicle route information generation method of some embodiments of the present disclosure, efficient vehicle route planning and optimization are achieved, computational efficiency is improved, and computing resources are saved to a certain extent. Specifically, traditional vehicle route planning methods often have the following problem: during the iterative solution, the parameters must be set to a certain number of rounds, and this number of rounds must be iterated to obtain the final minimum number of trips. Therefore, each time a virtual vehicle is added and then an iterative solution is performed, the iterative solution must be completed for a certain number of rounds. If a certain number of virtual vehicles is added, it is considered the minimum number of virtual vehicles required; if no result is obtained, the solution is iterated again by adding more vehicles. This method is relatively cumbersome and involves a large amount of repeated calculations, resulting in a waste of computing resources. Based on this, the multi-path vehicle route information generation method of some embodiments of the present disclosure first pre-processes the initial data based on a preset first on-time rate, obtaining an effective and targeted pre-processing result, clearly distinguishing the inserted vehicle route information from the uninserted delivery node list, and providing accurate basic data for subsequent delivery arrangements. Next, when the list of uninserted delivery nodes is not empty, the uninserted delivery nodes are rationally assigned and inserted into the virtual vehicle based on the list of uninserted delivery nodes and the virtual vehicle parameter data. This generates virtual vehicle data and determines initial vehicle routing information, effectively preventing missed delivery nodes and ensuring the integrity of the delivery schedule. This initial vehicle routing information is then used as the current vehicle routing information. Based on a preset second on-time rate and the current vehicle routing information, an optimization step is executed to obtain a single-round optimization result. After the optimization process completes, if the single-round optimization result does not meet the preset vehicle optimization conditions or meets the preset resource constraints, the single-round optimization result is determined as the vehicle routing information. This method, based on whether the single-round optimization result meets the preset vehicle optimization conditions or meets the preset resource constraints, rationally determines the final vehicle routing information, avoiding over-optimization or under-optimization and more efficiently utilizing computing resources. In summary, the vehicle path information generation method disclosed in the present invention can effectively solve the problems in traditional methods, realize efficient planning and optimization of vehicle paths, and provide a more high-quality and reliable solution for fields such as logistics and distribution, with high application value and practical significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other features, advantages, and aspects of the various embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0014] Figure 1is a flow chart of some embodiments of a method for generating multi-path vehicle path information according to the present disclosure;

[0015] Figure 2 is a schematic structural diagram of some embodiments of a multi-path oriented vehicle path information generation device according to the present disclosure;

[0016] Figure 3 It is a structural diagram of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0017] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0018] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other.

[0019] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0020] It should be noted that the modifications of "one" and "multiple" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, they should be understood as "one or more".

[0021] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0022] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0023] Figure 1 The process 100 of some embodiments of the multi-path oriented vehicle path information generation method according to the present disclosure is shown. The multi-path oriented vehicle path information generation method includes the following steps:

[0024] Step 101: pre-processing the initial data based on a preset first punctuality rate to obtain a pre-processing result.

[0025] In some embodiments, the execution entity (e.g., a computing device) may pre-process the initial data based on a preset first on-time rate to obtain a pre-processing result. The on-time rate may be the percentage of the number of delivery nodes whose arrival times fall within the time window corresponding to each delivery node. The preset first on-time rate may be set to 100% to ensure that all delivery nodes meet their time window constraints (i.e., arrival times fall within the time window of the delivery node) during pre-processing, thereby avoiding irreparable lateness during subsequent optimization. The initial data includes initial vehicle configuration data and delivery node data. The initial vehicle configuration data may be a collection of vehicle information for each existing actual vehicle when performing delivery without the influence of external factors, and may include relevant data such as the actual number of vehicles, the maximum load capacity of each vehicle, the operating time window, and the driving speed. The delivery node data may include a collection of delivery nodes and demand data for each delivery node to be delivered. A delivery node may be a location to which the goods to be delivered need to be delivered. A delivery node may be a location corresponding to a customer to whom the goods need to be delivered. The above-mentioned demand data set may include the geographical coordinates of each distribution node, the cargo demand weight of each distribution node, the time window of each distribution node (i.e., the earliest and latest delivery time), and the service time related data of each distribution node. The above-mentioned preprocessing results include the inserted vehicle path information and the list of uninserted distribution nodes. The above-mentioned inserted vehicle path information may include each trip corresponding to each actual vehicle after preprocessing, each distribution node successfully inserted into each trip, and the working time and corresponding load corresponding to each trip after preprocessing. The above-mentioned uninserted distribution node list may include each distribution node that has not been successfully inserted into any trip corresponding to any actual vehicle. In practice, the above-mentioned execution entity may sort the distribution nodes according to the urgency of the time window (such as ascending order of latest delivery) or spatial aggregation (density on geographical coordinates) so as to give priority to distribution nodes with strict constraints. Then, a greedy insertion algorithm can be used to process each delivery node in the sorted order. First, any delivery node is inserted into any existing trip corresponding to any actual vehicle. When inserting the node into any trip, the trip is subject to capacity verification and time window verification according to a pre-set first on-time rate of 100%. If the trip inserted into the delivery node meets these capacity and time window verifications, the delivery node is successfully inserted into the trip. The corresponding operating hours and load capacity are also updated. Each delivery node can be a location where goods need to be delivered by a vehicle. A trip can correspond to at least one delivery node. A trip can be information that represents the vehicle's travel path.The vehicle's driving path information can be represented as the process of the vehicle carrying the required goods for all distribution nodes corresponding to a trip, departing from a starting point (such as a warehouse), completing the delivery to each distribution node corresponding to the trip, and returning to the starting point. A vehicle can correspond to at least one trip. The capacity verification can be performed by verifying that the sum of the loads of the distribution nodes corresponding to the trip is less than or equal to the vehicle's maximum load. If the sum of the loads of the distribution nodes corresponding to any trip is less than or equal to the vehicle's maximum load, the trip satisfies the capacity verification; otherwise, the trip fails to meet the capacity verification. The time window verification can be performed by verifying that the arrival time of each distribution node in the trip is within the time window corresponding to each distribution node. If the arrival time of each distribution node in the trip is within the time window corresponding to each distribution node, the trip passes the time window verification; otherwise, the trip fails to meet the time window verification. The working time corresponding to the trip can be the time spent executing the trip, which can include the vehicle's round-trip time and the total service time of each distribution node corresponding to the trip. The load corresponding to the above trip can be the sum of the cargo demand weights of each distribution node corresponding to the trip. In addition, if any distribution node cannot be inserted into any existing trip corresponding to any actual vehicle, a new trip can be added for any actual vehicle to insert the distribution node into the newly added trip. After the new trip is added, the working time of the newly added trip needs to be verified. If the newly added trip meets the above working time verification, a new trip is added for the actual vehicle and the above distribution node is inserted into the newly added trip. Finally, the existing trips corresponding to each actual vehicle are updated. The above working time verification can be to check whether the remaining working time of the vehicle is greater than or equal to the time required for the new trip. If the difference between the working time window of any vehicle and the total time of all existing trips of the vehicle is greater than or equal to the time required for the new trip, it indicates that the new trip meets the above working time verification. Otherwise, it indicates that the new trip does not meet the above working time verification. If any delivery node fails to meet the aforementioned capacity verification and time window verification when inserted into any existing trip, and fails to meet the aforementioned working time verification when inserted into any new trip, then the delivery node will be integrated into the aforementioned list of uninserted delivery nodes. Then, after traversing all delivery nodes (i.e., preprocessing is completed), the existing trips corresponding to each actual vehicle at this time will be used as the trips corresponding to each actual vehicle after preprocessing, and the working time and load corresponding to each existing trip will be used as the working time and load corresponding to each trip after preprocessing. The trips corresponding to each actual vehicle after preprocessing, the delivery nodes successfully inserted into each trip, and the working time and load corresponding to each trip after preprocessing will be determined as the inserted vehicle route information.Finally, the inserted vehicle path information and the uninserted delivery node list can be obtained, which is the preprocessing result.

