Intelligent garage management method and system for collaborative dynamic guidance

Abstract

The invention discloses an intelligent garage management method and system for collaborative dynamic guidance, and aims to improve the parking resource utilization rate and the guidance efficiency. According to the method, a parking request containing user preferences and vehicle attributes is received, and a target parking lot is recommended and reserved. After the vehicle enters the parking lot, the system collects real-time data through the Internet of Things, and generates dynamic situation information in the parking lot by combining historical data and utilizing a space-time fusion prediction model. Based on this, the system executes multi-target dynamic optimal parking space distribution, and plans a driving path with the lowest comprehensive passing cost. In the guiding process, the system carries out continuous monitoring, and parking space and path re-planning is triggered immediately in case of congestion or parking space occupation and other abnormalities. The corresponding system comprises a request processing module, a parking lot sensing module, a data fusion and situation prediction module, a dynamic parking space distribution module, a real-time path planning module, a state monitoring and re-planning module, a billing updating module and the like, automatic and intelligent management from reservation, guiding to parking lot leaving billing is achieved through the cloud, and the parking efficiency and the user experience are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent parking garage management, and in particular to a collaborative dynamic guidance intelligent parking garage management method and system. Background Technology

[0002] With the acceleration of urbanization and the rapid growth of motor vehicle ownership, "parking difficulties" have become a common problem that plagues urban traffic management and residents' daily travel.

[0003] Traditional parking lot management is crude, relying mainly on manual guidance and static signage. As a result, car owners often face the dilemma of "difficulty in finding a parking space, slow passage, and chaotic management" after entering the garage. They need to spend a lot of time blindly searching for an empty parking space, which not only reduces the parking experience but also exacerbates traffic congestion and safety hazards inside the garage.

[0004] To address these challenges, smart parking systems have emerged and developed rapidly. Existing technological solutions have evolved primarily along several lines: First, by deploying sensor networks (such as ultrasonic, geomagnetic, and video sensors) to detect parking space occupancy in real time and disseminate information about available spaces to drivers through parking guidance screens and indicator lights, achieving basic "parking guidance." Second, by combining IoT and mobile internet technologies, allowing users to remotely search for and reserve parking spaces via mobile applications, and navigate within the parking lot using electronic maps or indicator lights. Third, by introducing more advanced image recognition and artificial intelligence algorithms, attempting to more accurately identify and schedule vehicles. For example, some solutions analyze historical vehicle images to assess parking space availability or use algorithms to dynamically optimize garage layout and vehicle access order.

[0005] However, existing intelligent parking garage management systems still have many limitations and have failed to achieve true collaborative dynamic guidance. Specifically:

[0006] The system is isolated, with most functional modules operating relatively independently. Parking space detection, route guidance, reverse vehicle search, and payment are often implemented by different subsystems, resulting in fragmented data flow and making it difficult to form a closed-loop collaborative optimization. For example, parking space allocation rarely considers real-time traffic flow, user personalized preferences (such as proximity to elevators or charging stations), and the future occupancy probability of the parking space, leading to recommended parking spaces that may not be globally optimal.

[0007] Static or semi-static guidance lacks dynamic adaptability, and most existing guidance strategies are based on static parking space status data at a certain moment. Once the planning is completed, the system struggles to adjust in real time to rapidly changing traffic conditions within the parking lot (such as sudden congestion or vehicle encounters). Studies have shown that incorrect road selection can consume a significant amount of driver time, and existing systems lack mechanisms for predicting and responding to dynamic traffic flow.

[0008] The data utilization is superficial, lacking situational awareness and predictive capabilities. The multi-dimensional data collected by the system (such as parking space status, vehicle speed, and historical patterns) has not been deeply integrated and analyzed. It has failed to build the ability to predict future short-term parking space occupancy and traffic congestion within the parking lot. This results in the system only being able to react in the "present," unable to conduct forward-looking planning and scheduling to prevent congestion or improve parking space turnover.

[0009] The existing solutions do not adequately consider user and vehicle characteristics. They often treat vehicles as homogeneous objects, primarily focusing on "availability" when allocating parking spaces, while neglecting vehicle size, model, and the user's personalized needs. Furthermore, the end-to-end personalized service chain from off-site to on-site is not yet fully established. For example, it lacks the ability to intelligently recommend the most convenient parking lots and spaces based on the user's final destination.

[0010] In summary, current intelligent parking garage management systems have significant shortcomings in terms of collaborative dynamic guidance.

[0011] Therefore, there is an urgent need for a new intelligent parking garage management method and system that can integrate multi-source information, achieve on-site and off-site collaboration, and make continuous optimization decisions based on real-time dynamic situation, so as to fundamentally improve parking efficiency, optimize user experience, and maximize the utilization of parking garage resources. Summary of the Invention

[0012] To address the significant shortcomings of existing intelligent parking garage management methods and systems in terms of collaborative dynamic guidance, a collaborative dynamic guidance intelligent parking garage management method and system is proposed.

[0013] A collaboratively dynamically guided intelligent parking garage management method includes the following steps:

[0014] S1. Receive a parking request from the user's terminal, and based on the destination information in the request, obtain information on multiple candidate parking lots in the vicinity;

[0015] S2. Based on user preference information and the information of the candidate parking lots, determine and reserve a target parking lot for the user;

[0016] S3. Obtain real-time multi-dimensional status data within the target parking lot, and perform fusion analysis based on the data and historical data to generate and continuously update dynamic status information within the parking lot that characterizes future parking space occupancy and traffic conditions;

[0017] S4. In response to a vehicle entering the target parking lot, the optimal parking space is dynamically allocated to the vehicle based on multi-objective decision-making, taking into account the dynamic situation information in the parking lot, vehicle attributes, and user preferences.

