Intelligent vehicle cooperative parking method and system, electronic device and readable storage medium

Abstract

This application provides an intelligent vehicle cooperative parking method, system, electronic device, and readable storage medium, relating to the fields of intelligent transportation and vehicle-to-everything (V2X) technology. The method includes: receiving parking space status information packets broadcast from other vehicles via near-field communication; acquiring environmental perception information around the vehicle using onboard environmental perception sensors; generating and maintaining a local dynamic map based on the parking space status information packets and environmental perception information using a distributed dynamic map construction algorithm; and planning a driving path to the target parking space based on the dynamic map using a cooperative optimization path planning algorithm. This application, through vehicle-to-vehicle direct communication and information fusion, enables self-organized cooperative parking for vehicles in indoor parking environments without a central node or public network coverage. This effectively overcomes information silos, reduces redundant searches and channel congestion, lowers system construction costs, and improves parking efficiency and system robustness.

Description

Technical Field

[0001] This invention relates to the fields of intelligent transportation and vehicle networking technology, and in particular to an intelligent vehicle cooperative parking method, system, electronic device, and readable storage medium. Background Technology

[0002] With the continuous growth of car ownership, problems such as drivers having difficulty finding available parking spaces, traffic congestion, and increased energy consumption are becoming increasingly prominent in complex indoor environments such as large underground parking lots and multi-story parking garages.

[0003] To address these issues, the industry has proposed several intelligent parking guidance solutions, which can be broadly categorized into three types: First, centralized guidance systems based on infrastructure, which utilize ultrasonic sensors or video detectors deployed throughout the parking lot, combined with a central server for data processing and parking space status updates. Second, mobile applications based on cloud platforms, which use cellular networks to upload vehicle location and parking information to the cloud, and then receive guidance instructions from the cloud. Third, "vehicle-road-cloud" collaborative solutions that combine roadside units (RSUs), which collect information through roadside equipment and distribute it to vehicles.

[0004] However, the aforementioned existing technologies have significant limitations in practical applications: First, GPS signals fail indoors, and commercial-grade indoor positioning technologies such as Bluetooth beacons and Ultra-Wideband (UWB) are expensive to deploy and maintain, making widespread application difficult. Second, each vehicle's independent parking search creates information silos, preventing the sharing of detected parking space status, leading to multiple vehicles repeatedly searching the same area, exacerbating congestion and energy waste. Furthermore, ultrasonic or video-based parking space detection systems require comprehensive sensor deployment and a central server, resulting in high system construction costs, difficulties in expansion and modification, and system paralysis if the central node fails. Additionally, unstable cellular network (4G / 5G) signals in areas such as underground parking lots prevent cloud-based parking applications from functioning reliably. As for the "vehicle-road-cloud" collaborative architecture, relying on roadside units or a central server for data aggregation and decision-making fails to address how vehicles can collaboratively complete parking tasks through self-organization under extreme conditions without a central node or reliable public network coverage. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide an intelligent vehicle collaborative parking method, system, electronic device and readable storage medium to solve the technical problems of repeated searches, channel congestion caused by information silos, system fragility and high transformation costs caused by infrastructure dependence.

[0006] In a first aspect, embodiments of the present invention provide an intelligent vehicle cooperative parking method, comprising: Receive parking space status information packets broadcast from other vehicles via near-field communication; The vehicle obtains environmental perception information about the vehicle's surroundings through onboard environmental perception sensors. Based on the parking space status information package and the environmental perception information, a distributed dynamic map construction algorithm is executed to generate and maintain a local dynamic map; wherein, the dynamic map contains status information of at least one parking space; Based on the dynamic map, a collaborative optimization path planning algorithm is executed to plan a driving route to the target parking space.

[0007] In one implementation, the execution of the distributed dynamic map construction algorithm includes: calculating a fusion weight for each received parking space status information packet; the fusion weight is determined based on the source, confidence level, and timeliness of the parking space status information packet; and according to the fusion weight, fusing the parking space status information in the parking space status information packet into the corresponding parking space record in the local dynamic map.

[0008] In one implementation, when the environmental perception information acquired by the vehicle conflicts with the parking space status recorded in the dynamic map, the dynamic map is updated based on the vehicle's perception information, and a high-confidence parking space status information packet is generated and broadcast.

[0009] In one implementation, the step of executing a collaborative optimization path planning algorithm based on the dynamic map to plan a driving route to the target parking space includes: constructing a dynamic cost map, digitizing the parking area into a topological map, and calculating the dynamic comprehensive cost for each travel edge; selecting parking spaces in an vacant state from the local dynamic map as candidate targets; for each candidate target, searching for the path with the minimum comprehensive cost on the dynamic cost map; and selecting the path with the minimum cost as the driving route to the target parking space.

[0010] In one implementation, the formula for calculating the dynamic comprehensive cost is:

[0011] in, For the passing edge (from node) i To the node j The dynamic comprehensive cost of ) The physical length of the passing edge. This refers to the distance weighting coefficient. This represents the real-time traffic density of the access edge. This represents the congestion penalty coefficient. For nodes j The proportion of nearby parking spaces in unknown condition. Penalty coefficient for unknown regions; As a synergistic utility factor, This is the reward coefficient.

