A parking space pile body matching suspension type charging pile scheduling method and system

Abstract

A parking stall pole matching suspension type charging pile scheduling method and system, in the method, first, the target position and priority are obtained by analyzing the charging request, and the position and state information of the current mobile pile and idle pile are obtained. Through comparison, if the time consumed by a mobile pile to reach a new target position is shorter than that of an idle pile, and the priority of the new request is higher than the original priority, the mobile pile is set as a relay charging pile. The relay charging pile changes course to go to the new target position, and the idle pile relays to the original target position of the relay charging pile. Finally, the space-time conflict of the scheme is detected based on global data, and if there is no conflict, the route is directly executed, and if there is a conflict, the mobile pile is adjusted to avoid the conflict, and the mobile pile is controlled again after generating a corrected route. The application is used to reduce the time consumed by the charging pile to move to the position, and improve the overall operation efficiency of the charging equipment in the parking lot.

Description

Technical Field

[0001] This application belongs to the field of new energy vehicle charging technology, and in particular relates to a method and system for scheduling suspended charging piles that match parking spaces and charging pile bodies. Background Technology

[0002] With the increasing popularity of new energy vehicles, parking lots typically install charging stations to meet users' charging needs. However, traditional fixed charging stations occupy a large amount of space, and charging spaces are often permanently tied to regular parking spaces. During peak parking hours, users need to spend a considerable amount of time searching for a parking space with an available charging station, reducing the efficiency of parking and charging.

[0003] Chinese invention patent publication number CN117207812A, entitled "A Suspended Car Charging Pile," discloses a charging pile body that is suspended, with external, lateral, and vertical travel tracks providing a travel path for the charging pile body. This related technology allows the charging pile body to move along an upper support beam, adjusting the position of the charging pile housing according to the location of the vehicle's charging port, and adjusting the height via an electric telescopic rod, thus facilitating the user to extend the power supply cable and insert the charging gun into the car's charging port.

[0004] During peak hours in parking lots of large shopping malls or office buildings, when multiple new energy vehicles park in different spaces within a short period and successively initiate charging requests, multiple suspended charging piles need to move simultaneously on a shared overhead track network to reach their respective target vehicle locations. Because the overhead track network includes multiple intersections and shared sections, the trajectories of multiple charging piles moving along the tracks to their respective target parking spaces can easily converge on the same track sections at the same time. This physical overlap of trajectories can cause congestion and mutual interference among the charging piles on the tracks. Furthermore, many scheduling strategies adopt a first-come, first-served mechanism. When faced with emergency charging due to low battery levels or requests from high-priority users, it is difficult to dynamically adjust the allocated mobile charging piles, leading to slow response times for high-priority tasks. At the same time, forced reallocation can easily exacerbate the spatiotemporal conflicts within the track network, increasing the time required for charging piles to move and positioning, and reducing the overall operating efficiency of charging equipment in the parking lot. Summary of the Invention

[0005] This application provides a method and system for scheduling suspended charging piles that match parking spaces, which reduces the time spent moving and positioning charging piles and improves the overall operating efficiency of charging equipment in parking lots.

[0006] In the first aspect, this application provides a method for scheduling suspended charging piles that matches parking spaces and charging pile bodies, which parses the received charging request to obtain the target location of the target parking space and the target scheduling priority of the charging request.

[0007] Obtain the current position, original target position, original scheduling priority, and initial position of idle charging piles in the idle area of ​​each mobile charging pile in the top track network that is in motion;

[0008] Calculate the first estimated time for an idle charging pile to travel from its initial position to its target position, and the second estimated time for each mobile charging pile to travel from its current position to its target position.

[0009] If there is at least one target mobile charging pile with a corresponding second estimated time less than the first estimated time and a target scheduling priority higher than its original scheduling priority, the target mobile charging pile will be identified as a relay charging pile.

[0010] The destination of the relay charging pile is updated to the target location and a first driving route is generated; the destination of the idle charging pile is updated to the original target location and a second driving route is generated, thus obtaining the first pre-scheduling scheme.

[0011] Based on the global spatiotemporal occupancy data of the top orbital network, spatiotemporal conflict detection is performed on the first pre-scheduling scheme;

[0012] If the result of the spatiotemporal conflict detection is that there is no conflict, then control the relay charging pile and the idle charging pile to move according to the first driving route and the second driving route respectively.

[0013] If the spatiotemporal conflict detection result indicates the existence of a conflict, the first pre-scheduling scheme is adjusted to generate a corrected route, and the relay charging pile and idle charging pile are moved according to the corrected route.

[0014] By analyzing charging request priorities and calculating the estimated time for idle and mobile charging piles to reach the target location, when a mobile charging pile has a shorter arrival time and a higher priority request, it is designated as a relay charging pile. Simultaneously, an idle charging pile takes over to the original target location. This prioritizes the response speed of high-priority requests, reducing waiting time for urgent or high-priority users. Furthermore, by performing conflict detection on the pre-scheduling scheme based on global spatiotemporal occupancy data and coordinating avoidance adjustments to generate corrected routes when conflicts exist, the system can proactively identify and handle potential trajectory overlaps that may occur when multiple charging piles are operating simultaneously in the top track network. This reduces the risk of collisions and track congestion during charging pile movement, thereby improving the overall scheduling efficiency and operational safety of suspended charging piles in complex track networks.

[0015] In conjunction with some implementations of the first aspect, in some implementations, spatiotemporal conflict detection is performed on the first pre-scheduling scheme based on global spatiotemporal occupancy data of the top orbital network, specifically including:

[0016] Extract the first spatial node sequence and the corresponding first estimated time window of the first driving route, and the second spatial node sequence and the corresponding second estimated time window of the second driving route;

[0017] Extract the sequence of occupied space nodes and the corresponding occupied time window of the remaining operating charging piles, excluding relay charging piles and idle charging piles, from the global spatiotemporal occupancy data.

[0018] The spatial coordinates of the first spatial node sequence and the second spatial node sequence are compared with the occupied spatial node sequence to filter out the overlapping track segments with coordinate intersection.

[0019] Identify the relative travel direction of relay charging piles or idle charging piles with the corresponding other operating charging piles on the overlapping track section. The relative travel direction includes the same-direction overtaking direction and the opposite-direction meeting direction.

[0020] The dynamic safety time interval is configured according to the track sections where the relative travel directions coincide, and the dynamic safety time interval for the opposite meeting direction is greater than the dynamic safety time interval for the same overtaking direction.

[0021] If the time difference between the first or second estimated time window and the occupied time window in the overlapping orbit segment is less than the dynamic safety time interval, a spatiotemporal conflict detection result with conflict is generated.

[0022] When the time difference is not less than the dynamic safety time interval, a spatiotemporal conflict detection result without conflict is generated.

[0023] By extracting the spatial node sequences and predicted time windows of each driving route and comparing them with the occupied spatiotemporal data of other operating charging piles, overlapping track sections with intersections can be accurately identified. Furthermore, by identifying the relative driving directions of charging piles in overlapping sections and configuring differentiated dynamic safety time intervals for oncoming and same-direction overtaking, the conflict determination criteria are made more consistent with actual physical motion laws. This spatiotemporal comparison method, which dynamically adjusts the safety time interval based on specific driving directions, improves the accuracy of spatiotemporal conflict detection and reduces misjudgments or excessive avoidance caused by using a single fixed time threshold. While ensuring track operation safety, it also improves the space utilization of the top track network and the smoothness of concurrent scheduling of multiple charging piles.

[0024] In conjunction with some implementations of the first aspect, in some implementations, the first pre-scheduling scheme is adjusted for cooperative avoidance to generate a corrected route, specifically including:

[0025] Identify the conflicting track nodes and their corresponding conflicting charging stations;

[0026] Calculate the first time delay cost that causes the conflicting charging station to execute a hovering waiting strategy outside the conflicting track node; calculate the second time delay cost that causes the conflicting charging station to execute a spatial detour strategy around the conflicting track node.

[0027] If both the first time delay cost and the second time delay cost are greater than the preset efficiency tolerance threshold, the destination of the relay charging pile in the first pre-scheduling scheme will be updated to the original target location, and the destination of the idle charging pile will be updated to the target location.

[0028] The driving route is regenerated based on the updated destination as a corrected route;

[0029] If the first or second time delay cost is not greater than a preset efficiency tolerance threshold, a strategy with a smaller time delay cost is selected to modify the corresponding route in the first pre-scheduling scheme, thereby generating a corrected route.

[0030] By adopting the above technical solution, when making coordinated avoidance adjustments, the impact of different avoidance actions on the overall scheduling progress can be quantitatively assessed by calculating the time delay costs of the hovering and waiting strategies and spatial detour strategies implemented by conflicting charging piles. By comparing these two costs with a preset efficiency tolerance threshold, the relay scheduling plan is promptly cancelled and the original destination allocation is restored when both costs are too high. Conversely, when the costs are controllable, a strategy with lower time consumption is selected to generate a corrected route. This dynamic adjustment mechanism based on quantitative cost assessment and threshold circuit breaking reduces prolonged local track congestion or excessively long single-task times caused by forced avoidance, reduces scheduling time losses in complex conflict scenarios, and improves the rationality of path planning and overall operational efficiency of the system when facing high-density conflicts.