[0018] S5. Based on the dynamic situation information in the field and vehicle attributes, dynamically plan the driving path for the vehicle from its current position to the optimal parking space;

[0019] S6. Send the optimal parking space information and the driving route to the user terminal for guidance, and dynamically adjust the guidance strategy according to environmental changes during the guidance process;

[0020] S7. Update parking space status and complete billing operation.

[0021] Furthermore, in step S3, the fusion analysis is achieved by modeling the spatial topology of the parking lot to capture spatial dependencies and combining historical and real-time time series data analysis to capture temporal patterns.

[0022] Furthermore, the step S4 of dynamically allocating the optimal parking space for the vehicle based on multi-objective decision-making includes: selecting candidate parking spaces that meet the physical constraints of the vehicle from the currently available parking spaces;

[0023] A comprehensive evaluation value is calculated for each candidate parking space by calculating a weighted average. The comprehensive evaluation value includes at least the estimated travel time based on real-time and predicted traffic flow, the walking distance to the user's preferred facility, the future vacancy probability of the parking space based on the dynamic situation information in the site, and the matching degree between the parking space attributes and the user's preferences.

[0024] The parking space with the highest overall evaluation value is assigned as the optimal parking space.

[0025] Furthermore, in step S5, when dynamically planning the driving path for the vehicle, the path cost assessment comprehensively considers the static physical length of the road segment, the travel time cost based on real-time sensor data, the predicted congestion impact based on the dynamic situation information in the field, and the travel difficulty coefficient based on vehicle attributes.

[0026] Furthermore, in step S6, dynamically adjusting the guidance strategy according to environmental changes includes: continuously monitoring traffic flow and target parking space status during vehicle travel; if the congestion level of the planned route exceeds a specified threshold or the parking space status is abnormal, a replanning process is triggered, using the vehicle's current position as a new starting point, and re-executing parking space allocation and route planning based on the latest environmental data.

[0027] A collaborative dynamic guidance intelligent parking garage management system for implementing the aforementioned method, characterized in that it includes:

[0028] The request processing and recommendation module, deployed in the cloud, is used to execute steps S1 and S2;

[0029] The in-park sensing module is deployed in a distributed manner inside the parking lot to execute step S3;

[0030] The data fusion and situation prediction module, deployed in the cloud, is used to execute step S3;

[0031] The dynamic parking space allocation module, deployed in the cloud, is used to execute step S4;

[0032] The real-time path planning module, deployed in the cloud, is used to execute step S5;

[0033] The status monitoring and replanning module, deployed in the cloud, is used to execute step S6;

[0034] The billing and status update module, deployed in the cloud, is used to execute step S7;

[0035] The request processing and recommendation module, the on-site perception module, the data fusion and situation prediction module, the dynamic parking space allocation module, the real-time route planning module, the status monitoring and replanning module, and the billing and status update module interact with each other and transmit instructions through the network to collaboratively complete the dynamic guidance and management of the entire process from receiving user requests to completing billing.

[0036] Furthermore, the request processing and recommendation module includes a user interface unit and a recommendation engine unit;

[0037] The user interface unit is used to interact with the user terminal to receive structured parking request data packets and transmit the data packets to the recommendation engine unit;

[0038] The recommendation engine unit is used to calculate and determine the target parking lot based on the destination information, user preferences and multi-source external data in the data packet, and send the reservation result to the user terminal through the user interface unit;

[0039] The data fusion and situation prediction module includes a sensor network unit and a predictive analysis unit;

[0040] The sensor network units are distributed and deployed in the parking lot to collect multi-dimensional status data in real time;

[0041] The predictive analysis unit is used to receive and fuse real-time data and historical data uploaded by the sensor network unit, and generate dynamic situation information in the field through a spatiotemporal fusion predictive model.

[0042] The dynamic parking space allocation module and the real-time route planning module include a decision-making unit and a route calculation unit;

[0043] The decision-making unit is used to receive the dynamic situation information in the field, vehicle attributes and user preferences, and output the optimal parking space allocation result based on the multi-objective decision-making model.

[0044] The path calculation unit is used to receive the optimal parking space information, the real-time location of the vehicle and the dynamic situation information in the field, and calculate the driving path based on the dynamic cost map.

[0045] The status monitoring and replanning module includes an anomaly detection unit and a replanning trigger unit;

[0046] The anomaly detection unit is used to continuously receive feedback from the sensor network and vehicle positioning data to monitor traffic flow and parking space status.

[0047] The replanning triggering unit is used to trigger the decision unit and the path calculation unit to re-execute the calculation when the anomaly detection unit determines that the replanning conditions are met.

[0048] Furthermore, the sensor network unit in the field sensing module includes:

[0049] The parking space status sensing subunit is deployed in each parking space to detect the occupancy status of the parking space and perform license plate recognition;

[0050] Traffic flow sensing subunits are deployed on key routes to collect data on traffic volume, vehicle speed, and queue length.

[0051] The environment and event perception subunit is used to collect facility status and safety event data;

[0052] The vehicle positioning beacon subunit is deployed on-site to provide real-time positioning data for vehicles.

[0053] Furthermore, the real-time route planning module plans the driving route in the following way:

[0054] A dynamic cost map is constructed, in which the cost of each road segment integrates static physical length, travel time based on real-time sensor data, predicted congestion impact based on the dynamic situation information in the field, and travel difficulty coefficient based on vehicle attributes.

[0055] The D*Lite algorithm is used to search for the optimal path from the vehicle's current position to the optimal parking space on the dynamic cost map.

