Intelligent assisted parking system based on unmanned guided vehicle

Abstract

This invention discloses a smart assisted parking system based on an unmanned guided vehicle (UAV), comprising a layered system architecture: a data layer, which provides basic data support for system operation, including storing parking lot map information, parking space occupancy information, and vehicle information within the lot; a system function layer, which analyzes and processes the information provided by the data layer and generates corresponding parking guidance control strategies, including a system operation mode decision module, a UAV task allocation module, a parking space occupancy statistics module, an abnormal state handling module, and a guidance path planning module; and an equipment layer, which executes the control commands generated by the system function layer and collects on-site information, including the UAV, intelligent gate, and server. This system offers advantages such as low-modification and rapid deployment of parking lots, full-scenario adaptive and precise vehicle guidance, significantly shortening parking space search time, and reducing costs while increasing efficiency, all relying on the UAV and cloud-based scheduling.

Description

Technical Field

[0001] This invention relates to the field of parking system technology, and more specifically, to an intelligent assisted parking system based on an unmanned guided vehicle. Background Technology

[0002] Currently, urban parking problems have become a major challenge for modern urban development. Statistics show that approximately 30% of traffic congestion in cities is caused by searching for parking spaces, which not only wastes drivers' time but also leads to increased fuel consumption and exacerbates urban air pollution. Traditional parking lots have long faced common bottlenecks: difficulty in parking, difficulty in finding a car, and difficulty in management. Drivers often drive around the lot, and parking spaces are used unevenly. Management relies heavily on manual inspections, resulting in low turnover efficiency and high operating costs. While there are many intelligent upgrades on the market, they generally require underground wiring and ceiling installation to add extra equipment, resulting in long construction cycles, high investment, and difficulties in quickly adjusting the layout to adapt to changing business needs.

[0003] In existing technologies, new smart parking lots divide the site into several contiguous parking areas and install turnstiles at the main entrance and exit. Parking information is collected by installing cable-pulley monitoring components at each of these contiguous parking areas. Each monitoring component consists of a parallel support cable stretched between two supports, a pull rope, and a camera that can move back and forth along the cable. This solution uses the camera to collect real-time parking space occupancy information and transmits the data back to a control box. The control box then aggregates the information and dynamically displays available spaces on the entrance display screen, thereby reducing the time drivers spend searching for parking spaces.

[0004] Static information display: The empty parking space map is only projected onto the fixed display screen at the entrance. Drivers still need to memorize and judge the route on their own. There is no continuous guidance in the parking lot and the car-finding behavior chain is not closed. During peak hours, it is still easy to go the wrong way and fight for a parking space. If drivers are not familiar with the parking lot, they still need to spend a certain amount of time to find the given empty parking space.

[0005] Camera coverage and mechanical wear: The camera mechanism relies on the reciprocating tension of steel wire ropes. The cables are subjected to tension for a long time, and exposure to sun and rain will cause them to elongate and cause wear on the pulley system. The parking lot requires frequent maintenance during long-term operation and management.

[0006] High deployment and renovation costs: Each group of connected parking spaces requires supporting frames and steel cables, relying on the existing parking lot's civil engineering structure. Garages with insufficient clearance or excessively dense columns are difficult to renovate. No parking space-level allocation strategy: The system only "detects empty spaces" without "assigning specific parking spaces," making it impossible to avoid competition and conflicts when multiple vehicles enter simultaneously, resulting in limited improvements in experience and efficiency. Poor functional scalability: With a fixed hardware architecture, adding features such as reverse vehicle search, charging space reservation, or VIP reservation requires additional cameras and wiring, drastically increasing marginal costs.

[0007] The automated parking allocation and navigation intelligent parking system features a "combined track network" suspended from the top of the parking lot. The track is composed of straight, cross-shaped, and curved sections, and is equipped with a "mobile hoisting mechanism" capable of horizontal and vertical movement. Each parking space is equipped with a vehicle identification sensor. Upon entry, the wireless central controller automatically allocates an available space and dispatches an "intelligent guidance mechanism" to move along the track in front of the vehicle. This guidance mechanism integrates a wireless microcontroller, a height-adjustable and rotating LED screen, and a buzzer. While the vehicle is in motion, it displays a personalized arrow and remaining distance in the driver's line of sight, achieving one-to-one dynamic guidance and avoiding congestion within the parking lot.

[0008] The track network is complex and costly: the top needs to be covered with straight, cross, and arc tracks as well as mobile hoisting mechanisms, requiring a large amount of steel, expansion bolts, and transmission components, and the amortization cost per parking space is much higher than that of the ground solution.

[0009] The mechanical mechanism has many potential failure points: the servo motor-lead screw-pulley-steel wheel multi-stage transmission means that any blockage in any link will cause the entire guide chain to fail; moreover, high-altitude maintenance requires a lifting platform, which makes maintenance difficult and causes long downtime.

[0010] Power supply and communication reliability: The mobile screen uses a track design, and long-term swaying and vibration can easily cause poor contact of the power slip ring. After a power outage and restart, the network needs to be reconnected, which can lead to boot interruption and blockage.

[0011] Poor adaptability to different scenarios: The top track cannot be installed in low-ceilinged areas (<2.2m), irregular ramps, or mechanical multi-level parking garages; outdoor parking lots face the risk of track failure due to water accumulation, snow accumulation, and icing. Furthermore, if there is an expansion need, the number of tracks and vehicles must be increased proportionally, resulting in a linear increase in marginal costs.

[0012] In summary, at least one of the following technical problems exists: The existing smart parking systems involve large-scale demolition and construction, resulting in long deployment cycles and high costs, and cannot quickly cover existing parking lots.

[0013] Traditional static indicator lights and geomagnetic solutions can only provide "area-level" information and cannot provide precise "one-on-one" guidance to car owners, resulting in long periods of driving around the internal roads of the parking lot when parking spaces are scarce.

[0014] Most existing solutions use fixed sensors and guide signs. Once the parking space layout is adjusted or a temporary event is held, the existing solutions cannot be flexibly adapted and require rewiring and modification.

[0015] The existing system lacks a unified control mechanism that dynamically adjusts the guidance granularity and guidance strategy based on the overall state. Summary of the Invention

[0016] The main objective of this invention is to provide an intelligent assisted parking system based on an unmanned guided vehicle, so as to solve at least one technical problem in the prior art.