[0031] In some embodiments, in conjunction with some implementations of the first aspect, after making cooperative avoidance adjustments to the first pre-scheduling scheme to generate a corrected route, the method further includes:

[0032] Traverse all track connections in the top track network that link the target location to the original target location, and generate a set of candidate correction routes;

[0033] If the candidate corrected route set is empty, or if the estimated arrival time of all candidate routes in the candidate corrected route set is greater than the preset scheduling failure threshold, it is determined that a corrected route cannot be generated.

[0034] Cancel the destination update operation for both the relay charging pile and the idle charging pile, so that the relay charging pile can return to its original destination location;

[0035] Select a new backup charging station from the available area and bind the destination of the backup charging station to the target location;

[0036] A third driving route is generated for the backup charging pile, and the track access permission of the backup charging pile is upgraded to the preset first level;

[0037] Based on the preset first-level track access permission, a yield command is sent to the other operating charging piles on the third driving route to control the other operating charging piles to stop their current tasks and move to the track closest to the other operating charging piles;

[0038] Control the backup charging station to move along the third driving route to the target location.

[0039] By adopting the above technical solution, when the candidate correction route set is empty or the expected time is too long, the original relay operation can be canceled and a new backup charging pile can be selected, thus timely mitigation and activation of the emergency response mechanism. By enhancing the track access permission of the backup charging pile and sending yield instructions to other operating charging piles on its route, other piles can stop their current tasks and move to the nearest track to give way, thereby opening up a low-obstruction passage for the backup charging pile. This reduces the risk of high-priority charging requests being shelved for a long time in extreme congestion or deadlock conditions, and improves the scheduling robustness and task completion rate of the system under complex operating conditions.

[0040] In conjunction with some implementations of the first aspect, in some implementations, based on a preset first-level track access permission, a yield command is sent to the remaining charging stations on the third travel route to control the remaining charging stations to stop their current tasks and move to the track closest to them, specifically including:

[0041] Obtain the current location of all other operating charging stations along the third driving route;

[0042] In the top track network, identify multiple avoidance branch tracks that connect to the third driving route, and obtain the current number of available berths for each avoidance branch track;

[0043] For each of the remaining operating charging piles, the multiple avoidance branch tracks are sorted in order of proximity to the current location to obtain the corresponding candidate avoidance sequence;

[0044] Traverse the candidate avoidance sequence and determine the first avoidance branch track with a current available berth quantity greater than zero and not occupied by other charging piles as the track closest to the other operating charging piles;

[0045] Decrement the number of currently available berths on the nearest track by one and generate a give-way instruction containing the coordinates of the nearest track.

[0046] Send a yield command to the other operating charging stations to control them to stop their current task and move to the nearest track.

[0047] By adopting the above technical solution, when clearing passageways for high-priority backup charging piles, the current positions of other operating charging piles are obtained, and the number of connected bypass branch tracks and available berths is determined. These are then sorted and filtered in order of distance from nearest to farthest, accurately locating the nearest available branch track with parking conditions. By assigning the determined nearest track to the corresponding other operating charging pile and simultaneously deducting the number of available berths, targeted yielding instructions are generated to control its yielding to the nearest available track. This shortens the movement distance and time spent by other operating charging piles in finding yielding positions, reduces the interference of large-scale yielding actions on the overall operation of the top track network, and improves the scheduling response speed of high-priority backup charging piles and the overall traffic efficiency of the track network.

[0048] In some implementations of the first aspect, the physical connection status between the charging pile at the target location and the corresponding vehicle charging port is obtained.

[0049] If the duration of a period without detected physical connection reaches a preset resource idle threshold, the scheduling status of the charging pile will be updated from target locked to detached, while maintaining the validity of the original charging request order.

[0050] Receive emergency charging requests from nearby parking spaces and obtain the location of the nearby parking spaces;

[0051] If the track distance between the nearby location and the target location is less than the preset nearest scheduling threshold, the destination of the charging pile in the free state will be updated to the nearby location.

[0052] Control the charging pile that is in a free state to move to a nearby location in response to an emergency charging request;

[0053] Simultaneously select a replacement charging station from the idle area and bind the destination of the replacement charging station to the target location;

[0054] Generate a compensation driving route for the replacement charging station and control the replacement charging station to move along the compensation driving route to the target location.

[0055] By monitoring the physical connection status between charging piles and vehicle charging ports at the target location, the system updates the charging pile status to an detached state when there is no connection for an extended period, thus promptly releasing unused charging pile resources. Upon receiving an emergency charging request from a nearby parking space and provided the track distance meets the proximity requirement, the system directly dispatches the detached charging pile to the nearest location to respond, shortening the physical movement distance for emergency tasks. Simultaneously, a replacement charging pile is selected from the idle area and dispatched to the original target location to ensure the effective execution of the original order. This reduces the resource idle rate of charging piles due to users not plugging in their devices in time, improving the system's response speed to emergency charging requests and the overall utilization rate of charging equipment.

[0056] In conjunction with some implementation methods of the first aspect, in some implementation methods, a compensation driving route for the replacement charging station is generated, and the replacement charging station is controlled to move along the compensation driving route to the target location, specifically including:

[0057] Obtain the departure route of a charging pile in a free state to a nearby location, and extract the departure track segment in the departure route that is connected to the target location;

[0058] Based on the current and target locations of the replacement charging piles, multiple candidate compensation routes are generated in the top track network;

[0059] Extract the entry track segment connected to the target location from each candidate compensation route;

[0060] In the case of a target entry section that does not spatially intersect with the exit section, the candidate compensation route corresponding to the target entry section is determined as the compensation route.

[0061] Given that all entry track sections have spatial intersections with exit track sections, determine the candidate compensation route with the shortest expected travel time;

[0062] Insert yield-and-wait markers at nodes upstream of the spatial intersection in the candidate compensation routes to generate compensation driving routes;

[0063] Control the replacement charging station to move along the compensated driving route to the target location, and control the replacement charging station to perform a hovering action when the yield waiting sign is triggered.

[0064] By adopting the above technical solution, when controlling the replacement charging pile to move to the target location, the spatial intersection of the departure track segment of the detached charging pile and the entry track segment of each candidate compensation route is extracted and compared. The route without intersection is selected first as the compensation route, which can physically isolate the two running trajectories. When an intersection is unavoidable, the route with the shortest expected time is selected and a yielding waiting sign is inserted upstream of the intersection. This causes the replacement charging pile to perform a hovering action at the key node to wait for the detached charging pile to leave first. This refined route planning and node control based on trajectory intersection analysis reduces the risk of congestion or collision between the replacement charging pile and the detached charging pile in the local area around the target location, and improves the safety of the coordinated operation of multiple charging piles under complex intersection conditions.

[0065] Secondly, embodiments of this application provide a parking space-pile matching suspended charging pile scheduling system, which includes: one or more processors and a memory; the memory is coupled to one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and one or more processors call the computer instructions to cause the system to perform the method described in the first aspect and any possible implementation of the first aspect.

[0066] Thirdly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a system, cause the system to perform the method described in the first aspect and any possible implementation thereof.

[0067] Fourthly, embodiments of this application provide a computer program product that, when run on a system, causes the system to execute the method described in any possible implementation of the first aspect.

[0068] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0069] 1. This application provides a method for scheduling suspended charging piles by matching parking spaces with charging pile bodies. By analyzing the priority of charging requests and calculating the estimated time for idle charging piles and mobile charging piles to reach the target location, when a mobile charging pile has a shorter time and a higher priority for a new request, it is designated as a relay charging pile, while the idle charging pile takes over to the original target location. This prioritizes the response speed of high-priority requests and reduces the waiting time for urgent or high-priority users. Furthermore, based on global spatiotemporal occupancy data, conflict detection is performed on the pre-scheduling scheme, and collaborative avoidance adjustments are made to generate corrected routes when conflicts exist. This method can identify and handle potential trajectory overlaps that may occur when multiple charging piles are running simultaneously in the top track network, reducing the collision risk and track congestion probability during charging pile movement. This improves the overall scheduling efficiency and operational safety of suspended charging piles in complex track networks.

[0070] 2. This application provides a method for scheduling suspended charging piles that match parking spaces. By monitoring the physical connection status between the charging pile and the vehicle's charging port at the target location, the method updates the charging pile status to an detached state when there is no connection for an extended period, thus releasing ineffectively occupied charging pile resources in a timely manner. When an emergency charging request is received from a nearby parking space and the track distance meets the proximity requirement, the detached charging pile is directly dispatched to the nearby location to respond, shortening the physical movement distance for emergency tasks. Simultaneously, a replacement charging pile is selected from the idle area and dispatched to the original target location to ensure the effective execution of the original order. This reduces the resource idle rate of charging piles caused by users not plugging in their devices in time, and improves the system's response speed to emergency charging requests and the overall utilization rate of charging equipment. Attached Figure Description

[0071] Figure 1 This is a flowchart illustrating a method for scheduling suspended charging piles that matches parking spaces in an embodiment of this application.

