Parking guidance control system and method based on unmanned aerial vehicle assistance

Abstract

The invention discloses a parking guidance control system and method based on unmanned aerial vehicle assistance. The parking guidance control system comprises a vehicle-mounted unmanned aerial vehicle, a main control unit and a man-machine interaction module. A user triggers a parking auxiliary guiding function through the man-machine interaction module, and the main control unit drives the vehicle-mounted unmanned aerial vehicle to lift off to detect a parking lot area after receiving the parking auxiliary guiding function triggered by the user. And parking guide data is sent to the main control unit according to the detected data, and the main control unit displays the parking guide data to a user through the man-machine interaction module. According to the invention, rapid identification, parking space state analysis and optimal path planning of the parking spaces in the surrounding area by the unmanned aerial vehicle are realized.

Description

A parking guidance control system and method based on unmanned aerial vehicle (UAV) assistance Technical Field

[0001] This invention relates to the field of intelligent vehicle control, and in particular to a parking guidance control method and system based on unmanned aerial vehicle (UAV) assistance. Background Technology

[0002] Against the strategic backdrop of the integrated development of low-altitude economy and intelligent transportation, industrial-grade drones, as airborne perception and execution units, are gradually becoming a key support for improving the intelligence level of scenarios such as urban parking management, large-scale station operation and maintenance, and emergency response through their efficient collaborative capabilities with ground-based intelligent mobile platforms. New energy pickup trucks, with their superior cargo space, external power discharge capabilities, and good road passability, provide ideal mobile bases for drone take-off, landing, recovery, charging, and communication relay, and are gradually demonstrating platform potential in applications such as logistics delivery and outdoor operations. Therefore, combining the high-altitude perception advantages of drones with the mobility and energy replenishment capabilities of new energy pickup trucks to construct an "air-ground" collaborative parking space search and autonomous navigation system is expected to significantly alleviate practical bottlenecks in complex parking environments, such as obstructed vision, delayed parking space status perception, and insufficient dynamic path planning capabilities.

[0003] In the existing technological system, parking navigation solutions mainly rely on three typical forms: First, site management systems based on fixed parking space sensors (see Chinese patent CN107230377A) statically monitor parking space status by deploying geomagnetic or ultrasonic detectors within the parking lot. This mode is suitable for enclosed areas with well-developed infrastructure, but struggles to cover unstructured areas. Second, autonomous parking systems based on vehicle-mounted vision and ultrasonic sensors rely on vehicle-mounted peripheral sensing hardware to achieve near-range parking space identification and parking control. Their sensing range is limited, making it unable to support large-scale parking space searches. While these solutions have promoted parking automation at different levels, they generally suffer from limited levels of intelligence, weak environmental adaptability, and insufficient system functionality.

[0004] Existing parking navigation solutions heavily rely on pre-installed high-precision maps and fixed parking lot signage, essentially limiting autonomous parking capabilities to indoor and outdoor parking lots with well-developed digital infrastructure. This makes them ill-suited for open-area parking scenarios without map coverage or lane markings. While vehicle-mounted sensor-based autonomous parking systems possess some near-field parking space recognition and entry capabilities, their perception range is limited to the vehicle's surrounding field of vision, hindering rapid discovery of large areas of available parking spaces and global path optimization. Although V2X-based collaborative guidance solutions can provide some real-time parking information, they still heavily depend on on-site equipment deployment and communication network coverage, significantly reducing their applicability in non-cooperative environments such as outdoor areas, temporary parking lots, or underground garages without infrastructure support. Crucially, existing systems generally lack real-time perception and response capabilities to dynamic obstacles, weather changes, and ground clearance, failing to perform online replanning and robust control in response to sudden changes in parking space status or environmental conditions. This results in severely compromised reliability and safety in complex real-world scenarios. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a parking guidance and control method and system based on unmanned aerial vehicle (UAV) assistance. This system enables UAVs to quickly identify parking spaces in the surrounding area, analyze parking space status, and plan the optimal route. With the help of vehicle-machine collaborative control, the system enables vehicles to autonomously navigate to vacant parking spaces. It effectively solves the problems of limited field of vision, untimely updates of parking space information, lack of route information in navigation software in large parking lots, and parking difficulties in multi-obstacle environments that exist in traditional parking systems.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A parking guidance and control system based on drone assistance includes a vehicle-mounted drone, a main control unit, and a human-machine interaction module;

[0008] The user triggers the parking assistance guidance function through the human-computer interaction module. After receiving the user's trigger of the parking assistance guidance function, the main control unit drives the vehicle-mounted drone to take off and detect the parking area. Based on the detection data, the drone sends parking guidance data to the main control unit, which then displays it to the user through the human-computer interaction module.