[0017] To achieve the above objectives, according to one aspect of the present invention, a smart assisted parking system based on an unmanned guided vehicle is provided, comprising: a hierarchical system architecture, specifically: The data layer provides basic data support for system operation, including storing parking lot map information, parking space occupancy information, and vehicle information within the parking lot. The system functional layer is used to analyze and process the information provided by the data layer and generate corresponding parking guidance control strategies. The system functional layer includes: a system operation mode decision module, a guidance vehicle task allocation module, a parking space occupancy statistics module, abnormal state handling, and guidance path planning. The device layer, which is used to execute control commands generated by the system function layer and collect on-site information, includes unmanned guided vehicles, intelligent gates, and servers.

[0018] Preferably, the parking lot map information is used to describe the road structure, parking space distribution and traffic relationship of the parking lot; the parking space occupancy information is used to reflect the real-time or periodic occupancy status of each parking space; and the vehicle information in the parking lot is used to characterize the identity and status of vehicles entering or driving in the parking lot.

[0019] Preferably, the system operation mode decision module is used to determine the system operation mode based on the current operation status of the parking lot; the guide vehicle task allocation module is used to allocate guidance or inspection tasks to the unmanned guide vehicle according to the operation mode; the parking space occupancy statistics module is used to perform statistical analysis on parking space occupancy information; the abnormal state handling module is used to handle abnormal situations such as vehicle deviation and parking space changes during the guidance process; and the guidance path planning module is used to generate the guidance path for the unmanned guide vehicle.

[0020] Preferably, in the system operation mode decision module, in the patrol mode, the server periodically sends a full-field patrol instruction to the unmanned guided vehicle, the unmanned guided vehicle completes the collection and transmission of the parking space status of the entire field, and the server refreshes the parking space occupancy database. In the targeted inspection mode, the server divides the parking lot into several macro areas, assigns the macro area with the highest vacancy rate to the entering vehicles, and sends the targeted inspection task of the corresponding macro area to the unmanned guided vehicle. The unmanned guided vehicle only confirms the parking spaces with uncertain status in the target area. In the one-to-one guidance mode, the server calculates the nearest available parking space and the optimal route for each newly entering vehicle, and sends the license plate, target parking space, and route triple to the designated unmanned guidance vehicle. The unmanned guidance vehicle completes the one-to-one guidance and reports the status in real time.

[0021] Preferably, in the abnormal state handling module, the unmanned guided vehicle determines whether a guidance deviation has occurred by monitoring the relative position relationship, driving direction and expected arrival time of the guided vehicle and itself. If the guided vehicle continues to deviate from the guidance path within a preset time window, the unmanned guided vehicle determines that the guidance task has been abnormally terminated and sends the data back to the server. The server releases the temporary binding relationship between the vehicle and the guided vehicle, releases the corresponding guidance resources and updates the parking space occupancy status.

[0022] Preferably, the hardware structure of the unmanned guided vehicle includes a lidar, a camera, a communication antenna, a motion chassis and a control circuit, and may also include at least one of a screen, indicator lights and a speaker; The lidar is used for environmental perception, navigation, and real-time obstacle avoidance; the camera is used for obstacle type identification, distance estimation of guided vehicles, and parking space occupancy and license plate recognition; the communication antenna is used to enhance wireless communication stability; the motion chassis provides mobility; and the control circuit realizes motion control and data processing of the unmanned guided vehicle.

[0023] Preferably, the unmanned guided vehicle includes: The system provides full-field positioning by integrating lidar, inertial measurement unit and wheel odometer to achieve real-time positioning. During system initialization, a global map is built by manual scanning or importing maps. Subsequently, environmental matching and pose correction are performed based on the electronic map. Automatic navigation plans a driving route based on the positioning results, dynamically adjusts the route to meet safety constraints, and achieves autonomous movement and route following; Obstacle recognition and classification response: Differentiated response strategies are adopted for pedestrians, pets, temporary roadblocks, and parked vehicles. When detours are not possible, an anomaly is reported to the server. Vehicle recognition and distance estimation: The rear-view camera identifies the characteristics of the guided vehicle, estimates the relative distance in real time, and dynamically adjusts the driving speed. Parking space status recognition uses cameras to detect the occupancy status of parking spaces, identifies the license plate of the vehicle occupying the space, and uploads the information to the server. It communicates bidirectionally, continuously interacts with the server, receives instructions and task information, and sends back its own status, environmental awareness results, and task progress.

[0024] Preferably, the hardware structure of the intelligent barrier gate includes a protective shell, a barrier gate body, a screen, and a license plate recognition camera; The license plate recognition camera collects image information of vehicles entering and leaving the venue and identifies the license plate as a unique identifier for the vehicle. The barrier gate controls the order of vehicle passage; the screen displays the number of remaining parking spaces and guidance strategy prompts; the protective casing protects the internal control circuit.

[0025] Preferably, the intelligent barrier gate includes: After vehicle information is uploaded and license plate recognition is completed, the vehicle's entry and exit times are recorded and uploaded to the server. The server then completes the entry and exit registration, parking duration calculation, and parking space data update. Voice and display prompts, based on the operating mode and guidance strategy issued by the server, output prompt information through the screen and voice broadcast module, including prompts for self-selection when there are plenty of seats, prompts for recommended areas when there are moderate seats, and prompts for follow-up guidance when there are few seats. It communicates and interacts with the server, establishes a network connection, uploads vehicle identification information, and receives guidance strategies and display instructions.

[0026] Preferably, the server can be expanded with incremental function modules, including at least one of a reverse vehicle search assistance module, a priority guidance module, and a dedicated parking space management module; The reverse vehicle search assistance module dispatches an idle unmanned guide vehicle to the vicinity of the target parking space to guide the car owner to find their car based on the vehicle parking location information. The priority guidance module assigns a parking space closer to the marked vehicle in a designated area, and prioritizes one-to-one guidance when the occupancy rate is high. The dedicated parking space management module distinguishes parking space types and allocates and guides vehicles only to parking spaces of the matching type based on their attributes.