[0072] Figure 2 This is another flowchart illustrating a method for scheduling suspended charging piles that matches parking spaces in an embodiment of this application.

[0073] Figure 3 This is a schematic diagram of the physical device structure of a suspended charging pile scheduling system that matches parking space piles, as provided in an embodiment of this application. Detailed Implementation

[0074] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0075] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0076] The following example is used in conjunction with Figure 1 The present application describes a method for scheduling suspended charging piles that matches parking spaces in its embodiments:

[0077] Please see Figure 1This is a flowchart illustrating a method for scheduling suspended charging piles that matches parking spaces in an embodiment of this application.

[0078] S101. Parse the received charging request to obtain the target location of the target parking space and the target scheduling priority of the charging request;

[0079] A charging request refers to a data packet sent by a user through a mobile terminal or in-vehicle system, indicating a need for power replenishment; a target parking space refers to the specific physical parking space where the vehicle that issued the charging request is currently parked or will soon be parked; a target location refers to the specific spatial coordinates of the target parking space in the three-dimensional coordinate system of the top track network; and target scheduling priority refers to the weight of the task processing order assigned by the system to the charging request based on factors such as the user's urgency, membership level, or remaining battery power.

[0080] This can be achieved by constructing a request processing module based on rule parsing and mapping. The system receives JSON-formatted charging request data packets sent by the front end and uses a built-in data parser to extract parking space identification codes and user account information from the data packets. The system accesses a pre-established parking space mapping database and converts the extracted parking space identification codes into target locations in a three-dimensional coordinate system through key-value pair matching. Simultaneously, the system extracts membership level parameters from the user account information and the current vehicle's remaining battery power parameters carried in the request, and calculates a comprehensive score using a preset multi-dimensional weighted scoring algorithm. The system inputs this comprehensive score into a priority mapping table for lookup operations, mapping the score to a specific target scheduling priority value, and finally outputs the target location and target scheduling priority to the system's scheduling memory.

[0081] This can also be achieved using natural language processing and machine learning classification models. The system constructs a large model specifically for intent recognition and parameter extraction. The underlying structure of this model employs a bidirectional long short-term memory network combined with a self-attention mechanism. The training set for this model consists of historical charging request text records and corresponding manually labeled location coordinates and priority levels. The test set consists of independently partitioned portions of historical request data. The system uses received charging requests containing text descriptions or unstructured data as input data to the model. Internally, the model converts the text into vectors through a word embedding layer and extracts semantic features through hidden layers, ultimately outputting structured target location coordinates and target scheduling priority levels.

[0082] S102. Obtain the current position, original target position, original scheduling priority, and initial position of idle charging piles in the idle area of ​​each mobile charging pile in the top track network that is in motion.

[0083] The top track network refers to the physical track system deployed above the parking lot for suspending and guiding the movement of charging piles; the movement state refers to the operating mode in which the charging pile is currently moving on the track to perform a charging task; the mobile charging pile refers to the suspended charging equipment in a moving state; the current position refers to the real-time spatial coordinates of the mobile charging pile at the current moment; the original destination position refers to the destination coordinates corresponding to the task currently being performed by the mobile charging pile; the original scheduling priority refers to the urgency weight of the task currently being performed by the mobile charging pile; the idle area refers to the area in the top track network specifically designated for parking equipment without assigned tasks; the idle charging pile refers to the charging equipment parked in the idle area waiting for task assignment; the initial position refers to the specific parking coordinates of the idle charging pile in the idle area.

[0084] This can be achieved by combining a distributed sensor network with real-time database queries. The system uses RFID readers and ultra-wideband positioning base stations deployed at various nodes of the top-level track network to collect reflected signals from the surface tags of each mobile charging pile and idle charging pile in real time at millisecond frequencies. After triangulation, the system obtains their current and initial positions. Simultaneously, the system accesses the central scheduling database via an internal high-speed local area network, retrieving the task record table for each mobile charging pile based on its unique device number. The system extracts the original target location coordinates and the recorded original scheduling priority value from this task record table, and then structurally concatenates this multi-source data, storing it in the state cache pool for the current scheduling cycle.

[0085] Alternatively, this can be achieved through IoT communication protocols and proactive reporting from edge computing nodes. Each charging pile integrates a high-precision servo motor encoder and an indoor positioning module. The main control chip of the charging pile sends telemetry messages containing its own status to the system's edge computing nodes at fixed time intervals via a message queue telemetry transmission protocol. The system subscribes to and parses these message topics through the edge computing nodes, directly reading the current location coordinates, task destination coordinates (i.e., original target location), and task priority (i.e., original scheduling priority) of the mobile charging pile from messages in mobile mode. Simultaneously, the system reads the initial location coordinates of idle charging piles from messages in standby mode, completing the collection and aggregation of all necessary status data.

[0086] S103. Calculate the first estimated time for an idle charging pile to reach the target location from its initial position, and the second estimated time for each mobile charging pile to reach the target location from its current position.

[0087] The first estimated time refers to the time, calculated theoretically by the system, required for an idle charging pile to travel from its initial position along the track network to its target position; the second estimated time refers to the time, calculated theoretically by the system, required for each mobile charging pile to travel from its current real-time position along the track network to its target position.

[0088] This can be achieved using a graph-based shortest path algorithm combined with a kinematic model. The system abstracts the top track network as a directed graph data structure containing nodes and edges, where the edge weights are set to the physical distance between adjacent nodes. The system uses Dijkstra's algorithm to plan the shortest distance path from the initial position to the target position for each idle charging pile, and the shortest distance path from the current position to the target position for each mobile charging pile. Subsequently, the system retrieves the physical parameters of the charging piles, including rated speed, maximum acceleration, and turning deceleration ratio. By performing piecewise kinematic integration on the shortest distance path, the theoretical time required to travel along that path is obtained, thus yielding the first and second estimated travel times.

[0089] Deep neural networks can also be used for time estimation. The system constructs a large-scale time prediction model based on a multilayer perceptron. Its underlying structure includes an input layer, multiple fully connected hidden layers, and an output layer. The hidden layers employ nonlinear activation functions to fit complex track operation time patterns. The training set for this model consists of the starting and ending coordinates of charging piles, the real-time congestion index of the track, and the corresponding actual travel time from historical scheduling logs. The test set is an independently partitioned subset of historical logs. The system inputs the initial and target positions of idle charging piles into the model, and the model outputs the first estimated time. The system inputs the current and target positions of each mobile charging pile into the model, and the model outputs the second estimated time. Both the input and output data completely correspond to the parameters required in this step.

[0090] S104. If there is at least one target mobile charging pile with a corresponding second estimated time less than the first estimated time and a target scheduling priority higher than its original scheduling priority, the target mobile charging pile shall be identified as a relay charging pile.

[0091] A target mobile charging station refers to a candidate device that meets specific time and priority conditions among all mobile charging devices; a relay charging station refers to a specific charging device that is ultimately selected by the system and requires interruption of the current task to respond to a new charging request.

[0092] This can be achieved by constructing a multi-condition logical judgment and sorting filtering module. The system traverses the data structure of all charging piles in the moving state in memory, extracting the second estimated time and original scheduling priority for each mobile charging pile. The system first performs a first-level numerical comparison to filter out a set of candidate devices whose second estimated time is strictly less than the first estimated time. Then, it performs a second-level logical comparison in the candidate device set to determine whether the target scheduling priority is greater than the original scheduling priority, eliminating devices that do not meet the conditions, and retaining the remaining ones as target mobile charging piles. If there are multiple target mobile charging piles, the system sorts them in ascending order according to the value of the second estimated time, selects the first one (i.e., the shortest time), and modifies its internal status flag to a relay charging pile.

[0093] Alternatively, a comprehensive evaluation algorithm based on a weighted matrix can be used. The system establishes a two-dimensional evaluation vector for each mobile charging station. The first element of the vector is the time difference (first estimated time minus second estimated time), and the second element is the priority difference (target scheduling priority minus original scheduling priority). The system sets hard filtering conditions: both the time difference and the priority difference must be greater than zero. Stations meeting these conditions are identified as target mobile charging stations. For these target mobile charging stations, the system uses a preset normalization function to map these two differences to the same order of magnitude (zero to one), and then uses preset weighting coefficients to perform a weighted summation to obtain a relay compatibility score. The system compares the scores of all target mobile charging stations and identifies the device with the highest score as the relay charging station.

[0094] S105, update the destination of the relay charging pile to the target location and generate the first driving route, update the destination of the idle charging pile to the original target location and generate the second driving route, and obtain the first pre-scheduling scheme.

[0095] The destination refers to the physical space coordinates that the system sends to the charging pile control module, indicating that the charging pile must eventually reach; the first travel route refers to the specific sequence of track nodes planned by the system for the relay charging pile, from its current location to the new target location; the second travel route refers to the specific sequence of track nodes planned by the system for the idle charging pile, from its initial location to the original target location; the first pre-scheduling scheme refers to a comprehensive scheduling plan initially generated by the system that includes collaborative movement paths for multiple devices.