[0056] The beneficial effects of this invention are as follows: First, the collaborative dynamic guidance intelligent parking garage management method breaks down data barriers between the internal and external modules of the parking lot, enabling the integration and collaborative decision-making of user preferences, parking resources, and real-time on-site conditions, thereby improving the efficiency of regional traffic operation. Second, through modeling and analysis of parking lot spatial topology and time-series data, the system achieves an enhanced capability, moving from passively displaying current available spaces to proactively predicting future short-term parking space occupancy and traffic congestion. Based on this, the system introduces a multi-objective optimization model, which can dynamically allocate parking spaces to each vehicle and plan the optimal path that integrates static and real-time traffic costs and predicted congestion impacts, significantly shortening the space search time. More importantly, the system possesses a dynamic replanning mechanism, capable of recalculating in real time when path congestion or parking space anomalies are detected, ensuring the robustness and reliability of the guidance strategy. Attached Figure Description

[0057] Figure 1 This is an overview diagram of the method flow of an embodiment of the present invention;

[0058] Figure 2 This is a flowchart illustrating the dynamic parking space allocation decision-making process according to an embodiment of the present invention.

[0059] Figure 3 This is a flowchart illustrating the real-time path planning process according to an embodiment of the present invention.

[0060] Figure 4 This is a flowchart of the monitoring and dynamic replanning mechanism in an embodiment of the present invention;

[0061] Figure 5 This is a schematic diagram of the system module architecture and data flow in an embodiment of the present invention. Detailed Implementation

[0062] To make the objectives and technical solutions of the present invention clearer, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings and examples.

[0063] like Figure 1 As shown, a collaboratively dynamically guided intelligent parking garage management method includes the following seven steps:

[0064] S1. Receive a parking request from the user terminal, and based on the destination information in the request, obtain information on multiple candidate parking lots in the vicinity; more preferably, the user terminal is mainly applied to mobile intelligent terminals, such as dedicated applications (APPs) on smartphones and tablets, or in-vehicle applications integrated into the in-vehicle infotainment system (central control screen). For intelligent connected vehicles with vehicle-to-everything (V2X) communication capabilities, the user terminal function can also be integrated into the on-board unit (OBU).

[0065] More preferably, the parking request is a structured data packet that includes at least:

[0066] 1. Vehicle identification: License plate number.

[0067] 2. Vehicle attributes: brand, model (e.g., sedan, SUV), size (length, width, height), power type (gasoline vehicle, pure electric vehicle, plug-in hybrid).

[0068] 3. Destination Information: The name of the target location entered by the user or the latitude and longitude coordinates obtained by selecting a point on the map.

[0069] 4. User preferences: Optional preference tags, such as "Need charging station", "Need accessible parking", "Priority to elevator / exit", "Lowest expected cost", etc.

[0070] 5. Estimated arrival time: The time can be manually entered by the user or estimated by the system based on real-time location and traffic conditions.

[0071] S2. Based on user preference information and the information of the candidate parking lots, determine and reserve a target parking lot for the user; more preferably, the user preference information is a personalized profile built by the system for the user, including:

[0072] 1. Static preferences: Long-term needs explicitly specified by users during registration or settings, such as "new energy vehicle owners" (who need charging stations by default) and "people with mobility impairments" (who need accessible parking spaces by default).

[0073] 2. Dynamic preferences: Temporary requirements specified in a single request, such as wanting to "be near Gate 3 of the mall" for this shopping trip.

[0074] 3. Historical behavioral preferences: The system uncovers implicit preferences by analyzing users' historical parking records, such as frequently choosing to park on a certain restaurant floor on weekends.

[0075] S3. Obtain real-time multi-dimensional status data within the target parking lot, and perform fusion analysis based on the data and historical data to generate and continuously update dynamic status information within the parking lot that characterizes future parking space occupancy and traffic conditions;

[0076] Furthermore, the real-time multidimensional status data is continuously collected through an IoT sensor network deployed within the parking lot, primarily including:

[0077] 1. Parking space level data: Occupancy status of each parking space (vacant / occupied / reserved), parking space type (regular, charging, accessible, large vehicle), and physical dimensions.

[0078] 2. Traffic flow data: Real-time vehicle speed, traffic volume, and queue length for each lane and intersection.

[0079] 3. Event data: Alarm events such as abnormal parking, wrong-way driving, congestion, and accidents.

[0080] 4. Facility status data: occupancy / idle / fault status of charging piles, current operating status of elevators / escalators and the number of people waiting (estimated).

[0081] 5. Environmental data: Video surveillance streams from key locations, used for visual analysis and verification.

[0082] S4. In response to a vehicle entering the target parking lot, based on multi-objective decision-making, dynamically allocate the optimal parking space for the vehicle, taking into account the dynamic situation information within the parking lot, vehicle attributes, and user preferences (included in the parking request data packet);

[0083] The dynamic situational information within the field is a structured predictive data output by the fusion analysis, mainly including:

[0084] 1. Parking space occupancy probability field: A matrix covering all parking spaces, where each element represents the probability that the parking space will be occupied in a specific time period in the future (e.g., the next 10 minutes).

[0085] 2. Traffic travel time prediction graph: A weighted directed graph covering all main roads and intersections, with the weight of each edge being the predicted average future travel time.

[0086] 3. Congestion Risk Index Map: Identifies areas and levels of potential future congestion.

[0087] S5. Based on the dynamic situation information in the field and vehicle attributes, dynamically plan the driving path for the vehicle from its current position to the optimal parking space;

[0088] S6. Send the optimal parking space information and the driving route to the user terminal for guidance, and dynamically adjust the guidance strategy according to environmental changes during the guidance process;

[0089] S7. Update parking space status and complete billing operation.

[0090] Based on steps S1-7 above, preferably, in step S3, the fusion analysis is achieved through a "spatiotemporal fusion prediction model." The spatiotemporal fusion prediction model includes the following:

[0091] 1. Spatial Modeling: The parking lot is abstracted as a topological graph, where nodes represent parking spaces, lane intersections, and facility points, and edges represent connecting paths and are assigned weights (such as length and width). Graph Neural Networks (GNNs) are used to capture spatial dependencies (such as how congestion in one area can spread to adjacent lanes).

