A parking lot virtual green wave guidance method and related device

Abstract

The application discloses a parking lot virtual green wave guiding method and related equipment, the core is to construct a space-time resource map, according to which the path is planned and the time window is reserved to form a virtual green wave band, and instructions are issued to vehicle terminals and intelligent gates, while the vehicle position deviation is monitored, and the local re-planning is triggered when the threshold is exceeded. The scheme accurately solves the defects of the prior art: through the space-time resource map, the global space-time resource is overall planned, the exclusive time window is reserved for the vehicle, the multi-vehicle path conflict is eliminated from the root, and the short board of lacking global vision and active scheduling is made up; the virtual green wave band in space and time is used to replace the static road network navigation, the vehicle is actively guided to pass in order, the micro traffic congestion is effectively solved, and the in-parking lot passing efficiency is greatly improved; the real-time monitoring and local re-planning mechanism can quickly respond to sudden conditions such as vehicle failure, avoid congestion diffusion, guarantee the stability of traffic flow, and finally realize the seamless and non-stop passing of the vehicle from the entrance to the parking space.

Description

Technical Field

[0001] This application relates to the field of intelligent parking guidance technology, and more specifically, to a virtual green wave guidance method and related equipment for parking lots. Background Technology

[0002] With the continuous growth of car ownership and the expanding scale of urban parking lots, especially underground parking lots with their complex network of passageways and dense intersections, the efficiency of vehicle passage after entry has become a pressing issue. Without effective guidance, problems such as congestion at intersections, difficulty in passing other vehicles, and prolonged searching for parking spaces can easily occur, impacting user experience and reducing the overall operational efficiency of parking lots. Therefore, developing efficient internal parking guidance technologies is of significant necessity.

[0003] Currently, existing parking guidance technologies mainly fall into two categories: one is basic guidance schemes, which end guidance once the vehicle enters the parking lot, or only provide parking space location indications through static signs, without involving dynamic route planning. This forces vehicles to search for their target parking space in complex passageways, and the process is prone to exacerbating congestion due to route misjudgment. The other is indoor navigation schemes, which can plan the shortest path from the user's current location to the target parking space, but are essentially passive road network navigation, focusing only on route planning rather than traffic management, and lacking control over the overall traffic flow of the parking lot. When multiple vehicles simultaneously navigate using this scheme, driving conflicts are very likely to occur at path intersections, failing to solve micro-level traffic congestion problems.

[0004] In other words, the main shortcomings of existing technologies are: First, they lack a global perspective and proactive scheduling capabilities. Existing solutions do not manage the spatial and temporal resources of parking lots in a coordinated manner and cannot avoid path conflicts between multiple vehicles in advance. Second, they are positioned as road network navigation rather than traffic management, focusing only on the length of the path and not considering the coordination of vehicle travel time, making it difficult to alleviate local congestion within the passageway. Third, the response mechanism is sluggish. When faced with emergencies such as vehicle breakdowns blocking passageways, they cannot quickly adjust navigation strategies, which can easily lead to the expansion of the congestion area. These shortcomings collectively highlight the urgent need to develop new proactive parking lot scheduling technologies.

[0005] Based on this, this application proposes a virtual green wave guidance scheme for parking lots to address the shortcomings of existing technologies and ensure parking lot order. Summary of the Invention

[0006] This application provides a virtual green wave guidance method and related equipment for parking lots. By constructing a spatiotemporal resource map to plan routes and reserving time windows to form virtual green wave belts, instructions are sent to vehicle terminals and smart gates. At the same time, vehicle position deviations are monitored and local replanning is triggered when the threshold is exceeded. This can eliminate multi-vehicle path conflicts at the root, solve micro-traffic congestion to improve traffic efficiency, quickly respond to emergencies to ensure traffic flow stability, and ultimately achieve seamless and uninterrupted passage of vehicles from the entrance to the parking space.

[0007] A virtual green wave guidance method for parking lots includes:

[0008] The parking lot access network is modeled as a directed graph, the access is divided into spatial units of a preset length, the time axis is divided into uniform time slices, and the combination of the spatial units and the time slices is used as the basic unit to construct a spatiotemporal resource map representing the spatiotemporal resource occupancy status of the parking lot.

[0009] Based on the spatiotemporal resource map, the planned path of the vehicle from the starting point to the target parking space is searched and determined in the time extension map, and a time window is reserved for the vehicle to pass through each spatial unit on the planned path, forming a virtual green wave for the vehicle to pass through.

[0010] The information of the virtual green wave is converted into control commands and sent to the vehicle terminal and the smart gates along the route.

[0011] The system monitors the spatiotemporal deviation between the actual and planned positions of vehicles in real time. If the determination result based on the spatiotemporal deviation vector exceeds a preset threshold, the vehicle is marked as deviating and a local replanning mechanism is triggered to reallocate spatiotemporal resources to the deviating vehicle and generate a new virtual green wave band.

[0012] Optionally, a path resource pre-allocation mechanism is employed during the search process of the time-spreading graph;

[0013] The path resource pre-allocation mechanism includes:

[0014] During the search process, candidate spatiotemporal units are temporarily reserved and updated to the temporary reserved resource list;

[0015] Once all time-space units along the entire path from the starting point to the target parking space have been successfully temporarily reserved, the temporary reservation will be converted into a formal reservation.

[0016] If the pre-allocation of the entire path fails, all spatiotemporal units in the temporary pre-allocated resource list are released, and a resource rollback operation is performed.

[0017] Optionally, the process of constructing the spatiotemporal resource map includes:

[0018] The channel is abstracted as a directed graph, where vertices represent channel intersections or key decision points, and edges represent channel segments connecting vertices. Each edge is discretized into multiple continuous and equal-length spatial units.

[0019] Divide the future preset time range into multiple consecutive time slices of equal length;

[0020] Each spatial unit is combined with each time slice to form a spatiotemporal unit;

[0021] The status information of each spatiotemporal unit is invoked, and a spatiotemporal resource map is constructed based on the status information of all the spatiotemporal units. The status information includes at least the real-time occupancy status and the reservation vehicle identifier.

[0022] Optionally, the cost function for searching and determining the planned path in the time-spread graph is:

[0023]

[0024] In the formula, The actual cost from the starting point to the current spatiotemporal node n is calculated by combining the weights of travel time, energy consumption, and path comfort. The estimated cost from the current spatiotemporal node n to the target parking space is an acceptable estimate in the time dimension to ensure the optimality of the search.

[0025] Optionally, the real-time monitoring of the vehicle's spatiotemporal deviation, if the determination result based on the spatiotemporal deviation vector exceeds a preset threshold, is marked as a vehicle deviating from its course, triggering a local replanning mechanism, including:

[0026] Real-time monitoring and calculation of the time deviation between the actual arrival time and the planned arrival time of a vehicle at a certain spatiotemporal node, as well as the spatial deviation between the actual position of the vehicle and the corresponding point on the planned path, form a spatiotemporal deviation vector;

[0027] If the magnitude of the spatiotemporal deviation vector exceeds a preset threshold, or if either dimension of the time deviation or the spatial deviation exceeds the corresponding preset independent threshold, it is determined as a trajectory deviation and marked as a deviating vehicle.

[0028] Centered on the deviating vehicle, a corresponding spatiotemporal influence domain is determined. The spatiotemporal influence domain includes the spatiotemporal unit currently occupied and reserved by the deviating vehicle, as well as the spatiotemporal unit reserved by other vehicles that cause conflict.

[0029] For all vehicles within the spatiotemporal influence domain, spatiotemporal resources are collaboratively reallocated based on task priority and current state.

[0030] Optionally, the process of collaboratively reallocating spatiotemporal resources includes:

[0031] Compare the task priorities of the deviating vehicle with those of vehicles with resource conflicts within the spatiotemporal influence domain;

[0032] If the task priority of the deviating vehicle is higher than that of the resource conflicting vehicle, then the spacetime resources with occupancy conflicts will be allocated to the deviating vehicle, and an emergency replanning will be initiated for the resource conflicting vehicle.