[0096] Route planning can be implemented using a heuristic search algorithm. The system sends instructions to the underlying control module of the relay charging station, overwriting the destination coordinates in its internal registers with the target location. Simultaneously, using the A* pathfinding algorithm, it searches for a first route containing a series of continuous track nodes on a gridded track map, starting from the current location of the relay charging station and ending at the target location. Similarly, the system sets the destination coordinates of the idle charging station to the original target location and generates a second route using the same A* pathfinding algorithm. The system packages the datasets of these two routes and the corresponding charging station identification codes into a structured JSON file, which is then stored in the scheduling queue as the first pre-scheduling scheme.

[0097] Alternatively, a fast matching and generation technique based on a pre-defined path dictionary can be used. The system maintains a large static path lookup table in memory, which pre-calculates and records the optimal static path between any two key nodes in the top track network. The system extracts node pairs from the current location of the relay charging pile to the target location, and node pairs from the initial location of the idle charging pile to the original target location. The system directly retrieves the corresponding path sequences from the path lookup table through a hash mapping lookup operation, and after smoothing by Bézier curves, uses them as the first and second driving routes, respectively. The system integrates these two routes into a global scheduling object, and then aggregates them to generate the first pre-scheduling scheme.

[0098] S106. Based on the global spatiotemporal occupancy data of the top track network, perform spatiotemporal conflict detection on the first pre-scheduling scheme;

[0099] Based on the global spatiotemporal occupancy data of the top orbital network, spatiotemporal conflict detection is performed on the first pre-scheduling scheme, specifically including:

[0100] Extract the first spatial node sequence and its corresponding first estimated time window for the first driving route, and the second spatial node sequence and its corresponding second estimated time window for the second driving route; extract the occupied spatial node sequence and its corresponding occupied time window for the remaining operating charging piles (excluding relay charging piles and idle charging piles) from the global spatiotemporal occupancy data; compare the spatial coordinates of the first spatial node sequence and the second spatial node sequence with the occupied spatial node sequence to filter out overlapping track segments with coordinate intersections; identify the relative driving direction of the relay charging pile or idle charging pile on the overlapping track segment with the corresponding remaining operating charging piles, including the same-direction overtaking direction and the opposite-direction meeting direction; configure the corresponding dynamic safety time interval for the overlapping track segment according to the relative driving direction, with the dynamic safety time interval corresponding to the opposite-direction meeting direction being greater than the dynamic safety time interval corresponding to the same-direction overtaking direction; if the time difference between the first estimated time window or the second estimated time window and the occupied time window on the overlapping track segment is less than the dynamic safety time interval, generate a spatiotemporal conflict detection result with conflict; if the time difference is not less than the dynamic safety time interval, generate a spatiotemporal conflict detection result without conflict.

[0101] Global spatiotemporal occupancy data refers to the collection of spatial location information that all equipment in the track network will occupy in various time periods in the future; spatiotemporal conflict detection refers to the process of determining whether the planned trajectories of multiple devices overlap at the same time and in the same space; the first spatial node sequence and the second spatial node sequence are the sets of coordinate points after discretization of the first and second travel routes, respectively; the first and second expected time windows are the time intervals during which the equipment is expected to arrive at and leave these nodes; the remaining operating charging piles refer to other working equipment besides relay and idle equipment; the occupied spatial node sequence and the occupied time window are the spatiotemporal resources reserved by other equipment; overlapping track sections refer to the track sections where different routes intersect or share in physical space; relative travel direction refers to the motion vector relationship between two devices in the overlapping section; same-direction pursuit direction refers to two devices traveling in the same direction; opposite-direction encounter direction refers to two devices traveling towards each other; dynamic safety time interval refers to the minimum time difference set to prevent collisions, with larger intervals for opposite-direction travel due to higher risk; spatiotemporal conflict detection result refers to the final conclusion determining whether there is a risk of collision.

[0102] This can be achieved by constructing a computational engine based on multidimensional array traversal and vector dot product. The system parses the first and second driving routes into discrete sequences of first and second spatial nodes, and calculates the first and second estimated time windows for reaching each node using kinematic formulas. The system retrieves the occupied spatial node sequences and occupied time windows of other operating charging piles from the central cache. Overlapping track segments are identified through array intersection operations. The system performs dot product operations using vectors representing driving directions. If the dot product result is greater than zero, it is identified as a same-direction pursuit direction, and a smaller dynamic safety time interval is allocated; if the dot product result is less than zero, it is identified as an opposite-direction meeting direction, and a larger dynamic safety time interval is allocated. Finally, the system calculates the absolute time difference between the estimated time window and the occupied time window, compares the time difference with the dynamic safety time interval one by one, and outputs a spatiotemporal conflict detection result if the difference is less than the interval; otherwise, it outputs a spatiotemporal conflict detection result if there is no conflict.

[0103] Conflict prediction can also be achieved using a spatiotemporal graph convolutional neural network model. The system constructs a large model specifically for spatiotemporal conflict detection, whose underlying structure includes graph convolutional layers and one-dimensional convolutional layers in the time dimension. The training set consists of comprehensive feature data of multi-track node sequences, time windows, relative driving directions, and dynamic safety time intervals from historical scheduling data that show both conflicted and non-conflicted events. The test set consists of retained independent trajectory samples. The system input includes the extracted first spatial node sequence, first predicted time window, second spatial node sequence, second predicted time window, and the remaining occupied spatial node sequences and occupied time windows of operating charging piles. Internally, the model automatically extracts features from overlapping track segments and combines them with dynamic safety time interval weights assigned based on relative driving directions for high-dimensional feature comparison. The final output classification result is the spatiotemporal conflict detection result indicating whether a conflict exists or not.

[0104] S107. If the result of the spatiotemporal conflict detection is that there is no conflict, then control the relay charging pile and the idle charging pile to move according to the first driving route and the second driving route respectively.

[0105] "No conflict" means that the system, through rigorous mathematical calculations or model predictions, confirms that all equipment in the pre-scheduled plan will maintain a safe distance in both time and space during future operation; "movement" refers to the process by which the charging pile equipment, after receiving instructions, physically displaces itself on the top track network using its own power system.

[0106] This can be achieved through a low-level instruction-issuing mechanism based on a programmable logic controller (PLC). The system converts the first and second travel routes into low-level machine control code containing specific motor speeds, steering angles, and running times. Subsequently, the system sends these control codes via industrial Ethernet to the PLCs integrated within the relay charging piles and idle charging piles, respectively. After parsing the code, the controllers drive the servo motors to operate according to the set parameters, thereby controlling the relay charging piles and idle charging piles to move along their respective routes on the top track network.

[0107] Alternatively, a trajectory tracking control algorithm based on closed-loop feedback can be used. The system inputs the first and second driving routes as reference trajectories into the microprocessors of the relay charging pile and the idle charging pile. During movement, the charging pile uses its onboard laser rangefinder and inertial measurement unit to acquire its actual position and speed in real time. The microprocessor runs a proportional-integral-derivative control algorithm to continuously calculate the deviation between the actual position and the reference trajectory and dynamically adjusts the output torque of the drive wheels to ensure that the relay charging pile and the idle charging pile move precisely along the predetermined route on the track network.

[0108] S108. If the result of the spatiotemporal conflict detection is that a conflict exists, the first pre-scheduling scheme is adjusted to generate a corrected route, and the relay charging pile and the idle charging pile are controlled to move according to the corrected route.

[0109] If the spatiotemporal conflict detection result indicates the existence of a conflict, the first pre-scheduling scheme is adjusted to collaboratively avoid conflicts in order to generate a corrected route, specifically including:

[0110] The process involves identifying the conflicting track nodes and their corresponding charging piles; calculating the first time delay cost for the charging piles to hover and wait outside the conflicting track nodes; calculating the second time delay cost for the charging piles to bypass the conflicting track nodes and execute a spatial detour strategy; if both the first and second time delay costs are greater than a preset efficiency tolerance threshold, updating the destination of the relay charging piles in the first pre-scheduling plan to the original target location, and updating the destination of the idle charging piles to the target location; regenerating a driving route based on the updated destination as a corrected route; if neither the first nor the second time delay cost is greater than the preset efficiency tolerance threshold, selecting the strategy with the smaller time delay cost to modify the corresponding route in the first pre-scheduling plan, generating a corrected route. Finally, controlling the movement of the relay charging piles and idle charging piles according to the corrected route.

[0111] Collaborative avoidance adjustment refers to the process of modifying the original pre-scheduling plan to eliminate the risk after a collision risk is detected; corrected route refers to the final driving path that is adjusted to ensure safety; conflict track node refers to the specific physical location where spatiotemporal overlap is expected; conflict charging pile refers to the equipment that is expected to collide at this node; hovering waiting strategy refers to the scheme of stopping the equipment at a safe distance and waiting for the road to clear; first time delay cost refers to the additional time consumed by performing hovering waiting; spatial detour strategy refers to the scheme of allowing the equipment to bypass the conflict node and take other paths; second time delay cost refers to the additional time consumed by performing spatial detour; preset efficiency tolerance threshold is the maximum allowable delay time limit configured by the system, derived from the service quality standard set by the system administrator, and serves as a benchmark for measuring whether the cost is acceptable.