[0026] Step 102, in response to the uninserted delivery node list being not empty, inserting each delivery node in the uninserted delivery node list into each virtual vehicle according to the uninserted delivery node list and the virtual vehicle parameter data to obtain virtual vehicle data, and determining the inserted vehicle path information and the virtual vehicle data as initial vehicle path information.

[0027] In some embodiments, the execution entity may, in response to the above-mentioned uninserted distribution node list being not empty, insert each distribution node in the above-mentioned uninserted distribution node list into each virtual vehicle according to the above-mentioned uninserted distribution node list and the virtual vehicle parameter data, obtain virtual vehicle data, and determine the above-mentioned inserted vehicle path information and the above-mentioned virtual vehicle data as the initial vehicle path information. Among them, each distribution node in the above-mentioned uninserted distribution node list corresponds to each virtual vehicle one by one. The above-mentioned virtual vehicle can be an additional vehicle created temporarily, and has the same configuration data as the actual vehicle (such as maximum load, working time window, driving speed and other related data). The above-mentioned virtual vehicle parameter data can be the configuration data of the virtual vehicle. The above-mentioned virtual vehicle data may include each virtual vehicle created, each trip corresponding to each virtual vehicle, each distribution node corresponding to each virtual vehicle, and the working time and corresponding load corresponding to each virtual vehicle trip. In practice, when the list of uninserted delivery nodes is not empty, the execution entity can create different trips for each delivery node in the uninserted delivery node list based on the geographic coordinates, cargo demand weight, time window, and service duration corresponding to each delivery node in the uninserted delivery node list, for delivery to each delivery node in the uninserted delivery node list. Furthermore, corresponding virtual vehicles can be created for each created trip. This allows the execution entity to obtain the trips corresponding to each virtual vehicle, the delivery nodes corresponding to each virtual vehicle, the operating time corresponding to each virtual vehicle trip, and the corresponding load. Then, the execution entity performs the aforementioned capacity verification and time window verification on each trip corresponding to each virtual vehicle. When each trip corresponding to each virtual vehicle passes the aforementioned capacity verification and time window verification, the execution entity determines the virtual vehicle, the trips corresponding to each virtual vehicle, the delivery nodes corresponding to each virtual vehicle, and the operating time corresponding to each virtual vehicle trip as virtual vehicle data. Finally, the inserted vehicle route information and the virtual vehicle data are determined as initial vehicle route information.

[0028] Optionally, the above execution entity may further perform the following steps:

[0029] In response to any uninserted distribution node in the uninserted distribution node list failing to be inserted into the virtual vehicle, a prompt message indicating the abnormality is output, and the step of generating the initial vehicle path information is interrupted. The prompt message indicating the abnormality can be a data information indicating an abnormality in the calculation, which can control the step of generating the initial vehicle path information. In practice, when the capacity verification and the time window verification are performed on each trip corresponding to each virtual vehicle, if there is any trip corresponding to the virtual vehicle that fails the capacity verification or the time window verification, a prompt message indicating the abnormality is output. When the execution subject recognizes the prompt message indicating the abnormality, the step of generating the initial vehicle path information is interrupted, and the distribution node corresponding to the trip that failed the verification and the prompt message indicating the abnormality can be output for subsequent processing.

[0030] Optionally, the above execution entity may further perform the following steps:

[0031] In response to the uninserted delivery node list being empty, the inserted vehicle path information is determined as the initial vehicle path information. In practice, if the uninserted delivery node list is empty, i.e., no delivery nodes are integrated into the uninserted delivery node list during the preprocessing step, the execution entity may determine the inserted vehicle path information as the initial vehicle path information.

[0032] Step 103: Using the determined initial vehicle path information as the current vehicle path information, and based on the preset second punctuality rate and the current vehicle path information, perform the following optimization steps:

[0033] Step 1031 , performing a single-round optimization process on the current vehicle path information to obtain a single-round optimization process result.

[0034] In some embodiments, the execution entity may perform a single-round optimization process on the current vehicle path information to obtain a single-round optimization process result.

[0035] In some optional implementations of some embodiments, the execution entity may perform a single-round optimization process on the current vehicle path information through the following steps to obtain a single-round optimization process result:

[0036] Step 1: Perform the removal process in the above-mentioned single-round optimization process on the current vehicle path information to obtain a queue of nodes to be assigned and the vehicle path information after the removal. The above-mentioned single-round optimization process includes the above-mentioned removal process, insertion process, and disturbance process. The above-mentioned queue of nodes to be assigned may include at least one delivery node corresponding to the removed trip. The above-mentioned vehicle path information after the removal may be the current vehicle path information with any trip removed. In practice, if the above-mentioned current vehicle path information includes virtual vehicle data, the above-mentioned process may preferentially remove the trip corresponding to any virtual vehicle from the virtual vehicle data in the above-mentioned current vehicle path information to obtain the vehicle path information after the removal. Then, the delivery node corresponding to the removed trip is integrated into the above-mentioned queue of nodes to be assigned. If the above-mentioned current vehicle path information does not include virtual vehicle data or all trips corresponding to each virtual vehicle in the virtual vehicle data have been removed, the above-mentioned execution entity may remove any trip corresponding to any actual vehicle from the above-mentioned current vehicle path information and integrate at least one delivery node corresponding to the removed trip into the above-mentioned queue of nodes to be assigned.

[0037] Step 2: Perform the above insertion process on the queue of nodes to be assigned and the vehicle path information after removal to obtain the vehicle path information after insertion.

[0038] Step three: perform the above-mentioned disturbance processing on the vehicle path information after the insertion to obtain the single-round optimization processing result.