[0033] If the task priority of the deviating vehicle is not higher than that of the resource conflicting vehicle, then the deviating vehicle is instructed to stop and wait at its current location or in a safe area ahead until the spacetime resource with the conflicting occupancy becomes available.

[0034] A virtual green wave guidance system for parking lots, comprising:

[0035] The resource map construction module is used to model the parking lot access network as a directed graph, divide the access into spatial units of a preset length, divide the time axis into uniform time slices, and construct a spatiotemporal resource map representing the spatiotemporal resource occupancy status of the parking lot using the combination of the spatial units and the time slices as the basic unit.

[0036] The virtual green wave planning module is used to search and determine the planned path of the vehicle from the starting point to the target parking space in the time extension map based on the spatiotemporal resource map, and to reserve a time window for the vehicle to pass through each spatial unit on the planned path, thereby forming a virtual green wave for the vehicle to pass through.

[0037] The control command issuing module is used to convert the information of the virtual green wave band into control commands and issue them to the vehicle terminal and the smart gates along the route.

[0038] The local replanning module is used to monitor the spatiotemporal deviation between the actual position and the planned position of a vehicle in real time. If the judgment result based on the spatiotemporal deviation vector exceeds a preset threshold, it is marked as a deviating vehicle, triggering the local replanning mechanism to reallocate spatiotemporal resources for the deviating vehicle and generate a new virtual green wave band.

[0039] A virtual green wave guidance device for parking lots includes a memory and a processor;

[0040] The memory is used to store programs;

[0041] The processor is configured to execute the program to implement the various steps of the parking lot virtual green wave guidance method as described in any of the preceding claims.

[0042] A readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the parking lot virtual green wave guidance method as described in any of the preceding claims.

[0043] A computer program product includes a computer program that, when executed by a processor, performs the steps of the parking lot virtual green wave guidance method as described in any of the preceding claims.

[0044] As can be seen from the above technical solutions, the parking lot virtual green wave guidance method and related equipment provided in this application include: modeling the parking lot channel network as a directed graph, dividing it into spatial units of a preset length and uniform time slices, and constructing a spatiotemporal resource map representing the spatiotemporal resource occupancy status; searching for paths in the time extension graph based on the map, and reserving passage time windows for vehicles in each spatial unit of the planned path to form a virtual green wave; converting the green wave information into control commands and sending them to the vehicle terminal and smart gate; monitoring vehicle position deviation in real time, and marking the deviating vehicle and triggering local replanning if the deviation exceeds a threshold, and reallocating spatiotemporal resources to generate a new green wave.

[0045] This application addresses the shortcomings of existing technologies in a targeted manner, with the following specific effects: Addressing the lack of a global perspective and proactive scheduling in existing technologies, this application integrates and visualizes the spatial and temporal resources of parking lot access through a spatiotemporal resource map, achieving unified management of global spatiotemporal resources. It reserves a dedicated passage time window for each vehicle, fundamentally preventing time overlap among multiple vehicles in the same spatial unit and eliminating path conflicts. Addressing the shortcomings of existing technologies in non-traffic management, the virtual green wave system does not merely provide static paths but plans dynamic passage schemes that coordinate space and time for vehicles, proactively guiding vehicles to pass in an orderly manner according to their reserved times. This effectively solves the micro-level traffic congestion problem within the parking lot, achieving seamless and uninterrupted passage from the entrance to the parking space, significantly improving the efficiency of traffic flow within the parking lot. Addressing the shortcomings of slow response and inability to handle emergencies, this application monitors the spatiotemporal deviation between the actual and planned positions of vehicles in real time. Once the deviation exceeds a preset threshold, the deviating vehicle is immediately marked and a local replanning is triggered. This quickly reallocates spatiotemporal resources to the deviating vehicle, enabling timely responses to emergencies such as vehicle malfunctions blocking access lanes, preventing congestion from spreading, and ensuring stable traffic flow in the parking lot. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0047] Figure 1 This is a flowchart of a virtual green wave guidance method for parking lots disclosed in an embodiment of this application;

[0048] Figure 2This is a schematic diagram of a virtual green wave guidance system for parking lots disclosed in an embodiment of this application;

[0049] Figure 3 This is a hardware structure block diagram of a parking lot virtual green wave guidance device disclosed in an embodiment of this application. Detailed Implementation

[0050] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0051] This application can be used in a wide variety of general-purpose or special-purpose computing device environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor devices, distributed computing environments including any of the above devices, etc.

[0052] The following section introduces the solution proposed in this application. The technical solution is as follows, please refer to the text below for details.

[0053] Figure 1 This is a flowchart of a virtual green wave guidance method for parking lots disclosed in an embodiment of this application.

[0054] like Figure 1 As shown, the method may include:

[0055] Step S1: Model the parking lot access network as a directed graph, divide the access into spatial units of a preset length, divide the time axis into uniform time slices, and construct a spatiotemporal resource map representing the spatiotemporal resource occupancy status of the parking lot using the combination of the spatial units and the time slices as the basic unit.

[0056] Specifically, when modeling the parking lot access network as a directed graph, the intersection points of the access points within the parking lot are used as nodes of the directed graph, and the one-way or two-way access segments connecting two intersection points are used as edges of the directed graph. The direction of the edges is consistent with the actual allowed passage direction of the access points. The basic attributes of each edge, such as access length, width, and speed limit, are also labeled in the directed graph. When dividing the access points into spatial units of preset length, the preset length needs to be set in conjunction with the body length and safe distance of common vehicles in the parking lot, ensuring that each spatial unit can only accommodate one vehicle for safe stopping or passage, avoiding conflicts caused by multiple vehicles occupying excessively long units and increased computational complexity due to excessively short units. When dividing the time axis into uniform time slices, the duration of each time slice is calculated and determined based on the access speed limit and the length of the spatial unit, ensuring that a vehicle can completely pass through a spatial unit within one time slice. When constructing a spatiotemporal resource map using a combination of spatial units and time slices as the basic unit, the map is presented in the form of a two-dimensional table or a three-dimensional model. The horizontal dimension is the spatial unit number, and the vertical dimension is the time slice number. The occupancy status of the corresponding spatiotemporal resources is marked in each basic unit. The reserved status is associated with the reserved vehicle identifier, and the occupied status is associated with the currently occupied vehicle identifier. This map can intuitively present the resource usage of each spatial unit in any time slice, providing global data support for subsequent route planning and resource reservation.

[0057] Step S2: Based on the spatiotemporal resource map, search and determine the planned path of the vehicle from the starting point to the target parking space in the time extension map, and reserve a time window for the vehicle to pass through each spatial unit on the planned path, forming a virtual green wave for the vehicle to pass through.

[0058] Specifically, when searching and determining the planned path based on the spatiotemporal resource map in the time-extended graph, the time-extended graph is a dynamic graph model formed by incorporating the time dimension on the directed graph constructed in step S1. Its nodes are combinations of spatial units and time slices, corresponding one-to-one with the basic units of the spatiotemporal resource map. Edges represent the movement of a vehicle from a certain spatial unit and time slice node to the corresponding node of an adjacent spatial unit in a subsequent time slice (e.g., if a vehicle is located in spatial unit A in time slice t1, it can move to spatial unit B adjacent to A in time slice t2, corresponding to an edge from "A-t1" to "B-t2"). The search process uses an algorithm suitable for dynamic path planning, aiming to reach the spatial unit corresponding to the target parking space in the shortest total time from the starting spatial unit. The objective function is to select the optimal planned path by ensuring that all spatial units traversed by the path are in an idle state. When reserving a time window for a vehicle to pass through each spatial unit on the planned path, the time window is one or more consecutive time slices, specifically determined by the vehicle's passage time through the spatial unit. This ensures that vehicles have sufficient time to pass without consuming additional resources. The reservation operation is achieved by marking the status of the corresponding time slice of the spatial unit traversed by the planned path as reserved and associating it with the vehicle's identifier in the spatiotemporal resource map, preventing other vehicles from occupying the space. A virtual green wave is formed through the above reservation operation, ensuring that each spatial unit on the planned path has its own dedicated passage time. When vehicles pass through each spatial unit sequentially according to the time window, they do not need to wait for other vehicles, forming a seamless passage sequence similar to a green wave.