[0092] 2. Time series analysis: Train historical data with deep learning (such as LSTM, Transformer models) to learn the periodic and trend patterns of parking space occupancy and traffic flow.

[0093] 3. Fusion: Input the real-time collected multidimensional data into the trained spatiotemporal model to predict the occupancy probability of each parking space and the passage time of each route in the next 5-15 minutes, and generate a dynamically updated situation map.

[0094] This preferred solution achieves prediction of the future state of the parking lot by establishing a spatial topology model and integrating time-series data analysis. It correlates discrete sensor data, enabling the system not only to know about available parking spaces but also to predict the likelihood of spaces becoming available and the probability of congestion. This allows for advance parking space allocation and route planning, fundamentally avoiding guidance failures and secondary congestion caused by information lag.

[0095] Based on steps S1-7 above, such as Figure 2 As shown, preferably, the step S4 of dynamically allocating the optimal parking space for the vehicle based on multi-objective decision-making includes: selecting candidate parking spaces that meet the vehicle's physical constraints from the currently available parking spaces;

[0096] A comprehensive evaluation value is calculated for each candidate parking space by calculating a weighted average. The comprehensive evaluation value includes at least the estimated travel time based on real-time and predicted traffic flow, the walking distance to the user's preferred facility, the future vacancy probability of the parking space based on the dynamic situation information in the site, and the matching degree between the parking space attributes and the user's preferences.

[0097] Furthermore, dynamic allocation is an online, real-time optimization process:

[0098] 1. Trigger: Triggered when a vehicle enters the parking lot entrance.

[0099] 2. Filtering: From all available parking spaces, first filter out those that do not meet the physical size and basic type of the vehicle (e.g., non-charging vehicles cannot occupy charging spaces) to form an initial candidate set.

[0100] 3. Scoring: For each candidate parking space, a multi-objective decision-making model is used to calculate a comprehensive score. Model inputs include: real-time vehicle location, dynamic situational information, valid user preferences, and static attributes of the parking space. A comprehensive score is calculated through weighted summation (weights are configurable).

[0101] 4. Decision: Select the parking space with the highest score, immediately update its status to "locked" in the system, and distribute the allocation result through the network.

[0102] The parking space with the highest overall evaluation value is assigned as the optimal parking space.

[0103] The technical advantage of this preferred solution lies in its ability to achieve global optimization that comprehensively considers traffic efficiency (time), user convenience (walking distance), resource utilization (probability of future vacancy), and personalized needs (preference matching). This ensures that the allocated parking spaces are not only physically available, but are optimally selected based on overall cost and user experience.

[0104] Based on steps S1-7 above, such as Figure 3 As shown, preferably, in step S5, when dynamically planning the driving path for the vehicle, the path cost evaluation comprehensively considers the static physical length of the road segment, the travel time cost obtained based on real-time sensor data, the predicted congestion impact based on the dynamic situation information in the field, and the travel difficulty coefficient based on vehicle attributes.

[0105] The technical advantage of this preferred solution lies in the fact that the planned path is no longer the shortest geometric path on the map, but rather the "spatiotemporally optimal path" with the lowest overall travel cost in the current and near future. It can proactively avoid real-time congestion points and predicted congestion areas, and guide large vehicles and other special vehicles away from narrow sections. This reduces the ineffective driving and waiting time of vehicles within the site, effectively manages traffic flow, reduces the risk of accidents, and improves overall traffic efficiency.

[0106] like Figure 4 As shown, preferably, in step S6, dynamically adjusting the guidance strategy according to environmental changes includes: continuously monitoring the traffic flow and target parking space status during vehicle travel; if the congestion level of the planned route exceeds a specified threshold or the parking space status is abnormal, a replanning process is triggered, using the vehicle's current position as a new starting point, and re-executing parking space allocation and route planning based on the latest environmental data.

[0107] Guidance is achieved through multimodal human-computer interaction:

[0108] 1. Mobile App Guidance: Display a dynamically updated map of the venue on the user's mobile app, highlight the planned route with highlighted lines, and provide text and voice navigation instructions (such as "Turn left at the next intersection to go to Zone B").

[0109] 2. On-site facility guidance: The system controls the indicator lights on the parking spaces (such as the green light flashing for reserved parking spaces) and displays the direction of the target parking area and the number of remaining parking spaces on the diversion LED screens at key intersections.

[0110] 3. In-vehicle terminal guidance: For vehicles with vehicle connectivity, route information can be directly sent to the in-vehicle navigation system.

[0111] The core of dynamic adjustment is continuous monitoring and trigger-based replanning.

[0112] Monitoring: The system tracks vehicle locations in real time (via Bluetooth beacon triangulation or camera identification) and monitors traffic flow along the planned route as well as the status of target parking spaces.

[0113] Triggering conditions: The system is triggered immediately when (a) the congestion index of the path ahead exceeds the threshold, or (b) the target parking space is unexpectedly occupied by another vehicle (“space grabbing”).

[0114] Replanning action: The system interrupts the current guidance instructions, takes the vehicle's latest real-time location as the starting point, and combines the latest dynamic situation information to re-execute S4 (possibly allocating a new parking space nearby) and S5 (planning a new route), and pushes the new parking space and route information to the user terminal in real time to update the guidance instructions.

[0115] like Figure 5 As shown, a collaborative dynamic guidance intelligent parking garage management system for implementing a collaborative dynamic guidance intelligent parking garage management method includes: a request processing and recommendation module, deployed in the cloud, for executing steps S1 and S2;

[0116] The in-park sensing module is deployed in a distributed manner inside the parking lot to execute step S3;

[0117] The data fusion and situation prediction module, deployed in the cloud, is used to execute step S3;

[0118] The dynamic parking space allocation module, deployed in the cloud, is used to execute step S4;

[0119] The real-time path planning module, deployed in the cloud, is used to execute step S5;

[0120] The status monitoring and replanning module, deployed in the cloud, is used to execute step S6;

[0121] The billing and status update module, deployed in the cloud, is used to execute step S7;

[0122] The request processing and recommendation module, the on-site perception module, the data fusion and situation prediction module, the dynamic parking space allocation module, the real-time route planning module, the status monitoring and replanning module, and the billing and status update module interact with each other and transmit instructions through the network to collaboratively complete the dynamic guidance and management of the entire process from receiving user requests to completing billing.

