Smart parking lot system and automatic parking system and method

Abstract

The invention discloses an intelligent parking lot system and an automatic parking system and method, and relates to the field of intelligent parking lots. The automatic parking system comprises a first UWB tag module, a laser radar, an industrial personal computer and an automatic driving controller. The first UWB tag module is arranged at the roof of a vehicle, the first UWB tag module receives a plurality of UWB positioning messages currently broadcasted by a site where the vehicle is located, the laser radar scans the site environment to obtain point cloud data of the current site, and the industrial personal computer positions the position of the vehicle according to the plurality of currently broadcasted UWB positioning messages and determines coordinates of the vehicle in the site currently. And a grid occupation map of the site is constructed according to the point cloud data of the current site, an idle parking space is identified from the grid occupation map of the site, an automatic parking track is planned, and the automatic parking track is adjusted in real time, so that the vehicle automatically runs to the idle parking space. According to the invention, the parking precision and safety can be improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent parking lots, and in particular to a smart parking lot system, an automatic parking system and a method. Background Art

[0002] The current automatic parking technology mainly relies on on-board cameras, ultrasonic radar and other sensors to realize automatic parking of vehicles. Although automatic parking technology has been applied to some models equipped with L2-level assisted driving systems, most systems still require drivers to monitor during the parking process, and most car manufacturers install automatic parking functions in high-end models, or open them to users through "hardware pre-embedded + OTA (Over-the-Air Technology) software paid activation / upgrade", which has the problem of complex operation.

[0003] Existing automatic parking technologies mainly rely on ultrasonic radars or cameras, which often cause inaccurate positioning due to signal interference or line of sight obstruction in complex or multi-source random interference environments (such as underground parking lots in large urban shopping malls or commercial buildings). They also cannot fully perceive the parking environment, limiting the overall effectiveness of automatic parking technology. Summary of the invention

[0004] The purpose of this application is to provide a smart parking system, an automatic parking system and a method, which can improve parking accuracy and safety.

[0005] To achieve the above objectives, this application provides the following solutions.

[0006] In a first aspect, the present application provides an automatic parking system, comprising: a first UWB tag module, a laser radar, an industrial computer and an automatic driving controller; the first UWB tag module is arranged on the roof of the vehicle; the first UWB tag module and the laser radar are respectively connected to the industrial computer, and the industrial computer is connected to the automatic driving controller; after the vehicle receives the automatic parking command, the first UWB tag module is used to receive a plurality of UWB positioning messages currently broadcast by the site where the vehicle is located; the laser radar is used to scan the site environment and obtain point cloud data of the current site; the industrial computer is used to locate the vehicle according to the plurality of UWB positioning messages currently broadcast, and determine the current coordinates of the vehicle in the site; a point cloud map of the current site is constructed according to the point cloud data of the current site, and a Lanelet2 map of the current site is determined according to the point cloud map of the current site, and the Lanelet2 map of the current site, the point cloud map of the current site and the current coordinates of the vehicle in the site are integrated to obtain a grid occupancy map of the current site; the industrial computer is also used to Identify vacant parking spaces from the grid occupancy map of the current site, and plan an automatic parking trajectory based on the grid occupancy map of the current site, with the current coordinates of the vehicle in the site as the starting point and the vacant parking space as the end point, and then control the vehicle to start according to the planned automatic parking trajectory; during the automatic parking process of the vehicle, the laser radar is used to scan the site environment in real time to obtain real-time point cloud data of the site; the industrial computer is also used to locate the vehicle according to multiple UWB positioning messages received in real time by the first UWB tag module, and determine the real-time coordinates of the vehicle in the site; construct a real-time point cloud map of the site according to the real-time point cloud data of the site obtained by the laser radar, and determine the real-time Lanelet2 map of the site according to the real-time point cloud map of the site, and fuse the real-time Lanelet2 map, the real-time point cloud map and the real-time coordinates of the vehicle in the site to obtain the real-time grid occupancy map of the site; the industrial computer is also used to plan or adjust the automatic parking trajectory in real time based on the real-time grid occupancy map, so that the vehicle automatically runs to the vacant parking space.