[0112] The system can analyze conflict detection results, accurately locate conflicting track nodes with overlapping time windows, and mark participating relay charging piles or idle charging piles as conflicting charging piles. The system calculates the time required for a conflicting charging pile to decelerate to zero and wait for its occupancy to end at its preceding position, as the first time delay cost. Simultaneously, it uses a pathfinding algorithm to avoid the conflicting track node and finds a suboptimal path, calculating the difference between the suboptimal path's time and the original time as the second time delay cost. The system reads a preset efficiency tolerance threshold from the configuration file and uses a logic comparator to determine if both costs are greater than the threshold. If so, a database rollback operation is performed, restoring the relay charging pile's destination to its original target position, pointing the idle charging pile to the target position, and re-invoking the pathfinding module to generate a corrected route. If a cost is not greater than the threshold, the system compares the two values, selects the action sequence corresponding to the strategy with the smaller value to overwrite the original route, generates a corrected route, and issues movement control commands.

[0113] This can also be achieved using deep reinforcement learning and multi-agent collaborative models. The system constructs a multi-agent deep deterministic policy gradient model, with an underlying structure including an Actor network and a Critic network. The training set consists of a large number of simulated scheduling scenarios with conflict nodes and corresponding delay penalty and reward data, while the test set contains retained complex conflict scenarios. The system inputs are conflicting track nodes, conflicting charging pile status, and a preset efficiency tolerance threshold. The model automatically evaluates the first time delay cost of the hovering waiting strategy and the second time delay cost of the spatial detour strategy. When the model evaluation finds that both costs are greater than the preset efficiency tolerance threshold, it outputs a destination exchange replanning instruction to generate a corrected route; when there is a cost that meets the threshold condition, the model directly outputs the adjusted trajectory node sequence corresponding to the strategy with the smaller cost as the corrected route. Finally, the system converts the corrected route into underlying drive signals to control the movement of relay charging piles and idle charging piles.

[0114] In the above embodiments, by parsing the charging request priority and calculating the estimated time for idle charging piles and mobile charging piles to reach the target location, when a mobile charging pile has a shorter time and a higher priority for the new request, it is designated as a relay charging pile, while the idle charging pile takes over to the original target location. This prioritizes the response speed of high-priority requests and reduces the waiting time for urgent or high-priority users. Simultaneously, by performing conflict detection on the pre-scheduling scheme based on global spatiotemporal occupancy data and coordinating avoidance adjustments to generate corrected routes when conflicts exist, the system can identify and handle potential trajectory overlaps that may occur when multiple charging piles are operating simultaneously in the top track network. This reduces the collision risk and track congestion probability of charging piles during movement, thereby improving the overall scheduling efficiency and operational safety of suspended charging piles in complex track networks.

[0115] Furthermore, in the above embodiment, after step S109 adjusts the first pre-scheduling scheme to generate a corrected route through cooperative avoidance, the system traverses all track connectivity paths in the top track network that connect the target location and the original target location, generating a set of candidate corrected routes. A track connectivity path refers to a combination of continuous track segments in the physical track topology that can connect the starting point and the ending point, such as a mainline track, auxiliary branch track, or a circular track. The set of candidate corrected routes refers to all possible alternative route combinations found by the system through a search algorithm.

[0116] This can be implemented using a depth-first search algorithm in graph theory combined with a topological matrix. The system abstracts the top track network as an adjacency matrix data structure, where nodes represent track intersections and edges represent track segments. The system uses the target location and the original target location as the search endpoints, and the current locations of the relay charging piles and idle charging piles as the starting points, running the depth-first search algorithm. During the search, the system uses a stack structure to record the currently visited node sequence and backtracks when encountering dead ends or visited nodes. When the search reaches the target node, the node sequence in the stack is extracted as a track connection path. The system repeats this process until all possible connected branches have been traversed, and all extracted path data is structured and encapsulated, stored in an in-memory list, thereby generating a set of candidate corrected routes.

[0117] Path generation can also be achieved using a large graph neural network model. The system constructs a large-scale path generation model based on a graph attention network, whose underlying structure includes multiple layers of graph attention layers and fully connected decoding layers. The training set of this model consists of historical track network topology snapshots and the corresponding set of all manually labeled feasible connected paths, while the test set consists of independently partitioned topology variant data. The system transforms the real-time topology graph of the current top track network into graph node feature vectors and edge weight matrices, which, along with the start and end coordinates, are input into the model. Internally, the model dynamically evaluates the connectivity importance of each track node through an attention mechanism and directly outputs multiple connected node sequences through the decoding layer. The system receives these node sequence data output by the model, uses them as track connected paths, and aggregates them to generate a set of candidate corrected routes.

[0118] If the candidate route set is empty, or if the estimated arrival time of all candidate routes in the candidate route set exceeds the preset scheduling failure threshold, the system determines that a corrected route cannot be generated. Estimated arrival time refers to the theoretical time required for the charging station to complete the entire journey, calculated by the system based on the physical length of each candidate route, speed limits, and current track congestion. The preset scheduling failure threshold is the maximum allowable delay time set by the system administrator based on the quality of service agreement or historical user tolerance data. The preset scheduling failure threshold serves as a benchmark for assessing the acceptableness of the estimated arrival time; only routes are considered valid if the estimated arrival time is less than or equal to this threshold.

[0119] The system cancels the destination update operation for both the relay charging pile and the idle charging pile, so that the relay charging pile can resume its original destination location.

[0120] The system selects a backup charging station from the idle area and binds the destination of the backup charging station to the target location. A backup charging station refers to a redundant charging device parked in the idle area that is not currently assigned any tasks and is in normal working order, such as a fully charged mobile energy storage station or an idle station that has just completed maintenance. Binding means establishing a strong association between the unique identifier of the charging station and the target spatial coordinates in the system's task scheduling mapping table.

[0121] This can be implemented using a first-in, first-out (FIFO) queue and a memory-mapped table. The system maintains a queue of idle devices in memory. All standby charging stations parked in the idle area and functioning normally are listed in this queue according to the time they entered the idle area. When a new selection is needed, the system performs a dequeue operation, popping a device number from the head of the queue as the selected standby charging station. Subsequently, the system creates a new task record in the central scheduling memory database, using the popped standby charging station number as the primary key and the target location's coordinates as the attribute value. The system updates the state machine flag of the standby charging station, changing it from an idle state to a task execution state, thus completing the binding operation between the destination and the target location.

[0122] This can also be achieved using a large-scale resource allocation model based on reinforcement learning. The system constructs a large-scale resource allocation model based on a deep Q-network, whose underlying structure includes a state input layer, a multilayer perceptron hidden layer, and an action value output layer. The training set of this model consists of device selection records and corresponding task completion efficiency reward values ​​under different idle zone distribution states in historical scheduling. The test set consists of virtual simulation idle zone state data. The system's input data includes the location coordinates, battery status, and target location coordinates of all backup charging piles within the current idle zone. Internally, the model evaluates the long-term benefits of selecting different backup charging piles and outputs a backup charging pile number with the highest action value. The system receives this output number, selects the corresponding backup charging pile, and completes the binding by calling the system API interface to write the target location coordinates into the task register of that charging pile.

[0123] The system generates a third travel route for the backup charging pile and elevates its track access permission to the preset first level. The third travel route refers to a sequence of physical track nodes specifically planned by the system for the backup charging pile, leading from its current location in the available space to the target location. Track access permission refers to the priority attribute assigned by the system to different operating devices for occupying right-of-way in the track network. The preset first level refers to the highest right-of-way level defined in the system's permission dictionary; this level of permission is strictly higher than the normal level of permission for regular tasks.

[0124] This can be achieved using a heuristic search algorithm and permission flag overriding. The system obtains the current coordinates of the backup charging station and the target location coordinates, and inputs them into the path planning engine equipped with the A* satellite pathfinding algorithm. The algorithm uses the physical distance of the track network as a heuristic function to search for the shortest sequence of nodes avoiding static obstacles in the gridded map, which serves as the third driving route. Next, the system accesses the attribute table of the backup charging station in the central control database and locates the track access permission field. The system executes an update statement, overriding the default value of this field to the highest value code representing the preset first level. The system packages the generated third driving route and the updated permission code into a control command and sends it to the underlying controller of the backup charging station.

[0125] Alternatively, a deep neural network path generation model combined with a permission configuration module can be used. The system constructs a large-scale path generation model based on a long short-term memory network, with an underlying structure including a sequence input layer, a recurrent hidden layer, and a fully connected output layer. The training set consists of historical best driving trajectory sequence data, and the test set consists of independently partitioned start-endpoint pairs. The system's input data consists of the start coordinates and target location coordinates of the backup charging pile. Internally, the model directly outputs a smooth sequence of third driving route nodes through sequence inference. After receiving this output data, the system calls the permission management microservice interface to send a permission escalation request to the authentication center. After verifying the validity of the request, the authentication center modifies the permission level field in the session token of the backup charging pile in the memory cache to the preset first level, and returns the updated token along with the third driving route to the scheduling main program.