[0027] The technical solution of this invention has the following technical effects: This invention effectively addresses the pain points of existing smart parking systems—large-scale demolition and construction, long deployment cycles, and high costs—through a layered architecture and lightweight deployment design with unmanned guided vehicles and unified cloud scheduling. It allows for rapid upgrades of existing parking lots without large-scale civil engineering modifications. By adaptively switching between multiple operating modes based on parking space occupancy rates and utilizing the function reuse mechanism of unmanned guided vehicles, it achieves efficient inspections during off-peak hours and precise one-to-one guidance during peak hours, significantly reducing the average time drivers spend searching for parking spaces during peak periods and improving parking lot turnover efficiency. The system is highly flexible and scalable, adaptable to various parking scenarios. Adjustments to parking space layouts require no hardware modifications; adaptation is achieved quickly through software configuration alone. It also supports functional expansions such as reverse vehicle search and parking space reservation. The modular design and simplified operation and maintenance scheme significantly reduce deployment and long-term maintenance costs, avoiding the high-frequency maintenance issues of traditional solutions. Balancing user experience, operational efficiency, and economic benefits, it possesses significant value for large-scale application. Attached Figure Description

[0028] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 A schematic diagram of the overall architecture of the intelligent assisted parking system based on an unmanned guided vehicle according to the present invention is shown; Figure 2 It shows Figure 1 The overall operation logic diagram of the intelligent assisted parking system based on unmanned guided vehicles; Figure 3 It shows Figure 1 The initial operation logic view of the intelligent assisted parking system based on unmanned guided vehicles; Figure 4 It shows Figure 1 The logic diagram of vehicle entry operation in the cruising mode of the intelligent assisted parking system based on unmanned guided vehicles; Figure 5 It shows Figure 1 The logical view of parking space information acquisition in the cruising mode of the intelligent assisted parking system based on unmanned guided vehicles; Figure 6 It shows Figure 1 The logic diagram for vehicle departure processing in the cruising mode of the intelligent assisted parking system based on unmanned guided vehicles; Figure 7 It shows Figure 1 The operational logic diagram of the vehicle entry mode of the intelligent assisted parking system based on unmanned guided vehicles; Figure 8 It shows Figure 1 The logic diagram of parking space information acquisition in the directional inspection mode of the intelligent assisted parking system based on unmanned guided vehicles; Figure 9 It shows Figure 1 The logical view of vehicle departure processing in the directional inspection mode of the intelligent assisted parking system based on unmanned guided vehicles; Figure 10 It shows Figure 1 The one-to-one guidance mode vehicle entry operation logic view of the intelligent assisted parking system based on unmanned guided vehicles; Figure 11 It shows Figure 1 A one-to-one guidance mode vehicle departure operation logic view of the intelligent assisted parking system based on unmanned guided vehicles; Figure 12 It shows Figure 1 A structural view of an autonomous guided vehicle (RV) in a smart assisted parking system based on an RV. Figure 13 It shows Figure 1 The image shows a structural view of an intelligent gate for a smart assisted parking system based on unmanned guided vehicles.

[0029] The above figures include the following reference numerals: 1. LiDAR; 2. Protective housing; 3. Front-view camera; 4. Side-view camera; 5. Chassis; 6. Interactive screen; 7. Communication antenna; 8. Rear-view camera; 9. Barrier gate; 10. Screen; 11. License plate recognition camera. Detailed Implementation

[0030] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0031] like Figures 1 to 13 As shown, this embodiment of the invention provides a smart assisted parking system based on an unmanned guided vehicle, comprising: a layered system architecture, specifically: a data layer, which provides basic data support for system operation, including storing parking lot map information, parking space occupancy information, and vehicle information within the lot; a system function layer, which analyzes and processes the information provided by the data layer and generates corresponding parking guidance control strategies, including: a system operation mode decision module, a guided vehicle task allocation module, a parking space occupancy statistics module, abnormal state handling, and guidance path planning; and an equipment layer, which executes the control commands generated by the system function layer and collects on-site information, including an unmanned guided vehicle, an intelligent barrier gate 9, and a server.

[0032] This invention relates to an unmanned guided vehicle: a small mobile robot equipped with SLAM, path planning, and active obstacle avoidance functions, capable of environmental mapping and vehicle guidance. An intelligent assisted parking system and device: includes the unmanned guided vehicle, a server, and an intelligent barrier gate. Cloud scheduling: a microservice cluster deployed in the cloud, responsible for real-time parking space management, global path optimization, and task dispatch. Lightweight deployment: can be quickly launched and operational without adjusting the overall layout of the parking lot or structurally modifying the ground and ceiling.

[0033] This invention addresses the problems of "large-scale demolition and construction, long cycles, and high costs" in upgrading existing parking lots to be smart. It also addresses the shortcomings of existing solutions, which mostly provide only static "area-level" prompts and rely on fixed-track guidance equipment that can only plan relatively fixed paths. This invention, considering both cost and benefits, proposes a "low-construction, modular, and cloud-based unified scheduling" smart assisted parking system. The goal is to complete the smart transformation of medium to large parking lots in a short time, providing drivers with more intuitive parking guidance at a lower cost, significantly reducing the time drivers spend searching for parking spaces during peak hours. Simultaneously, the entire system introduces multi-mode operation control logic based on parking space occupancy status, achieving dynamic scheduling and efficiency optimization of guidance resources. Furthermore, the modular design allows for quick adjustments and re-launch when parking space layouts are temporarily changed, achieving "install-and-use, modify-and-configure" full-area coverage and precise parking space-level navigation.

[0034] The overall architecture of the system of this invention is as follows: Figure 1 As shown, the unmanned guided vehicle parking guidance system of the present invention adopts a layered system architecture, including a terminal layer, a system function layer, and a data layer. The data layer provides basic data support for system operation, and includes parking lot map information, parking space occupancy information, and vehicle information within the parking lot. Specifically, the parking lot map information describes the road structure, parking space distribution, and traffic relationships of the parking lot; the parking space occupancy information reflects the real-time or periodic occupancy status of each parking space; and the vehicle information within the parking lot identifies and identifies vehicles entering or driving within the parking lot. The system function layer analyzes and processes the information provided by the data layer and generates corresponding parking guidance control strategies. The system function layer includes a system operation mode decision module, a guided vehicle task allocation module, a parking space occupancy statistics module, an abnormal state handling module, and a guidance path planning module. The system comprises several modules: a system operation mode decision module to determine the system operation mode based on the current parking lot status; a guide vehicle task allocation module to assign guidance or inspection tasks to the unmanned guide vehicle according to the operation mode; a parking space occupancy statistics module to perform statistical analysis of parking space occupancy information; an abnormal status handling module to handle abnormal situations such as vehicle deviation and parking space changes during guidance; and a guidance path planning module to generate the guidance path for the unmanned guide vehicle. The equipment layer executes the control commands generated by the system function layer and collects on-site information. This equipment layer includes the unmanned guide vehicle, the intelligent barrier gate 9, and a server. Specifically, the unmanned guide vehicle performs vehicle guidance, parking space inspection, or status detection tasks; the intelligent barrier gate 9 controls vehicle entry and exit and collects vehicle information; and the server carries the computational logic and data processing functions of the system function layer.