[0007] Optionally, the automatic parking system also includes: a second UWB tag module, a third UWB tag module, a fourth UWB tag module and a fifth UWB tag module; the second UWB tag module, the third UWB tag module, the fourth UWB tag module and the fifth UWB tag module are respectively arranged at the two headlights and the two taillights of the vehicle in a one-to-one correspondence; the second UWB tag module, the third UWB tag module, the fourth UWB tag module and the fifth UWB tag module are all connected to the automatic driving controller; when the vehicle is working in the keyless entry and start mode, the first UWB tag module, the second UWB tag module, the third UWB tag module, the fourth UWB tag module and the fifth UWB tag module all receive the ranging message from the key end, and send the ranging message to the automatic driving controller; the automatic driving controller is used to calculate the position of the key end, and unlock the vehicle according to the unlocking instruction sent by the key end at the position, or start the vehicle according to the start instruction sent by the key end at the position; the key end is a mobile phone or a car key.

[0008] In the second aspect, the present application provides a smart parking system, which includes: multiple UWB base stations, a mobile charging vehicle and the aforementioned automatic parking system; multiple UWB base stations are arranged at different locations in the parking lot; the parking lot is provided with a parking and charging area; multiple UWB base stations are used to broadcast multiple UWB positioning messages to a first UWB tag module in the automatic parking system; the automatic parking system is used to control the vehicle to automatically move to an empty parking space according to the received multiple UWB positioning messages; when the empty parking space is an empty parking space in the parking and charging area, after the automatic parking system controls the vehicle to automatically move to the empty parking space in the parking and charging area, the UWB base station is used to use the first UWB The tag module receives the parking coordinates of the vehicle located by the industrial computer, and obtains the current coordinates of the mobile charging vehicle at the same time; the UWB base station is also used to control the mobile charging vehicle to move to the location of the vehicle according to the current coordinates of the mobile charging vehicle and the parking coordinates of the vehicle; after the mobile charging vehicle moves to the location of the vehicle, the UWB base station is also used to transmit the charging port opening instruction to the industrial computer through the first UWB tag module; the industrial computer is used to open the charging port cover of the vehicle according to the charging port opening instruction, and feedback the information that the charging port cover has been opened to the UWB base station through the first UWB tag module; the UWB base station is also used to control the mobile charging vehicle to perform the gun insertion operation and start charging the vehicle after receiving the information that the charging port cover has been opened.

[0009] Optionally, the smart parking system also includes: a parking lot host; multiple UWB base stations are connected to the parking lot host; the industrial computer in the automatic parking system is used to transmit the real-time coordinates of the vehicle in the venue to the parking lot host through a first UWB tag module and multiple UWB base stations; the parking lot host stores a digital twin parking lot virtual model; the parking lot host is used to locate the vehicle position in real time in the digital twin parking lot virtual model according to the real-time coordinates of the vehicle in the venue, and display it in the digital twin parking lot virtual model in the form of a virtual vehicle model.

[0010] Optionally, the smart parking system also includes: a contactless payment system; the contactless payment system is used to receive payment information generated by the parking lot host through the first UWB tag module, and deduct parking fees from the account bound between the user and the parking lot according to the payment information.