[0126] Based on the preset first-level track access permission, a yield command is sent to the remaining charging piles on the third travel route to control them to stop their current tasks and move to the track closest to them. Specifically, this includes: obtaining the current position of each remaining charging pile on the third travel route; identifying multiple yielding branch tracks connected to the third travel route in the top track network and obtaining the current number of available parking spaces on each yielding branch track; and for each remaining charging pile, sorting the multiple yielding branch tracks in order of proximity to its current position. The system retrieves the corresponding candidate avoidance sequence; iterates through the candidate avoidance sequence, identifying the first avoidance branch track with a greater than zero number of currently available berths and not occupied by other charging piles as the track closest to the remaining operating charging piles; it decrements the number of currently available berths on the identified closest track by one and generates a yield instruction containing the coordinates of the closest track; it sends the yield instruction to the corresponding remaining operating charging piles to control them to stop their current tasks and move to the nearest track; the yield instruction is a mandatory control message issued by the system to low-privilege devices, requiring them to immediately interrupt their current actions and perform an avoidance operation. An avoidance branch track refers to an auxiliary track section set up beside the main track for temporary equipment parking to clear the main road, such as a maintenance turnout or a passing loop. The number of currently available berths refers to the number of currently vacant physical spaces on the avoidance branch track that can accommodate charging piles. The candidate avoidance sequence is a list formed by arranging all available avoidance tracks according to spatial distance.

[0127] This can be achieved through a logic control flow based on spatial geometry calculations and memory state updates. The system extracts the spatial coordinate range of the third driving route and retrieves the current positions of all other operating charging piles within this range from the real-time database using a spatial query statement. The system reads the track topology map, extracts all avoidance branch track nodes directly connected to the third driving route, and queries the sensor status table to obtain the current number of available berths for each branch track. For each other operating charging pile, the system calculates the straight-line distance between its current position and the entrance of each avoidance branch track using the Euclidean distance formula, and uses a quicksort algorithm to sort the distances in ascending order to generate a candidate avoidance sequence. The system iterates through this sequence using a loop structure, reads the number of berths and the occupancy flag, and finds the first track that meets the conditions as the nearest track. The system decrements the number of berths for this track in memory, concatenates the track coordinates to generate a yield instruction message, and sends it to the corresponding other operating charging pile via the transmission control protocol, triggering its interrupt routine and causing it to enter the track.

[0128] Alternatively, a large-scale multi-agent collaborative allocation model can be used. The system constructs a large-scale allocation model based on a combination of graph convolutional networks and reinforcement learning. The underlying structure includes a graph feature extraction layer and a multi-agent action output layer. The training set consists of a large amount of track network topology data with congestion scenarios, equipment location data, and corresponding optimal avoidance allocation schemes. The test set consists of randomly generated congestion scenarios. The system's input data includes the current position of each other operating charging pile on the third driving route, the coordinates of multiple connected avoidance branch tracks, and the number of currently available parking spaces on each branch track. Internally, the model constructs this data into a spatiotemporal graph structure, extracts distance and capacity features through graph convolution, and directly outputs a matching avoidance branch track as the nearest track for each other operating charging pile. The system receives the matching results output by the model, automatically deducts the number of currently available parking spaces for the corresponding track from the database, and calls the instruction encapsulation module to generate a yielding instruction containing the coordinates of that track. Finally, the system broadcasts the instruction to the corresponding other operating charging piles through the wireless radio frequency communication module, forcing them to stop their original tasks and perform avoidance movement.

[0129] Control the backup charging station to move along the third driving route to the target location.

[0130] In the above embodiments, when the candidate correction route set is empty or the expected time is too long, the original relay operation can be canceled and a new backup charging pile can be selected to promptly stop losses and activate the emergency response mechanism. By increasing the track access permission of the backup charging pile and sending a yield instruction to other operating charging piles on its route, other piles can stop their current tasks and move to the nearest track to give way, thereby opening up a low-obstruction passage for the backup charging pile. This reduces the risk of high-priority charging requests being shelved for a long time in extreme congestion or deadlock conditions, and improves the scheduling robustness and task completion rate of the system under complex operating conditions.

[0131] The above embodiments detail the dynamic relay, conflict handling, and forced avoidance mechanisms under extreme conditions during the charging pile's journey to the target location, ensuring that the charging pile can reach its destination efficiently and safely. However, in practical applications, after the charging pile successfully arrives at the target parking space, users often fail to connect the charging gun for various reasons. This not only results in the ineffective occupation of allocated resources but also affects the charging experience of other users in the vicinity. The following will combine... Figure 2 Another method for scheduling suspended charging piles with matching parking space charging piles in the embodiments of this application is described:

[0132] Please see Figure 2 This is another flowchart illustrating a method for scheduling suspended charging piles that matches parking spaces in an embodiment of this application.

[0133] S201. Obtain the physical connection status between the charging pile at the target location and the corresponding vehicle charging port;

[0134] The physical connection status refers to the objective fact that a mechanical lock and an electrical communication handshake have been established between the charging gun head of the suspended charging pile and the charging interface at the vehicle end; in the context of system scheduling, the target location and the charging pile respectively represent the endpoint coordinates of the task and the physical entity executing the task.

[0135] This can be achieved based on underlying hardware sensors and IoT communication protocols. The system integrates a high-precision microswitch and proximity sensor inside the charging gun of the charging pile. When the user inserts the charging gun into the vehicle's charging port, the mechanical pressure triggers the microswitch to close, generating a high-level electrical signal. The microcontroller unit inside the charging pile polls the voltage level of this pin in real time. Once a high level is detected, the physical connection status is determined to be true. Subsequently, the microcontroller unit encapsulates this status data into a lightweight message queue telemetry transmission protocol data packet and sends it in real time to the backend server of the cloud dispatch system via the built-in wireless communication module and mobile cellular network. The system backend receives and parses the data packet, updating the corresponding connection status field of the charging pile in the memory database.

[0136] Alternatively, this can be achieved through message parsing of the charging communication handshake protocol. The system utilizes the controller area network bus communication module on the charging pile's main control board. When the charging gun is inserted into the vehicle, the charging pile's main control board automatically sends a handshake request message to the vehicle's battery management system. If the vehicle responds normally, both parties will begin exchanging parameter messages such as the maximum allowable charging voltage and the current battery state of charge. The system monitors the controller area network bus data stream inside the charging pile in real time through an edge computing gateway. Once it intercepts the parameter configuration message successfully exchanged between the two parties, it confirms that the physical connection has been established at the software protocol level. The edge computing gateway then converts this confirmation signal into a standardized application layer interface call and reports it to the central dispatch system to complete the status acquisition.

[0137] S202. If the duration of the period during which no physical connection is detected reaches a preset resource idle threshold, the scheduling status of the charging pile is updated from the target locked status to the free status, while maintaining the validity of the original charging request order.

[0138] The preset resource idle threshold refers to the maximum time that the system allows a charging pile to remain in an unconnected state after arriving at its destination. This value is derived from the statistical upper limit of the average user plug-in preparation time in historical operational data. The scheduling status refers to the task affiliation attribute of the charging pile at the system logic level. The target locked status means that the charging pile belongs exclusively to the current parking space order and cannot be called by other tasks. The detached status means that the charging pile is physically stationary but has logically released its exclusive rights and can be reassigned at any time. The order validity means that the original user's charging request is still retained in the system's pending queue.

[0139] This can be achieved by constructing a timer and state machine module based on a time wheel algorithm. After confirming that the charging pile has arrived at the target location, the system registers a countdown task for the charging pile in the time wheel data structure in memory. The total countdown duration is set to a preset resource idle threshold. During the countdown, the system continuously monitors the physical connection status obtained in step S201. If the status changes to connected, the timed task is deregistered; if the countdown reaches zero and the status is still disconnected, the time wheel triggers a timeout callback function. This callback function performs a database update operation, changing the scheduling status field in the charging pile status table from target locked to detached. Simultaneously, the callback function performs a skip operation on the original charging request's order record in the order management table, without modifying its pending service status, thereby maintaining the validity of the original charging request order.

[0140] This can also be achieved using a streaming data processing engine for event window analysis. The system deploys a streaming computing framework, using the charging pile's "arrival event" and "connection event" as real-time data stream inputs. The system opens a sliding time window for each charging pile, with the window length strictly equal to a preset resource idle threshold. The window starts timing when an "arrival event" enters the data stream. If no corresponding "connection event" is matched in the data stream before the sliding window ends, the streaming computing framework outputs a "timeout" derivative event. The system's business logic layer subscribes to this derivative event. Upon receiving the event, it calls the application programming interface of the state management microservice to overwrite the charging pile's scheduling state in memory as detached. Simultaneously, the business logic layer does not send any cancellation or suspension commands to the order database, naturally maintaining the validity of the original charging request order.

[0141] S203, Receive an emergency charging request from a nearby parking space and obtain the location of the nearby parking space;

[0142] "Adjacent parking space" refers to other parking spaces that are physically close to the target location and located in the same power supply zone or on the same track branch; "emergency charging request" refers to a charging task with a high priority identifier, such as a request for a car to anchor due to extremely low battery or a request issued by a high-level member; "adjacent location" refers to the specific spatial coordinates of the adjacent parking space in the three-dimensional coordinate system of the top track network.