[0039] In some optional implementations of some embodiments, the execution entity may perform the above insertion processing on the queue of nodes to be assigned and the removed vehicle path information through the following steps to obtain the inserted vehicle path information:

[0040] Step 1: Perform preliminary insertion processing on the queue of nodes to be assigned and the removed vehicle path information to obtain preliminary insertion results. The preliminary insertion results include a list of failed insertions and the preliminary inserted vehicle path information. In practice, the execution entity may insert a delivery node from the queue of nodes to be assigned into any trip from the removed vehicle path information to obtain the vehicle path information for the delivery node. If the removed vehicle path information includes virtual vehicle data, the delivery node will be preferentially inserted into any trip corresponding to any actual vehicle. The trip for which the delivery node is inserted is then subjected to the aforementioned capacity verification. If the trip for which the delivery node is inserted meets the aforementioned capacity verification, the percentage of delivery nodes within the time window corresponding to each delivery node in the vehicle path information for which the delivery node is inserted is compared to determine whether the percentage of the total number of delivery nodes is greater than or equal to the preset second on-time rate. If the obtained on-time rate of the vehicle path information for which the delivery node is inserted is greater than or equal to the preset second on-time rate, the preliminary insertion of the delivery node is successful. The delivery node is then removed from the queue of nodes to be assigned. If there are still delivery nodes in the queue of nodes to be assigned, the next delivery node is inserted into the vehicle path information into which the delivery node was inserted, thus completing the insertion of all delivery nodes in the queue of nodes to be assigned. Conversely, if the delivery node fails to be inserted into any trip, it is transferred to an insertion failure list. If there are still delivery nodes in the queue of nodes to be assigned, the next delivery node is inserted into the vehicle path information after being removed, and a preliminary insertion is performed according to the processing method for that delivery node. Ultimately, a failure insertion list and preliminary inserted vehicle path information are obtained.

[0041] Step 2: In response to the fact that the insertion failure list in the above preliminary insertion result is not empty, the above preliminary insertion result is ejected and inserted to obtain the vehicle path information after insertion. The vehicle path information after insertion may include the vehicle path information after ejection and the insertion failure list. In practice, for any distribution node in the above insertion failure list, the execution entity may use the Ejection Search method in GES to insert the distribution node into any trip corresponding to any vehicle in the above preliminary inserted vehicle path information (the trip corresponding to the actual vehicle is inserted first) to obtain the trip ejected and inserted into the distribution node. The trip ejected and inserted into the distribution node is then subjected to the above capacity verification and the above time window verification. When the trip ejected and inserted into the distribution node satisfies the above capacity verification and the above time window verification and the vehicle path information ejected and inserted into the distribution node meets the preset second punctuality rate, it indicates that the ejection and insertion of the distribution node is successful. The ejected distribution nodes generated by the ejection insertion are integrated into the insertion failure list. In this way, continue to eject any other distribution node in the above-mentioned insertion failure list into the vehicle path information of the distribution node that was ejected, so that the insertion of each distribution node in the above-mentioned insertion failure list can be achieved. On the contrary, if the distribution node fails to be ejected and inserted into any trip, the distribution node will be transferred to the insertion failure list. When there are still distribution nodes in the above-mentioned insertion failure list, the next distribution node will be ejected and inserted into the vehicle path information that was initially inserted, and the ejection insertion will be performed according to the ejection insertion processing method of the distribution node. On the contrary, if the distribution node fails to complete the ejection insertion into any trip, when there are still distribution nodes in the insertion failure list, the next distribution node will be ejected and inserted into the vehicle path information that was initially inserted, and the ejection insertion will be performed according to the processing method of the distribution node. Finally, the insertion failure list and the ejection-inserted vehicle path information, that is, the vehicle path information after insertion, can be obtained.

[0042] Step 3: In response to the insertion failure list in the preliminary insertion result being empty, the preliminary insertion result is determined as the vehicle route information after insertion. In practice, when the preliminary insertion result does not include any delivery nodes for which preliminary insertion failed, that is, each delivery node in the queue of nodes to be assigned is inserted into the vehicle route information after removal, the preliminary insertion result is determined as the vehicle route information after insertion.

[0043] In some optional implementations of some embodiments, the execution entity may perform the disturbance processing on the inserted vehicle path information through the following steps to obtain a single-round optimization processing result, including:

[0044] The first step is to generate a set of status indicators based on the inserted vehicle route information. This set of status indicators includes status indicator information for each trip. This status indicator information may include the number of delivery nodes in a trip, the number of time window overlaps between adjacent delivery nodes, the geographic distance between consecutive nodes, and the ratio of the total weight of the corresponding cargo to the vehicle's maximum load. In practice, for each trip in the inserted vehicle route information, the execution entity can retrieve the geographic coordinates, cargo demand weight, and time window of each delivery node from the delivery node data. First, the number of time window overlaps between adjacent delivery nodes in each trip can be calculated based on the time window of each delivery node. Second, the geographic distance between consecutive delivery nodes can be calculated based on the geographic coordinates of each delivery node in each trip. Next, the ratio of the sum of the cargo demand weights for each delivery node in each trip to the vehicle's maximum load can be calculated. Finally, the number of delivery nodes, the number of time window overlaps between adjacent delivery nodes, the geographic distance between consecutive delivery nodes, and the ratio of the total weight of the cargo to the vehicle's maximum load corresponding to each trip can be determined as the status indicator information for each trip. Finally, the status indicator information corresponding to each of the above trips is integrated into a status indicator set.

[0045] In the second step, based on the above-mentioned state indicator set, the above-mentioned vehicle path information after insertion is dynamically disturbed and a disturbance type result is obtained. The above-mentioned disturbance type result can be, but is not limited to, one of the following: path-level disturbance type, node-level disturbance type, and parallel disturbance type. In practice, the above-mentioned execution subject can determine the path-level disturbance type as the disturbance type result in response to determining that the above-mentioned vehicle path information after insertion meets the preset path-level disturbance type selection condition. The above-mentioned preset path-level disturbance type selection condition can be that the number corresponding to the trips with a large number of distribution nodes in the above-mentioned vehicle path information after insertion is greater than the trip number threshold, or the number corresponding to the trips with a large ratio of the total weight of the goods to the maximum load of the vehicle is greater than the above-mentioned trip number threshold. In response to determining that the above-mentioned vehicle path information after insertion meets the preset node-level disturbance type selection condition, the node-level disturbance type is determined as the disturbance type result. The pre-set node-level perturbation type selection condition can be that the number of trips with a large number of overlaps in the time windows of adjacent delivery nodes in the inserted vehicle path information is greater than the trip number threshold, or the number of trips with a large geographical distance between consecutive nodes is greater than the trip number threshold. In response to determining that the inserted vehicle path information satisfies both the pre-set path-level perturbation type selection condition and the pre-set node-level perturbation type selection condition, the parallel perturbation type is determined as the perturbation type result. For example, the execution entity can pre-define respective judgment thresholds corresponding to the number of delivery nodes, the number of overlaps in the time windows of adjacent delivery nodes, the geographical distance between consecutive nodes, and the ratio of the total weight of the corresponding cargo to the maximum load of the vehicle. When the number of delivery nodes of any trip is greater than the corresponding predefined threshold, it can be judged that the number of delivery nodes of this trip is large; when the number of time window overlaps of adjacent delivery nodes of any trip is greater than the corresponding predefined threshold, it can be judged that the number of time window overlaps of adjacent delivery nodes of this trip is large; when the geographical location distance of consecutive nodes of any trip is greater than the corresponding predefined threshold, it can be judged that the geographical location distance of consecutive nodes of this trip is large; when the ratio of the total weight of the corresponding goods to the maximum load of the vehicle volume of any trip is greater than the corresponding predefined threshold, it can be judged that the ratio of the total weight of the corresponding goods to the maximum load of the vehicle volume of this trip is large. When the above-mentioned inserted vehicle path information meets the above-mentioned preset path-level disturbance type selection conditions, the path-level disturbance type is determined as the disturbance type result. When the above-mentioned inserted vehicle path information meets the preset node-level disturbance type selection conditions, the node-level disturbance type is determined as the disturbance type result. When the above-mentioned inserted vehicle path information meets both the above-mentioned preset path-level disturbance type selection conditions and the preset node-level disturbance type selection conditions, the parallel disturbance type is determined as the disturbance type result.