[0009] The vehicle-mounted drone has a built-in communication unit, which is connected to the main control unit or the user's mobile terminal, or the vehicle-mounted drone's communication unit is connected to the main control unit or the user's mobile terminal through a cloud platform.

[0010] The communication unit of the vehicle-mounted drone communicates directly with the main control unit / user mobile terminal as the main link mode. Only when communication abnormalities occur in the main link mode or the communication delay does not meet the set requirements will it switch to the protection link mode. The protection link mode means that the communication unit of the vehicle-mounted drone communicates with the main control unit / user mobile terminal through the cloud platform.

[0011] The vehicle-mounted drone is equipped with a photosensitive element. After the vehicle-mounted drone takes off, the drone's camera is activated, and the light signal from the photosensitive element is acquired. Based on the light signal, the camera switches to night mode. In night mode, the vehicle-mounted drone's control system controls the camera's ISO sensitivity and the supplementary light intensity of the supplementary light system.

[0012] A parking guidance control method based on drones includes a user triggering a parking guidance function through a human-machine interaction module. After receiving the user's triggering of the parking guidance function, the main control unit drives the vehicle-mounted drone to take off and detect the parking area. Based on the detected image data, the drone sends parking guidance data to the main control unit, which then displays the data to the user through the human-machine interaction module.

[0013] The main control unit responds to the user's parking space search command, drives the vehicle-mounted drone to take off and take aerial photos of the parking lot, and dynamically generates a two-stage "location search-path exploration" mission strategy: the first stage controls the drone to fly to the target area to perform a wide-area scan; the second stage instructs the drone to perform reverse low-altitude verification of the connecting path from the optimal empty parking space to the current position of the vehicle.

[0014] Vacancy location and reverse path verification include: After the UAV identifies and selects a qualified vacant parking space through its onboard visual sensors, the system automatically triggers the path verification engine: The UAV takes off from the coordinates of the vacant parking space and, from the perspective of simulating vehicle driving, performs a reverse exploration flight along the internal road network of the parking lot towards the current location of the user's vehicle; During the flight, it integrates visual SLAM and RTK-GPS data to build a topology map of the passable area in real time, dynamically avoids fixed and moving obstacles, and records the coordinate sequence of all key path inflection points and decision points, ultimately generating a navigation path that has been verified by the entity and is accessible to the vacant parking space.

[0015] During the mission, the vehicle-mounted drone flew to the target parking lot to conduct a high-altitude panoramic scan. Using visual AI algorithms, it identified and selected compliant empty parking spaces in real time. It then hovered directly above the optimal empty parking space and recorded its precise coordinates. Afterward, the drone took off from the empty parking space and, using a low-altitude reverse flight mode to simulate a vehicle's perspective, explored the path along the parking lot's internal roads towards the user's vehicle's current location. It constructed a real-time topology map of the passable area, dynamically avoided fixed and moving obstacles, and recorded the coordinate sequence of all key path inflection points. Finally, it sent the final navigation path to the vehicle's navigation system for display, guiding the vehicle to the empty parking space.

[0016] Based on the verified path sequence, navigation instructions are generated to guide the user's vehicle to the target empty space in an orderly manner along the waypoint coordinates. The drone keeps leading the flight in front of the vehicle and monitors the path status in real time. When an obstacle or an empty parking space is detected, an alarm is issued through the vehicle's infotainment system. After reaching an empty parking space, the mission ends, and the drone automatically returns and lands on the roof platform.

[0017] When the vehicle-mounted drone takes off to scan and photograph the parking lot, the environmental adaptation module is activated. The environmental adaptation module continuously monitors environmental data throughout the mission. If the light intensity is detected to be too low, the drone is controlled to turn on night mode to enhance the image acquisition effect. If dynamic or static obstacles are identified in the path, the drone is controlled to perform hovering or obstacle avoidance actions. If the real-time image shows that the target parking space is occupied, the process of finding a new vacant parking space is triggered.

[0018] A charging port is set on the vehicle-mounted drone support platform. After the vehicle-mounted drone lands on the support platform on the roof of the vehicle, the main control unit locks the drone to land. After the drone lands and is locked, the charging mode is intelligently selected based on the drone's current power level, battery health status, subsequent task priority, and the vehicle's own energy reserves. The charging modes include fast charging and standard charging.