[0035] The overall operating logic of this invention system is based on three modules: an unmanned guided vehicle, a server, and an intelligent gate 9. It is divided into three operating modes and an initial deployment mode based on real-time parking space occupancy rates. The server compares the real-time parking space occupancy rate with multiple preset operating threshold ranges and switches the system to the corresponding operating mode based on the comparison results. These modes include, but are not limited to, patrol mode, directional inspection mode, and one-to-one guidance mode. The switching of operating modes in the system is controlled by the server: the server continuously updates the overall parking space occupancy rate in real-time or periodically and determines the occupancy rate threshold range based on the latest occupancy rate. A parking space occupancy rate of less than 40% is considered low, 40% to 70% is considered medium, and greater than 70% is considered high. When the occupancy rate crosses adjacent threshold intervals, and when multiple consecutive statistical results indicate that the parking space occupancy rate consistently crosses adjacent threshold intervals, the server triggers a switch to the corresponding operating mode. After triggering the operating mode switch, the server performs mode adjustment according to the following rules: For assigned but not yet completed guidance tasks, the original operating mode is maintained until the current guidance task is completed or abnormally terminated, avoiding guidance interruption due to mode switching; for idle unattended guided vehicles, their scheduling strategy and task allocation method are immediately updated according to the new operating mode; for vehicles newly entering the parking lot after the mode switch, they are uniformly processed and guided according to the new operating mode. Through the above methods, a smooth switch of operating modes under different parking space occupancy states is achieved, avoiding frequent system jitter, while ensuring the continuity and stability of the parking guidance process.

[0036] The specific descriptions of each mode of this invention are as follows: Initialization (Initial Deployment): Before the parking lot opens, the unmanned guided vehicle performs a "circling patrol mode." This involves the unmanned guided vehicle circling the parking lot once according to the optimal route provided by the server, ensuring it passes every parking space. Simultaneously, the vehicle uses the onboard front-view camera 3 and side-view camera 4 to determine the occupancy status of parking spaces via YOLO. If a parking space is occupied, the corresponding license plate is identified via OCR, and then the parking space occupancy information and vehicle information are uploaded to the server. After receiving the latest occupancy matrix from the guided vehicle, the server refreshes the database to complete initialization. The database content includes the parking space occupancy status and the license plates of the occupied parking spaces. Patrol Mode (Low parking space occupancy rate, such as less than 40%): Vehicle Entry: When a vehicle arrives at the smart gate 9, the smart gate 9 identifies the license plate, records the entry time, and uploads the vehicle information to the server. Afterwards, the smart gate 9 screen 10 displays "Sufficient parking spaces available, please choose a parking space" and raises the barrier to allow passage, without further voice or text guidance. Parking Information Acquisition: After entering the parking lot, vehicles can choose a suitable parking space on their own. To obtain their parking location information, the server periodically (e.g., every 10 minutes, depending on the size of the parking lot) broadcasts a "full-area patrol" command to all guide vehicles, requesting an update of the entire parking space table. Upon receiving the command, the unmanned guide vehicle enters "circling patrol mode," the principle of which is explained in the previous section on "circling patrol mode." This allows the server to refresh its database after receiving the latest occupancy matrix from the guide vehicles. Vehicle Departure: When the vehicle leaves on its own, the smart barrier gate 9 and license plate recognition camera 11 recognize the license plate and record the departure time. The system then uploads the vehicle information to the server, obtains the vehicle's entry time and payment from the server, and displays it on screen 10. After the vehicle completes payment, the barrier gate opens to allow passage, and the server updates the corresponding parking space data based on the uploaded information. Targeted inspection mode (medium parking space occupancy rate, such as greater than 40% but less than 70%): Vehicle entry: When a vehicle arrives at the smart gate 9, the license plate recognition camera 11 of the smart gate 9 recognizes the license plate and records the entry time, then uploads the vehicle information. The server divides the parking lot into several "macro areas" according to the latest occupancy matrix, calculates the vacancy rate of each area, assigns the "macro area with the highest vacancy rate" to the newly entered vehicle, and sends the area ID to the smart gate 9. After receiving the returned "suggested area", the smart gate 9 prompts on the screen 10 "Please go to area A / B, there are more vacant spaces" and raises the barrier to allow passage. Parking Information Acquisition: After entering the parking lot, the vehicle is highly likely to choose a recommended parking space in the area. At this time, the server issues a "directional inspection" task to the guide vehicle: only scan the unidentified parking spaces in this area to reduce the patrol mileage. After receiving the "directional inspection" instruction from the server, the unmanned guide vehicle switches to "fixed-point information collection mode," goes to the target macro area, and reconfirms the parking spaces that were not scanned or whose status was unclear by the front-view camera 3 and side-view camera 4. After updating the parking space status of the area and sending it back to the server, it then returns to standby or continues to the next round of inspection. Vehicle Departure: The process is the same as the patrol mode when the vehicle leaves.Vehicle deviation from guidance anomaly handling mechanism: When a target vehicle deviates from the guidance path of the unmanned guidance vehicle, the guidance vehicle monitors the relative position of the target vehicle to itself, its direction of travel, or its expected arrival time to determine whether the guided vehicle has deviated from its guidance path. If the guidance vehicle detects that the target vehicle continues to deviate from the guidance path within a preset time window, it determines that the current guidance task has abnormally terminated and sends the anomaly information back to the server. The server then releases the temporary binding relationship between the vehicle and the guidance vehicle, releasing the corresponding guidance resources. Simultaneously, the server updates the parking space occupancy status based on subsequent perception results, enabling the system to maintain operational stability and efficient utilization of guidance resources even when vehicle behavior is uncertain. One-to-one guidance mode (high parking space occupancy rate, such as greater than 70%): Vehicle entry: When a vehicle arrives at the smart gate 9, the license plate recognition camera 11 of the smart gate 9 recognizes the license plate and records the entry time, and then uploads the vehicle information. After receiving the information, the server calculates the "nearest available space + optimal path" for each new vehicle in real time, and sends the (license plate, target parking space, path) triplet to the designated guidance vehicle. After the smart gate 9 receives the server's return "there is a guidance vehicle to pick up", it prompts on the screen 10 "Please follow the unattended guidance vehicle ahead to the parking space" and simultaneously provides a voice reminder. Then the gate is raised to allow passage, and the vehicle is allowed to stop briefly outside the gate 9 to wait for the guidance vehicle to arrive. One-to-one guidance: Upon receiving a server instruction, the guide vehicle switches to "active guidance mode," first moving to wait for the target vehicle in front of the smart gate 9. After confirming the vehicle is following, the entire route map is displayed on the interactive screen 6. During the journey, it integrates LiDAR 1 and a forward-facing camera 3 for real-time obstacle avoidance, responding to people, pets, and temporary roadblocks (sound and light alerts, detours, or stopping). Simultaneously, the rear-view camera 8 calculates the distance to the vehicle in front in real time and dynamically adjusts the speed to ensure the vehicle doesn't lose track. Meanwhile, the server records the "vehicle-guide vehicle" binding relationship and continuously monitors the entire area: if the target parking space is occupied by someone else or a temporary obstacle appears on the path, the server immediately recalculates and pushes a new path to the guide vehicle. Upon reaching the target parking space, the guide vehicle reports "task completed" to the server, the vehicle-guide vehicle binding is released, and the guide vehicle returns to the standby area or accepts the next task. The server then releases the "vehicle-guide vehicle" binding relationship and updates the parking space data. Vehicle departure: The process is the same as when the occupancy rate is <40%.