[0012] In one embodiment, the method further includes: after determining the target parking space, generating an intent information packet containing the target parking space and / or the planned route, and broadcasting the intent information packet to other vehicles via near-field communication.

[0013] In one embodiment, the method further includes: during the driving process, when it is detected that the target parking space is occupied, or when an intent information packet from another vehicle is received indicating that it will arrive at the same target parking space faster, triggering the re-execution of the collaborative optimization path planning algorithm to reselect the target parking space and plan the route.

[0014] In one embodiment, the method further includes: activating the system and connecting to a distributed self-organizing network composed of other vehicles when the vehicle enters the target area.

[0015] In one implementation, the parking space status information package includes at least the parking space identifier, status, confidence level, timestamp, and lifetime.

[0016] In one embodiment, the method further includes: after the vehicle parks in the target parking space, confirming the parking is complete through an onboard environmental perception sensor, updating the parking space status in the local dynamic map to occupied, and broadcasting the final parking space status information packet.

[0017] Secondly, embodiments of the present invention also provide an intelligent vehicle cooperative parking system, the system comprising: The near-field communication module is used to receive parking space status information packets broadcast from other vehicles via near-field communication. The local environment perception module is used to acquire environmental perception information around the vehicle through onboard environment perception sensors. The map generation and maintenance module is used to execute a distributed dynamic map construction algorithm based on the parking space status information package and the environmental perception information to generate and maintain a local dynamic map; wherein the dynamic map contains status information of at least one parking space; The route planning module is used to execute a collaborative optimization route planning algorithm based on the dynamic map to plan a driving route to the target parking space.

[0018] Thirdly, embodiments of the present invention also provide an electronic device, including a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method described in any one of the first aspects.

[0019] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing computer-executable instructions, which, when invoked and executed by a processor, cause the processor to implement the method described in any one of the first aspects.

[0020] This invention provides an intelligent vehicle cooperative parking method, system, electronic device, and readable storage medium. The method includes: receiving parking space status information packets broadcast from other vehicles via near-field communication; acquiring environmental perception information around the vehicle via an onboard environmental perception sensor; generating and maintaining a local dynamic map by executing a distributed dynamic map construction algorithm based on the parking space status information packets and the environmental perception information; and planning a driving path to the target parking space by executing a cooperative optimization path planning algorithm based on the dynamic map.

[0021] This application addresses the problems in existing technologies, such as high system construction costs, difficulties in modification and expansion, high risk of central node failure, and instability of cellular network signals, which arise in parking environments without a central node or reliable public network coverage. These problems stem from vehicles' reliance on infrastructure such as roadside units (RSUs), cloud servers, or full-coverage sensors. Furthermore, the application addresses the technical issues of information silos formed by the inability of vehicles to share information, leading to repeated searches by multiple vehicles, exacerbating traffic congestion, and wasting energy. This application achieves the technical effect of efficient parking search through self-organizing network collaboration under extreme conditions where GPS signals fail and there is no infrastructure support.

[0022] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0023] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0024] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0025] Figure 1 This is an architecture diagram of an intelligent vehicle cooperative parking system provided in an embodiment of the present invention; Figure 2A flowchart illustrating an intelligent vehicle cooperative parking method provided in an embodiment of the present invention; Figure 3 A flowchart of a distributed dynamic map construction algorithm provided in an embodiment of the present invention; Figure 4 A schematic diagram of collaborative path planning cost provided in an embodiment of the present invention; Figure 5 An example diagram illustrating the workflow of a vehicle from entering the parking lot to parking, provided as an embodiment of the present invention; Figure 6 A timing diagram for vehicle-to-vehicle collaboration via intent broadcasting is provided as an embodiment of the present invention. Figure 7 A structural diagram of another intelligent vehicle cooperative parking system provided in an embodiment of the present invention; Figure 8 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Currently, to address the parking difficulties in complex indoor environments such as large underground parking lots and multi-story parking garages, existing technologies mainly fall into three categories: First, centralized parking guidance systems based on full-coverage sensors and central servers. These require the deployment of numerous ultrasonic or video detectors within the parking lot, resulting in high construction costs and system paralysis if the central node fails. Second, cloud-based parking applications relying on cellular networks upload vehicle location and parking information to the cloud for guidance. However, unstable cellular signals in areas like underground parking lots prevent reliable operation. Third, the "vehicle-road-cloud" collaborative solution proposed in recent years aggregates information and distributes it to vehicles through roadside units (RSUs), but still relies on infrastructure and public network support. A common drawback of these existing technologies is their reliance on fixed infrastructure (sensor networks, central servers, roadside units) or public network coverage, rendering them inoperable in indoor parking environments without central nodes or reliable network signals. Meanwhile, the lack of direct information sharing mechanisms between vehicles creates information silos, leading to multiple vehicles repeatedly searching in the same area, exacerbating traffic congestion and energy waste. Other technologies involve vehicle-road cooperation, but none of them have solved the problem of vehicles coordinating to complete parking tasks through self-organization under completely decentralized conditions without infrastructure support.

[0028] Based on this, the present application provides an intelligent vehicle collaborative parking method, system, electronic device and readable storage medium, which, through vehicle-to-vehicle direct communication and information collaborative integration, enables vehicles to self-organize collaborative parking in indoor parking environments without central nodes or public network coverage, effectively overcoming information silos, reducing redundant searches and channel congestion, reducing construction costs, and improving parking efficiency and system robustness.