[0059] Step S3: Convert the information of the virtual green wave into control commands and send them to the vehicle terminal and the smart gates along the route.

[0060] Specifically, when converting virtual green wave information into control commands, the virtual green wave information includes the spatial unit sequence of the planned route, the passage time window corresponding to each spatial unit, and the suggested speed for vehicles passing through each spatial unit. The suggested speed is calculated based on the time window and the length of the spatial unit to ensure vehicles pass on time. The converted control commands use a standardized data format, including a command identifier, a unique vehicle identifier, and the command's effective time. When sent to the vehicle terminal, it is achieved through the wireless communication network deployed within the parking lot. After receiving the command, the vehicle terminal guides the driver to follow the virtual green wave through a graphical navigation interface and voice prompts. When sent to the smart gates along the route, the command includes the gate's opening and closing times. After receiving the command, the smart gates automatically open or close according to the preset time, avoiding vehicles waiting at the gates and ensuring the continuity of green wave passage.

[0061] Step S4: Monitor the spatiotemporal deviation between the actual and planned positions of the vehicle in real time. If the judgment result based on the spatiotemporal deviation vector exceeds a preset threshold, it is marked as a deviating vehicle, triggering a local replanning mechanism to reallocate spatiotemporal resources for the deviating vehicle and generate a new virtual green wave band.

[0062] Specifically, when monitoring the spatiotemporal deviation between a vehicle's actual location and its planned location in real time, multi-source positioning technology is used to obtain the vehicle's actual location. For example, positioning base stations or Bluetooth beacons are deployed in underground parking lots. The vehicle terminal calculates its real-time location by receiving positioning signals. The planned location is the spatial unit position that the vehicle should be in the current time slice within a virtual green wave band. The spatiotemporal deviation is quantified by constructing a spatiotemporal deviation vector, which includes spatial deviation components and temporal deviation components. When the judgment result based on the spatiotemporal deviation vector exceeds a preset threshold, the vehicle is marked as deviating. The preset threshold is set according to the parking lot's traffic safety requirements. When any component of the spatiotemporal deviation vector exceeds the corresponding threshold, or the weighted sum of two components exceeds the comprehensive threshold, the vehicle is determined to be in a deviated state and its identifier is marked. The process of triggering a local replanning mechanism to reallocate spatiotemporal resources for vehicles that have deviated from their designated paths and generate new virtual green wave bands is as follows: The scope of local replanning is limited to the spatial unit where the vehicle is currently located and its multiple adjacent spatial units. Based on the spatiotemporal resource map, idle spatial units within the current and subsequent time slices within this scope are selected. The same path search algorithm as in step S2 is used to replan short-distance paths. Unoccupied time windows are reserved for each spatial unit on the new path, thereby generating new virtual green wave bands. Time slices with smaller deviations from the original time windows are selected first to reduce the impact on other vehicles. At the same time, the status of the corresponding basic unit in the spatiotemporal resource map is updated, and the new virtual green wave band is converted into control commands and sent to the vehicle terminal and the newly added smart gates along the route to ensure that vehicles quickly return to an orderly passage state.

[0063] The process of constructing the spatiotemporal resource map may specifically include:

[0064] The channel is abstracted as a directed graph, where vertices represent channel intersections or key decision points, and edges represent channel segments connecting vertices. Each edge is discretized into multiple continuous and equal-length spatial units.

[0065] Divide the future preset time range into multiple consecutive time slices of equal length;

[0066] Each spatial unit is combined with each time slice to form a spatiotemporal unit;

[0067] The status information of each spatiotemporal unit is invoked, and a spatiotemporal resource map is constructed based on the status information of all the spatiotemporal units. The status information includes at least the real-time occupancy status and the reservation vehicle identifier.

[0068] Specifically, in the channel abstraction and spatial unit partitioning stage, the parking lot channel network needs to be transformed into a directed graph model. The vertices of the directed graph correspond to channel intersections and key decision points within the parking lot. Channel intersections include various forks in the road and the junctions of circular channels; key decision points include parking space entrances / exits (nodes where vehicles need to turn from the channel to enter a parking space), smart gate installation locations (nodes where vehicles need to pass or make brief stops), and the starting points of narrow channel sections (nodes where oncoming traffic needs to be coordinated in advance). Each vertex needs to be assigned a unique identifier and its actual geographic coordinates recorded. The coordinate accuracy needs to be adapted to the subsequent vehicle positioning requirements to ensure accurate spatial location determination. The edges of the directed graph correspond to channel segments connecting two vertices, and the direction of the edge needs to be consistent with the actual allowed traffic direction of the channel: if the channel segment only allows one-way traffic, the edge has a one-way attribute; if the channel segment supports two-way traffic, it corresponds to two one-way edges with opposite directions. At the same time, each edge needs to be labeled with its own basic attributes such as length, width, and speed limit requirements, providing a basis for subsequent spatial unit partitioning. Subsequently, each edge is discretized into multiple continuous spatial units of equal length. The length of the spatial unit must match the size of the mainstream vehicles in the parking lot and reserve sufficient safety distance to ensure that each spatial unit can only accommodate one vehicle to safely stop or pass through. Furthermore, there is no overlap or gap between the spatial units. The specific geographical range of each spatial unit can be calculated by the starting coordinates of the edge and the length of the spatial unit, thus establishing a unified standard for subsequent determination of spatiotemporal resource occupancy.

[0069] Secondly, in the time-slice segmentation stage, a future pre-set time range needs to be determined. This range should be set with reference to the average time it takes for vehicles to travel from the entrance to the target parking space during peak hours, and additional time should be reserved to handle unforeseen circumstances, covering the entire process requirements for vehicle passage. Then, this future pre-set time range is divided into multiple consecutive time slices of equal length. The duration of each time slice needs to be derived by combining the speed limit of the passage and the length of the spatial unit, ensuring that vehicles can completely pass through a spatial unit within a single time slice. This avoids situations where time slices are too short, requiring vehicles to occupy the same spatial unit across multiple time slices, or where time slices are too long, resulting in reduced utilization of spatiotemporal resources. Each time slice needs to be assigned a unique time sequence identifier, and there should be no time interval between adjacent time slices, forming a continuous timeline. This provides a unified granularity for subsequent time-dimensional management of spatiotemporal resources.

[0070] Next, in the spatiotemporal unit combination stage, a one-to-one correspondence between spatial units and time slices is adopted to construct spatiotemporal units. Using a two-dimensional matrix as the basic carrier, all spatial units are arranged horizontally, grouped and arranged in an orderly manner according to their respective passage segments, ensuring that spatial units within the same passage segment are presented collectively. All time slices are arranged vertically, sequentially according to time, consistent with the timeline order. Each cell in the matrix is ​​an independent spatiotemporal unit, and each spatiotemporal unit must be assigned a unique identifier that clearly reflects its corresponding spatial unit and time slice. This ensures that each spatiotemporal unit corresponds to a unique spatial location and time slice combination, achieving grid-based management of parking lot spatiotemporal resources. This avoids confusion between different spatiotemporal resources and facilitates quick subsequent queries of the status of a specific spatial unit in a particular time slice, or the overall status of all spatial units within a specific time slice.