[0037] This invention relates to an unmanned guided vehicle. The hardware structure of the unmanned guided vehicle includes: a lidar 1, a front-view camera 3, a side-view camera 4, a rear-view camera 8, a communication antenna 7, a chassis 5, and its internal control circuitry. An external protective shell 2 is provided. Depending on the scenario and requirements, additional hardware devices such as an interactive screen 6, indicator lights, and speakers can be added to expand its functionality. The lidar 1 is used to perceive the environment and provide navigation and real-time obstacle avoidance capabilities; the front-view camera 3, side-view camera 4, and rear-view camera 8 provide more multi-dimensional data for determining obstacle types and distances to the guided vehicle; the communication antenna 7 extends to the outside of the protective shell 2 to enhance the stability of the signal link underground; the protective shell 2 protects the internal control circuitry; the chassis 5 provides mobility; and optional hardware devices such as the interactive screen 6, indicator lights, and speakers provide more diverse extended functions.

[0038] The unmanned guided vehicle described in this invention possesses the following core functions: 1. Global Positioning: The unmanned guided vehicle achieves real-time positioning within the parking lot environment by fusing multi-source sensor information obtained from LiDAR 1, Inertial Measurement Unit (IMU), and wheeled odometer. During system initialization, the guided vehicle is manually controlled to scan the environment to obtain a global map or directly import an existing map. Upon entering the parking lot, the unmanned guided vehicle performs environmental matching based on the electronic map sent by the server and continuously corrects its own pose information during driving, thereby achieving stable positioning throughout the entire area without external positioning base stations, providing basic pose data for subsequent navigation and guidance tasks. 2. Automatic Navigation: Based on the global positioning results, the unmanned guided vehicle can autonomously plan its driving route and complete automatic navigation according to the target location or guidance path sent by the server. During navigation, the unmanned guided vehicle dynamically adjusts its driving path based on the environmental information perceived in real time by LiDAR 1 and forward-looking camera 3. Under the premise of meeting safety constraints, it travels along the optimal or suboptimal path to the target area, entrance waiting point, or location near the target parking space, thereby achieving autonomous movement, chassis 5 movement, and path following. 3. Obstacle Recognition and Classification Response: During operation, the unmanned guided vehicle continuously perceives its surroundings using LiDAR 1, a front-view camera 3, and side-view cameras 4, identifying and classifying detected obstacles. Depending on the type of obstacle, the unmanned guided vehicle adopts differentiated response strategies, such as slowing down or stopping and issuing audio-visual cues for pedestrians or pets; detouring around temporary roadblocks or parked vehicles; and reporting abnormal states to the server when unable to autonomously handle obstacles. These methods ensure the safety and continuity of the guidance process. 4. Vehicle Recognition and Distance Estimation: In one-to-one guidance mode, the unmanned guided vehicle uses a rear-view camera 8 to identify the license plate or vehicle feature information of the guided vehicle to confirm its identity. Simultaneously, based on visual ranging or multi-sensor fusion, the unmanned guided vehicle estimates the relative distance between itself and the guided vehicle in real time and dynamically adjusts its speed accordingly to ensure stable following and avoid the risk of losing track or rear-end collisions. 5. Parking Space Occupancy Recognition and License Plate Recognition of Occupied Vehicles: In patrol or directional inspection mode, the unmanned guided vehicle uses the front-view camera 3 and side-view cameras 4 to detect parking spaces along the route one by one to determine whether a parking space is occupied. When an occupied parking space is detected, the license plate information of the parked vehicle is further identified, and the parking space number, occupancy status, and license plate information are uploaded to the server to update the overall parking space occupancy matrix. 6. Two-Way Communication with the Server: The unmanned guided vehicle establishes a wireless communication network through the communication antenna 7 to maintain a continuous connection with the server, used to receive operating mode instructions, guidance tasks, target parking space information, and path information. Simultaneously, it transmits its own status information, environmental perception results, parking space detection results, and task execution progress back to the server in real time or periodically to support cloud-based judgment and scheduling decisions regarding the overall operating status.