[0019] The advantages of this invention are: it enables the drone to quickly identify parking spaces in the surrounding area, analyze the parking space status, and plan the optimal route. With the help of vehicle-machine collaborative control, it enables the vehicle to autonomously navigate to an empty parking space. It mainly solves the problems of limited field of vision, untimely updates of parking space information, lack of route information in navigation software in large parking lots, and difficulty in parking in multi-obstacle environments that exist in traditional parking systems.

[0020] This invention significantly improves parking navigation efficiency and user experience in urban environments. Its technical effects are reflected in the following aspects: The aerial visual perception system based on a drone platform breaks through the field-of-view limitations of traditional vehicle-mounted sensors and fixed cameras. Through autonomous location finding and reverse path verification mechanisms, it provides users with real-time, field-verified information on available parking spaces and accessible routes, making it particularly suitable for traditional navigation blind spots such as large open-air parking lots and unfamiliar areas. Furthermore, by having the drone explore the route first, it avoids users' vehicles ineffectively entering areas with no parking spaces or impassable paths, saving significant search time. The adoption of a dual-stage "location finding-route exploration" task planning and dual-mode communication data link ensures the reliability and stability of the entire process from finding available spaces to guiding routes. Compared with the traditional cyclical detour search method, this invention significantly reduces vehicle fuel consumption and emissions, and improves parking efficiency. By enhancing the adaptability to ambient light, GPS interference, and dynamic obstacles, it ensures the successful execution of guidance tasks in different parking environments and the safety of the drone itself. Full-process automated control greatly reduces the user's operational burden, achieving a fundamental shift from "people finding parking spaces" to "parking spaces finding people." Attached Figure Description

[0021] The following is a brief explanation of the contents of each of the accompanying drawings and the markings in the drawings:

[0022] Figure 1 is a schematic diagram of the overall system architecture of this application;

[0023] Figure 2 is a flowchart of the parking space search process in this application;

[0024] Figure 3 is a flowchart of the UAV road exploration process in this application;

[0025] Figure 4 is the UAV environment adaptive decision diagram of this application;

[0026] Figure 5 is a schematic diagram of the UAV decision-making process in this application. Detailed Implementation

[0027] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and the description of the preferred embodiments.

[0028] This embodiment addresses a rapid parking guidance control scheme for parking spaces in a parking lot. Specifically, it provides a control scheme for rapid navigation and guidance of parking spaces for vehicles equipped with vehicle-mounted drones. The drone-assisted parking guidance control system includes a vehicle-mounted drone, a main control unit, and a human-machine interface module. The vehicle-mounted drone has a camera that can capture video within the parking lot area to locate available parking spaces. The human-machine interface module is an onboard host system that uses its screen to activate the auxiliary guidance function and provide feedback on guidance and navigation routes, thus achieving parking guidance control. Specifically, the user triggers the parking auxiliary guidance function through the human-machine interface module. Upon receiving this trigger, the main control unit drives the vehicle-mounted drone to take off and probe the parking area. Based on the probed data, the drone sends parking guidance data to the main control unit, which then displays it to the user through the human-machine interface module.

[0029] The vehicle-mounted drone is mounted on a vehicle with a corresponding drone landing platform. A charging port is set on the drone support platform. After the drone lands on the support platform on the roof, the main control unit locks the drone in place. Once the drone is locked in place, the system intelligently selects the charging mode based on the drone's current battery level, battery health, subsequent mission priorities, and the vehicle's own energy reserves. The charging modes include both fast charging and standard charging.

[0030] The vehicle-mounted drone includes the drone and cameras, communication units, photosensitive elements, etc. installed on the drone. The vehicle-mounted drone has a built-in communication unit, which is connected to the main control unit or the user's mobile terminal, or the vehicle-mounted drone's communication unit is connected to the main control unit or the user's mobile terminal through a cloud platform.

[0031] The communication unit of the vehicle-mounted drone communicates directly with the main control unit / user mobile terminal as the main link mode. Only when communication abnormalities occur in the main link mode or the communication delay does not meet the set requirements will it switch to the protection link mode. The protection link mode means that the communication unit of the vehicle-mounted drone communicates with the main control unit / user mobile terminal through the cloud platform, realizing the purpose of guiding and controlling the drone through a dual-link connection to the vehicle.

[0032] The vehicle-mounted drone is equipped with a photosensitive element. After the vehicle-mounted drone takes off, the drone's camera is activated, and the light signal from the photosensitive element is acquired. Based on the light signal, the camera switches to night mode. In night mode, the vehicle-mounted drone's control system controls the camera's ISO sensitivity and the supplementary light intensity of the supplementary light system to improve the video clarity of the parking lot captured by the drone at night or in low-light conditions, which facilitates subsequent guidance and control.