[0046] In the third step, in response to the disturbance type being a path-level disturbance type, the vehicle path information after insertion is subjected to path-level disturbance processing to obtain the vehicle path information after disturbance. Since the disturbance processing is only for each trip in the vehicle path information after insertion, if the vehicle path information after insertion includes an insertion failure list, the insertion failure list will not be processed. In practice, the execution entity may use a predefined path swap operator and a predefined path remove operator to perform disturbance processing on the vehicle path information after insertion to reduce the number of distribution nodes in each trip or the ratio of the total weight of the corresponding goods to the maximum load of the vehicle, thereby obtaining the vehicle path information after disturbance. The predefined path swap operator may be set to randomly swap any trip corresponding to two different vehicles, i.e., replace the vehicles corresponding to the two trips. The path swap operator may be set to remove a portion of consecutive distribution nodes from any trip corresponding to any vehicle to any trip corresponding to another vehicle. After the predefined path swap operator and the predefined path remove operator are used to perform disturbance processing on each trip in the inserted vehicle path information, the disturbed vehicle path information can be obtained.

[0047] In the fourth step, in response to the disturbance type being a node-level disturbance type, the vehicle path information after insertion is subjected to node-level disturbance processing to obtain the vehicle path information after disturbance. In practice, the execution subject may use a predefined point swap operator to perform disturbance processing on the vehicle path information after insertion to reduce the number of overlapping time windows of adjacent distribution nodes or the geographical location distance between consecutive nodes to obtain the vehicle path information after disturbance. The predefined point swap operator may be set to randomly select two trips and randomly swap the distribution nodes in the two trips one by one. After the predefined point swap operator is used to perform disturbance processing on each trip in the vehicle path information after insertion, the vehicle path information after disturbance may be obtained.

[0048] In step 5, in response to the disturbance type being a parallel disturbance type, the following steps are performed:

[0049] In a first sub-step, path-level perturbation processing and node-level perturbation processing are performed on the inserted vehicle path information to obtain parallel perturbation vehicle path information. In practice, the execution entity may perform path-level perturbation processing and node-level perturbation processing on the inserted vehicle path information. After performing path-level perturbation processing and node-level perturbation processing on each trip in the inserted vehicle path information, parallel perturbation vehicle path information is obtained.

[0050] The second sub-step involves cross-checking the vehicle path information after the parallel perturbations to generate conflict information. This conflict information may include delivery nodes that have been processed by both the path-level perturbation process and the node-level perturbation process. In practice, the execution entity can verify and extract the delivery nodes that have been perturbed by both the path-level perturbation process and the node-level perturbation process to perform conflict detection on the vehicle path information after the parallel perturbations. Consequently, each extracted delivery node is determined as conflict information.

[0051] The third sub-step is to generate the disturbed vehicle path information based on the preset conflict resolution information and the aforementioned conflict information. The preset conflict resolution information may represent a predefined conflict resolution rule. The predefined conflict resolution rule may prioritize path-level disturbance processing, with node-level disturbance processing retained only when there is no conflict. In practice, the execution entity may restore each delivery node included in the conflict information to its state before node-level disturbance processing in the parallel disturbed vehicle path information according to the predefined conflict resolution rule represented by the preset conflict resolution information, thereby obtaining the disturbed vehicle path information.

[0052] Step 6: Generate a single-round optimization result based on the obtained perturbed vehicle path information. In practice, if the inserted vehicle path information includes a list of failed insertions, the execution entity may determine the failed insertion list and the obtained perturbed vehicle path information as the single-round optimization result. If the inserted vehicle path information does not include the list of failed insertions, the obtained perturbed vehicle path information may be determined as the single-round optimization result.

[0053] The above-mentioned first to sixth steps serve as an inventive point of the embodiment of the present disclosure, which solves the technical problem that "the disturbance processing in the existing vehicle path optimization scheme only focuses on the local optimization of a single dimension (such as nodes), lacks a comprehensive evaluation of the multi-dimensional state of vehicle distribution and targeted disturbance processing, resulting in the inability to achieve full-process optimization from path planning to vehicle load and time window constraints." The reasons why it is difficult to meet the needs of efficient vehicle path optimization in the existing technology are as follows: the existing vehicle path optimization scheme has deficiencies in the comprehensive evaluation of multi-dimensional states, the accuracy of disturbance processing, and the implementation of full-process optimization. It is difficult to effectively adapt to the complex and changeable path planning needs under different distribution scenarios. At the same time, there are defects in the balanced utilization of vehicle resources and the improvement of distribution efficiency, resulting in poor distribution timeliness and the inability to fully meet the needs of efficient vehicle path optimization. If the above factors are solved, it can be achieved to improve the efficiency of vehicle path optimization, balance the utilization of vehicle resources, and achieve full-process performance optimization. To achieve this, the present disclosure employs a vehicle routing optimization method based on multi-dimensional state assessment and dynamic perturbation processing. The specific steps include: comprehensively and quantitatively assessing the status of each vehicle delivery trip by generating a set of state indicators containing information such as the number of delivery nodes, the number of overlapping time windows between adjacent nodes, the geographic distance between consecutive nodes, and the ratio of total cargo weight to the vehicle's maximum load; intelligently selecting path-level, node-level, or parallel perturbation types based on predefined trigger conditions based on the state indicator set; accurately perturbing the vehicle routing information using corresponding swap operators, remove operators, and other operators for different perturbation types; further cross-checking and conflict resolution for parallel perturbation types to generate a stable optimized routing result; and comprehensively inserting a failure list (if any) to generate a single-round optimization result. This allows for full-process optimization, from route planning to vehicle load and time window constraints. This allows for more accurate perception of the multi-dimensional status of each trip, flexible selection of perturbation processing methods, and optimization of various aspects of vehicle routing optimization, improving delivery efficiency and vehicle resource utilization, thereby meeting the needs of vehicle routing optimization.