[0123] Preferably, the request processing and recommendation module includes a user interface unit for interacting with the user terminal and receiving parking requests, and a recommendation engine unit for calculating parking lot recommendation scores based on multi-source information. The user interface unit acts as a bridge for data exchange between the system and various user terminals. It provides services in the form of an application programming interface (API). For individual users, the carrier is mainly a smartphone application (APP) or a mini-program integrated into WeChat or Alipay; for in-vehicle scenarios, interaction is conducted through the embedded client of the in-vehicle infotainment system or an in-vehicle unit conforming to the vehicle-to-everything (V2X) communication protocol. The structured parking request data packet received by this unit specifically includes: vehicle identification (such as license plate number), vehicle attributes (brand, model, dimensions, power type such as fuel / pure electric / plug-in hybrid), destination information (target location name or latitude and longitude coordinates), user preferences (such as tags like "need charging station," "priority to elevator," "lowest expected cost," etc.), and estimated arrival time.

[0124] Recommendation Engine Unit: This unit is the core decision-making unit of the module. It accesses multiple external data services, including geocoding services from map service providers (such as Gaode and Baidu) and data from city-level parking platforms. Internally, it maintains a static parking information database, storing the basic attributes of each networked parking lot (location, total number of parking spaces, floor structure, facility distribution, and fee rules). Upon receiving a request, the engine executes the following process: First, based on the destination information, it searches the database for multiple candidate parking lots in the vicinity; second, through the real-time data interface of each parking lot, it obtains information such as the real-time number of available parking spaces and the estimated queue time at the entrance; finally, based on a configurable multi-objective scoring model, it comprehensively evaluates the matching degree between each candidate parking lot and the user's request. Scoring factors include: static facility matching degree (whether there are charging piles), spatial accessibility (driving distance and time), destination convenience (walking distance from the parking lot to the destination), economy (fee rate), and real-time load status. The parking lot with the highest score is determined as the "target parking lot," and this unit then calls the parking system's reservation interface to lock in the parking space qualification, and returns the reservation success information and initial navigation guidance information to the user through the user interface unit.

[0125] The in-parking perception module is deployed in a distributed manner within each parking lot, responsible for comprehensively and in real-time collecting multi-dimensional state data of the physical world within the parking lot. The in-parking perception module includes a sensor network unit and a data aggregation and preprocessing unit.

[0126] Sensor Network Unit: This unit consists of heterogeneous sensor nodes and data acquisition hardware distributed throughout the parking lot. Specifically, it includes:

[0127] Parking space status sensing subunit: Deploy video parking space cameras or geomagnetic sensors on each parking space to detect the "vacant," "occupied," or "reserved" status of the parking space in real time. The video cameras can also perform license plate recognition to associate specific vehicles.

[0128] Traffic flow perception subunit: Wide-angle surveillance cameras are deployed at key locations such as main roads, intersections, and ramps, with built-in video analysis algorithms to calculate traffic flow, average vehicle speed, and vehicle queue length in real time; millimeter-wave radar is deployed in areas with complex visual conditions to provide more stable target detection and speed measurement.

[0129] Environment and event perception subunit: Deploy passenger flow statistics cameras in elevator lobbies and stairwells; integrate smart meters and communication modules into charging piles to report their working status (idle / charging / fault); automatically detect safety events such as abnormal parking, wrong-way driving, and pedestrian intrusion by analyzing the global video stream.

[0130] Vehicle positioning beacon subunit: Bluetooth beacons (iBeacon / Eddystone) or ultra-wideband (UWB) positioning base stations are regularly deployed on the ceiling of the parking lot to provide real-time location information with meter-level accuracy for vehicles entering the parking lot.

[0131] Data Aggregation and Preprocessing Unit: This unit typically runs on a local edge computing server within the parking lot. It receives raw data streams from all sensor subunits via wired (Ethernet) or wireless (industrial Wi-Fi) networks. Its core functions include data cleaning (filtering out noise and outliers), timestamp synchronization, format standardization, and local caching of the processed structured status data (e.g., "Parking space 023 in Zone B: Status = Occupied, License Plate = A12345, Time = T"). Simultaneously, it uploads the data to the cloud-based data fusion and situation prediction module in real time via 4G / 5G CPE or fiber optic leased lines.

[0132] The data fusion and situation prediction module includes a sensor network unit for real-time collection of parking space occupancy, vehicle traffic and facility usage status data, and a predictive analysis unit for fusion analysis of the data and generation of dynamic situation information within the site. The data fusion and situation prediction module is deployed on a cloud-based high-performance computing platform and is responsible for transforming massive amounts of real-time and historical data into insights into future situations.

[0133] Data Storage and Management Unit: This unit is responsible for the persistent storage and organization of system data. It adopts a hybrid database architecture: a time-series database (such as InfluxDB) is used for efficient storage and querying of timestamped sequence data generated by sensors (such as changes in parking space status and traffic flow); a relational database (such as MySQL) is used to store relational data such as parking lot structure information, user profiles, and event logs; and an object storage service is used to store unstructured data such as video clips and images. Historical data mainly includes long-term accumulated parking space occupancy rates, traffic flow patterns, and event statistics.

[0134] Predictive Analysis Unit: This unit is the core algorithm unit of the module, realizing "fusion analysis". Internally, it runs a spatiotemporal fusion prediction model, which operates through the following steps:

[0135] Spatial topology modeling: The physical structure of the parking lot is abstracted as a graph structure. Nodes represent key locations such as parking spaces, lane intersections, and elevator entrances; edges represent connecting paths and are assigned attributes such as length, width, and direction. This graph is used to encode spatial dependencies.