[0033] In this embodiment, the parking guidance control method based on drone assistance includes a user triggering a parking assistance guidance function through a human-computer interaction module. After receiving the user's triggering of the parking assistance guidance function, the main control unit drives the vehicle-mounted drone to take off and detect the parking area. Based on the detected image data, the main control unit sends parking guidance data to the main control unit, which then displays it to the user through the human-computer interaction module.

[0034] The main control unit responds to the user's parking space search command, drives the vehicle-mounted drone to take off and take aerial photos of the parking lot, and dynamically generates a two-stage "location search-path exploration" mission strategy: the first stage controls the drone to fly to the target area to perform a wide-area scan; the second stage instructs the drone to perform reverse low-altitude verification of the connecting path from the optimal empty parking space to the current position of the vehicle.

[0035] Vacancy location and reverse path verification include: After the UAV identifies and selects a qualified vacant parking space through its onboard visual sensors, the system automatically triggers the path verification engine: The UAV takes off from the coordinates of the vacant parking space and, from the perspective of simulating vehicle driving, performs a reverse exploration flight along the internal road network of the parking lot towards the current location of the user's vehicle; During the flight, it integrates visual SLAM and RTK-GPS data to build a topology map of the passable area in real time, dynamically avoids fixed and moving obstacles, and records the coordinate sequence of all key path inflection points and decision points, ultimately generating a navigation path that has been verified by the entity and is accessible to the vacant parking space.

[0036] During the mission, the vehicle-mounted drone flew to the target parking lot to conduct a high-altitude panoramic scan. Using visual AI algorithms, it identified and selected compliant empty parking spaces in real time. It then hovered directly above the optimal empty parking space and recorded its precise coordinates. Afterward, the drone took off from the empty parking space and, using a low-altitude reverse flight mode to simulate a vehicle's perspective, explored the path along the parking lot's internal roads towards the user's vehicle's current location. It constructed a real-time topology map of the passable area, dynamically avoided fixed and moving obstacles, and recorded the coordinate sequence of all key path inflection points. Finally, it sent the final navigation path to the vehicle's navigation system for display, guiding the vehicle to the empty parking space.

[0037] Based on the verified path sequence, navigation instructions are generated to guide the user's vehicle to the target empty space in an orderly manner along the waypoint coordinates. The drone keeps leading the flight in front of the vehicle and monitors the path status in real time. When an obstacle or an empty parking space is detected, an alarm is issued through the vehicle's infotainment system. After reaching an empty parking space, the mission ends, and the drone automatically returns and lands on the roof platform.

[0038] When the vehicle-mounted drone takes off to scan and photograph the parking lot, the environmental adaptation module is activated. The environmental adaptation module continuously monitors environmental data throughout the mission. If the light intensity is detected to be too low, the drone is controlled to turn on night mode to enhance the image acquisition effect. If dynamic or static obstacles are identified in the path, the drone is controlled to perform hovering or obstacle avoidance actions. If the real-time image shows that the target parking space is occupied, the process of finding a new vacant parking space is triggered.

[0039] A charging port is set on the vehicle-mounted drone support platform. After the vehicle-mounted drone lands on the support platform on the roof of the vehicle, the main control unit locks the drone to land. After the drone lands and is locked, the charging mode is intelligently selected based on the drone's current power level, battery health status, subsequent task priority, and the vehicle's own energy reserves. The charging modes include fast charging and standard charging.

[0040] In a preferred embodiment, the vehicle-mounted drone takes off and captures video footage to identify available parking spaces in the parking lot. A visual AI algorithm identifies and filters compliant available spaces in real time. The drone then flies directly above the optimal available parking space, hovers, and records its precise coordinates. The distance between the optimal parking space and the vehicle's current location is calculated. If the distance is greater than a set distance, a path verification engine is activated. The drone takes off from the parking space coordinates and, simulating a vehicle's perspective, performs a reverse exploration flight along the parking lot's internal road network towards the user's vehicle's current location. It records the coordinate sequence of all key path inflection points and decision points during the flight, ultimately generating a verified navigation path to the available parking space. This provides a navigation path for vehicles to reach distant parking spaces, facilitating timely arrival at the optimal parking space. Otherwise, if the distance is less than or equal to a set distance threshold, the path verification engine is disabled. The vehicle-mounted drone then uses lights and a horn to emit a warning sound, guiding the user to quickly reach the optimal parking space. When the distance is short, the location guidance signal can be sent directly to the user through the speaker and lighting system built on the drone. The user can quickly reach the empty parking space, reducing the possibility that the best empty parking space is occupied and improving the parking success rate.