[0071] Finally, in the status information retrieval and map construction phase, the acquisition methods and content of spatiotemporal unit status information are first clarified. Status information should at least include real-time occupancy status and reserved vehicle identifiers. Real-time occupancy status is collected through multi-source sensing devices deployed within the parking lot: ultra-wideband positioning base stations can monitor the vehicle's stay and movement status within the spatial unit in real time; infrared sensors at the entrance and exit of the passage can determine whether a vehicle has entered or left the spatial unit; the opening and closing status of the smart gate can help confirm the time it takes for a vehicle to pass through the passage segment. The collection frequency of the sensing devices must be synchronized with the duration of the time slice to ensure the timeliness of the status information. Real-time occupancy status is divided into three categories: idle, meaning there are no vehicles in the spatiotemporal unit; occupied, meaning there are vehicles staying or passing through the spatiotemporal unit; and pending release, meaning the vehicle is about to leave the spatiotemporal unit, and the next time slice will transition to an idle state. Reserved vehicle identifiers are derived from the reservation records in the previous route planning module. When a vehicle completes route planning and reserves a specific spatiotemporal unit, the system automatically associates the vehicle's unique identifier (such as license plate number or vehicle terminal identifier) ​​with the corresponding spatiotemporal unit. If a spatiotemporal unit has no reservation record, it is marked as unreserved. The system employs a real-time retrieval and incremental update mechanism for status information: Each time the system completes a real-time occupancy status collection, it simultaneously retrieves the reserved vehicle identifier for the corresponding spatiotemporal unit, updating only the information of spatiotemporal units whose status has changed, thus reducing data processing load. Based on the status information of all spatiotemporal units, a visualized spatiotemporal resource map is constructed: In the 3D map, spatial units with different statuses are distinguished by modules of different colors, allowing for intuitive identification of statuses such as idle, occupied, and reserved. Hovering the mouse over any module displays the time slice identifier, real-time occupancy status, and reserved vehicle identifier for that spatiotemporal unit. In the 2D dynamic table, each row corresponds to a spatial unit, and each column corresponds to a time slice. The cells clearly label the status information with text, and the system supports filtering by status type and jumping to different time slices. This facilitates real-time monitoring of parking lot spatiotemporal resource usage by staff and provides accurate and real-time resource status data support for subsequent vehicle route planning and local replanning.

[0072] As can be seen from the above technical solutions, the parking lot virtual green wave guidance method and related equipment provided in this application include: modeling the parking lot channel network as a directed graph, dividing it into spatial units of a preset length and uniform time slices, and constructing a spatiotemporal resource map representing the spatiotemporal resource occupancy status; searching for paths in the time extension graph based on the map, and reserving passage time windows for vehicles in each spatial unit of the planned path to form a virtual green wave; converting the green wave information into control commands and sending them to the vehicle terminal and smart gate; monitoring vehicle position deviation in real time, and marking the deviating vehicle and triggering local replanning if the deviation exceeds a threshold, and reallocating spatiotemporal resources to generate a new green wave.

[0073] This application addresses the shortcomings of existing technologies in a targeted manner, with the following specific effects: Addressing the lack of a global perspective and proactive scheduling in existing technologies, this application integrates and visualizes the spatial and temporal resources of parking lot access through a spatiotemporal resource map, achieving unified management of global spatiotemporal resources. It reserves a dedicated passage time window for each vehicle, fundamentally preventing time overlap among multiple vehicles in the same spatial unit and eliminating path conflicts. Addressing the shortcomings of existing technologies in non-traffic management, the virtual green wave system does not merely provide static paths but plans dynamic passage schemes that coordinate space and time for vehicles, proactively guiding vehicles to pass in an orderly manner according to their reserved times. This effectively solves the micro-level traffic congestion problem within the parking lot, achieving seamless and uninterrupted passage from the entrance to the parking space, significantly improving the efficiency of traffic flow within the parking lot. Addressing the shortcomings of slow response and inability to handle emergencies, this application monitors the spatiotemporal deviation between the actual and planned positions of vehicles in real time. Once the deviation exceeds a preset threshold, the deviating vehicle is immediately marked and a local replanning is triggered. This quickly reallocates spatiotemporal resources to the deviating vehicle, enabling timely responses to emergencies such as vehicle malfunctions blocking access lanes, preventing congestion from spreading, and ensuring stable traffic flow in the parking lot.

[0074] In some embodiments of this application, a path resource pre-occupancy mechanism is adopted in step S2 during the search process of the time extension graph. The core purpose of this mechanism is to avoid candidate spatiotemporal units being occupied by other parallel path search tasks during the path search process, which would cause the finally determined path to be unusable due to spatiotemporal resource conflicts, thereby improving the success rate and reliability of path planning.

[0075] The path resource pre-allocation mechanism may specifically include:

[0076] During the search process, candidate spatiotemporal units are temporarily reserved and updated to the temporary reserved resource list;

[0077] Once all time-space units along the entire path from the starting point to the target parking space have been successfully temporarily reserved, the temporary reservation will be converted into a formal reservation.

[0078] If the pre-allocation of the entire path fails, all spatiotemporal units in the temporary pre-allocated resource list are released, and a resource rollback operation is performed.

[0079] Specifically, in the temporary pre-allocation phase, when the search algorithm selects a candidate spatiotemporal unit that meets the criteria in the temporal expansion graph—meaning that the spatiotemporal unit is currently idle and can form a continuous path with the selected preceding spatiotemporal units—the system immediately performs a temporary pre-allocation operation on that candidate spatiotemporal unit. Temporary pre-allocation is not a formal resource lock; rather, it is achieved by marking the temporary pre-allocation status of the spatiotemporal unit within the system and adding it to the temporary pre-allocation resource list. The temporary pre-allocation resource list needs to record the unique identifier of each temporarily pre-allocated spatiotemporal unit, the trigger time of the pre-allocation operation, and the identifier of the path search task to which it belongs, in order to distinguish the pre-allocated resources of different search tasks and avoid confusion. The key to this phase is real-time performance; that is, once a candidate spatiotemporal unit enters the selection range of the search algorithm and meets the idle condition, pre-allocation is immediately performed to prevent other search tasks from preemptively occupying it, reserving resource space for the complete search of the current path.

[0080] Secondly, in the formal reservation conversion phase, as the search algorithm starts from the origin, it gradually filters and temporarily reserves candidate spatiotemporal units until it successfully covers all the spatiotemporal units required for the entire path from the origin to the target parking space. This means that each spatiotemporal unit on the entire path has been marked as temporarily reserved, and a continuous, uninterrupted passage sequence can be formed between the spatiotemporal units. At this point, the system triggers the formal reservation conversion process. The system updates the status of all spatiotemporal units in the temporarily reserved resource list from temporarily reserved to formally reserved, and associates these spatiotemporal units with the vehicle identifiers corresponding to the currently planned path. Simultaneously, it updates the spatiotemporal resource map. The updated spatiotemporal resource map marks these formally reserved spatiotemporal units as reserved, ensuring that other path search tasks can no longer reserve or reserve them, thus completely locking the spatiotemporal resources of the entire path and guaranteeing the subsequent passage of vehicles according to the planned path.

[0081] Finally, during the resource rollback phase, if any candidate space-time unit cannot be temporarily reserved during the path search process—for example, if the unit has already been temporarily reserved or formally booked by another search task, or if an unforeseen event causes it to become occupied, preventing the formation of a complete path from the starting point to the target parking space (i.e., the entire path reservation fails)—the system will immediately perform a resource rollback operation. At this time, the system will iterate through the temporary reservation resource list, releasing all space-time units marked as temporarily reserved, clearing their temporary reservation markers, restoring them to an idle state, and simultaneously clearing the temporary reservation resource list. This ensures that these space-time resources that have not formed a complete path can re-enter the resource pool for use by other path search tasks, preventing long-term resource idleness due to partial reservation failures and ensuring efficient utilization of parking lot space-time resources.

[0082] When searching for and determining the planned path in the time-extended graph, a cost function is used to measure the merits of different candidate paths in order to select the optimal path. The core function of this cost function is to comprehensively consider the actual consumption and the estimated consumption of the path, so as to ensure that the search results are both efficient and in line with the actual traffic needs.

[0083] The cost function for determining the planned path in the time-spread graph is:

[0084]

[0085] In the formula, The actual cost from the starting point to the current spatiotemporal node n is calculated by combining the weights of travel time, energy consumption, and path comfort. The estimated cost from the current spatiotemporal node n to the target parking space is an acceptable estimate in the time dimension to ensure the optimality of the search.