[0039] This invention relates to an intelligent barrier gate 9, comprising a protective shell 2, a barrier gate 9, a screen 10, and a license plate recognition camera 11. The screen 10 displays the remaining number of parking spaces and indicates whether there are unattended vehicles available for entry; the license plate recognition camera 11 collects vehicle license plate information; the barrier gate 9 controls the sequential passage of vehicles; and the protective shell 2 protects the internal control circuitry. The intelligent barrier gate 9 of this invention possesses the following core functions: 1. Vehicle license plate recognition: The intelligent barrier gate 9 collects image information of vehicles entering or leaving the parking area through the license plate recognition camera 11 and recognizes the vehicle's license plate to obtain a unique identifier for the vehicle. The recognized license plate information serves as vehicle identity data for subsequent processes such as parking timekeeping, guidance strategy matching, and fee settlement. 2. Vehicle information upload: After the vehicle completes license plate recognition, the intelligent barrier gate 9 records the vehicle's entry or exit time and uploads the vehicle information to a server. The server uses this information to complete vehicle entry and exit registration, parking duration calculation, and updates the corresponding parking space data, thus forming a complete closed loop of vehicle entry and exit data. 3. Voice and Screen 10 Information Display: Based on the current operating mode or guidance strategy returned by the server, the intelligent barrier gate 9 outputs corresponding prompts to the driver on the screen 10 and the voice broadcast module. These prompts include, but are not limited to: prompting the driver to choose a parking spot independently when there are sufficient parking spaces; prompting the driver to proceed to the recommended parking area when there are moderately many parking spaces; and prompting the driver to wait and follow the unmanned guided vehicle to the target parking space when parking spaces are scarce. This intuitive prompting reduces the driver's decision-making burden. 4. Server Communication: The intelligent barrier gate 9 establishes a communication connection with the server via the network to upload vehicle identification information, receive guidance strategies, and display instructions. The server triggers the corresponding scheduling process based on the information uploaded by the intelligent barrier gate 9, making the entrance module the unified trigger node for vehicles entering the parking lot within the system.

[0040] This invention presents a system operation logic based on adaptive switching of multiple operating modes according to parking space occupancy rate. Using the overall parking space occupancy rate as a unified scheduling basis, the system continuously assesses the parking lot's operational status through a server and switches between different system operation modes under different occupancy rate ranges, forming a complete, continuous, and scalable closed-loop operation logic. This logic includes at least a patrol mode, a directional inspection mode, and a one-to-one guidance mode. By setting occupancy rate threshold ranges and a stability determination mechanism, it achieves smooth switching between modes, avoiding frequent system jitter or guidance interruptions. This logic enables the system to automatically adjust the guidance mode and resource input intensity based on the parking lot's occupancy rate status, ensuring guidance effectiveness while reducing unnecessary equipment actions and computational load, balancing parking efficiency and resource consumption during peak and off-peak periods.

[0041] This invention presents a hierarchical system architecture with unified server decision-making and distributed execution by unmanned guided vehicles (UAVs). The architecture centralizes the overall parking lot status assessment, operation mode decision-making, task allocation, and path planning on the server side, while delegating navigation, obstacle avoidance, information collection, and guidance execution to the UAVs. Key aspects of this architecture include: the server acting as the global control hub, determining scheduling strategies based on parking space occupancy information, vehicle entry and exit information, and UAV status; the UAVs, equipped with a LiDAR 1, a front-view camera 3, a side-view camera 4, a rear-view camera 8, a communication antenna 7, a chassis 5, and an interactive screen 6, only needing general mobile robot capabilities and corresponding sensors to be reused as inspection equipment, guidance equipment, or information collection terminals in different operating modes; and the intelligent barrier gate 9 serving as the unified trigger node for vehicle entry, decoupling the physical entrance / exit behavior from the server scheduling logic.

[0042] This invention avoids large-scale fixed hardware deployment inside parking lots, giving the system inherent scalability and maintainability.

[0043] Another key aspect of the unmanned guided vehicle (UAV) functional reuse mechanism design in this invention is that the same UAV assumes different functional roles in different operating modes, achieving a high degree of hardware resource reuse. The hardware includes a LiDAR (1), protective casing (2), front-view camera (3), side-view camera (4), chassis (5), interactive screen (6), communication antenna (7), and rear-view camera (8). In low parking space occupancy, the UAV primarily performs periodic patrols, collecting and updating parking space occupancy status information. In medium parking space occupancy, the UAV switches to a directional inspection role, only confirming areas with changing occupancy and spaces with unclear status. In high parking space occupancy, the UAV is scheduled as a one-to-one guidance device, directly participating in vehicle path guidance and accelerating the search for available parking spaces. This functional reuse mechanism avoids the traditional design approach of "one function corresponding to one set of fixed equipment," significantly reducing system deployment costs and improving equipment utilization.

[0044] This invention provides a secure operation mechanism for dynamic binding and abnormal release of guidance resources. In a one-to-one guidance mode, it introduces a temporary binding mechanism between the vehicle and the unmanned guidance vehicle. A logical association is established between the vehicle's identity, the target parking space, and the guidance vehicle via a server. An interactive screen 6 displays guidance information, while a LiDAR 1 and a rear-view camera 8 ensure continuous monitoring, guaranteeing the uniqueness and controllability of the guidance process. Simultaneously, this invention also includes an abnormal detection and release mechanism. When the guided vehicle deviates from the predetermined path, fails to follow as expected, or the guidance conditions change, the system automatically releases the binding relationship, releases the guidance resources, and updates the parking space occupancy status. This ensures the system maintains overall operational stability even under complex and uncertain parking behaviors, reduces the risk of problems during long-term operation, and improves robustness under various conditions.

[0045] This invention features rapid deployment and highly compatible design without requiring large-scale civil engineering modifications. At the system design level, it emphasizes "lightweight deployment," with key points including: not relying on heavy infrastructure such as geomagnetic sensors, fixed camera arrays, or overhead tracks; using unmanned guided vehicles equipped with LiDAR 1, a front-view camera 3, a side-view camera 4, and a rear-view camera 8 for mobile inspection, replacing fixed parking space detection equipment; and adapting to different parking lot layouts through server logic configuration rather than physical modifications. Therefore, this invention is applicable to various scenarios such as underground parking lots, surface parking lots, and ramp parking lots, and can be quickly relaunched through software database updates when parking space layouts are adjusted or operations are disrupted. This invention aims to protect a comprehensive technical solution for a rapidly deployable smart parking system based on mobile guidance devices for existing parking lots, rather than being limited to a specific site type.