[0041] In another preferred embodiment, when the distance exceeds a set distance threshold, the path verification engine is activated for path planning. If the distance is too far and the drone has left the optimal parking space, parking will fail. To reduce this, when the drone captures video of the parking lot, a visual AI algorithm is used to identify and filter compliant empty parking spaces in real time. The optimal empty parking space is selected based on the distance between the empty space and the vehicle. This involves selecting multiple closest compliant empty parking spaces as candidate spaces, and sorting them in order of distance priority, from farthest to closest to the vehicle. The drone hovers directly above the farthest candidate space and records its precise coordinates. Then, the drone takes off from the coordinates of the farthest candidate space and, simulating the perspective of a vehicle, travels along the internal road network of the parking lot according to the sorted candidate spaces. The system flies to the next available parking space, looping until it reaches the last and closest available parking space. Then, it flies towards the vehicle's location from the nearest available parking space, performing a reverse exploration flight from the user's current location. It records the coordinate sequence of all critical path inflection points and decision points during the flight, ultimately generating a verified navigation path that connects multiple available parking spaces. When driving along this path, the system approaches each available parking space in order of increasing distance. This interconnected planning avoids the time-consuming process of replanning a single available parking space if it is occupied during path verification. By planning multiple parking spaces at once, and linking them according to their distance, the system can reach multiple available spaces, significantly reducing the likelihood of all spaces being occupied and improving the reliability and accuracy of the guidance.

[0042] The guidance scheme in this embodiment belongs to the field of collaborative technology of new energy pickup trucks and unmanned systems. Specifically, it involves a parking space finding and navigation system and method based on the collaboration of UAVs and ground mobile platforms. The system integrates a UAV dispatch cabin, an airborne sensing unit, and an adaptive communication module into the new energy pickup truck to realize the rapid identification of parking spaces in the surrounding area, analysis of parking space status, and optimal path planning by the UAV. With the help of vehicle-machine collaborative control, the vehicle can autonomously navigate to an empty parking space. It mainly solves the problems of limited field of vision, untimely updates of parking space information, lack of path information in navigation software in large parking lots, and difficulty in parking in multi-obstacle environments in traditional parking systems.

[0043] Addressing the core pain points of inefficient parking space searching for drivers in urban environments and the inability of traditional navigation systems to detect real-time parking availability and route feasibility, this solution constructs an intelligent parking guidance system that coordinates "drones, vehicles, and cloud platforms." Through a four-level linkage mechanism, it achieves a closed loop of parking space discovery and route verification.

[0044] 1. Collaborative Detection and Decision-Making System:

[0045] The main control unit responds to the user's location search command by fusing drone aerial imagery, real-time vehicle GPS coordinates, parking lot GIS layers, and user vehicle size data in real time. Based on this, it dynamically generates a two-stage "location search-pathfinding" task strategy: the first stage controls the drone to fly to the target area for wide-area scanning; the second stage instructs the drone to perform reverse low-altitude verification of the connecting path from the optimal empty parking space to the vehicle's current location.

[0046] 2. Vacancy location and path reverse verification mechanism:

[0047] Once the drone identifies and selects a suitable vacant parking space using its onboard visual sensors, the system automatically triggers the path verification engine. The drone takes off from the coordinates of the vacant parking space and, simulating the perspective of a vehicle, performs a reverse exploration flight along the internal road network of the parking lot towards the user's current location. During the flight, it integrates visual SLAM and RTK-GPS data to construct a real-time topology map of the passable area, dynamically avoids fixed and moving obstacles, and records the coordinate sequences of all critical path inflection points and decision points. Ultimately, it generates a physically verified navigation path that leads to the vacant parking space.

[0048] Typical application scenario: In large multi-story open-air parking lots, drones can quickly locate the vacant space closest to the entrance and automatically explore an optimal path that avoids obstacles such as internal construction barriers and temporary parking.

[0049] 3. Dual-modal data link communication:

[0050] Constructing a hierarchical communication architecture to ensure control and data reliability in complex urban scenarios:

[0051] Main link mode (high-speed data mode): The drone establishes a direct communication link with the user's vehicle / mobile terminal to transmit high-definition panoramic images and real-time video streams, allowing users to intuitively confirm available space information;

[0052] Protective Link Mode (Strong Anti-interference Mode): Drone → Cloud Platform → Vehicle-to-Vehicle (V2V) When the drone is severely obstructed by buildings, causing the direct connection signal to deteriorate, it automatically switches to a communication mode relayed through the cloud platform to prioritize the low-latency and high-reliability transmission of critical commands (such as waypoint coordinates and path status).