[0011] In a third aspect, the present application provides an automatic parking method, comprising: after a vehicle receives an automatic parking instruction, receiving multiple UWB positioning messages currently broadcast by a site where the vehicle is located through a first UWB tag module; obtaining point cloud data of the current site; locating the vehicle position according to the multiple UWB positioning messages currently broadcast, and determining the current coordinates of the vehicle in the site; constructing a point cloud map of the current site, and determining the Lanelet2 map of the current site based on the point cloud map of the current site; fusing the Lanelet2 map of the current site, the point cloud map of the current site, and the current coordinates of the vehicle in the site to obtain a grid occupancy map of the current site; identifying an empty parking space from the grid occupancy map of the current site, and based on the grid occupancy map of the current site, Taking the current coordinates of the vehicle in the venue as the starting point and the vacant parking space as the end point, an automatic parking trajectory is planned, and then the vehicle is controlled to start according to the planned automatic parking trajectory; during the automatic parking process of the vehicle, the vehicle position is located according to multiple UWB positioning messages received in real time by the first UWB tag module, and the real-time coordinates of the vehicle in the venue are determined; the real-time point cloud data of the venue is acquired to construct a real-time point cloud map of the venue, and the real-time Lanelet2 map of the venue is determined according to the real-time point cloud map of the venue; the real-time Lanelet2 map, the real-time point cloud map and the real-time coordinates of the vehicle in the venue are integrated to obtain the real-time grid occupancy map of the venue; based on the real-time grid occupancy map, the automatic parking trajectory is adjusted in real time to make the vehicle automatically run to the vacant parking space.

[0012] Optionally, if multiple vehicles receive automatic parking instructions at the same time, the automatic parking trajectory is adjusted in real time, specifically including: defining an object with a moving speed greater than a speed threshold as a dynamic interferer; the object includes pedestrians and vehicles; predicting the trajectory of the dynamic interferer, and projecting the dynamic interferer in the form of a rectangular outline into a Frenet coordinate system established based on a reference path at each predicted trajectory point; the reference path is the automatic parking trajectory planned by the vehicle; the horizontal axis of the Frenet coordinate system represents the reference path, and the vertical axis of the Frenet coordinate system represents the distance; determining the shortest distance between the rectangular outline point on the predicted trajectory and the horizontal axis; if the shortest distance is less than the automatic parking path safety distance, it is determined that the dynamic interferer invades the automatic parking trajectory planned by the vehicle; projecting the dynamic interferer at the trajectory point where the shortest distance is located into the ST diagram, and using a dynamic programming algorithm to plan a speed curve that avoids the dynamic interferer in the ST diagram; the horizontal axis of the ST diagram represents time, and the vertical axis of the ST diagram represents the reference path; performing secondary smoothing on the speed curve to obtain an optimal speed curve that avoids the dynamic interferer.

[0013] According to the specific embodiments provided in this application, this application discloses the following technical effects.

[0014] The present application provides a smart parking system, an automatic parking system and a method. A first UWB tag module is arranged on the roof of a vehicle, and uses ultra-wideband positioning technology to achieve centimeter-level precise positioning. It can maintain high accuracy even in a complex environment with interference from dynamic obstacles, thereby significantly improving the accuracy of parking. In addition, during the automatic parking process of the vehicle, the environment of the site is perceived in real time by receiving multiple UWB positioning messages broadcast by the site in real time, providing comprehensive environmental information of the site for automatic parking, and achieving safer parking operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 This is a layout diagram of a UWB tag module provided in one embodiment of the present application.

[0017] Figure 2 A schematic diagram of node arrangement provided in one embodiment of the present application.

[0018] Figure 3 A schematic diagram of a parking process provided in another embodiment of the present application.

[0019] Figure 4 A schematic diagram of the docking and charging process provided in another embodiment of the present application.

[0020] Figure 5 A schematic diagram of a flow chart of an automatic parking method provided in accordance with an embodiment of the present application.

[0021] Reference numerals: first PEPS node-1, second PEPS node-2, third PEPS node-3, fourth PEPS node-4, fifth PEPS node-5. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0023] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0024] The goal of Automated Valet Parking (AVP) is to completely replace the driver in the parking scenario, realize the unmanned driving function of the vehicle, and complete the parking task. Similar to the manual parking behavior process, the driver is replaced by sensors for environmental perception, system decision-making, and vehicle control. The typical complete valet parking system used by autonomous valet parking consists of smart parking lot infrastructure, valet parking host vehicle system, cloud and interactive terminals. Smart parking lots should have perception and communication functions. Perception means that smart parking lots rely on various sensors such as cameras, ultrasonic radars, and valet parking servers configured in the parking lot to perceive valet parking host vehicles, other vehicles, pedestrians, and other targets; communication means that real-time communication between the vehicle end and the field end can be achieved.