[0046] This invention employs a scalable design concept that decouples the system's operational logic from its hardware implementation. By decoupling the system's operational logic and scheduling strategy from the specific hardware implementation, it can be expanded to include functions such as reverse vehicle search, parking space reservation, charging parking space management, and VIP guidance. Furthermore, by adding an interactive screen 6 to the guiding vehicle, more information can be displayed to the driver; and by adding speakers and indicator lights, the prompts are made more prominent. This design ensures the system's long-term evolution capability, constructing a smart parking system architecture with a unified scheduling logic at its core, supporting the overlay of multiple business functions.

[0047] This invention adopts an architecture of "unmanned guided vehicle + unified cloud scheduling" to achieve low-construction modification, rapid deployment, and refined parking guidance for medium and large parking lots. It has the following significant advantages in terms of functionality, efficiency, cost, and adaptability: Short deployment cycle and low construction cost: The core equipment of the system is a mobile unmanned guided vehicle (LiDAR 1, protective shell 2, front-view camera 3, side-view camera 4, chassis 5, interactive screen 6, communication antenna 7, rear-view camera 8) and a cloud service platform. It eliminates the need for large-scale deployment of tracks, power cables, or fixed sensors. Only an intelligent barrier gate 9 with a license plate recognition camera 11 and network access equipment need to be installed at the parking lot entrance. Deployment and operation can be completed quickly, significantly reducing modification costs compared to traditional solutions. One-to-one dynamic navigation significantly shortens parking search time: Through real-time cloud scheduling and high-precision positioning, the unmanned guided vehicle can assign the nearest available parking space to each new car and guide it to the target parking space, reducing the average parking search time during peak hours from 10-15 minutes to less than 90 seconds, improving parking lot turnover and customer satisfaction. High flexibility and scalability: When parking space layouts are adjusted, temporarily closed, or events are held, there is no need for rewiring or modification. Simply update the parking space distribution in the cloud to adapt to the site and achieve rapid resumption of operations. Simple operation and maintenance, low long-term costs: The unmanned guided vehicle adopts a modular design. Key components such as LiDAR 1, front-view camera 3, side-view camera 4, rear-view camera 8, communication antenna 7, chassis 5, and interactive screen 6 can be quickly replaced. Routine maintenance only requires software updates and battery management, avoiding the high-frequency maintenance problems of traditional tracks and mechanical transmission devices. In summary, this invention not only breaks through the limitations of existing smart parking systems in terms of technical implementation ("large-scale construction and demolition, long deployment cycle, and lack of flexibility"), but also has large-scale application value in terms of economic benefits and user experience.

[0048] The core idea of ​​the "rapidly deployable smart parking system" proposed in this invention is as follows: A guidance device capable of autonomous movement under chassis 5, equipped with LiDAR 1 and various cameras, serves as a dynamic information carrier. Through a simplified cloud-terminal architecture, it completes parking space status collection, path calculation, and one-to-one parking guidance, achieving low-cost, short-cycle, and highly flexible smart upgrades without damaging the existing building structure. From this perspective, it is difficult to find other technical paths that can simultaneously meet the three major goals of "low civil engineering renovation," "modular construction," and "one-to-one guidance." Local alternatives include: fixed sensors + indicator lights: These can detect parking spaces and provide area guidance, but each parking space requires a geomagnetic sensor and camera, resulting in long wiring construction, high maintenance costs, and only area-specific guidance without one-to-one real-time navigation; top-mounted track-type robots: These allow for movable guidance via a suspended track, but require significant modifications and are costly, with secondary construction needed for parking space adjustments; full-coverage video + large screen: High-position cameras monitor parking spaces, requiring high network connectivity and still relying on drivers to find their own spaces, resulting in low efficiency; ground markings + simple robots: These rely on fixed tracks, require regular ground maintenance, and lack flexibility. The comparison shows that the mobile unmanned guided vehicle of this invention, combined with cloud-based scheduling, requires no major modifications and achieves rapid deployment, accurate navigation, and flexible adaptation. Its overall cost and effectiveness are superior to existing solutions.

[0049] This invention utilizes an unmanned guided vehicle parking guidance system to provide orderly and efficient guidance even with dynamic changes in parking spaces, reducing congestion in finding parking spots. When traffic is concentrated in large shopping malls, hospitals, and transportation hubs, the server dispatches available guided vehicles. These vehicles use LiDAR (LiDAR 1) for positioning and navigation, front-view cameras (3) and side-view cameras (4) to collect parking space data, and an interactive screen (6) provides route prompts. Data is updated after the vehicle guides the driver to a parking space. Changes to the parking space layout only require cloud-based configuration updates, which the guided vehicle can then adapt to.

[0050] In another embodiment, the present invention expands the functionality of the intelligent assisted parking system without changing the architecture, only adding software modules: 1) Reverse car search: When the car owner initiates a car search, the server dispatches a guide vehicle to the vicinity of the parking space to guide the car search, reusing the existing LiDAR 1 navigation, camera perception, and interactive screen 6 prompt functions; 2) Reservation / priority guidance: The server marks exclusive parking spaces, and after verification by the license plate recognition camera 11, the guide vehicle is dispatched with priority; 3) Dedicated parking space management: Distinguish between ordinary and charging parking spaces, the server matches and issues them, and the guide vehicle can perform navigation as usual.

[0051] Based on the above description, the present invention achieves the following technical effects: The present invention utilizes the lidar 1, protective shell 2, front-view camera 3, side-view camera 4, chassis 5, interactive screen 6, communication antenna 7, rear-view camera 8 of the unmanned guided vehicle, along with a cloud-based layered architecture, to achieve lightweight transformation and upgrade of existing parking lots; relying on the adaptive switching mode of parking space occupancy rate, it achieves efficient inspection and precise one-to-one guidance during peak and off-peak hours, significantly shortening the time for finding parking spaces; the system is flexible and expandable, adaptable to multiple scenarios, and supports iterative functions such as reverse vehicle search; the modular design simplifies operation and maintenance, reduces costs and increases efficiency, and has high promotional value.