[0086] In the cost function expression, the actual cost from the starting point to the current spatiotemporal node is a comprehensive quantification of various consumptions incurred by the vehicle during its journey. Travel time primarily considers factors such as the speed limit of the passage between the current spatiotemporal node and the previous spatiotemporal node, and the length of the spatial unit, reflecting the actual travel time of the vehicle on this path. Energy consumption incorporates factors such as vehicle type, passage gradient, and changes in travel speed, reflecting the energy cost of vehicle travel. Path comfort weights are set based on factors such as the smoothness of the path and historical congestion records; paths with fewer turns, gentler angles, and lower historical congestion rates have higher comfort weights. The calculation of the actual cost adjusts the weighting of each factor according to the operational needs of the parking lot to ensure that the actual cost accurately reflects the overall performance of the path.

[0087] The estimated cost from the current spatiotemporal node to the target parking space is a forward-looking estimate of the time required for a vehicle to travel from the current spatiotemporal node to the target parking space. This estimate must be acceptable, meaning the estimated cost should not exceed the actual cost of the vehicle's journey from the current spatiotemporal node to the target parking space. This is to avoid overly optimistic estimates that could cause the search process to miss better paths. The calculation of the estimated cost mainly references information such as the idle status of spatiotemporal units surrounding the target parking space in the spatiotemporal resource map, the overall speed limit of the passage from the current spatiotemporal node to the target parking space, and the average cost of similar paths in historical traffic data. This ensures that the estimated cost is both valuable for reference and does not deviate from reality, thereby assisting the search algorithm in quickly focusing on potential optimal paths and improving search efficiency.

[0088] Overall, this cost function, by combining actual and estimated costs, enables the search algorithm to take into account both the current consumption and potential future consumption of a path in the time-expanded graph. It quickly selects planned paths that are short in travel time, low in energy consumption, high in comfort, and free from resource conflicts, providing a high-quality path foundation for the subsequent formation of virtual green waves.

[0089] In some embodiments of this application, the process of step S4, which involves real-time monitoring of the spatiotemporal deviation between the vehicle's actual position and its planned position, and marking it as deviating from the vehicle if the determination result based on the spatiotemporal deviation vector exceeds a preset threshold, thereby triggering a local replanning mechanism, is described. Specifically, this process may include:

[0090] Step S41: Monitor and calculate in real time the time deviation between the actual arrival time and the planned arrival time of the vehicle at a certain spatiotemporal node, as well as the spatial deviation between the actual position of the vehicle and the corresponding point on the planned path, to form a spatiotemporal deviation vector.

[0091] Specifically, through real-time monitoring and calculation, a spatiotemporal deviation vector characterizing vehicle trajectory deviation is formed, providing a quantitative basis for subsequent deviation judgment. First, spatiotemporal nodes correspond to key spatiotemporal units on the vehicle's planned path (such as turning points in channel segments, spatiotemporal units corresponding to intelligent gates, etc.), which are the system's preset deviation monitoring benchmarks. In actual monitoring, obtaining the vehicle's actual arrival time relies on multi-source sensing devices within the parking lot: such as ultra-wideband positioning base stations deployed at spatiotemporal nodes, which can capture in real-time the time when a vehicle enters the spatial range corresponding to that node; or infrared sensor arrays within the channel, which calculate the actual time of arrival at the spatiotemporal node by analyzing the timing of the vehicle triggering the sensors. The planned arrival time, on the other hand, originates from the virtual green wave data generated by the system in the early stages, i.e., the start time of the reserved passage time slice corresponding to that spatiotemporal node in the planned path, which the system can directly retrieve from the path planning records.

[0092] The calculation of time deviation is achieved by comparing the actual arrival time with the planned arrival time. If the actual arrival time is earlier than the planned arrival time, the deviation is considered to be ahead; if it is later than the planned arrival time, it is considered to be delayed. Both must be recorded with clear attribute identifiers and deviation levels. The calculation of spatial deviation uses the geographical coordinates of corresponding points on the planned path as a benchmark. These corresponding points are either the center location of the spatial unit associated with the spatiotemporal node or a system-preset path reference point. By comparing the actual location coordinates uploaded in real-time by the vehicle terminal with the coordinates of this benchmark point, it is determined whether the vehicle has deviated from the spatial range of the planned path. The degree of deviation is qualitatively described as "slight deviation" or "severe deviation." For example, if the vehicle remains within the adjustable range around the planned spatial unit, it is considered a slight deviation; if it exceeds this range, it is considered a severe deviation. Finally, the time and spatial deviations are combined as two dimensions to form a spatiotemporal deviation vector. This vector must simultaneously record the attributes and degrees of both deviations to ensure that the impact of deviations can be comprehensively considered in subsequent judgments.

[0093] Step S42: If the magnitude of the spatiotemporal deviation vector exceeds a preset threshold, or if either dimension of the time deviation or the spatial deviation exceeds the corresponding preset independent threshold, it is determined as a trajectory deviation and marked as a deviating vehicle.

[0094] Specifically, by setting dual judgment conditions, the system ensures accurate identification of vehicles deviating from their planned path and avoids misjudgments or omissions. First, for the condition that the magnitude of the spatiotemporal deviation vector exceeds a preset threshold, the system evaluates the combined impact of time and spatial deviations using pre-defined logical rules. If the combined effect prevents the vehicle from entering subsequent spatiotemporal units as originally planned (e.g., excessive time delay causing subsequent units to be occupied, or excessive spatial deviation preventing return to the original path), then the magnitude of the judgment vector exceeds the threshold. This condition primarily considers the overall impact of the deviation, preventing situations where a single-dimensional deviation does not reach an independent threshold but carries a high overall risk from being overlooked.

[0095] Secondly, the condition that either the time deviation or the spatial deviation exceeds its corresponding preset independent threshold is a strict control over a single deviation dimension: the independent threshold for time deviation is set for situations that seriously affect subsequent traffic flow, such as time delays that may cause vehicles to miss all subsequent reserved time and space units, or early arrivals that may cause time overlap and conflicts with vehicles ahead; the independent threshold for spatial deviation is for situations where fine-tuning is not possible to return to the original path, such as when a vehicle deviates too far from the planned path and enters the spatial range of other channels, leaving no room for adjustment. When either dimension meets this condition, regardless of the deviation of the other dimension, it is directly judged as a trajectory deviation.

[0096] If either of the two judgment conditions is met, the system will immediately mark the vehicle as a deviating vehicle. The marking operation needs to be updated to the system's vehicle status list simultaneously to provide a clear target object for the subsequent determination of the spatiotemporal influence domain. At the same time, the original virtual green wave command issuance for the vehicle will be suspended to prevent it from continuing to travel along the wrong path.

[0097] Step S43: Taking the deviating vehicle as the center, determine the corresponding spatiotemporal influence domain. The spatiotemporal influence domain includes the spatiotemporal unit currently occupied and reserved by the deviating vehicle, as well as the spatiotemporal units reserved by other vehicles that cause conflict.

[0098] Specifically, the spatiotemporal range affected by the deviating vehicle is delineated to provide a clear operational boundary for subsequent collaborative resource allocation, avoiding excessive replanning scope that could disrupt global traffic flow. Firstly, the first type of spatiotemporal unit to be included in the impact domain is the spatiotemporal unit currently occupied and reserved by the deviating vehicle: the currently occupied unit is the current time slice unit corresponding to the spatial unit the vehicle was using when it deviated, which the system can directly obtain through real-time occupancy status records; the reserved units are all spatiotemporal units in the original planned path that the vehicle has not subsequently used, which need to be filtered from the reserved records on the spatiotemporal resource map to ensure that no subsequent resources originally planned for the vehicle are missed.

[0099] The second category to be included are spatiotemporal units reserved by other vehicles that cause conflicts. These include spatiotemporal units reserved by other vehicles that overlap in time or space with the spatiotemporal units currently occupied or reserved by other vehicles. For example, a unit currently occupied by a vehicle may overlap with a time slice of the same spatial unit reserved by another vehicle, or a subsequent unit reserved by a vehicle may overlap in time slices with other vehicles' reserved units. The system can filter out conflicting units and their corresponding other vehicles by querying the associated vehicle identifiers of all reserved units in the spatiotemporal resource map and comparing them with the relevant units of the vehicle that is deviating from the reservation.