[0053] 4. Environmental Adaptation Strategy:

[0054]

[0055] This intelligent parking guidance system consists of two core modules: a user vehicle mobility platform and an inspection drone. Through intelligent empty space recognition, reverse path verification, and vehicle-machine collaborative control, it achieves efficient parking guidance in urban parking environments. When a task is initiated, the user sends a location search command and injects vehicle size parameters through the vehicle's infotainment system. Based on parking lot GIS information, real-time traffic data, and drone range constraints, the onboard main control unit dynamically generates a three-stage task strategy: "wide-area scanning - precise positioning - path verification," simultaneously determining the drone's initial search coordinates.

[0056] During the mission, the drone flew to the target parking lot to conduct a high-altitude panoramic scan, using visual AI algorithms to identify and select compliant empty parking spaces in real time. It then hovered directly above the optimal empty parking space and recorded its precise coordinates. Afterward, the drone took off from the empty parking space and, using a low-altitude reverse flight mode to simulate a vehicle's perspective, explored a path along the parking lot's internal roads towards the user's vehicle's current location. It constructed a real-time topology map of the passable area, dynamically avoided fixed and moving obstacles, and recorded the coordinate sequence of all critical path inflection points. The user's vehicle then proceeded slowly or waited in a safe area according to the guidance.

[0057] The onboard main control unit integrates images, positioning, obstacle data transmitted from the drone, and the vehicle's real-time location to dynamically assess path feasibility, traffic risks, and environmental changes (such as sudden obstacles or deteriorating lighting conditions). This triggers dynamic path optimization, communication mode switching (direct / relay), or a re-exploration mechanism. The environment adaptation module responds hierarchically through a rules engine. For example, when a moving obstacle is detected, it controls the drone to hover and wait, and provides real-time steering and speed suggestions to the user's vehicle.

[0058] During the guidance phase, the system generates navigation instructions based on the verified path sequence, guiding the user's vehicle to the target parking space in an orderly manner along the waypoint coordinates. The drone maintains a leading flight ahead of the vehicle and monitors the path status in real time, issuing immediate alarms in case of any abnormalities. After the mission is completed, the drone automatically returns and lands on the vehicle's roof platform, completing precise repositioning through UWB / vision assistance. The onboard main control unit updates the local parking lot dynamic map and completes energy replenishment and data preparation for subsequent tasks, achieving fully automated closed-loop guidance from parking location to parking. The specific implementation method is as follows:

[0059] 1. Initialization and Task Planning: As shown in Figure 1, the system mainly consists of two core modules: a pickup truck mobile platform 10 and an inspection drone 20. At task startup:

[0060] S101: Users can issue parking space search commands through the pickup truck's emulator system, and simultaneously input priority parameters such as vehicle size and time requirements;

[0061] S102: The main control unit 11 calls up the parking lot's geographical information and real-time road data, and dynamically generates a three-stage task strategy of "scan-positioning-verification" based on the battery endurance of the drone 20, the camera's field of view, and the actual size of the user's vehicle, combined with ambient light and weather conditions.

[0062] S103: The main control unit 11 determines the initial scanning path and hovering detection altitude of the UAV based on the distribution of parking lot entrances and the internal road network structure, and generates a reverse path tracing waypoint recording instruction;

[0063] S104: The system initializes the UAV visual recognition algorithm, obstacle avoidance logic and environment adaptation module, and prepares to perform the tasks of empty parking space recognition and path feasibility verification.

[0064] 2. Vacancy search and confirmation:

[0065] S201: As shown in the flowchart in Figure 2, the drone 20 flies to the center of the target area and turns on all cameras to collect images of the entire area.

[0066] S202: The UAV onboard controller 21 processes the image, detects and filters out available parking spaces that can accommodate the user's car, and selects the optimal available parking space based on factors such as the distance from the user.

[0067] S203: Determine if the parking space is available (e.g., whether there are obstacles or if it is already occupied). If it is unavailable, ignore the parking space and re-filter; if it is available, fly directly above the parking space and hover.

[0068] S204: Information such as images of available parking spaces can be sent back to the vehicle's infotainment system for user confirmation. Once the user is satisfied, the road exploration mode can be activated.

[0069] 3. Path exploration and verification:

[0070] S301: As shown in the flowchart in Figure 3, the drone starts from an empty parking space and flies at low altitude towards the vehicle, sensing the path in real time.

[0071] S302: During flight, prioritize the widest, most lane-like path.