[0136] Temporal pattern learning: Using deep learning models such as Long Short-Term Memory Network (LSTM) or Transformer, historical time series data is trained to learn the periodicity (such as weekday morning and evening rush hours), trend and event correlation of parking space occupancy and traffic congestion.

[0137] Real-time fusion and prediction: Multi-dimensional state data uploaded in real time by the on-site perception module (as observation input) is injected into the trained model. The model combines the correlation of the spatial map and the learned temporal patterns to perform rolling predictions and output dynamic situation information on the site. This information is specifically represented as: a parking space occupancy probability matrix (predicting the probability of each parking space being occupied in the next 5-15 minutes), a channel travel time prediction map (predicting the travel time of each path segment in the future), and a congestion risk heat map (identifying areas and levels of potential future congestion).

[0138] The dynamic parking space allocation module and the real-time route planning module include a decision-making unit for performing optimized parking space allocation based on the situation information, vehicle attributes and user preferences, and a route calculation unit for planning driving routes based on the situation information and vehicle attributes.

[0139] Decision-making unit: This unit is the core of the allocation logic execution. Its workflow is as follows:

[0140] Candidate set generation: First, quickly filter out the set of parking spaces that meet the physical constraints of the vehicle from all parking spaces. The constraints include: the size of the parking space (length, width, height) must accommodate the outline of the vehicle; the type of parking space matches the vehicle demand (e.g., only electric vehicles can be assigned charging parking spaces); and the current status of the parking space is "vacant".

[0141] Multi-objective comprehensive scoring: For each candidate parking space, a comprehensive evaluation value is calculated. The scoring function uses a weighted sum model: Score = w1*F1 + w2*F2 + w3*F3 + w4*F4. Where:

[0142] F1 (Estimated Travel Time): Calculates the estimated time to reach the parking space based on the vehicle's current location, real-time traffic flow, and travel time prediction map in the situation prediction map.

[0143] F2 (Walking distance to preferred facility): Calculates the shortest path distance from the parking space to the user's preferred facility (such as a specific elevator entrance).

[0144] F3 (Probability of future parking space vacancy): Taken from the complement of the probability of future parking space occupancy in the situation information (1 - occupancy probability), priority is given to allocating parking spaces that are more likely to remain vacant in the future in order to improve turnover rate.

[0145] F4 (Parking Space Attribute Matching): Quantifies the degree of fit between parking space attributes (such as charging pile power, whether it is an accessible parking space, and the quietness of the area) and user preferences (static and dynamic).

[0146] w1, w2, w3, and w4 are dynamically configurable weight coefficients.

[0147] Optimal decision: Select the parking space with the highest comprehensive evaluation value and immediately mark its status as "allocated and locked" through the internal system interface.

[0148] Path Calculation Unit: This unit is responsible for calculating the optimal path in a dynamic environment. Its workflow is as follows:

[0149] Dynamic cost map construction: Using a static topology map of the parking lot as the base map, and combining real-time sensor data (current traffic speed) and dynamic situational information within the parking lot (predicted travel time and congestion risk), a comprehensive travel cost is dynamically assigned to each road segment (edge) in the map. The cost calculation integrates static physical length, real-time travel time cost, predicted congestion impact, and a travel difficulty coefficient based on vehicle attributes (such as large vehicles needing to avoid narrow curves).

[0150] Optimal path search: The A* algorithm or its dynamic version, D*Lite (both path planning algorithms), is used to search on a constructed dynamic cost map. Starting from the vehicle's real-time location (provided by a positioning beacon) and ending at the assigned optimal parking space, it seeks the path with the minimum overall toll cost. The output is a series of ordered lane nodes and turning commands.

[0151] The monitoring and replanning module includes an anomaly monitoring unit for continuously monitoring traffic flow and parking space status, and a replanning triggering unit for triggering the reallocation of parking spaces or the planning of routes when certain conditions are met.

[0152] Anomaly Detection Unit: This unit continuously monitors two key aspects: First, it tracks in real time whether a vehicle deviates from the planned path or stops abnormally, using data from the vehicle positioning beacon unit; second, it monitors whether the traffic congestion index on the planned path exceeds the threshold and whether the target parking space is unexpectedly occupied, using real-time data streams from the sensor network unit.

[0153] Replanning Trigger Unit: This unit is immediately triggered when the anomaly detection unit detects any of the following situations: sudden severe congestion ahead of the planned path; the target parking space is occupied by other vehicles; the vehicle deviates significantly from the path and cannot automatically return. After triggering, this unit sends an interrupt signal to the system core, and using the vehicle's latest real-time position as the new starting point, carrying the latest environmental data, it re-invokes the decision-making unit of the dynamic parking space allocation module and the path calculation unit of the real-time path planning module to generate a new "parking space, path" combination scheme, and delivers it to the guidance execution unit.

[0154] The billing and status update module is responsible for the closed-loop operation of parking services and the synchronous maintenance of system resource status.

[0155] Billing and Status Update Unit: This unit is the endpoint of the business logic. It works deeply with the site perception module. Parking begins when a vehicle is confirmed to have entered an assigned parking space (through video recognition or changes in geomagnetic sensor status combined with location information). Parking ends when the vehicle leaves (through exit gate recognition or changes in parking space status). The unit automatically calculates the fee based on parking duration, parking space type (e.g., different rates for charging spaces), and preset fee rules, and deducts the fee through an integrated third-party payment gateway (e.g., WeChat Pay, Alipay). Simultaneously, this unit is responsible for updating key status changes such as parking space status (from "occupied" to "vacant") and vehicle departure events to the system's central database, ensuring global consistency.