[0143] This can be implemented based on an interactive architecture between the front-end application layer and the back-end spatial database. The system receives Hypertext Transfer Protocol (HTTP) request messages sent by users via mobile clients through an application server deployed in the cloud. The system parses the message header and body, extracting the business field containing the "emergency" identifier and the user's current nearest parking space number, thus receiving an emergency charging request. Subsequently, the system connects to the back-end spatial relational database, executes Structured Query Language (SCL) commands, and performs an exact match query using the nearest parking space number as the index key. The database returns the 3D coordinate data bound to that parking space, which the system reads and instantiates into a nearest location variable in memory for direct use by the subsequent scheduling algorithm module.

[0144] Intent recognition and coordinate mapping can also be achieved using a large-scale natural language processing (NLP) model. The system constructs a large-scale NLP model based on the Transformer architecture, whose underlying structure includes a multi-head attention mechanism layer and a feedforward neural network layer. The training set of this model consists of historical user-sentence transcription data containing parking space descriptions and urgent emotions, along with corresponding manually labeled emergency status tags and precise coordinates. The test set consists of retained independent request texts. The system inputs the received unstructured text or speech-transcribed text from users into the model. Internally, the model extracts semantic features and understands the context, directly outputting a structured parsing result. This output includes a Boolean value identifier for the emergency charging request and the corresponding coordinates of nearby locations. The system directly receives this output data to complete the request reception and location acquisition.

[0145] S204. If the track distance between the nearby location and the target location is less than the preset nearest scheduling threshold, the destination of the charging pile in the free state will be updated to the nearby location.

[0146] Track distance refers to the total path length that a charging pile needs to travel along the actual physical route of the top track network from one node to another; the preset nearest scheduling threshold is the maximum length standard set by the system to judge whether the distance is close enough, which is derived from a comprehensive calculation of the track network topology density and the tolerable waiting time for emergency tasks.

[0147] This can be achieved through graph theory-based shortest path search and memory object overwriting. The system abstracts the top track network as a weighted undirected graph, where nodes are track intersections and edge weights are the actual physical lengths. The system extracts the current target location coordinates of the detached charging pile and the coordinates of newly acquired neighboring locations. The system calls Dijkstra's algorithm to calculate the shortest path from the target location to the neighboring location in the weighted undirected graph and accumulates the weights of all edges on this path to obtain the precise track distance. The system compares this track distance with a preset nearest-neighbor scheduling threshold read from the global configuration file. If the track distance is less than the preset nearest-neighbor scheduling threshold, the system locates the control object of the detached charging pile in memory and directly overwrites its internal destination attribute field with the coordinate data of the neighboring location.

[0148] This can also be achieved using a pre-computed distance matrix cache and fast lookup matching. During the initialization phase, the system performs offline calculations on the track distances between all pairs of parking space nodes in the top track network and stores the results in a high-speed in-memory database based on key-value pairs, forming a global distance matrix. When this step is executed, the system concatenates the node identifier of the target location with the node identifiers of neighboring locations to form a query key, and directly reads the corresponding track distance value from the high-speed in-memory database with constant time complexity. The system uses a logic checker to compare this value with a preset nearest-neighbor scheduling threshold. If the threshold is less than the threshold, the system publishes a destination change event through the internal message bus. Upon receiving the change instruction, the device management microservice that subscribed to this event updates the destination field of the charging pile to the nearest location at the database level.

[0149] S205. Control the charging pile in a free state to move to a nearby location in response to an emergency charging request;

[0150] This can be achieved based on the generation and distribution mechanism of underlying motion control commands. Based on the updated proximity locations and the topology of the top track network, the system plans a specific driving trajectory from the current target location to a nearby location. The system decomposes this trajectory into a series of underlying motion control commands containing target speed, acceleration, and steering angle. These command sequences are packaged and sent via industrial Ethernet to the motion controller integrated inside the charging pile, which is in a detached state. After parsing the commands, the motion controller outputs pulse signals to drive the servo motor of the charging pile, controlling the charging pile to move smoothly along the track to a nearby location, thus physically responding to emergency charging requests.

[0151] Alternatively, autonomous navigation control based on target coordinates can be used. Instead of performing detailed trajectory decomposition, the system directly encapsulates the three-dimensional coordinates of nearby locations into a target guidance data packet, which is then transmitted to the detached charging pile via a wireless radio frequency network. Upon receiving these coordinates, the charging pile's internal control computer activates its built-in autonomous navigation algorithm. The charging pile utilizes its onboard LiDAR and track marker sensors to perceive its surroundings and its own position in real time, autonomously calculating the optimal velocity vector to reach the nearest location and dynamically adjusting the output torque of its drive wheels. Relying on its intelligent sensing and control capabilities, the charging pile autonomously moves to the nearest location, responding to emergency charging requests.

[0152] S206. Simultaneously select a replacement charging station from the idle area and bind the destination of the replacement charging station to the target location;

[0153] When previously allocated equipment is requisitioned for an emergency mission, the system must immediately activate a compensation mechanism to fill the resource gap at the target location. Parallel processing can minimize the original user's waiting time, ensuring that when the original user is ready to plug in their weapon, the new equipment is already in place or on its way.

[0154] S207. Generate a compensation driving route for the replacement charging pile and control the replacement charging pile to move along the compensation driving route to the target location.

[0155] The system generates a compensation driving route for the replacement charging station and controls the replacement charging station to move along the compensation driving route to the target location, specifically including:

[0156] The system acquires the departure route of a detached charging pile to a nearby location and extracts the departure track segment connected to the target location within the departure route. Based on the current and target locations of the replacement charging pile, multiple candidate compensation routes are generated in the top track network. The entry track segment connected to the target location is extracted from each candidate compensation route. If a target entry track segment does not spatially intersect with the departure track segment, the candidate compensation route corresponding to the target entry track segment is determined as the compensation route. If all entry track segments spatially intersect with the departure track segments, the candidate compensation route with the shortest estimated travel time is determined. A yielding waiting marker is inserted at the node upstream of the spatial intersection in the candidate compensation route to generate the compensation route. The system controls the replacement charging pile to move along the compensation route to the target location and controls the replacement charging pile to perform a hovering action when the yielding waiting marker is triggered.

[0157] The compensation route refers to the entirely new path planned for the replacement charging pile to reach the target location; the departure route is the path taken by the original free-floating charging pile to leave the target location; the departure track section and the entry track section are the necessary end track sections when leaving and entering the target location, respectively; the candidate compensation route consists of multiple alternative paths; spatial intersection refers to the overlapping area of ​​two routes in physical space; the target entry track section is the entry path that does not overlap with the departure path; the yielding and waiting indicator is the logical command for the control equipment to pause; the hovering action is the state in which the equipment stops moving on the track.

[0158] This can be implemented using computational geometry and spatiotemporal topology analysis algorithms. The system calls the path planning module to obtain the departure route of the detached charging pile, extracting the first track segment directly connected to the target location as the departure track segment. The system uses the K-shortest path algorithm to generate multiple candidate compensation routes from the replacement charging pile to the target location, and extracts the last connected track segment of each as the entry track segment. The system uses the line segment intersection algorithm in computational geometry to determine whether each entry track segment and departure track segment have overlapping coordinates. If a target entry track segment with no intersection is found, the corresponding candidate route is directly designated as the compensation driving route. If all entry segments have intersections, the system calculates the theoretical travel time of each candidate route and selects the route with the shortest travel time. The system inserts a yield waiting flag code into the data structure of the physical node preceding the intersection region in this route, generating the final compensation driving route. The system sends the route to the underlying controller of the replacement charging station. The controller drives the motor to move. When it reads the yielding waiting flag code, the controller actively cuts off the motor drive and applies the brakes to perform a hovering action. It waits for the intersection area to be cleared before continuing to the target location.

[0159] Alternatively, a large-scale path generation model combining graph neural networks and reinforcement learning can be used. The system constructs a large model based on a combination of spatiotemporal graph convolutional networks and deep Q-networks. The underlying structure includes graph convolutional layers for extracting track topology features and fully connected layers for outputting action sequences. The training set consists of a large amount of track network data, equipment location data, and corresponding manually labeled safety avoidance routes and hovering nodes, including scenarios of entering and exiting collisions. The test set consists of randomly generated complex intersection scenarios. The system's input data includes the departure route of the detached charging pile, the departure track segment, the current location of the replacement charging pile, the target location, and multiple generated candidate compensation routes and corresponding entry track segments. Internally, the model automatically performs high-dimensional feature extraction and collision risk assessment on spatial intersections. If there are no intersection entry segments, the model directly outputs the candidate route corresponding to that segment; if all intersections exist, the model comprehensively evaluates the time consumption and risk, outputs a candidate route with the shortest expected time, and automatically adds yield waiting markers to the upstream nodes of the intersection in the output node sequence, thereby generating a compensation driving route. The system receives the output data, converts it into a low-level control message, and sends it to the replacement charging station. The replacement charging station moves according to the message, and when it parses the yielding waiting flag, it triggers an internal interrupt service routine, controls the servo motor to lock, and performs a hovering action.