[0052] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A smart assisted parking system based on an unmanned guided vehicle, characterized in that, This includes a layered system architecture, specifically: The data layer provides basic data support for system operation, including storing parking lot map information, parking space occupancy information, and vehicle information within the parking lot. The system functional layer is used to analyze and process the information provided by the data layer and generate corresponding parking guidance control strategies. The system functional layer includes: a system operation mode decision module, a guidance vehicle task allocation module, a parking space occupancy statistics module, abnormal state handling, and guidance path planning. The device layer, which is used to execute control commands generated by the system function layer and collect on-site information, includes unmanned guided vehicles, intelligent gates, and servers. 2.The unmanned guided vehicle based intelligent assisted parking system of claim 1, wherein, The parking lot map information is used to describe the road structure, parking space distribution and traffic relationship of the parking lot; the parking space occupancy information is used to reflect the real-time or periodic occupancy status of each parking space; and the vehicle information in the parking lot is used to characterize the identity and status of vehicles entering or driving in the parking lot. 3.The unmanned guided vehicle based intelligent assisted parking system of claim 1, wherein, The system operation mode decision module is used to determine the system operation mode based on the current operation status of the parking lot; the guide vehicle task allocation module is used to allocate guidance or inspection tasks to unmanned guide vehicles according to the operation mode. The parking space occupancy statistics module is used to perform statistical analysis on parking space occupancy information; the abnormal state handling module is used to handle abnormal situations such as vehicle deviation and parking space changes during the guidance process; and the guidance path planning module is used to generate the guidance path for the unmanned guided vehicle. 4.The unmanned guided vehicle based intelligent assisted parking system of claim 1, wherein, In the system operation mode decision module, under the patrol mode, the server periodically sends a full-field patrol command to the unmanned guided vehicle, the unmanned guided vehicle completes the collection and transmission of the parking space status of the entire field, and the server refreshes the parking space occupancy database. In the targeted inspection mode, the server divides the parking lot into several macro areas, assigns the macro area with the highest vacancy rate to the entering vehicles, and sends the targeted inspection task of the corresponding macro area to the unmanned guided vehicle. The unmanned guided vehicle only confirms the parking spaces with uncertain status in the target area. In the one-to-one guidance mode, the server calculates the nearest available parking space and the optimal route for each newly entering vehicle, and sends the license plate, target parking space, and route triple to the designated unmanned guidance vehicle. The unmanned guidance vehicle completes the one-to-one guidance and reports the status in real time. 5.The unmanned guided vehicle based intelligent assisted parking system of claim 1, wherein, In the abnormal state handling module, the unmanned guided vehicle determines whether a guidance deviation has occurred by monitoring the relative position, driving direction and expected arrival time of the guided vehicle and itself. If the guided vehicle continues to deviate from the guidance path within a preset time window, the unmanned guided vehicle determines that the guidance task has been abnormally terminated and sends the data back to the server. The server releases the temporary binding relationship between the vehicle and the guided vehicle, releases the corresponding guidance resources and updates the parking space occupancy status. 6.The unmanned guided vehicle based intelligent assisted parking system of claim 1, wherein, The hardware structure of the unmanned guided vehicle includes a lidar, a camera, a communication antenna, a motion chassis and a control circuit, and may also include at least one of a screen, indicator lights and a speaker; The lidar is used for environmental perception, navigation, and real-time obstacle avoidance; the camera is used for obstacle type identification, distance estimation of guided vehicles, and parking space occupancy and license plate recognition; the communication antenna is used to enhance wireless communication stability; the motion chassis provides mobility; and the control circuit realizes motion control and data processing of the unmanned guided vehicle.

7. The unmanned guided vehicle based intelligent assisted parking system according to claim 1, wherein, The unmanned guided vehicle includes: The system provides full-field positioning by integrating lidar, inertial measurement unit and wheel odometer to achieve real-time positioning. During system initialization, a global map is built by manual scanning or importing maps. Subsequently, environmental matching and pose correction are performed based on the electronic map. Automatic navigation plans a driving route based on the positioning results, dynamically adjusts the route to meet safety constraints, and achieves autonomous movement and route following; Obstacle recognition and classification response: Differentiated response strategies are adopted for pedestrians, pets, temporary roadblocks, and parked vehicles. When detours are not possible, an anomaly is reported to the server. Vehicle recognition and distance estimation: The rear-view camera identifies the characteristics of the guided vehicle, estimates the relative distance in real time, and dynamically adjusts the driving speed. Parking space status recognition uses cameras to detect the occupancy status of parking spaces, identifies the license plate of the vehicle occupying the space, and uploads the information to the server. It communicates bidirectionally, continuously interacts with the server, receives instructions and task information, and sends back its own status, environmental awareness results, and task progress. 8.The unmanned guided vehicle based intelligent assisted parking system of claim 1, wherein, The hardware structure of the intelligent barrier gate includes a protective shell, a barrier gate body, a screen, and a license plate recognition camera; The license plate recognition camera collects image information of vehicles entering and leaving the venue and identifies the license plate as a unique identifier for the vehicle. The barrier gate controls the order of vehicle passage; the screen displays the number of remaining parking spaces and guidance strategy prompts. The protective casing protects the internal control circuitry. 9.The unmanned guided vehicle based intelligent assisted parking system of claim 1, wherein, The intelligent barrier gate includes: After vehicle information is uploaded and license plate recognition is completed, the vehicle's entry and exit times are recorded and uploaded to the server. The server then completes the entry and exit registration, parking duration calculation, and parking space data update. Voice and display prompts, based on the operating mode and guidance strategy issued by the server, output prompt information through the screen and voice broadcast module, including prompts for self-selection when there are plenty of seats, prompts for recommended areas when there are moderate seats, and prompts for follow-up guidance when there are few seats. It communicates and interacts with the server, establishes a network connection, uploads vehicle identification information, and receives guidance strategies and display instructions. 10.The unmanned guided vehicle based intelligent assisted parking system of claim 1, wherein, The server can be expanded with incremental function modules, including at least one of a reverse vehicle search assistance module, a priority guidance module, and a dedicated parking space management module; The reverse vehicle search assistance module dispatches an idle unmanned guide vehicle to the vicinity of the target parking space to guide the car owner to find their car based on the vehicle parking location information. The priority guidance module assigns a parking space closer to the marked vehicle in a designated area, and prioritizes one-to-one guidance when the occupancy rate is high. The dedicated parking space management module distinguishes parking space types and allocates and guides vehicles only to parking spaces of the matching type based on their attributes.

Images