[0029] like Figure 1 The diagram illustrates the architecture of an intelligent vehicle cooperative parking system. Physically, the system consists of two types of entities: intelligent vehicle terminals (hereinafter referred to as "vehicle terminals") and optional lightweight parking space beacons. The vehicle terminal is the core of the system, installed on each participating vehicle; the parking space beacon is a low-cost auxiliary device that can be deployed next to a parking space. The system operates through a fundamental innovation in its network architecture—a fully decentralized self-organizing vehicle network (VANET)—where vehicles exchange information via near-field direct communication, jointly building and maintaining the global state.

[0030] The vehicle terminal is an integrated on-board unit, and its main internal components and connections are as follows: Near Field Communication (NFC) Module: This module employs hardware supporting vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication, such as a 3GPP-standard C-V2X (Cellular Vehicle-to-Everything) PC5 (ProSe Communication 5) interface module, or an IEEE 802.11p-based DSRC module. This module is responsible for broadcasting and receiving agreed-upon data packets and serves as the sole wireless interface for the vehicle to exchange information with its external environment.

[0031] Local environment perception module: This includes at least one forward-facing visual camera and one short-range millimeter-wave radar. The camera is used to identify parking lines, obstacles within the parking space, and to read parking beacon IDs; the radar is used to accurately measure the distance to nearby vehicles or objects. The data from both are fused within the processor.

[0032] Positioning and Inertial Navigation Unit: This unit includes a microelectromechanical system (MEMS) inertial measurement unit (IMU) and a vehicle controller area network (CAN) bus interface. The IMU provides triaxial acceleration and angular velocity data; the CAN bus interface receives the vehicle's own wheel speed pulse signals. This unit provides dead reckoning (DR) capabilities in the absence of GPS.

[0033] The onboard processing unit (ATU) is the "brain" of the terminal, typically an embedded computing device containing a central processing unit (CPU), a graphics processing unit (GPU), and memory. It is directly connected to the communication module, sensing module, and positioning unit via an internal bus (such as PCIe (Peripheral Component Interconnect Express) or USB (Universal Serial Bus)) or Ethernet, running the core distributed dynamic map building algorithm and collaborative optimization path planning algorithm of this application.

[0034] Human-machine interface: This is usually the vehicle's central touch screen, connected to the processing unit. It is used to receive the user's "start parking search" command and display the recommended navigation route, target parking space number, and real-time dynamic map in a graphical manner.

[0035] Composition and working principle of lightweight parking space beacons: A parking space beacon is a simple device that includes: Low-power communication modules, such as Bluetooth Low Energy (BLE) modules, periodically (e.g., once per second) broadcast a signal containing its unique ID.

[0036] Status trigger switch: Employs a low-power geomagnetic sensor. Its working principle is as follows: when there is no car, the sensor detects the Earth's background magnetic field; when a car parks, the metal body causes a significant change in the magnetic field. The sensor converts this change into a voltage level signal.

[0037] Microcontroller: Connects the communication module and the status trigger switch. Its operating logic is as follows: it continuously reads the switch's signal; when the signal indicates "car present," it controls the communication module to add an "OCCUPIED" status flag to its broadcast packet; when the signal indicates "no car present," it adds a "FREE" status flag. The parking space beacon itself does not possess any computational fusion or path planning capabilities; it only provides the most basic status perception and broadcasting.

[0038] To facilitate understanding of this embodiment, a detailed description of an intelligent vehicle cooperative parking method disclosed in this embodiment of the invention will be provided first. (See [link to relevant documentation]). Figure 2 The diagram shows a flowchart of a cooperative parking method for intelligent vehicles, which may include the following steps: S202 receives parking space status information packets broadcast from other vehicles via near-field communication.

[0039] Near-field communication (NFC) refers to short-range direct communication technologies in the vehicle-to-everything (V2X) field, such as the C-V2X PC5 interface (based on 3GPP Release 14 and above standards), DSRC (Dedicated Short-Range Communications), and IEEE 802.11p. Parking space status information packets at least include parking space identifier, status, confidence level, timestamp, and time-to-live (TTL). Parking space identifier (e.g., "B2-015"); status (idle / occupied / unknown); confidence level (between 0 and 1, indicating the credibility of the information source); timestamp (used for timeliness judgment); TTL (used for discarding expired data). Parking space status information may also include optional fields: source vehicle ID, location information, route information, etc.

[0040] For example, this can be achieved through a C-V2X PC5 interface module or DSRC module installed on the vehicle: when the vehicle enters the parking lot, its near-field communication module continuously listens to the preset channel and passively receives parking space status information packets periodically broadcast by other vehicles without establishing a connection. Each parking space status information packet contains at least the following fields: unique parking space identifier, vacancy or occupancy status, confidence level, timestamp, and lifespan.

[0041] For example, in the initial stage of a vehicle entering an underground parking garage, the vehicle can receive dozens of parking space status information packets broadcast by multiple vehicles that have entered the garage before it within a few seconds, which is used to quickly initialize the local dynamic map. During the journey, the vehicle continuously receives real-time updated parking space status information packets in order to detect changes in parking space status in a timely manner. This receiving process is completed entirely within the vehicle-to-vehicle direct communication range, without relying on cellular networks, roadside units or any central nodes, and is a one-sided execution behavior, without the need to establish handshake confirmation with other vehicles.