[0025] The valet parking master car is a vehicle with communication functions and basic intelligent driving functions, mainly including perception system, decision system and execution system. The valet parking master car should be able to communicate intelligently with the parking lot and receive the trajectory and control commands sent by the valet parking server; at the same time, the valet parking master car can execute the trajectory and emergency obstacle avoidance instructions sent by the valet parking server, and can rely on the vehicle's own sensors to achieve emergency obstacle avoidance parking in abnormal situations. The cloud has interactive front-end, data storage and computing functions. Users communicate with the cloud valet parking interactive front-end through mobile devices to view the vehicle parking status and control the start or cancellation of parking. The cloud processes user data and sends control commands to the valet parking server. The interactive terminal provides the user with an interactive environment, which is generally a smartphone or the central control screen of the valet parking master car. Users can enable valet parking, view the valet parking process, and receive takeover notifications from the cloud in abnormal situations when valet parking is executed.

[0026] Autonomous valet parking technology mainly relies on ultrasonic radar or camera. In complex or multi-source random interference environments (such as underground parking lots in large urban shopping malls or business buildings), inaccurate positioning is often caused by signal interference or line of sight obstruction. In view of this, in an exemplary embodiment, the present application embodiment provides an automatic parking system, including: a first UWB tag module, a laser radar, an industrial computer and an automatic driving controller. The first UWB tag module is set on the roof of the vehicle; the first UWB tag module and the laser radar are connected to the industrial computer, and the industrial computer is connected to the automatic driving controller.

[0027] After the vehicle receives the automatic parking command, the first UWB tag module is used to receive multiple UWB positioning messages currently broadcast by the site where the vehicle is located; the laser radar is used to scan the site environment and obtain the point cloud data of the current site; the industrial computer is used to locate the vehicle position according to the multiple UWB positioning messages currently broadcast, and determine the current coordinates of the vehicle in the site; construct a point cloud map of the current site according to the point cloud data of the current site, and determine the Lanelet2 map of the current site according to the point cloud map of the current site, and integrate the Lanelet2 map of the current site, the point cloud map of the current site and the current coordinates of the vehicle in the site to obtain the grid occupancy map of the current site; the industrial computer is also used to identify vacant parking spaces from the grid occupancy map of the current site, and based on the grid occupancy map of the current site, plan an automatic parking trajectory with the current coordinates of the vehicle in the site as the starting point and the vacant parking space as the end point, and then control the vehicle to start according to the planned automatic parking trajectory.

[0028] During the automatic parking process of the vehicle, the laser radar is used to scan the site environment in real time to obtain real-time point cloud data of the site; the industrial computer is also used to locate the vehicle position according to multiple UWB positioning messages received in real time by the first UWB tag module, and determine the real-time coordinates of the vehicle in the site; a real-time point cloud map of the site is constructed according to the real-time point cloud data of the site obtained by the laser radar, and a real-time Lanelet2 map of the site is determined according to the real-time point cloud map of the site, and the real-time Lanelet2 map, the real-time point cloud map and the real-time coordinates of the vehicle in the site are integrated to obtain a real-time grid occupancy map of the site; the industrial computer is also used to plan or adjust the automatic parking trajectory in real time based on the real-time grid occupancy map, so that the vehicle automatically runs to an empty parking space.