[0160] In the above embodiments, by monitoring the physical connection status between the charging pile and the vehicle charging port at the target location, the charging pile status is updated to an detached state when there is no connection for a long time, which can release the ineffectively occupied charging pile resources in a timely manner. When an emergency charging request is received from a nearby parking space and the track distance meets the proximity condition, the detached charging pile is directly dispatched to the nearby location to respond, shortening the physical movement distance of the emergency task. At the same time, a replacement charging pile is selected from the idle area and dispatched to the original target location to ensure the effective execution of the original order, reducing the resource idle rate of charging piles caused by users not plugging in in time, and improving the system's response speed to emergency charging requests and the overall utilization rate of charging equipment.

[0161] The system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 3 This is a schematic diagram of the physical device structure of a suspended charging pile scheduling system that matches parking space piles, provided in an embodiment of this application.

[0162] It should be noted that, Figure 3 The structure of the system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0163] like Figure 3As shown, the system includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes based on a program stored in Read-Only Memory (ROM) 302 or a program loaded from storage portion 308 into Random Access Memory (RAM) 303, such as executing the methods described in the above embodiments. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0164] The following components are connected to I / O interface 305: input section 306 including a camera, infrared sensor, etc.; output section 307 including a liquid crystal display (LCD) and speakers, etc.; storage section 308 including a hard disk, etc.; and communication section 309 including a network interface card such as a LAN (Local Area Network) card and a modem, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0165] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit (CPU) 301, it performs the various functions defined in the present invention.

[0166] It should be noted that the computer-readable medium shown in the embodiments of the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. The transmitted data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.

[0167] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0168] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the system described in the above embodiments; or it may exist independently and not assembled into the system. The storage medium carries one or more computer programs that, when executed by a processor of a system, cause the system to implement the methods provided in the above embodiments.

[0169] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0170] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0171] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0172] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for scheduling a suspension type charging pile matched with a parking pile body, characterized in that, include: Parse the received charging request to obtain the target location of the target parking space and the target scheduling priority of the charging request; Obtain the current position, original target position, original scheduling priority, and initial position of idle charging piles in the idle area of ​​each mobile charging pile in the top track network that is in motion; Calculate the first estimated time for the idle charging pile to travel from the initial position to the target position, and the second estimated time for each mobile charging pile to travel from the current position to the target position; If there is at least one corresponding target mobile charging pile whose second estimated time is less than the first estimated time and whose target scheduling priority is higher than its original scheduling priority, the target mobile charging pile is determined as a relay charging pile. The destination of the relay charging pile is updated to the target location and a first driving route is generated; the destination of the idle charging pile is updated to the original target location and a second driving route is generated, thus obtaining a first pre-scheduling scheme. Based on the global spatiotemporal occupancy data of the top track network, spatiotemporal conflict detection is performed on the first pre-scheduling scheme; If the result of the spatiotemporal conflict detection is that there is no conflict, then the relay charging pile and the idle charging pile are controlled to move according to the first driving route and the second driving route respectively; If the result of the spatiotemporal conflict detection is that a conflict exists, the first pre-scheduling scheme is adjusted to avoid conflict to generate a corrected route, and the relay charging pile and the idle charging pile are moved according to the corrected route.

2. The method according to claim 1, characterized in that, The spatiotemporal conflict detection of the first pre-scheduling scheme based on the global spatiotemporal occupancy data of the top orbital network specifically includes: Extract the first spatial node sequence of the first driving route and its corresponding first estimated time window, and the second spatial node sequence of the second driving route and its corresponding second estimated time window; Extract the sequence of occupied space nodes and the corresponding occupied time window of the remaining operating charging piles, excluding the relay charging piles and the idle charging piles, from the global spatiotemporal occupancy data; The first spatial node sequence and the second spatial node sequence are compared with the occupied spatial node sequence to filter out overlapping track segments with coordinate intersections. Identify the relative travel direction of the relay charging pile or the idle charging pile with the corresponding other operating charging piles on the overlapping track section, the relative travel direction including the same-direction chasing direction and the opposite-direction meeting direction; According to the relative travel direction, a corresponding dynamic safety time interval is configured for the overlapping track section, and the dynamic safety time interval corresponding to the opposite meeting direction is greater than the dynamic safety time interval corresponding to the same chasing direction. If the time difference between the first or second estimated time window and the occupied time window in the overlapping orbit segment is less than the dynamic safety time interval, a spatiotemporal conflict detection result with conflict is generated. If the time difference is not less than the dynamic safety time interval, a spatiotemporal conflict detection result without conflict is generated.

3. The method according to claim 1, characterized in that, The step of collaboratively avoiding adjustments to the first pre-scheduling scheme to generate a corrected route specifically includes: Identify the conflicting track nodes and their corresponding conflicting charging stations; Calculate the first time delay cost that causes the conflicting charging pile to execute a hovering waiting strategy outside the conflicting track node; calculate the second time delay cost that causes the conflicting charging pile to execute a spatial detour strategy around the conflicting track node; If both the first time delay cost and the second time delay cost are greater than the preset efficiency tolerance threshold, the destination of the relay charging pile in the first pre-scheduling scheme is updated to the original target location, and the destination of the idle charging pile is updated to the target location. The driving route is regenerated based on the updated destination as the corrected route; If the first time delay cost or the second time delay cost is not greater than the preset efficiency tolerance threshold, a strategy with a smaller time delay cost is selected to modify the corresponding route in the first pre-scheduling scheme, thereby generating the corrected route.

4. The method according to claim 1, characterized in that, After performing cooperative avoidance adjustments on the first pre-scheduling scheme to generate a corrected route, the method further includes: Traverse all track connections in the top track network that link the target location and the original target location to generate a set of candidate correction routes; If the candidate corrected route set is empty, or if the estimated arrival time of all candidate routes in the candidate corrected route set is greater than the preset scheduling failure threshold, it is determined that the corrected route cannot be generated. Cancel the destination update operation for the relay charging pile and the idle charging pile, so that the relay charging pile can resume going to the original target location; Select a new backup charging station from the available area and bind the destination of the backup charging station to the target location; A third driving route is generated for the backup charging pile, and the track access permission of the backup charging pile is upgraded to a preset first level; Based on the preset first level of track access permission, a yield instruction is sent to the other operating charging piles on the third driving route to control the other operating charging piles to stop their current tasks and move to the track closest to the other operating charging piles; Control the backup charging pile to move along the third driving route to the target location.

5. The method according to claim 4, characterized in that, Based on the preset first-level track access permission, sending a yield command to the remaining charging piles on the third travel route to control the remaining charging piles to stop their current tasks and move to the track closest to them specifically includes: Obtain the current location of each of the remaining operating charging piles on the third driving route; In the top track network, multiple avoidance branch tracks connected to the third driving route are identified, and the current number of available berths for each avoidance branch track is obtained. For each of the remaining operating charging piles, the multiple avoidance branch tracks are sorted in order of distance from the current location to the farthest point to obtain the corresponding candidate avoidance sequence; Traverse the candidate avoidance sequence and determine the first avoidance branch track with a number of currently available berths greater than zero and not occupied by other charging piles as the track closest to the remaining operating charging piles; Subtract one from the number of currently available berths for the nearest track and generate the give way instruction containing the coordinates of the nearest track; The yield command is sent to the other operating charging stations to control them to stop their current task and move to the nearest track.

6. The method according to claim 1, characterized in that, The method further includes: Obtain the physical connection status between the charging pile at the target location and the corresponding vehicle charging port; If the duration of the period during which the physical connection status is not detected reaches a preset resource idle threshold, the scheduling status of the charging pile will be updated from the target locked status to the free status, while maintaining the validity of the original charging request order. Receive an emergency charging request from a nearby parking space and obtain the location of the nearby parking space; If the track distance between the nearby location and the target location is less than a preset nearest-neighbor scheduling threshold, the destination of the charging pile in the free state will be updated to the nearby location. Control the charging pile in the free state to move to the nearby location in response to the emergency charging request; Simultaneously, a replacement charging station is selected from the idle area, and the destination of the replacement charging station is bound to the target location; Generate a compensation driving route for the replacement charging station, and control the replacement charging station to move along the compensation driving route to the target location.

7. The method according to claim 6, characterized in that, The process of generating a compensated driving route for the replacement charging station and controlling the replacement charging station to move along the compensated driving route to the target location specifically includes: Obtain the departure route of the charging pile in the free state to the nearby location, and extract the departure track segment in the departure route that is connected to the target location; Based on the current location of the replacement charging station and the target location, multiple candidate compensation routes are generated in the top track network; Extract the entry track segment connected to the target position from each of the candidate compensation routes; In the case where there is a target entry track section that does not spatially intersect with the departure track section, the candidate compensation route corresponding to the target entry track section is determined as the compensation driving route; If all the entry track sections have spatial intersection with the exit track sections, determine the candidate compensation route with the shortest expected time. Insert a yield-waiting marker at the node upstream of the spatial intersection in the candidate compensation route to generate the compensation driving route; Control the replacement charging pile to move along the compensated driving route to the target location, and control the replacement charging pile to perform a hovering action when the yielding waiting sign is triggered.

8. A suspended charging pile scheduling system for matching parking space charging piles, characterized in that, The system includes: One or more processors and a memory; the memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the system to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are executed on the system, the system performs the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is run on the system, the system performs the method as described in any one of claims 1-7.

Images