[0054] Optionally, the above execution entity may further perform the following steps:

[0055] In response to the single-round optimization processing result meeting the preset vehicle optimization condition, the current vehicle path information is updated according to the single-round optimization processing result, and the above-mentioned optimization step is continued according to the updated current vehicle path information. The preset vehicle optimization condition can be that when the single-round optimization processing result does not include the insertion failure list, the number of trips in the single-round optimization processing result is greater than the number of trips in the single-round optimization processing result obtained by the previous single-round optimization processing; when the single-round optimization processing result includes the insertion failure list, the number of trips in the single-round optimization processing result is greater than the number of trips in the single-round optimization processing result obtained by the previous single-round optimization processing, or the number of delivery nodes included in the insertion failure list in the single-round optimization processing result is greater than the number of delivery nodes included in the insertion failure list in the single-round optimization processing result obtained by the previous single-round optimization processing. In practice, when the single-round optimization processing result meets the preset vehicle optimization condition, the above-mentioned execution entity can use the trips corresponding to each vehicle, the delivery nodes corresponding to each vehicle, and the working time and load corresponding to each trip of each vehicle in the current single-round optimization processing result as the updated current vehicle path information and perform the single-round optimization processing again. If the result of the above single-round optimization process includes an insertion failure list, in the next single-round optimization process, each delivery node in the insertion failure list can be added to the queue of nodes to be allocated obtained by the removal process.

[0056] Step 1032 : In response to the single-round optimization processing result not meeting the preset vehicle optimization condition or the single-round optimization processing reaching the preset resource constraint condition, the single-round optimization processing result is determined as vehicle path information.

[0057] In some embodiments, the execution entity may determine the single-round optimization processing result as vehicle path information in response to the single-round optimization processing result not meeting the preset vehicle optimization condition or the single-round optimization processing reaching the preset resource constraint condition. The preset resource constraint condition may be the number of predefined single-round optimization processing times or the running time. In practice, when the single-round optimization processing result does not include an insertion failure list, when the number of trips in the single-round optimization processing result is less than or equal to the number of trips in the single-round optimization processing result obtained from the previous single-round optimization processing, the single-round optimization processing result is determined as vehicle path information. When the single-round optimization processing result includes an insertion failure list, when the number of trips in the single-round optimization processing result is less than or equal to the number of trips in the single-round optimization processing result obtained from the previous single-round optimization processing and the number of distribution nodes included in the insertion failure list in the single-round optimization processing result is less than or equal to the number of distribution nodes included in the insertion failure list in the single-round optimization processing result obtained from the previous single-round optimization processing, the single-round optimization processing result is determined as vehicle path information. When the preset number of single-round optimization processes or running time is reached, if the corresponding single-round optimization process at this moment is not completed, the corresponding single-round optimization process at this moment can be completed first to obtain the single-round optimization process result. Finally, the single-round optimization process result can be determined as the vehicle path information.

[0058] Optionally, the above execution entity may further perform the following steps:

[0059] The first step is to generate a scheduling information set corresponding to the vehicle path information based on the vehicle path information. The scheduling information set includes executable scheduling instructions. The executable scheduling instructions can be scheduling instructions that are executable on actual vehicles. In practice, the execution entity can convert the trips corresponding to the actual vehicles in the vehicle path information, the delivery nodes corresponding to each trip, and the geographic coordinates, cargo weight requirements, time windows, and other related data of each corresponding delivery node into executable scheduling instructions according to a predefined scheduling information structure (e.g., JSON). If the vehicle path information includes data corresponding to virtual vehicles or a failed insertion list, the above-mentioned method is used to convert the trips corresponding to the virtual vehicles, the delivery nodes corresponding to each trip, and the geographic coordinates, cargo weight requirements, time windows, and other related data of each corresponding delivery node into executable scheduling instructions according to a predefined scheduling information structure (e.g., JSON). The delivery nodes in the failed insertion list can be sent to a manual terminal for allocation and scheduling by a control person. Finally, the executable scheduling instructions are integrated into a scheduling information set, and the scheduling information set is stored in a database (e.g., MongoDB).

[0060] The second step is to generate a traffic condition assessment result based on the traffic data of the current road network obtained. The traffic data of the current road network may include the location and length of each congested road section in the current road network and the label information corresponding to each congested road section. The traffic data of the current road network may be obtained through a pre-set API. The traffic condition assessment result may include the congestion situation of the entire road network and the location and congestion level of each congested road section. In practice, the execution entity may use the pre-set AutoNavi Map API or TomTom Traffic API to obtain the length of each congested road section and the label information corresponding to each congested road section (such as unobstructed, light congestion, and heavy congestion), and replace the different congestion label information with congestion levels, such as unobstructed corresponding to a congestion level of 1, light congestion corresponding to a congestion level of 2, and heavy congestion corresponding to a congestion level of 3. The sum of the lengths of the above-mentioned congested road sections is divided by the length of the entire road network to obtain the congestion index of the congestion situation of the entire road network. If the congestion index is less than 0.2, the entire road network is considered to be unobstructed; if the congestion index is greater than or equal to 0.2 and less than 0.5, the entire road network is considered to be lightly congested; and if the congestion index is greater than or equal to 0.5, the entire road network is considered to be heavily congested. The congestion situation of the entire road network, as well as the location and congestion level of each congested road section, are then determined as the traffic condition assessment results.

[0061] The third step is to generate an optimized dispatch information set based on the traffic condition assessment results and the dispatch information set. In practice, the execution entity can use the AM-DRL algorithm to calculate the driving routes for each trip using the traffic condition assessment results and the executable dispatch instructions in the dispatch information set. The driving routes for each trip are then assigned to the executable dispatch instructions for each vehicle. This completes the optimization of the dispatch information set, resulting in an optimized dispatch information set.

[0062] The fourth step is to generate vehicle availability information based on the above-mentioned vehicle configuration data and the acquired current vehicle data. The acquired current vehicle data may be the vehicle information corresponding to each existing vehicle at the current moment. The vehicle information corresponding to each existing vehicle at the current moment may include relevant data such as the number, number (license plate number), maximum load, current location, and working time window of each existing vehicle. In practice, the execution entity may obtain the vehicle information corresponding to each existing vehicle at the current moment through the above-mentioned vehicle configuration data and real-time monitoring equipment (such as an on-board IoT device). The availability of each existing vehicle is then determined according to predefined vehicle conditions, i.e., existing vehicles that do not meet the predefined vehicle conditions are filtered out. The predefined vehicle conditions may include that the maximum load corresponding to the actual vehicle at the current moment is at least greater than the cargo demand weight of each distribution node, the current actual vehicle's location is at the starting point of the distribution trip (such as a warehouse), and the current actual vehicle's working time window meets at least the time window of any distribution node. Thus, the filtered actual vehicles and their corresponding vehicle information at the current moment can be obtained. Finally, the actual vehicles filtered above and their corresponding vehicle information at the current moment are integrated, and the integration result is determined as vehicle availability information.