[0072] S303: As shown in Figure 5, through real-time semantic segmentation and environmental analysis, it continuously determines whether the road ahead is clear. If an obstacle is detected, it is marked as a "temporary obstacle," and the system retreats to the nearest fork in the road, selecting another untried path.

[0073] S304: If the path is clear, fly along the path for a period of time and record key GPS waypoints.

[0074] S305: Repeat S302-S304 until the drone reaches the vehicle's current location, successfully generating a complete feasible route from the vehicle to the available parking space.

[0075] S306: Sends the generated complete route information to the vehicle's infotainment system for vehicle navigation.

[0076] 4. Environmental adaptive response:

[0077] S401: As shown in the decision tree in Figure 4, the environment adaptation module continuously monitors environmental data throughout the entire task.

[0078] S402: If low light intensity is detected, control the drone to turn on night mode to enhance image acquisition.

[0079] S403: If dynamic or static obstacles are detected in the path, control the drone to perform hovering or obstacle avoidance maneuvers.

[0080] S404: If the real-time image shows that the target parking space is occupied, the process of finding a new available parking space is triggered.

[0081] 5. Communication and Energy Security:

[0082] S501: Throughout the process, the data communication gateway (4G / 5G+C-V2X) of the pickup truck mobile platform 10 and the dual-mode communication link (direct + relay) of the drone 20 ensure the stability and reliability of control commands and data transmission.

[0083] S502: After the mission is completed or when needed, the drone can return to the pickup truck and recharge via the cabin power supply interface to prepare for subsequent missions.

[0084] 7. Pickup truck security system collaboration:

[0085] The high-capacity power battery system of the Pipi Truck 10 provides flexible and efficient energy replenishment for the UAV 20 through the V2L (Vehicle-to-Load) external discharge interface 14. This interface is designed to meet the different power requirements of fast charging and standard charging for the UAV (for example, fast charging can reach 3.5kW, and standard charging is about 0.8-1.5kW), and has precise voltage and current regulation and safety protection mechanisms (overvoltage, overcurrent, short circuit and real-time temperature monitoring).

[0086] After the drone lands and locks onto the ground, the main control unit 11 intelligently selects the charging mode based on the drone's current battery level, battery health status, subsequent task priorities (such as whether multiple area positioning tasks need to be performed continuously), and the pickup truck's own energy reserves.

[0087] Fast charging mode: Activated when the user urgently needs to continue using the device or when the drone's battery is severely low, using a larger current to quickly increase the battery level to the working threshold (e.g., 40%).

[0088] Standard charging mode: Activated during breaks in regular tasks or when there is no urgent need, it charges the battery to full capacity with a standard current to optimize battery life.

[0089] Maintenance mode: Based on data from the battery management system, equalization charging or shallow charging and discharging maintenance is initiated during idle periods to extend battery pack life.

[0090] During the charging process, the main control unit 11 monitors parameters such as the SOC of the pickup truck battery 10, the charging status of the drone battery 20, and the temperature of the interface 14 in real time to ensure that the charging process is safe and efficient, and dynamically adjusts the charging strategy or issues an alarm when necessary.

[0091] Obviously, the specific implementation of this invention is not limited to the above-described methods. Any non-substantial improvements made using the inventive concept and technical solution of this invention are within the protection scope of this invention.

Claims

1. A parking guidance and control system based on unmanned aerial vehicle (UAV) assistance, characterized in that: It includes a vehicle-mounted drone, a main control unit, and a human-machine interaction module; the user triggers the parking assistance guidance function through the human-machine interaction module. After receiving the user's trigger of the parking assistance guidance function, the main control unit drives the vehicle-mounted drone to take off and detect the parking area. Based on the detection data, the drone sends parking guidance data to the main control unit, which then displays it to the user through the human-machine interaction module.

2. The parking guidance and control system based on unmanned aerial vehicle (UAV) assistance as described in claim 1, characterized in that: The vehicle-mounted drone has a built-in communication unit. The communication unit of the vehicle-mounted drone is connected to the main control unit or the user's mobile terminal, or the communication unit of the vehicle-mounted drone is connected to the main control unit or the user's mobile terminal through a cloud platform. The communication unit of the vehicle-mounted drone communicates directly with the main control unit / user's mobile terminal as the main link mode. Only when the main link mode experiences communication abnormalities or the communication delay does not meet the set requirements will it switch to the protection link mode. The protection link mode means that the communication unit of the vehicle-mounted drone communicates with the main control unit / user's mobile terminal through the cloud platform.