[0042] For example, each parking space status information packet (ParkingDataPacket) must contain metadata for fusion, in addition to basic information: json {"parkingSpotId": "B2-015", / / Unique identifier for parking space} "status": "FREE", / / status: FREE / OCCUPIED / RESERVED "confidence": 0.92, / / Confidence level (0-1), determined by the source of information and the way it is perceived. "timestamp": 1648567890.123, / / High-precision timestamp "sourceVehicleId": "Veh_7A3", / / Source vehicle ID (if self-sensing) "ttl": 30, / / Time to live (seconds), requires reconfirmation if timeout occurs. "positionHint": {"x": 120.5, "y": 45.2, "zone": "B2"} / / Relative or absolute position reference} S204 acquires environmental perception information about the vehicle's surroundings through onboard environmental perception sensors.

[0043] Among them, vehicle-mounted environmental perception sensors refer to hardware devices installed on vehicles to collect information about the physical environment around the vehicle, including at least a visual camera and a short-range millimeter-wave radar. The visual camera is used to acquire image information, and the short-range millimeter-wave radar is used to accurately measure the distance, relative speed, and angle between the vehicle and surrounding objects. Environmental perception information refers to structured data collected and processed by the above sensors, including at least the existence of parking spaces, parking space status (vacant / occupied), unique identifier of the parking space, obstacle information, and relative positional relationship with nearby vehicles or objects.

[0044] For example, this is accomplished collaboratively by multiple types of environmental perception sensors integrated on the vehicle: a vision camera installed at the front of the vehicle collects image data in real time, and uses image recognition algorithms to identify parking space lines, obstacles within the parking space, and markings on parking space beacons to determine whether the parking space is vacant and obtain a unique parking space identifier; simultaneously, a short-range millimeter-wave radar continuously transmits and receives reflected signals to accurately measure the distance and relative speed between the vehicle and nearby vehicles, walls, or obstacles; the onboard processing unit synchronizes and aligns the multi-source data acquired by the camera and radar in time and space, and uses sensor fusion algorithms to comprehensively determine the parking space status and surrounding environmental information. For example, when the vehicle drives near a parking space, the camera identifies the parking space line and determines that there is no vehicle in the parking space, while the radar confirms that there are no obstacles in the area and the distance measurement value matches the characteristics of an vacant parking space, thus generating environmental perception information that includes the parking space identifier and the "vacant" status.

[0045] S206, based on parking space status information packets and environmental perception information, executes a distributed dynamic map construction algorithm to generate and maintain a local dynamic map.

[0046] The dynamic map contains the status information of at least one parking space. The distributed dynamic map construction algorithm refers to the process of integrating multi-source, potentially conflicting local information into a globally consistent parking space status map based on parking space status information packets and environmental perception information, using mechanisms such as weighted fusion, time decay, and conflict detection. The core of this algorithm lies in calculating the fusion weight through source weights, confidence levels, and timeliness factors, and handling information conflicts through a state accumulation decision mechanism to achieve decentralized collaborative perception. The local dynamic map refers to the data structure stored in the memory of each vehicle's onboard processing unit, used to record the status information of multiple parking spaces in the parking lot. Each parking space record includes at least a unique parking space identifier, current status (idle / occupied / unknown), fusion weight, and timestamp.

[0047] In one implementation, when the environmental perception information acquired by the vehicle conflicts with the parking space status recorded in the dynamic map, the dynamic map is updated based on the vehicle's perception information, and a high-confidence parking space status information packet is generated and broadcast.

[0048] For example, this step is executed independently by the vehicle's onboard processing unit: First, the processing unit creates an empty dynamic map in memory, using a grid-topology hybrid representation to structurally model the parking area. When a parking space status information packet is received, the processing unit extracts fields such as parking space identifier, status, confidence level, timestamp, and lifetime from the packet, and calculates fusion weights based on the source and timeliness of the information packet. If the parking space record does not exist in the dynamic map, it is stored directly; if it exists, weighted fusion is performed, and the cumulative weights of the "idle" and "occupied" states are compared to determine the final state. Meanwhile, the processing unit continuously receives environmental perception information output from the vehicle's environmental perception sensors. If the perception results conflict with the records in the fused map, the map is updated immediately based on the vehicle's perception, and a high-confidence information packet is generated and broadcast externally. In addition, the processing unit periodically traverses the local dynamic map, performs time decay calculations on outdated records, and removes records with too low weight or expired lifespans to ensure the freshness and accuracy of the map information. For example, within seconds of a vehicle entering a parking lot, an initial map covering a large area can be quickly constructed by fusing information packets broadcast by multiple vehicles that have entered the parking lot beforehand. During the journey, it continuously fuses newly received information and vehicle perception information to update the parking space status in real time.

[0049] For example, the onboard processing unit maintains a local map, LocalParkingMap. When a new information packet is received, it is fused according to the following rules: New parking space information entry: If the parking space is not recorded in the local map, it is directly stored, and the information source and time are recorded. Existing parking space information update: A weighted fusion and time decay model is used.