[0063] The fifth step is to dynamically schedule tasks for each actual vehicle based on the scheduling performance evaluation model using the optimized scheduling instruction set and vehicle availability information. The scheduling performance evaluation model can be a pre-trained spatiotemporal graph convolutional evaluator (STGCE) whose input is the optimized scheduling instruction set and vehicle availability information and whose output is a vehicle task allocation plan. In practice, the scheduling performance evaluation model can be a spatiotemporal graph convolutional evaluator (STGCE). The scheduling performance evaluation model can include an input layer, a spatiotemporal feature fusion layer, and a multi-task decision output layer. First, in the input layer of the scheduling performance evaluation model, a dynamic heterogeneous graph can be constructed based on real-time vehicle status data, task requirement data, and road network status data. The dynamic heterogeneous graph contains three types of nodes (vehicle nodes, task nodes, and intersection nodes) and four types of edges (vehicle-task allocation edges, task-task path sequence edges, intersection-intersection adjacency edges, and task-intersection spatial relationship edges). A trip may include multiple tasks. The weights of these edges can be calculated based on the real-time congestion level and the Euclidean distance between nodes. For example, the above execution entity receives the real-time status of vehicle V1 (position coordinates (116.40, 39.90), remaining load ratio 0.7, time window margin 120 minutes), the demand data of task T1 (delivery coordinates (116.42, 39.88), cargo weight 150kg, time window [9:00, 11:00]), and road network status data (intersection R1 congestion level 1, historical traffic speed 30km / h). Vehicle V1 is mapped to a vehicle node, task T1 is mapped to a task node, and intersection R1 is mapped to an intersection node. Task T1 is assigned to vehicle V1, and an assignment edge is established from vehicle V1 to task T1; the spatial distance between task T1 and intersection R1 is less than a threshold (500 meters), and a spatial relationship edge is established from task T1 to intersection R1. For calculating edge weights, for example, for the edge from vehicle V1 to task T1, the congestion level can be the congestion level of the area where task T1 is located (e.g., 1), and the distance can be the Euclidean distance from the vehicle location to the delivery coordinate (3.5 kilometers). The corresponding weight can then be calculated using the following formula: 1 / (0.1+1×3.5)≈0.27. This method yields a dynamic heterogeneous graph consisting of the three types of nodes and four types of edges corresponding to each vehicle. Second, in the spatiotemporal feature fusion layer, spatial convolution and temporal fusion can be performed on the dynamic heterogeneous graph. The spatial convolution independently aggregates neighborhood features for each of the four edge types, while the temporal fusion combines current and historical features through a gating mechanism. For example, the execution entity performing graph convolution operations on each of the four edge types in the dynamic heterogeneous graph may include: for vehicle-task assignment edges, aggregating vehicle node features onto the task node; and for task-intersection spatial relationship edges, aggregating intersection congestion features onto the task node.During gated fusion, the forget gate output of each task node, such as task node T1, can be calculated as 0.6 for the current feature vector [5.3, 3.1, 0.85, 32400, 39600, 1.39e-4, 2] and the update gate output of the previous feature vector [5.2, 3.0, 0.82, 32400, 39600, 1.39e-4, 1]. The fused feature can then be calculated as: (1 - 0.6) × historical feature vector + 0.6 × 0.8 = 0.4 × current feature vector + 0.48 × previous feature vector. This method yields a fused feature matrix encompassing all vehicle nodes, where each row corresponds to the fused feature vector of a vehicle node. Third, in the multi-task decision output layer, performance evaluation scores and vehicle execution commands can be generated based on the fused feature matrix. The predicted evaluation scores are calculated through global feature pooling, while the vehicle execution commands can be obtained through scheduling instructions. The above-mentioned vehicle execution command may include path planning parameters and task allocation parameters. The above-mentioned scheduling instruction may be a vehicle node-level classification result. For example, the above-mentioned execution subject embeds and executes the high-order nodes of the fusion feature matrix, including: global mean pooling to obtain the graph feature vector [0.82, -0.17, 0.75, 0.63, 0.15, 0.22]; outputting the prediction evaluation score of 0.85 through two layers of fully connected layers (128 dimensions to 64 dimensions and 64 dimensions to 1); performing node classification on the feature vector [0.82, -2.22, 0.7, 0.67, 0.12, 0.21] of each vehicle node, such as vehicle node V1, and outputting the action probability distribution [0.1, 0.8, 0.1] through the fully connected layer (128 dimensions to 64 dimensions and 64 dimensions to 3 dimensions); selecting the action corresponding to the maximum probability (such as a detour alternative route) as the scheduling instruction. Next, for each vehicle, such as vehicle V1's dispatch instruction (detour alternate route), real-time traffic data is combined to generate a vehicle execution command, including: avoid congested road section L1 (congestion level 2), adopt alternative routes R1, R3, and R5 (estimated time savings of 12 minutes), and update the task time window to [9:10, 11:10]. Finally, this dispatch instruction is sent to the vehicle terminal of the corresponding vehicle via JSON format. This method enables dynamic task scheduling for each actual vehicle.

[0064] Step 6: In response to receiving the execution status information for each path corresponding to the vehicle path information, dynamically adjust the scheduling performance evaluation model based on the execution status information for each path. This execution status information includes data such as actual arrival time, actual path length, and actual fuel consumption. In practice, after each vehicle completes its scheduling task, the execution entity can collect statistics on the execution status information for each path. Then, by calculating the proportion of each vehicle's execution status information for each path in the fused feature vector corresponding to each vehicle node in the fused feature matrix, indicator information for each path is obtained. The indicator information for each path is then organized to obtain the various indicator information corresponding to the vehicle path information. This indicator information includes, for example, punctuality rate, load utilization rate, and path cost coefficient. Each indicator information is then weighted to obtain an actual evaluation score. When the difference between the actual evaluation score and the predicted evaluation score exceeds a predefined evaluation score difference threshold, the scheduling performance evaluation model is adjusted to obtain an updated scheduling performance evaluation model for use in the next scheduling performance evaluation model iteration. For example, the collected performance information on each vehicle's completion of each route is converted into metrics, such as vehicle V1's actual arrival time (9:08), actual route length (15.3 km), and actual fuel consumption (2.1 L) on any route. By calculating the proportion of each route's performance information in the fused feature vector of each vehicle node, the corresponding metrics for each vehicle are obtained, such as 92% on-time rate, 85% load utilization, and 0.9 path cost coefficient. Then, after weighting each metric, the actual evaluation score is 0.82. The difference between the predicted evaluation score (e.g., 0.85) and the actual evaluation score of 0.82 is calculated to be 0.03. Since the predefined evaluation score difference threshold is 0.02, the learning rate can be set to 0.01. In the back-propagation update of the spatiotemporal feature fusion layer, if the difference between the predicted evaluation score (e.g., 0.85) and the actual evaluation score is less than or equal to 0.02, no adjustments to the scheduling performance evaluation model are required. According to this method, the scheduling performance evaluation model can be dynamically adjusted.