[0156] Application Cases (Reference) Figure 1-5 Before setting off, Mr. Wang opened the smart parking app on his phone and entered his destination, "City Center Shopping Mall." After obtaining the user's authorization, the app (user interface unit) automatically retrieved his vehicle information (license plate number, vehicle type: SUV, new energy vehicle) and common preferences ("Need charging station", "Prioritize proximity to elevator"). He clicked "Find parking space." The app sent this parking request (including destination, vehicle attributes, and user preferences) to the cloud.

[0157] Upon receiving the request, the cloud-based recommendation engine first searches for all connected parking lots within a 1-kilometer radius of the destination coordinates, retrieving candidate parking lot information, including: location, total number of parking spaces, current number of available spaces, availability of charging stations, pricing, and estimated walking distance from each mall entrance to the target shopping area. Next, the engine performs a comprehensive scoring based on Mr. Wang's preferences: the parking lot with ample available charging stations and the shortest walking distance from the parking garage elevator to the mall's core area receives the highest score. The system identifies this parking lot as the target parking lot, reserves a "charging space qualification" for him, and pushes the reservation information and navigation route to Mr. Wang's app.

[0158] Situational awareness and prediction within the field (corresponding to S3)

[0159] As Mr. Wang drove there, the sensor network units of the target parking lot were continuously working: video parking cameras identified the occupancy status of each parking space; wide-angle cameras analyzed the traffic flow speed on each main road; and the charging pile manager reported the real-time status of each charging pile. This real-time multi-dimensional status data (parking space status, traffic flow, charging pile status) was aggregated to the local edge server. The edge server then uploaded the processed data to the cloud in real time. The predictive analysis unit in the cloud's data fusion and situation prediction module began to work. It called up the parking lot's historical data from the past month (such as the parking space occupancy patterns in different areas at different times of the day) and combined it with the newly uploaded real-time data to run a spatial-temporal fusion prediction model. This model abstracted the parking lot structure into a topological map, used a graph neural network (GNN) to analyze spatial relationships (such as congestion in area B affecting area C), and used an LSTM time series prediction model to analyze temporal patterns. A few seconds later, it generated a dynamic situation information of the parking lot for the next 10 minutes: one was a heat map of parking space occupancy probability, showing which parking spaces were likely to be occupied soon; the other was a channel passage time prediction map, showing which paths were about to become congested.

[0160] Dynamic parking space allocation and real-time route planning (corresponding to S4 and S5)

[0161] Mr. Wang's car entered the parking lot entrance. The system immediately triggered the dynamic parking space allocation module. Its decision-making unit first filtered out large-sized parking spaces suitable for SUVs from all available charging spaces, forming a candidate parking space list. Then, it calculated a comprehensive evaluation value for each candidate parking space: based on Mr. Wang's real-time location and the predicted travel time in the situation information, it calculated the estimated travel time to each parking space; based on the parking lot map, it calculated the walking distance from each parking space to the mall elevator entrance; considering the parking space occupancy probability, it selected those parking spaces that were more likely to remain vacant in the future; finally, the parking space attribute (fast charging pile) perfectly matched Mr. Wang's preference of "needing a charging pile". Through weighted calculation, the system selected the parking space with the highest comprehensive score as the optimal parking space and immediately locked its status. Immediately afterwards, the path calculation unit of the real-time path planning module started. It planned a path on the parking lot digital map with the vehicle's current location as the starting point and the optimal parking space as the destination. The planning process considers not only distance but also a comprehensive set of factors: the actual length of the road segment, the current travel time based on real-time camera data, the predicted congestion impact based on situational information, and the increased difficulty of SUVs navigating narrow bends. Ultimately, it plans a route with the lowest overall travel cost.

[0162] Guided and dynamic replanning (corresponding to S6)

[0163] The optimal parking space (e.g., "Charging Space C12 in Zone D") and the planned route were immediately sent to Mr. Wang's app. The app interface displayed a dynamic map, guiding him with a blue highlighted line, while simultaneously providing a voice prompt: "Please go straight, turn left after 50 meters." At the same time, the guidance screens in the parking lot and the indicator light on parking space C12 in Zone D illuminated, indicating the direction. During the guidance process, the anomaly monitoring unit of the status monitoring and replanning module was continuously working. It tracked Mr. Wang's vehicle position via Bluetooth beacons and monitored the traffic flow ahead of the planned route. Suddenly, due to a brief vehicle malfunction ahead, the congestion index at an intersection on Mr. Wang's planned route spiked, exceeding the threshold. The replanning trigger unit was immediately activated. Starting from Mr. Wang's vehicle's latest position and considering the latest situation in the parking lot, the system instantly re-executed the parking space allocation and route planning, assigning him another nearby available charging space (C15) with similar conditions and planning a new route to bypass the congestion. The new guidance instructions were immediately updated on Mr. Wang's app.

[0164] Parking completed and billing (corresponding to S7): Mr. Wang successfully parked his car in the newly assigned parking space C15. The video parking camera above the space recognized the vehicle, and the system confirmed the parking status. After finishing his shopping, Mr. Wang clicked "One-Click Leave" on the app. The system automatically calculated the parking time and charging fee, and deducted the payment through contactless payment. The barrier gate automatically opened, and the status of parking space C15 was updated to "Idle," awaiting the next user.

[0165] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A collaboratively dynamically guided intelligent parking garage management method, characterized in that, Includes the following steps: S1. Receive a parking request from the user's terminal, and based on the destination information in the request, obtain information on multiple candidate parking lots in the vicinity; S2. Based on user preference information and the information of the candidate parking lots, determine and reserve a target parking lot for the user; S3. Obtain real-time multi-dimensional status data within the target parking lot, and perform fusion analysis based on the data and historical data to generate and continuously update dynamic status information within the parking lot that characterizes future parking space occupancy and traffic conditions; S4. In response to a vehicle entering the target parking lot, the optimal parking space is dynamically allocated to the vehicle based on multi-objective decision-making, taking into account the dynamic situation information in the parking lot, vehicle attributes, and user preferences. S5. Based on the dynamic situation information in the field and vehicle attributes, dynamically plan the driving path for the vehicle from its current position to the optimal parking space; S6. Send the optimal parking space information and the driving route to the user terminal for guidance, and dynamically adjust the guidance strategy according to environmental changes during the guidance process; S7. Update parking space status and complete billing operation.