[0050] Weight calculation: Weight W = W_source × W_confidence × e^(-λ × Δt); Wherein, W_source: Source weight. Self-vehicle perception (visual confirmation) has the highest weight (e.g., 1.0), followed by V2V information from high-reputation vehicles (e.g., 0.8), and anonymous parking space beacons have the lowest weight (e.g., 0.6). W_confidence: The confidence level inherent in the information packet. λ: Time decay factor. Δt is the difference between the current time and the timestamp of the information packet.

[0051] Status Decision: Calculate the cumulative weights of the "vacant" and "occupied" states for the same parking space. If the cumulative weight of a certain state exceeds a threshold (e.g., 0.7) and is higher than that of another state by a certain proportion (e.g., twice), then update the local map to that state; otherwise, mark it as "unknown state" and wait for further confirmation.

[0052] Conflict resolution mechanism: If the parking space status detected in real time by the vehicle's sensors (such as cameras) conflicts with the status recorded in the fused map, the real-time perception of the vehicle shall prevail, and a high-confidence information packet shall be generated and broadcast immediately to correct any possible group errors.

[0053] S208, based on a dynamic map, executes a collaborative optimization path planning algorithm to plan a driving route to the target parking space.

[0054] The collaborative optimization path planning algorithm, based on a dynamic cost map and the principle of maximizing group benefits, comprehensively evaluates geometric distance, traffic density, unknown area penalties, and collaborative utility rewards to obtain a driving path that minimizes individual parking costs and benefits group information gain. The target parking space refers to the specific parking space the vehicle plans to enter, ultimately selected from the candidate parking spaces with a "vacant" status and high confidence level chosen from the local dynamic map after evaluation by the collaborative optimization path planning algorithm, through comprehensive cost comparison. This parking space may be the geometrically closest vacant parking space, or it may be a non-nearest parking space with the optimal total cost after comprehensively considering traffic congestion and the value of exploring unknown areas.

[0055] For example, this step is initiated by the onboard processing unit after receiving the driver's parking search command: First, the processing unit digitizes the parking area into a topological graph structure, with intersections as nodes and lanes as edges. Based on the parking space status information recorded in the local dynamic map, it calculates the dynamic comprehensive cost for each passage edge. This cost comprehensively considers the physical length of the edge, real-time traffic density, the proportion of unknown parking spaces in the traversed area, and the collaborative utility factor. Subsequently, the processing unit selects parking spaces with a "vacant" status and high confidence from the local dynamic map as candidate targets. For each candidate target, it uses an improved A-Star algorithm to search for the path with the minimum comprehensive cost on the dynamic cost map.

[0056] For example, when multiple paths lead to available parking spaces, the algorithm might choose a path that is geometrically longer but passes through unknown areas, yields a cooperative utility reward, and avoids congested areas, rather than the shortest path. After the search is complete, the algorithm compares the combined costs of all candidate paths and selects the combination of paths with the lowest total cost as the final planning result. During the journey, the processing unit continuously monitors changes in the target parking space status and receives intent information packets from other vehicles. If it detects that the target parking space is occupied or that another vehicle's intent indicates it will arrive at the same space faster, it immediately triggers replanning, reselects a target from the remaining candidate parking spaces, and plans a new path, ensuring the system's dynamic adaptability and fault tolerance.

[0057] It should be noted that this application can achieve real-time parking space map construction and efficient parking route planning in closed parking lot environments with limited satellite signals and poor network coverage, based on vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) near-field communication through distributed collaboration. Compared with the shortcomings of existing technologies, the measures taken and the beneficial effects produced by this application can be seen in Table 1 below:

[0058] Table 1 In one implementation, a distributed dynamic map construction algorithm is executed, including: calculating a fusion weight for each received parking space status information packet; the fusion weight is determined based on the source, confidence level, and timeliness of the parking space status information packet; and according to the fusion weight, merging the parking space status information in the parking space status information packet into the corresponding parking space record in the local dynamic map.

[0059] In one implementation, a collaborative optimization path planning algorithm is executed based on a dynamic map to plan a driving route to the target parking space, including: constructing a dynamic cost map, digitizing the parking area into a topological map, and calculating the dynamic comprehensive cost for each travel edge; selecting parking spaces in an vacant state from the local dynamic map as candidate targets; for each candidate target, searching for the path with the minimum comprehensive cost on the dynamic cost map; and selecting the path with the minimum cost as the driving route to the target parking space.

[0060] For example, the parking lot is digitized as a grid map or topology map. The cost of each node (path point) or passage edge is determined by a combination of dynamic factors. The formula for calculating the dynamic comprehensive cost is:

[0061] in, For the passing edge (from node) i To the node j The dynamic comprehensive cost of ) The physical length of the passing edge. This refers to the distance weighting coefficient. This represents the real-time traffic density of the access edge. This represents the congestion penalty coefficient. For nodes j The proportion of nearby parking spaces in unknown condition. Penalty coefficient for unknown regions; As a synergistic utility factor, This is the reward coefficient.

[0062] In one implementation, the method further includes: after determining the target parking space, generating an intent information packet containing the target parking space and / or the planned route, and broadcasting the intent information packet to other vehicles via near-field communication.

[0063] For example, the vehicle uses an improved A* or D Lite algorithm to search for a path to the currently evaluated optimal candidate parking space on a dynamic cost map. Once the planning is complete, the vehicle broadcasts its "driving intention," including: {"intendedPath": [node sequence], "targetSpot": "B2-015", "eta": 60}.