[0100] Finally, the two types of spatiotemporal units mentioned above are integrated to form a spatiotemporal influence domain centered on the deviating vehicle. The influence domain must clearly mark all spatiotemporal unit identifiers and associated vehicles (deviating vehicles and conflicting vehicles) it contains, and record the current status of each unit.

[0101] Step S44: For all vehicles within the spatiotemporal influence domain, coordinate the reallocation of spatiotemporal resources based on task priority and current status.

[0102] Specifically, by combining task priority with the current status of vehicles, efficient and coordinated allocation of resources within the affected area is achieved, balancing the passage needs of different vehicles and reducing the impact of replanning on overall traffic flow. First, task priority determination must consider multiple factors: for example, vehicles with urgent needs have higher priority than ordinary vehicles; vehicles that have traveled a considerable distance in the parking lot and are close to their target parking space have higher priority than vehicles that have just entered and have a longer remaining path; vehicles with no violations or deviation records have higher priority than vehicles that have previously deviated from their trajectory. The system has a pre-set priority determination rule base that can automatically assign priority levels to all vehicles within the affected area based on their real-time attributes. These levels must be clearly distinguished and serve as the core basis for resource allocation.

[0103] The considerations for a vehicle's current status include: whether the vehicle is in motion (vehicles in motion should avoid frequent route adjustments to prevent the risk of sudden braking or lane changes), whether it has already occupied a portion of a time-space unit (occupied units should be prioritized for normal passage before adjusting subsequent units), and the complexity of the remaining path (vehicles with fewer branching paths have greater flexibility in adjustments). The system obtains the current status of each vehicle through real-time monitoring data and path planning records, supplementing the priority rules. For example, for two vehicles with the same priority, the vehicle in a stationary waiting state can adjust resources first, avoiding danger caused by adjustments made by a vehicle in motion.

[0104] During resource allocation, the system prioritizes replanning spatiotemporal resources for high-priority vehicles: from the idle spatiotemporal units within the influence domain, units that meet the vehicle's subsequent path requirements are selected and reserved to form a new local virtual green wave, ensuring its passage continuity. For medium- and low-priority vehicles, their original reserved units are adjusted based on the remaining idle resources and the planning results of high-priority vehicles. This may involve delaying or advancing their passage time slots, or fine-tuning their spatial paths to adjacent, conflict-free passage units. After allocation, the system immediately updates the status of all units within the influence domain on the spatiotemporal resource map and issues new resource allocation instructions to the corresponding vehicle terminals and smart gates along the route. Simultaneously, it notifies vehicle owners of the route adjustment information, ensuring all vehicles pass through in an orderly manner according to the new plan and quickly eliminating the risk of conflicts caused by vehicles deviating from the plan.

[0105] The process of collaboratively reallocating spatiotemporal resources includes:

[0106] Compare the task priorities of the deviating vehicle with those of vehicles with resource conflicts within the spatiotemporal influence domain;

[0107] If the task priority of the deviating vehicle is higher than that of the resource conflicting vehicle, then the spacetime resources with occupancy conflicts will be allocated to the deviating vehicle, and an emergency replanning will be initiated for the resource conflicting vehicle.

[0108] If the task priority of the deviating vehicle is not higher than that of the resource conflicting vehicle, then the deviating vehicle is instructed to stop and wait at its current location or in a safe area ahead until the spacetime resource with the conflicting occupancy becomes available.

[0109] Specifically, the system first performs a task priority comparison operation between the deviating vehicle and the resource-conflicting vehicle. Here, a resource-conflicting vehicle refers to a vehicle within the spatiotemporal influence domain whose reserved spatiotemporal unit overlaps with the currently occupied or originally reserved unit of the deviating vehicle. The system first needs to filter out all such resource-conflicting vehicles from the list of associated vehicles in the spatiotemporal influence domain to form a conflict vehicle list. Subsequently, the system retrieves the priority determination rule base, extracts the priority determination criteria for both the deviating vehicle and each conflicting vehicle, and automatically generates their priority levels based on the preset weight logic of the rule base. During the comparison process, the system needs to compare the priority levels of the deviating vehicle and the conflicting vehicle one by one to clarify the priority relationship between them. If there are multiple conflicting vehicles, a priority correspondence between the deviating vehicle and each conflicting vehicle needs to be established separately to provide a basis for subsequent differentiated resource allocation.

[0110] Secondly, for scenarios where the task priority of a deviating vehicle is higher than that of a resource-conflicting vehicle, a resource priority allocation and emergency replanning operation for conflicting vehicles is performed. In this case, the system needs to explicitly allocate the conflicting spatiotemporal resources to the deviating vehicle. Specific operations include: updating the reserved association markers of these conflicting units on the spatiotemporal resource map from conflicting vehicles to deviating vehicles, simultaneously locking these units to prevent other vehicles from occupying them; and immediately issuing an updated virtual green wave instruction to the deviating vehicle's terminal, clearly informing it that it can use the conflicting resource for passage. For resource-conflicting vehicles, the system needs to simultaneously activate the emergency replanning mechanism: compared to regular replanning, emergency replanning prioritizes retrieving unoccupied idle spatiotemporal units within the spatiotemporal influence domain, shortens the time spent on path search and resource reservation, quickly plans a new local path for the conflicting vehicle to avoid the original conflicting resource, and reserves the corresponding spatiotemporal unit; simultaneously, the system informs the conflicting vehicle owner of the reason for replanning and the new path information through prompts, guiding them to travel along the new path and avoiding stagnation due to resource allocation.

[0111] Finally, for scenarios where the task priority of the deviating vehicle is no higher than that of the resource-conflicting vehicle, the system performs a parking wait and resource status monitoring operation. The system first determines whether the deviating vehicle's current location meets parking conditions: if the current location is in the middle of the passage and may obstruct other vehicles, the system instructs the deviating vehicle to move to a safe area ahead. The safe area ahead must be a pre-set temporary parking spot within the parking lot, and the system will guide the vehicle there via navigation instructions. If the current location itself does not affect the passage of other vehicles, the vehicle is allowed to remain at the current location. During the parking wait, the system monitors the status of conflicting spatiotemporal resources in real time, continuously tracking whether the resource changes from reserved or occupied to available through a spatiotemporal resource map. Once a conflicting resource becomes available, the system immediately issues a command to the deviating vehicle to resume passage, simultaneously updating its virtual green wave to guide the vehicle to continue along the newly planned route. At the same time, to avoid excessively long waiting times for deviating vehicles, the system will push real-time waiting time estimates and resource status update prompts to the vehicle owner, improving the user experience.

[0112] This resource collaborative allocation process, through priority differentiation, not only ensures the passage efficiency of high-priority vehicles, but also avoids additional conflicts caused by low-priority vehicles deviating from their designated paths through reasonable parking waiting and emergency replanning, ultimately achieving rapid recovery and stable operation of traffic flow within the spatiotemporal influence domain.

[0113] The following describes a parking lot virtual green wave guidance system provided in the embodiments of this application. The parking lot virtual green wave guidance system described below can be referred to in correspondence with the parking lot virtual green wave guidance method described above.

[0114] See Figure 2 , Figure 2 This is a schematic diagram of a parking lot virtual green wave guidance system disclosed in an embodiment of this application.

[0115] like Figure 2 As shown, the parking lot virtual green wave guidance system may include:

[0116] The resource map construction module 110 is used to model the parking lot passage network as a directed graph, divide the passage into spatial units of a preset length, divide the time axis into uniform time slices, and construct a spatiotemporal resource map representing the spatiotemporal resource occupancy status of the parking lot using the combination of the spatial units and the time slices as the basic unit.