2. The intelligent parking garage management method with collaborative dynamic guidance according to claim 1, characterized in that: In step S3, the fusion analysis is achieved by modeling the spatial topology of the parking lot to capture spatial dependencies and combining historical and real-time time series data analysis to capture temporal patterns.

3. The intelligent parking garage management method with collaborative dynamic guidance according to claim 1, characterized in that, The step S4, which dynamically allocates the optimal parking space to the vehicle based on multi-objective decision-making, includes: selecting candidate parking spaces that meet the vehicle's physical constraints from the currently available parking spaces; A comprehensive evaluation value is calculated for each candidate parking space by calculating a weighted average. The comprehensive evaluation value includes at least the estimated travel time based on real-time and predicted traffic flow, the walking distance to the user's preferred facility, the future vacancy probability of the parking space based on the dynamic situation information in the site, and the matching degree between the parking space attributes and the user's preferences. The parking space with the highest overall evaluation value is assigned as the optimal parking space.

4. The intelligent parking garage management method with collaborative dynamic guidance according to claim 1, characterized in that, In step S5, when dynamically planning a driving path for a vehicle, the path cost assessment comprehensively considers the static physical length of the road segment, the travel time cost obtained based on real-time sensor data, the predicted congestion impact based on the dynamic situation information in the field, and the travel difficulty coefficient based on vehicle attributes.

5. The intelligent parking garage management method with collaborative dynamic guidance according to claim 1, characterized in that, In step S6, the dynamic adjustment of the guidance strategy according to environmental changes includes: continuously monitoring the traffic flow and target parking space status during vehicle travel; if the congestion level of the planned route exceeds a specified threshold or the parking space status is abnormal, a replanning process is triggered, using the vehicle's current position as a new starting point, and re-executing parking space allocation and route planning based on the latest environmental data.

6. A collaborative dynamic guidance intelligent parking garage management system for implementing the method of any one of claims 1 to 5, characterized in that, include: The request processing and recommendation module, deployed in the cloud, is used to execute steps S1 and S2; The in-park sensing module is deployed in a distributed manner inside the parking lot to execute step S3; The data fusion and situation prediction module, deployed in the cloud, is used to execute step S3; The dynamic parking space allocation module, deployed in the cloud, is used to execute step S4; The real-time path planning module, deployed in the cloud, is used to execute step S5; The status monitoring and replanning module, deployed in the cloud, is used to execute step S6; The billing and status update module, deployed in the cloud, is used to execute step S7; The request processing and recommendation module, the on-site perception module, the data fusion and situation prediction module, the dynamic parking space allocation module, the real-time route planning module, the status monitoring and replanning module, and the billing and status update module interact with each other and transmit instructions through the network to collaboratively complete the dynamic guidance and management of the entire process from receiving user requests to completing billing.

7. The collaborative dynamic guidance intelligent parking garage management system according to claim 6, characterized in that: The request processing and recommendation module includes a user interface unit and a recommendation engine unit; The user interface unit is used to interact with the user terminal to receive structured parking request data packets and transmit the data packets to the recommendation engine unit; The recommendation engine unit is used to calculate and determine the target parking lot based on the destination information, user preferences and multi-source external data in the data packet, and send the reservation result to the user terminal through the user interface unit; The data fusion and situation prediction module includes a sensor network unit and a predictive analysis unit; The sensor network units are distributed and deployed in the parking lot to collect multi-dimensional status data in real time; The predictive analysis unit is used to receive and fuse real-time data and historical data uploaded by the sensor network unit, and generate dynamic situation information in the field through a spatiotemporal fusion predictive model. The dynamic parking space allocation module and the real-time route planning module include a decision-making unit and a route calculation unit; The decision-making unit is used to receive the dynamic situation information in the field, vehicle attributes and user preferences, and output the optimal parking space allocation result based on the multi-objective decision-making model. The path calculation unit is used to receive the optimal parking space information, the real-time location of the vehicle and the dynamic situation information in the field, and calculate the driving path based on the dynamic cost map. The status monitoring and replanning module includes an anomaly detection unit and a replanning trigger unit; The anomaly detection unit is used to continuously receive feedback from the sensor network and vehicle positioning data to monitor traffic flow and parking space status. The replanning triggering unit is used to trigger the decision unit and the path calculation unit to re-execute the calculation when the anomaly detection unit determines that the replanning conditions are met.

8. The collaborative dynamic guidance intelligent parking garage management system according to claim 7, characterized in that, The sensor network unit in the field sensing module includes: The parking space status sensing subunit is deployed in each parking space to detect the occupancy status of the parking space and perform license plate recognition; Traffic flow sensing subunits are deployed on key routes to collect data on traffic volume, vehicle speed, and queue length. The environment and event perception subunit is used to collect facility status and safety event data; The vehicle positioning beacon subunit is deployed on-site to provide real-time positioning data for vehicles.

9. The collaborative dynamic guidance intelligent parking garage management system according to claim 7, characterized in that, The real-time route planning module plans the driving route in the following ways: A dynamic cost map is constructed, in which the cost of each road segment integrates static physical length, travel time based on real-time sensor data, predicted congestion impact based on the dynamic situation information in the field, and travel difficulty coefficient based on vehicle attributes. The D*Lite algorithm is used to search for the optimal path from the vehicle's current position to the optimal parking space on the dynamic cost map.

Images