[0064] Once nearby vehicles receive the information, they will mark the area involved in the route as "predicted high traffic" in the short term, thereby proactively avoiding or choosing off-peak times when planning their own routes, achieving non-conflict distributed collaboration.

[0065] In one implementation, the method further includes: during the driving process, when it is detected that the target parking space is occupied, or when an intent information packet from another vehicle is received indicating that it will arrive at the same target parking space faster, triggering the re-execution of the collaborative optimization path planning algorithm to reselect the target parking space and plan the route.

[0066] In one implementation, the method further includes activating the system and connecting to a distributed self-organizing network of other vehicles when the vehicle enters the target area.

[0067] In one implementation, the method further includes: after the vehicle parks in the target parking space, confirming the parking is complete through an onboard environmental perception sensor, updating the parking space status in the local dynamic map to occupied, and broadcasting the final parking space status information packet.

[0068] For example, such as Figure 3 The diagram shows a distributed dynamic map construction algorithm. After a vehicle enters the parking lot, the following steps can be performed: System initialization and map creation: When a vehicle enters the garage, the onboard processing unit creates an empty local dynamic map, LocalMap, in memory. This map uses a raster-topology hybrid representation.

[0069] Packet reception and preprocessing: The near-field communication module continuously monitors the channel. When a ParkingDataPacket is received, the on-board processing unit extracts all its fields.

[0070] Information validity verification: Check the Time-to-Live (TTL) field in the data packet. If it has expired, discard the packet and end the processing flow for this information; if it has not expired, proceed to the next step.

[0071] Calculate the information fusion weight W: Based on the source, confidence level, and timeliness of the information packets, calculate their weights for fusion. The formula for calculating weight W is: W = W_source × W_confidence × e^(-λ × Δt) Where: W_source is the source weight: if the information comes from direct identification by the vehicle's perception module, it takes the maximum value of 1.0; if it comes from V2V forwarding from other vehicle terminals, it takes 0.7; if it comes from a parking space beacon, it takes 0.5. W_confidence is the confidence level inherent in the data packet. λ is a preset time decay factor (e.g., 0.05). Δt is the difference between the current time and the timestamp in the data packet.

[0072] Map update decision: Check if a record corresponding to the parking space ID exists in the LocalMap. If it does not exist, create a new parking space record; if it exists, perform existing parking space information fusion and conflict detection.

[0073] New parking space record creation: Add the new parking space information (including status, location, and calculated weight W) as a new record to LocalMap, and then jump to dynamic map maintenance.

[0074] Existing parking space information fusion and conflict detection: The weight W of this information packet is added to the cumulative weight of its claimed state (e.g., "FREE"). Then, the cumulative weights of the "FREE" and "OCCUPIED" states are compared. If the cumulative weight of either state exceeds a preset threshold (e.g., 0.65) and exceeds the weight of the other state by at least 50%, the parking space state in the map is updated to this state. If the above conditions are not met, the parking space state is marked as "UNKNOWN".

[0075] Real-time conflict resolution based on vehicle perception: During the fusion process, if the information collected in real time by the vehicle perception module directly conflicts with the state recorded in the LocalMap, the processing unit immediately updates the map based on the vehicle perception result and generates a ParkingDataPacket with confidence=1.0 for external broadcasting, so as to achieve rapid self-correction of the system.

[0076] Dynamic map maintenance: The processing unit periodically traverses the LocalMap, calculates the time decay of the weights of all records (i.e., multiplies by e^(-λ × Δt)), and removes outdated records with weights below the lower limit (e.g., 0.1) or TTL timeouts to ensure the freshness of map information.

[0077] For example, such as Figure 4 The diagram illustrates a collaborative path planning cost. Based on a dynamic map, a collaborative optimization path planning algorithm is executed to plan a driving route to the target parking space. This may include the following steps: Construct a dynamic cost map: Digitize the parking lot as a topology graph (nodes represent intersections, edges represent passages). Calculate the dynamic comprehensive cost Cost(i, j) for each edge (from node i to node j) using the following formula: Cost(i, j) =α×Dist(i, j) +β×Density(i, j) +γ×Uncertainty(j)-δ×Utility(i, j) Where: Dist(i, j) is the physical length of the edge, and α is the distance weight coefficient. Density(i, j) is the real-time traffic density of the edge, predicted by analyzing multiple received IntentPackets, and β is the congestion penalty coefficient. Uncertainty(j) is the proportion of parking spaces in the "UNKNOWN" state in the vicinity of node j, and γ is the unknown area penalty coefficient. Utility(i, j) is the collaborative utility factor (an innovation of this algorithm). If edge (i, j) passes through or is close to the "UNKNOWN" area, this is a positive reward, and δ is the reward coefficient. This factor quantifies the behavior of exploring unknown areas and contributing group information into individual path benefits.

[0078] Filtering candidate parking spaces and calculating the optimal route: The onboard processing unit filters parking spaces with a status of "FREE" and high confidence from the LocalMap as candidate targets. For each candidate target, the improved A-Star algorithm is used to search for a path with the minimum total cost on the constructed dynamic cost map.