[0117] The virtual green wave planning module 120 is used to search and determine the planned path of the vehicle from the starting point to the target parking space in the time extension map based on the spatiotemporal resource map, and to reserve a time window for the vehicle to pass through each spatial unit on the planned path, thereby forming a virtual green wave for the vehicle to pass through.

[0118] The control command sending module 130 is used to convert the information of the virtual green wave into control commands and send them to the vehicle terminal and the smart gates along the route.

[0119] The local replanning module 140 is used to monitor the spatiotemporal deviation between the actual position and the planned position of a vehicle in real time. If the judgment result based on the spatiotemporal deviation vector exceeds a preset threshold, it is marked as a deviating vehicle, triggering the local replanning mechanism to reallocate spatiotemporal resources for the deviating vehicle and generate a new virtual green wave band.

[0120] As can be seen from the above technical solutions, the parking lot virtual green wave guidance method and related equipment provided in this application include: modeling the parking lot channel network as a directed graph, dividing it into spatial units of a preset length and uniform time slices, and constructing a spatiotemporal resource map representing the spatiotemporal resource occupancy status; searching for paths in the time extension graph based on the map, and reserving passage time windows for vehicles in each spatial unit of the planned path to form a virtual green wave; converting the green wave information into control commands and sending them to the vehicle terminal and smart gate; monitoring vehicle position deviation in real time, and marking the deviating vehicle and triggering local replanning if the deviation exceeds a threshold, and reallocating spatiotemporal resources to generate a new green wave.

[0121] This application addresses the shortcomings of existing technologies in a targeted manner, with the following specific effects: Addressing the lack of a global perspective and proactive scheduling in existing technologies, this application integrates and visualizes the spatial and temporal resources of parking lot access through a spatiotemporal resource map, achieving unified management of global spatiotemporal resources. It reserves a dedicated passage time window for each vehicle, fundamentally preventing time overlap among multiple vehicles in the same spatial unit and eliminating path conflicts. Addressing the shortcomings of existing technologies in non-traffic management, the virtual green wave system does not merely provide static paths but plans dynamic passage schemes that coordinate space and time for vehicles, proactively guiding vehicles to pass in an orderly manner according to their reserved times. This effectively solves the micro-level traffic congestion problem within the parking lot, achieving seamless and uninterrupted passage from the entrance to the parking space, significantly improving the efficiency of traffic flow within the parking lot. Addressing the shortcomings of slow response and inability to handle emergencies, this application monitors the spatiotemporal deviation between the actual and planned positions of vehicles in real time. Once the deviation exceeds a preset threshold, the deviating vehicle is immediately marked and a local replanning is triggered. This quickly reallocates spatiotemporal resources to the deviating vehicle, enabling timely responses to emergencies such as vehicle malfunctions blocking access lanes, preventing congestion from spreading, and ensuring stable traffic flow in the parking lot.

[0122] Optionally, a path resource pre-allocation mechanism is employed during the search process of the time-spreading graph;

[0123] The path resource pre-allocation mechanism includes:

[0124] During the search process, candidate spatiotemporal units are temporarily reserved and updated to the temporary reserved resource list;

[0125] Once all time-space units along the entire path from the starting point to the target parking space have been successfully temporarily reserved, the temporary reservation will be converted into a formal reservation.

[0126] If the pre-allocation of the entire path fails, all spatiotemporal units in the temporary pre-allocated resource list are released, and a resource rollback operation is performed.

[0127] Optionally, the process of constructing the spatiotemporal resource map includes:

[0128] The channel is abstracted as a directed graph, where vertices represent channel intersections or key decision points, and edges represent channel segments connecting vertices. Each edge is discretized into multiple continuous and equal-length spatial units.

[0129] Divide the future preset time range into multiple consecutive time slices of equal length;

[0130] Each spatial unit is combined with each time slice to form a spatiotemporal unit;

[0131] The status information of each spatiotemporal unit is invoked, and a spatiotemporal resource map is constructed based on the status information of all the spatiotemporal units. The status information includes at least the real-time occupancy status and the reservation vehicle identifier.

[0132] Optionally, the cost function for searching and determining the planned path in the time-spread graph is:

[0133]

[0134] In the formula, The actual cost from the starting point to the current spatiotemporal node n is calculated by combining the weights of travel time, energy consumption, and path comfort. The estimated cost from the current spatiotemporal node n to the target parking space is an acceptable estimate in the time dimension to ensure the optimality of the search.

[0135] Optionally, the real-time monitoring of the vehicle's spatiotemporal deviation, if the determination result based on the spatiotemporal deviation vector exceeds a preset threshold, is marked as a vehicle deviating from its course, triggering a local replanning mechanism, including:

[0136] Real-time monitoring and calculation of the time deviation between the actual arrival time and the planned arrival time of a vehicle at a certain spatiotemporal node, as well as the spatial deviation between the actual position of the vehicle and the corresponding point on the planned path, form a spatiotemporal deviation vector;

[0137] If the magnitude of the spatiotemporal deviation vector exceeds a preset threshold, or if either dimension of the time deviation or the spatial deviation exceeds the corresponding preset independent threshold, it is determined as a trajectory deviation and marked as a deviating vehicle.

[0138] Centered on the deviating vehicle, a corresponding spatiotemporal influence domain is determined. The spatiotemporal influence domain includes the spatiotemporal unit currently occupied and reserved by the deviating vehicle, as well as the spatiotemporal unit reserved by other vehicles that cause conflict.

[0139] For all vehicles within the spatiotemporal influence domain, spatiotemporal resources are collaboratively reallocated based on task priority and current state.

[0140] Optionally, the process of collaboratively reallocating spatiotemporal resources includes:

[0141] Compare the task priorities of the deviating vehicle with those of vehicles with resource conflicts within the spatiotemporal influence domain;

[0142] If the task priority of the deviating vehicle is higher than that of the resource conflicting vehicle, then the spacetime resources with occupancy conflicts will be allocated to the deviating vehicle, and an emergency replanning will be initiated for the resource conflicting vehicle.

[0143] If the task priority of the deviating vehicle is not higher than that of the resource conflicting vehicle, then the deviating vehicle is instructed to stop and wait at its current location or in a safe area ahead until the spacetime resource with the conflicting occupancy becomes available.

[0144] The parking lot virtual green wave guidance system provided in this application embodiment can be applied to parking lot virtual green wave guidance equipment. Figure 3 The hardware structure block diagram of the parking lot virtual green wave guidance device is shown. Figure 3 The hardware structure of a parking lot virtual green wave guidance device may include: at least one processor 1, at least one communication interface 2, at least one memory 3, and at least one communication bus 4;

[0145] In this embodiment of the application, the number of processor 1, communication interface 2, memory 3, and communication bus 4 is at least one, and processor 1, communication interface 2, and memory 3 communicate with each other through communication bus 4;

[0146] Processor 1 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0147] Memory 3 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device;

[0148] The memory stores a program, which the processor can call. The program is used for:

[0149] The parking lot access network is modeled as a directed graph, the access is divided into spatial units of a preset length, the time axis is divided into uniform time slices, and the combination of the spatial units and the time slices is used as the basic unit to construct a spatiotemporal resource map representing the spatiotemporal resource occupancy status of the parking lot.

[0150] Based on the spatiotemporal resource map, the planned path of the vehicle from the starting point to the target parking space is searched and determined in the time extension map, and a time window is reserved for the vehicle to pass through each spatial unit on the planned path, forming a virtual green wave for the vehicle to pass through.

[0151] The information of the virtual green wave is converted into control commands and sent to the vehicle terminal and the smart gates along the route.

[0152] The system monitors the spatiotemporal deviation between the actual and planned positions of vehicles in real time. If the determination result based on the spatiotemporal deviation vector exceeds a preset threshold, the vehicle is marked as deviating and a local replanning mechanism is triggered to reallocate spatiotemporal resources to the deviating vehicle and generate a new virtual green wave band.

[0153] Optionally, the refined and extended functions of the program can be referred to the above description.