3. A parking guidance and control system based on unmanned aerial vehicle (UAV) assistance as described in claim 1 or 2, characterized in that: The vehicle-mounted drone is equipped with a photosensitive element. After the vehicle-mounted drone takes off, the drone's camera is activated, and the light signal from the photosensitive element is acquired. Based on the light signal, the camera switches to night mode. In night mode, the vehicle-mounted drone's control system controls the camera's ISO sensitivity and the supplementary light intensity of the supplementary light system.

4. A parking lot guidance and control method based on unmanned aerial vehicle (UAV) assistance, characterized in that: This includes a parking assistance and guidance function triggered by the user through the human-computer interaction module. After receiving the user's trigger on the parking assistance and guidance function, the main control unit drives the vehicle-mounted drone to take off and detect the parking area. Based on the detected image data, the drone sends parking guidance data to the main control unit, which then displays the data to the user through the human-computer interaction module.

5. The parking guidance control method based on unmanned aerial vehicle (UAV) assistance as described in claim 4, characterized in that: The main control unit responds to the user's parking search command, drives the vehicle-mounted drone to take off and take aerial photos of the parking lot, and dynamically generates a two-stage "location search-path exploration" mission strategy: the first stage controls the drone to fly to the target area to perform wide-area scanning. In the second phase, the command drone performs reverse low-altitude verification of the connection path from the optimal empty parking space to the vehicle's current position.

6. A parking lot guidance and control method based on unmanned aerial vehicle (UAV) assistance as described in claim 5, characterized in that: Vacancy location and reverse path verification include: After the UAV identifies and selects a qualified vacant parking space through its onboard visual sensors, the system automatically triggers the path verification engine: The UAV takes off from the coordinates of the vacant parking space and, from the perspective of simulating vehicle driving, performs a reverse exploration flight along the internal road network of the parking lot towards the current location of the user's vehicle; During the flight, it integrates visual SLAM and RTK-GPS data to build a topology map of the passable area in real time, dynamically avoids fixed and moving obstacles, and records the coordinate sequence of all key path inflection points and decision points, ultimately generating a navigation path that has been verified by the entity and is accessible to the vacant parking space.

7. A parking lot guidance and control method based on unmanned aerial vehicle (UAV) assistance as described in any one of claims 4-6, characterized in that: During the mission, the vehicle-mounted drone flew to the target parking lot to conduct a high-altitude panoramic scan. Using visual AI algorithms, it identified and selected compliant empty parking spaces in real time. It then hovered directly above the optimal empty parking space and recorded its precise coordinates. Afterward, the drone took off from the empty parking space and, using a low-altitude reverse flight mode to simulate a vehicle's perspective, explored the path along the parking lot's internal roads towards the user's vehicle's current location. It constructed a real-time topology map of the passable area, dynamically avoided fixed and moving obstacles, and recorded the coordinate sequence of all key path inflection points. Finally, it sent the final navigation path to the vehicle's navigation system for display, guiding the vehicle to the empty parking space.

8. A parking lot guidance and control method based on unmanned aerial vehicle (UAV) assistance as described in any one of claims 4-6, characterized in that: Based on the verified path sequence, navigation instructions are generated to guide the user's vehicle to the target empty space in an orderly manner along the waypoint coordinates. The drone keeps leading the flight in front of the vehicle and monitors the path status in real time. When an obstacle or an empty parking space is detected, an alarm is issued through the vehicle's infotainment system. After reaching an empty parking space, the mission ends, and the drone automatically returns and lands on the roof platform.

9. A parking lot guidance and control method based on unmanned aerial vehicle (UAV) assistance as described in any one of claims 4-6, characterized in that: When the vehicle-mounted drone takes off to scan and photograph the parking lot, the environmental adaptation module is activated. The environmental adaptation module continuously monitors environmental data throughout the mission. If the light intensity is detected to be too low, the drone is controlled to turn on night mode to enhance the image acquisition effect. If dynamic or static obstacles are identified in the path, the drone is controlled to perform hovering or obstacle avoidance actions. If the real-time image shows that the target parking space is occupied, the process of finding a new vacant parking space is triggered.

10. A parking lot guidance control method based on unmanned aerial vehicle (UAV) assistance as described in any one of claims 4-6, characterized in that: A charging port is set on the vehicle-mounted drone support platform. After the vehicle-mounted drone lands on the support platform on the roof of the vehicle, the main control unit locks the drone to land. After the drone lands and is locked, the charging mode is intelligently selected based on the drone's current power level, battery health status, subsequent task priority, and the vehicle's own energy reserves. The charging modes include fast charging and standard charging.

Images