[0079] Optimal Decision and Intent Broadcasting: Compare the Total Cost of the routes corresponding to all candidate parking spaces, and select the objective-route combination with the minimum total cost as the final planning result. Subsequently, immediately generate and begin periodically broadcasting an IntentPacket containing information about this route.

[0080] Real-time replanning during driving: If any of the following situations occur during driving: (a) the perception module detects that the target parking space is already occupied; (b) an IntentPacket from another vehicle indicates that it will arrive at the same parking space faster, the processing unit immediately triggers replanning, returning to the steps of filtering candidate parking spaces and calculating the optimal route, reselecting the target from the remaining candidate parking spaces, and planning a new route. Figure 5 The diagram illustrates an example of a vehicle's workflow from entry to parking, thus ensuring the system's strong fault tolerance and dynamic adaptability.

[0081] For example, such as Figure 6 The diagram shown illustrates a sequence diagram of vehicle-to-vehicle coordination via intent broadcasting, combined with... Figure 5 The flowchart below illustrates the complete collaborative workflow of this application, using the example of a vehicle equipped with a vehicle terminal (hereinafter referred to as "this vehicle") entering a parking lot. The specific steps are as follows: System Activation and Network Access: The vehicle enters the underground parking garage entrance, where the GPS signal is lost. The driver clicks the "Smart Parking" button on the human-machine interface. The vehicle terminal powers on and activates, the near-field communication module begins monitoring the wireless channel, and connects to the distributed self-organizing network comprised of other vehicles in the garage.

[0082] Passive learning and map initialization: Within seconds, this vehicle receives several ParkingDataPackets (parking space status information packets) and IntentPackets (driving intention packets) broadcast from earlier-entering vehicles (such as vehicles A and B) via the near-field communication module. The onboard processing unit immediately invokes a distributed dynamic map building algorithm to fuse this information and quickly generate a wide-coverage initial local dynamic map, LocalMap.

[0083] Collaborative route planning decision: The onboard processing unit invokes a collaborative optimization route planning algorithm. Based on the current LocalMap, the algorithm calculates multiple routes to each candidate available parking space and their dynamic comprehensive costs. After comparison, the route to parking space "B2-015" is selected as the optimal solution with the lowest current comprehensive cost (although this route may not be the geometrically shortest, its total cost is higher due to the high congestion costs or lack of collaborative utility rewards of other routes).

[0084] Intent Broadcast and Navigation Initiation: After determining the target and route, the vehicle generates an IntentPacket containing the target parking space "B2-015" and the planned route, and begins periodically broadcasting it to the network via the near-field communication module to declare its intent. Simultaneously, the human-machine interface displays the navigation route to the driver, and the vehicle begins to travel along the planned route.

[0085] En route perception and real-time map updates: During driving, the vehicle's local environment perception module continuously operates. When it detects that parking space "B2-012" marked "UNKNOWN" on the map is actually vacant, the processing unit immediately updates the local map and broadcasts a high-confidence ParkingDataPacket, contributing new information to the group. When it receives an IntentPacket from oncoming vehicle C, predicting that the two will meet at the intersection ahead, the processing unit can slightly adjust its speed to achieve proactive avoidance and smooth passage.

[0086] Target Verification and Anomaly Handling: The vehicle is approaching the target parking space "B2-015". The perception module performs a final verification of the parking space status.

[0087] If the parking space is vacant (verification successful): proceed with the parking operation and final status confirmation steps.

[0088] If the parking space is already occupied (verification failed): This indicates that an unconnected vehicle parked in the space after the other vehicle's intention was broadcast. The processing unit immediately updates the parking space status to "OCCUPIED" and broadcasts it, while simultaneously triggering route replanning to reselect the optimal parking space and route based on the updated map.

[0089] Parking operation and final status confirmation: This vehicle has successfully parked in parking space "B2-015". The sensing module confirms the operation is complete.

[0090] Task closure and information synchronization: The onboard processing unit officially updates the parking space status in the LocalMap to "OCCUPIED" and broadcasts a final ParkingDataPacket with confidence=1.0. This provides the most reliable status change information to the entire vehicle network, completing the full closed loop of this parking search task from information perception and collaborative decision-making to result feedback, and providing the latest global status to other vehicles subsequently.

[0091] Based on the same concept, embodiments of the present invention provide an intelligent vehicle cooperative parking system, such as... Figure 7 The diagram shows another intelligent vehicle cooperative parking system 700, which includes: The near-field communication module 710 is used to receive parking space status information packets broadcast from other vehicles via near-field communication.

[0092] The local environment perception module 720 is used to acquire environmental perception information around the vehicle through the on-board environment perception sensor.

[0093] The map generation and maintenance module 730 is used to generate and maintain a local dynamic map by executing a distributed dynamic map construction algorithm based on parking space status information packets and environmental perception information; wherein the dynamic map contains status information of at least one parking space.

[0094] The route planning module 740 is used to execute a collaborative optimization route planning algorithm based on a dynamic map to plan a driving route to the target parking space.

[0095] In one implementation, the map generation and maintenance module 730 is further configured to calculate the fusion weight for each received parking space status information packet; the fusion weight is determined based on the source, confidence level, and timeliness of the parking space status information packet; and according to the fusion weight, the parking space status information in the parking space status information packet is fused into the corresponding parking space record in the local dynamic map.