[0065] The above-mentioned first to sixth steps serve as an inventive point of the embodiment of the present disclosure, which solves the technical problem that "the existing vehicle scheduling scheme is only based on static path information for scheduling, lacks comprehensive consideration of real-time traffic conditions and vehicle dynamic information and intelligent scheduling capabilities, resulting in the inability to achieve dynamic collaborative optimization from path planning to task allocation." The reasons why the existing technology is difficult to meet the needs of efficient vehicle dynamic scheduling are as follows: the existing vehicle scheduling scheme has deficiencies in real-time traffic data integration, accurate judgment of vehicle availability, and intelligent implementation of dynamic task scheduling, and it is difficult to effectively adapt to complex and changeable road conditions and vehicle status. At the same time, there are defects in the rationality of task allocation and scheduling accuracy, resulting in low scheduling efficiency and the inability to fully meet the needs of efficient vehicle dynamic scheduling. If the above factors are solved, it can be achieved to improve vehicle scheduling efficiency, enhance scheduling accuracy, and achieve dynamic collaborative optimization. To achieve this effect, the present disclosure adopts a vehicle path optimization method based on real-time traffic perception and intelligent scheduling. The specific steps include: converting vehicle path information into an executable scheduling instruction set and storing it to ensure the standardization and operability of the scheduling instruction; using real-time traffic data to generate traffic condition assessment results to provide accurate road condition basis for scheduling optimization; combining traffic conditions and scheduling instruction set, using the AM-DRL algorithm to optimize the scheduling information set to achieve dynamic adjustment of path planning; integrating vehicle configuration data with real-time vehicle data to generate vehicle availability information and accurately screen suitable vehicles; using a scheduling performance evaluation model, based on the optimized scheduling instruction set and vehicle availability information, dynamically scheduling actual vehicles to achieve intelligent task allocation and dynamic path planning; dynamically adjusting the scheduling performance evaluation model based on the execution status information of each path to continuously improve the accuracy and adaptability of the scheduling plan. In this way, dynamic collaborative optimization from path planning to task allocation can be achieved. This can fully integrate real-time traffic and vehicle dynamic information, make intelligent decisions on scheduling plans, optimize all aspects of vehicle scheduling, improve scheduling efficiency and accuracy, effectively cope with complex road conditions and vehicle state changes, and meet the needs of more efficient vehicle dynamic scheduling.

[0066] The above-mentioned embodiments of the present disclosure have the following beneficial effects: Through the multi-path vehicle route information generation method of some embodiments of the present disclosure, efficient vehicle route planning and optimization are achieved, computational efficiency is improved, and computing resources are saved to a certain extent. Specifically, traditional vehicle route planning methods often have the following problem: during the iterative solution, the parameters must be set to a certain number of rounds, and this number of rounds must be iterated to obtain the final minimum number of trips. Therefore, each time a virtual vehicle is added and then an iterative solution is performed, the iterative solution must be completed for a certain number of rounds. If a certain number of virtual vehicles is added, it is considered the minimum number of virtual vehicles required; if no result is obtained, the solution is iterated again by adding more vehicles. This method is relatively cumbersome and involves a large amount of repeated calculations, resulting in a waste of computing resources. Based on this, the multi-path vehicle route information generation method of some embodiments of the present disclosure first pre-processes the initial data based on a preset first on-time rate, obtaining an effective and targeted pre-processing result, clearly distinguishing the inserted vehicle route information from the uninserted delivery node list, and providing accurate basic data for subsequent delivery arrangements. Next, when the list of uninserted delivery nodes is not empty, the uninserted delivery nodes are rationally assigned and inserted into the virtual vehicle based on the list of uninserted delivery nodes and the virtual vehicle parameter data. This generates virtual vehicle data and determines initial vehicle routing information, effectively preventing missed delivery nodes and ensuring the integrity of the delivery schedule. This initial vehicle routing information is then used as the current vehicle routing information. Based on a preset second on-time rate and the current vehicle routing information, an optimization step is executed to obtain a single-round optimization result. After the optimization process completes, if the single-round optimization result does not meet the preset vehicle optimization conditions or meets the preset resource constraints, the single-round optimization result is determined as the vehicle routing information. This method, based on whether the single-round optimization result meets the preset vehicle optimization conditions or meets the preset resource constraints, rationally determines the final vehicle routing information, avoiding over-optimization or under-optimization and more efficiently utilizing computing resources. In summary, the vehicle path information generation method disclosed in the present invention can effectively solve the problems in traditional methods, realize efficient planning and optimization of vehicle paths, and provide a more high-quality and reliable solution for fields such as logistics and distribution, with high application value and practical significance.

[0067] Further references Figure 2 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a multi-path vehicle path information generation device. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0068] like Figure 2As shown, some embodiments of the multi-path vehicle path information generation device 200 include: a preprocessing unit 201, an insertion unit 202 and an optimization unit 203. The preprocessing unit 201 is configured to preprocess the initial data based on a preset first punctuality rate to obtain a preprocessing result, wherein the initial data includes initial vehicle configuration data and distribution node data, and the preprocessing result includes inserted vehicle path information and a list of uninserted distribution nodes; the insertion unit 202 is configured to, in response to the uninserted distribution node list being not empty, insert each distribution node in the uninserted distribution node list into each virtual vehicle according to the uninserted distribution node list and virtual vehicle parameter data to obtain virtual vehicle data, and insert the inserted vehicle path information and the inserted virtual vehicle parameter data into each virtual vehicle. The virtual vehicle data is determined as initial vehicle path information, wherein each delivery node in the above-mentioned uninserted delivery node list corresponds one-to-one to each virtual vehicle corresponding to the above-mentioned virtual vehicle data; the optimization unit 203 is configured to use the determined initial vehicle path information as the current vehicle path information, and based on the preset second punctuality rate and the current vehicle path information, perform the following optimization steps: perform a single-round optimization process on the current vehicle path information to obtain a single-round optimization process result; in response to the single-round optimization process result not meeting the preset vehicle optimization condition or the single-round optimization process meeting the preset resource constraint condition, determine the single-round optimization process result as the vehicle path information.

[0069] It is understood that the units described in the device 200 are similar to those described in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the device 200 and the units included therein, and will not be repeated here.