[0154] This application embodiment also provides a readable storage medium that can store a program suitable for execution by a processor, the program being used for:

[0155] The parking lot access network is modeled as a directed graph, the access is divided into spatial units of a preset length, the time axis is divided into uniform time slices, and the combination of the spatial units and the time slices is used as the basic unit to construct a spatiotemporal resource map representing the spatiotemporal resource occupancy status of the parking lot.

[0156] Based on the spatiotemporal resource map, the planned path of the vehicle from the starting point to the target parking space is searched and determined in the time extension map, and a time window is reserved for the vehicle to pass through each spatial unit on the planned path, forming a virtual green wave for the vehicle to pass through.

[0157] The information of the virtual green wave is converted into control commands and sent to the vehicle terminal and the smart gates along the route.

[0158] The system monitors the spatiotemporal deviation between the actual and planned positions of vehicles in real time. If the determination result based on the spatiotemporal deviation vector exceeds a preset threshold, the vehicle is marked as deviating and a local replanning mechanism is triggered to reallocate spatiotemporal resources to the deviating vehicle and generate a new virtual green wave band.

[0159] Optionally, the refined and extended functions of the program can be referred to the above description.

[0160] This application also provides a computer program product, including a computer program, wherein the computer program is executed by a processor using the following method:

[0161] The parking lot access network is modeled as a directed graph, the access is divided into spatial units of a preset length, the time axis is divided into uniform time slices, and the combination of the spatial units and the time slices is used as the basic unit to construct a spatiotemporal resource map representing the spatiotemporal resource occupancy status of the parking lot.

[0162] Based on the spatiotemporal resource map, the planned path of the vehicle from the starting point to the target parking space is searched and determined in the time extension map, and a time window is reserved for the vehicle to pass through each spatial unit on the planned path, forming a virtual green wave for the vehicle to pass through.

[0163] The information of the virtual green wave is converted into control commands and sent to the vehicle terminal and the smart gates along the route.

[0164] The system monitors the spatiotemporal deviation between the actual and planned positions of vehicles in real time. If the determination result based on the spatiotemporal deviation vector exceeds a preset threshold, the vehicle is marked as deviating and a local replanning mechanism is triggered to reallocate spatiotemporal resources to the deviating vehicle and generate a new virtual green wave band.

[0165] Optionally, the refined and extended functions of the program can be referred to the above description.

[0166] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0167] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0168] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A virtual green wave guidance method for parking lots, characterized in that, include: The parking lot access network is modeled as a directed graph, the access is divided into spatial units of a preset length, the time axis is divided into uniform time slices, and the combination of the spatial units and the time slices is used as the basic unit to construct a spatiotemporal resource map representing the spatiotemporal resource occupancy status of the parking lot. Based on the spatiotemporal resource map, the planned path of the vehicle from the starting point to the target parking space is searched and determined in the time extension map, and a time window is reserved for the vehicle to pass through each spatial unit on the planned path, forming a virtual green wave for the vehicle to pass through. The information of the virtual green wave is converted into control commands and sent to the vehicle terminal and the smart gates along the route. The system monitors the spatiotemporal deviation between the actual and planned positions of vehicles in real time. If the determination result based on the spatiotemporal deviation vector exceeds a preset threshold, the vehicle is marked as deviating and a local replanning mechanism is triggered to reallocate spatiotemporal resources to the deviating vehicle and generate a new virtual green wave band.

2. The method according to claim 1, characterized in that, A path resource pre-allocation mechanism is employed during the search process of the time-spreading graph. The path resource pre-allocation mechanism includes: During the search process, candidate spatiotemporal units are temporarily reserved and updated to the temporary reserved resource list; Once all time-space units along the entire path from the starting point to the target parking space have been successfully temporarily reserved, the temporary reservation will be converted into a formal reservation. If the pre-allocation of the entire path fails, all spatiotemporal units in the temporary pre-allocated resource list are released, and a resource rollback operation is performed.

3. The method according to claim 1, characterized in that, The process of constructing the spatiotemporal resource map includes: The channel is abstracted as a directed graph, where vertices represent channel intersections or key decision points, and edges represent channel segments connecting vertices. Each edge is discretized into multiple continuous and equal-length spatial units. Divide the future preset time range into multiple consecutive time slices of equal length; Each spatial unit is combined with each time slice to form a spatiotemporal unit; The status information of each spatiotemporal unit is invoked, and a spatiotemporal resource map is constructed based on the status information of all the spatiotemporal units. The status information includes at least the real-time occupancy status and the reservation vehicle identifier.

4. The method according to claim 1, characterized in that, The cost function for determining the planned path in the time-spread graph is: In the formula, The actual cost from the starting point to the current spatiotemporal node n is calculated by combining the weights of travel time, energy consumption, and path comfort. The estimated cost from the current spatiotemporal node n to the target parking space is an acceptable estimate in the time dimension to ensure the optimality of the search.

5. The method according to claim 1, characterized in that, The real-time monitoring of vehicle spatiotemporal deviation, if the determination result based on the spatiotemporal deviation vector exceeds a preset threshold, is marked as a deviating vehicle, triggering a local replanning mechanism, including: Real-time monitoring and calculation of the time deviation between the actual arrival time and the planned arrival time of a vehicle at a certain spatiotemporal node, as well as the spatial deviation between the actual position of the vehicle and the corresponding point on the planned path, form a spatiotemporal deviation vector; If the magnitude of the spatiotemporal deviation vector exceeds a preset threshold, or if either dimension of the time deviation or the spatial deviation exceeds the corresponding preset independent threshold, it is determined as a trajectory deviation and marked as a deviating vehicle. Centered on the deviating vehicle, a corresponding spatiotemporal influence domain is determined. The spatiotemporal influence domain includes the spatiotemporal unit currently occupied and reserved by the deviating vehicle, as well as the spatiotemporal unit reserved by other vehicles that cause conflict. For all vehicles within the spatiotemporal influence domain, spatiotemporal resources are collaboratively reallocated based on task priority and current state.

6. The method according to claim 5, characterized in that, The process of collaboratively reallocating spatiotemporal resources includes: Compare the task priorities of the deviating vehicle with those of vehicles with resource conflicts within the spatiotemporal influence domain; If the task priority of the deviating vehicle is higher than that of the resource conflicting vehicle, then the spacetime resources with occupancy conflicts will be allocated to the deviating vehicle, and an emergency replanning will be initiated for the resource conflicting vehicle. If the task priority of the deviating vehicle is not higher than that of the resource conflicting vehicle, then the deviating vehicle is instructed to stop and wait at its current location or in a safe area ahead until the spacetime resource with the conflicting occupancy becomes available.

7. A virtual green wave guidance system for parking lots, characterized in that, include: The resource map construction module is used to model the parking lot access network as a directed graph, divide the access into spatial units of a preset length, divide the time axis into uniform time slices, and construct a spatiotemporal resource map representing the spatiotemporal resource occupancy status of the parking lot using the combination of the spatial units and the time slices as the basic unit. The virtual green wave planning module is used to search and determine the planned path of the vehicle from the starting point to the target parking space in the time extension map based on the spatiotemporal resource map, and to reserve a time window for the vehicle to pass through each spatial unit on the planned path, thereby forming a virtual green wave for the vehicle to pass through. The control command issuing module is used to convert the information of the virtual green wave band into control commands and issue them to the vehicle terminal and the smart gates along the route. The local replanning module is used to monitor the spatiotemporal deviation between the actual position and the planned position of a vehicle in real time. If the judgment result based on the spatiotemporal deviation vector exceeds a preset threshold, it is marked as a deviating vehicle, triggering the local replanning mechanism to reallocate spatiotemporal resources for the deviating vehicle and generate a new virtual green wave band.

8. A virtual green wave guidance device for parking lots, characterized in that, Including memory and processor; The memory is used to store programs; The processor is used to execute the program to implement the various steps of the parking lot virtual green wave guidance method as described in any one of claims 1-6.

9. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements each step of the parking lot virtual green wave guidance method as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, The computer program is executed by the processor to perform the steps of the parking lot virtual green wave guidance method as described in any one of claims 1-6.

Images