[0096] In one implementation, the path planning module 740 is also used to construct a dynamic cost map, digitize the parking area into a topology map, and calculate the dynamic comprehensive cost for each travel edge; filter parking spaces in an vacant state from the local dynamic map as candidate targets; for each candidate target, search for the path with the minimum comprehensive cost on the dynamic cost map; and select the path with the minimum cost as the driving path to the target parking space.

[0097] The system provided in this embodiment of the invention has the same implementation principle and technical effects as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the system embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0098] This invention also provides an electronic device, such as... Figure 8 The diagram shows the structure of the electronic device, which includes a processor 81 and a memory 80. The memory 80 stores computer-executable instructions that can be executed by the processor 81. The processor 81 executes the computer-executable instructions to implement the above-mentioned intelligent vehicle cooperative parking method.

[0099] exist Figure 8 In the illustrated embodiment, the electronic device further includes a bus 82 and a communication interface 83, wherein the processor 81, the communication interface 83, and the memory 80 are connected via the bus 82.

[0100] The memory 80 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 83 (which can be wired or wireless), such as the Internet, wide area network, local area network, or metropolitan area network. The bus 82 may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus. The bus 82 can be divided into an address bus, a data bus, and a control bus. For ease of representation, Figure 8 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0101] Processor 81 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 81 or by software instructions. The processor 81 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this invention can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in the memory. The processor 81 reads the information in the memory and, in conjunction with its hardware, completes the steps of the intelligent vehicle cooperative parking method described in the foregoing embodiment.

[0102] This invention also provides a computer-readable storage medium storing computer-executable instructions. When these computer-executable instructions are invoked and executed by a processor, they cause the processor to implement the aforementioned intelligent vehicle cooperative parking method. For specific implementation details, please refer to the foregoing method embodiments, which will not be repeated here.

[0103] The computer program products of the intelligent vehicle cooperative parking method, device and electronic device provided in the embodiments of the present invention include a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.

[0104] Unless otherwise specifically stated, the relative steps, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0105] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0106] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0107] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for cooperative parking of intelligent vehicles, characterized in that, The method includes: Receive parking space status information packets broadcast from other vehicles via near-field communication; The vehicle obtains environmental perception information about the vehicle's surroundings through onboard environmental perception sensors. Based on the parking space status information package and the environmental perception information, a distributed dynamic map construction algorithm is executed to generate and maintain a local dynamic map; wherein, the dynamic map contains status information of at least one parking space; Based on the dynamic map, a collaborative optimization path planning algorithm is executed to plan a driving route to the target parking space.

2. The method according to claim 1, characterized in that, The execution of the distributed dynamic map construction algorithm includes: For each received parking space status information packet, its fusion weight is calculated; the fusion weight is determined based on the source, confidence level, and timeliness of the parking space status information packet. Based on the fusion weight, the parking space status information in the parking space status information package is fused into the corresponding parking space record in the local dynamic map.

3. The method according to claim 1, characterized in that, The method further includes: When the environmental perception information acquired by the vehicle conflicts with the parking space status recorded in the dynamic map, the dynamic map is updated based on the vehicle's perception information, and a high-confidence parking space status information packet is generated and broadcast.

4. The method according to claim 1, characterized in that, The step of executing a collaborative optimization path planning algorithm based on the dynamic map to plan a driving route to the target parking space includes: Construct a dynamic cost map, digitize the parking area into a topological map, and calculate the dynamic comprehensive cost for each passage edge; Select available parking spaces from the local dynamic map as candidate targets; For each candidate objective, search the path with the minimum overall cost on the dynamic cost map; Choose the route with the lowest cost as the driving route to the target parking space.

5. The method according to claim 4, characterized in that, The formula for calculating the dynamic comprehensive cost is as follows: in, For the passing edge (from node) i To the node j The dynamic comprehensive cost of ) The physical length of the passing edge. This refers to the distance weighting coefficient. This represents the real-time traffic density of the access edge. This represents the congestion penalty coefficient. For nodes j The proportion of nearby parking spaces with unknown status. Penalty coefficient for unknown regions; As a synergistic utility factor, This is the reward coefficient.

6. The method according to claim 1, characterized in that, The method further includes: After determining the target parking space, an intent information packet containing the target parking space and / or the planned route is generated, and the intent information packet is broadcast to other vehicles via near-field communication.

7. The method according to claim 1, characterized in that, The method further includes: During the journey, when the target parking space is detected to be occupied, or when the intention information packet of other vehicles is received indicating that they will arrive at the same target parking space faster, the collaborative optimization path planning algorithm is re-executed to reselect the target parking space and plan the route.

8. An intelligent vehicle cooperative parking system, characterized in that, The system includes: The near-field communication module is used to receive parking space status information packets broadcast from other vehicles via near-field communication. The local environment perception module is used to acquire environmental perception information around the vehicle through onboard environment perception sensors. The map generation and maintenance module is used to execute a distributed dynamic map construction algorithm based on the parking space status information package and the environmental perception information to generate and maintain a local dynamic map; wherein the dynamic map contains status information of at least one parking space; The route planning module is used to execute a collaborative optimization route planning algorithm based on the dynamic map to plan a driving route to the target parking space.

9. An electronic device, characterized in that, The method includes a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method according to any one of claims 1 to 7.

Images