[0070] Reference below Figure 3 , which shows a structural diagram of an electronic device 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0071] like Figure 3 As shown, the electronic device 300 may include a processing device 301 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. Various programs and data required for the operation of the electronic device 300 are also stored in the RAM 303. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0072] Typically, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as needed.

[0073] In particular, according to some embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from a network via the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are performed.

[0074] It should be noted that the computer-readable medium described in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or device. In some embodiments of the present disclosure, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0075] In some embodiments, the client and server can communicate using any currently known or future developed network protocol, such as HTTP (HyperText Transfer Protocol), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0076] The computer-readable medium may be included in the electronic device; or it may exist independently without being assembled into the electronic device. The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: pre-processes the initial data based on a preset first punctuality rate to obtain a pre-processing result, wherein the initial data includes initial vehicle configuration data and distribution node data, and the pre-processing result includes inserted vehicle path information and a list of uninserted distribution nodes; in response to the uninserted distribution node list being not empty, insert each distribution node in the uninserted distribution node list into each virtual vehicle according to the uninserted distribution node list and virtual vehicle parameter data to obtain virtual vehicle data, and The inserted vehicle path information and the virtual vehicle data are determined as initial vehicle path information, wherein each delivery node in the uninserted delivery node list corresponds one-to-one to each virtual vehicle corresponding to the virtual vehicle data; the determined initial vehicle path information is used as current vehicle path information, and based on a preset second punctuality rate and the current vehicle path information, the following optimization steps are performed: a single-round optimization process is performed on the current vehicle path information to obtain a single-round optimization process result; in response to the single-round optimization process result not meeting a preset vehicle optimization condition or the single-round optimization process meeting a preset resource constraint condition, the single-round optimization process result is determined as the vehicle path information.

[0077] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may 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 may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0078] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0079] The units described in some embodiments of the present disclosure may be implemented in software or in hardware. The units described may also be provided in a processor. For example, they may be described as: a processor comprising a preprocessing unit, an insertion unit, and an optimization unit. The names of these units do not, in some cases, constitute limitations on the units themselves. For example, the preprocessing unit may also be described as a "unit that preprocesses the initial data based on a preset first punctuality rate to obtain a preprocessing result."

[0080] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0081] Some embodiments of the present disclosure further provide a computer program product, including a computer program, which implements any of the above-mentioned multi-path-oriented vehicle path information generation methods when executed by a processor.

[0082] The above description is only an illustration of some preferred embodiments of the present disclosure and the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure.

Claims

1. A method for generating vehicle path information for multiple paths, comprising: Based on a preset first punctuality rate, preprocessing the initial data to obtain a preprocessing result, wherein the initial data includes initial vehicle configuration data and delivery node data, and the preprocessing result includes inserted vehicle path information and a list of uninserted delivery nodes; In response to the uninserted delivery node list being not empty, inserting each delivery node in the uninserted delivery node list into each virtual vehicle based on the uninserted delivery node list and virtual vehicle parameter data to obtain virtual vehicle data, and determining the inserted vehicle path information and the virtual vehicle data as initial vehicle path information, wherein each delivery node in the uninserted delivery node list corresponds one-to-one to each virtual vehicle corresponding to the virtual vehicle data; The determined initial vehicle path information is used as the current vehicle path information, and based on the preset second punctuality rate and the current vehicle path information, the following optimization steps are performed: Perform a single-round optimization process on the current vehicle path information to obtain a single-round optimization result; In response to the single-round optimization processing result not meeting the preset vehicle optimization condition or the single-round optimization processing reaching the preset resource limitation condition, the single-round optimization processing result is determined as the vehicle path information.

2. The method according to claim 1, wherein The method further comprises: In response to any uninserted delivery node in the uninserted delivery node list failing to be inserted into the virtual vehicle, outputting a prompt message indicating an abnormality, and interrupting the step of generating the initial vehicle path information.

3. The method according to claim 1, wherein The method further comprises: In response to the single-round optimization processing result reaching the preset vehicle optimization condition, the current vehicle path information is updated according to the single-round optimization processing result, and the optimization step is continued according to the updated current vehicle path information.

4. The method according to claim 1, wherein The method further comprises: In response to the uninserted delivery node list being empty, the inserted vehicle path information is determined as initial vehicle path information.

5. The method according to claim 1, wherein The single-round optimization processing is performed on the current vehicle path information to obtain the single-round optimization processing result, including: Performing a removal process in the single-round optimization process on the current vehicle path information to obtain a queue of nodes to be assigned and the vehicle path information after the removal, wherein the single-round optimization process includes the removal process, the insertion process, and the disturbance process; Performing the insertion process on the to-be-allocated node queue and the removed vehicle path information to obtain the inserted vehicle path information; The disturbance processing is performed on the inserted vehicle path information to obtain a single-round optimization processing result.

6. The method according to claim 4, wherein: The inserting process is performed on the to-be-assigned node queue and the removed vehicle path information to obtain the inserted vehicle path information, including: Performing preliminary insertion processing on the queue of nodes to be assigned and the removed vehicle path information to obtain a preliminary insertion result, wherein the preliminary insertion result includes an insertion failure list and preliminary inserted vehicle path information; In response to the insertion failure list in the preliminary insertion result being not empty, performing ejection insertion on the preliminary insertion result to obtain vehicle path information after insertion, wherein the vehicle path information after insertion includes the vehicle path information after ejection insertion and the insertion failure list; In response to the insertion failure list in the preliminary insertion result being empty, the preliminary insertion result is determined as the vehicle path information after insertion.

7. A multi-path oriented vehicle path information generation device, comprising: a preprocessing unit configured to preprocess the initial data based on a preset first punctuality rate to obtain a preprocessing result, wherein the initial data includes initial vehicle configuration data and delivery node data, and the preprocessing result includes inserted vehicle path information and a list of uninserted delivery nodes; an inserting unit configured to, in response to the uninserted delivery node list being not empty, insert each delivery node in the uninserted delivery node list into each virtual vehicle based on the uninserted delivery node list and virtual vehicle parameter data to obtain virtual vehicle data, and determine the inserted vehicle path information and the virtual vehicle data as initial vehicle path information, wherein each delivery node in the uninserted delivery node list corresponds one-to-one to each virtual vehicle corresponding to the virtual vehicle data; The optimization unit is configured to use the determined initial vehicle path information as the current vehicle path information, and based on a preset second punctuality rate and the current vehicle path information, perform the following optimization steps: use the determined initial vehicle path information as the current vehicle path information, and based on the preset second punctuality rate and the current vehicle path information, perform the following optimization steps: perform a single-round optimization process on the current vehicle path information to obtain a single-round optimization process result; in response to the single-round optimization process result not meeting a preset vehicle optimization condition or the single-round optimization process meeting a preset resource constraint condition, determine the single-round optimization process result as the vehicle path information.

8. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

9. A computer-readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images