[0029] UWB (Ultra Wide Band) technology will enable the implementation of L4 AVP. UWB, Bluetooth or Wi-Fi are essentially radio waves, but UWB is completely different from other wireless transmission technologies. It is a pulse-based wireless carrier communication technology that uses nanosecond non-sinusoidal narrow pulses for data transmission. It occupies a wide spectrum range (3.1GHz to 10.6GHz). In contrast, Bluetooth's spectrum range is only 2.402GHz to 2.480GHz; and Wi-Fi only has some very narrow spectrums at 2.4GHz and 5GHz respectively. Unlike the continuous wave signals of Bluetooth and Wi-Fi, UWB transmits pulse signals, which gives it two unique advantages: precise positioning and high security. Compared with other current communication technologies, UWB has strong anti-interference capabilities, low power and ultra-large bandwidth, and theoretically lower costs.

[0030] The first UWB tag module installed on the roof can receive UWB signals both inside and outside the car. The first UWB tag module receives high-precision positioning signals from the UWB base station at the site (e.g. parking lot), which increases the high-precision positioning capability of the vehicle when performing self-AVP. UWB technology can achieve more accurate vehicle positioning and automatic parking, greatly improving the efficiency and safety of parking.

[0031] In another exemplary embodiment of the present application, the automatic parking system further includes: a second UWB tag module, a third UWB tag module, a fourth UWB tag module, and a fifth UWB tag module. The second UWB tag module, the third UWB tag module, the fourth UWB tag module, and the fifth UWB tag module are respectively arranged at the two headlights and the two taillights of the vehicle in a one-to-one correspondence. The second UWB tag module, the third UWB tag module, the fourth UWB tag module, and the fifth UWB tag module are all connected to the automatic driving controller. When the vehicle operates in keyless entry and start mode, the first UWB tag module, the second UWB tag module, the third UWB tag module, the fourth UWB tag module and the fifth UWB tag module all receive the ranging message from the key end, and send the ranging message to the automatic driving controller; the automatic driving controller is used to calculate the position of the key end, and unlock the vehicle according to the unlocking instruction sent by the key end at the position, or start the vehicle according to the start instruction sent by the key end at the position; the key end is a mobile phone or a car key.

[0032] Figure 1 The layout of the UWB tag modules for this application. A total of four UWB tag modules are placed at the headlights and taillights: A2 UWB, A3 UWB, A4 UWB and A5 UWB, corresponding to the second UWB tag module, the third UWB tag module, the fourth UWB tag module and the fifth UWB tag module. A UWB tag module is placed on the roof: A1 UWB, corresponding to the first UWB tag module.

[0033] PEPS (Passive Entry & Passive Start) is a keyless vehicle entry system that uses the CCC3.0 specification as a theoretical guide. It uses the complementary advantages of UWB, BLE (Bluetooh Low Energy), and NFC (Near Field Communication) to achieve an interconnected third-generation digital key solution, providing a safer, more accurate, and smarter car key. By placing a UWB tag module on the roof, the digital key can be located and authenticated. The passive entry and start of the vehicle are controlled by the digital key on the mobile phone, which accurately locates the position of the digital key and enables accurate automatic unlocking of the vehicle. At the same time, keyless technology provides great convenience and flexibility for drivers. Through digital car key secret key management, car owners can grant other users access to their vehicles through the cloud.

[0034] There is no compromise in the functions and performance of PEPS and AVP. In PEPS mode, the first UWB tag module, the second UWB tag module, the third UWB tag module, the fourth UWB tag module and the fifth UWB tag module all work, while in AVP mode, they mainly rely on the first UWB tag module to work. Therefore, the location of the first UWB tag module can be used as the fifth PEPS node 5, and the fifth PEPS node 5 can also be called an AVP+PEPS node. The AVP+PEPS node works in both AVP mode and PEPS mode. The locations of the second UWB tag module, the third UWB tag module, the fourth UWB tag module and the fifth UWB tag module can be used as four PEPS nodes, respectively. Figure 2 The first PEPS node 1, the second PEPS node 2, the third PEPS node 3 and the fourth PEPS node 4 in the PEPS mode. In PEPS mode, the AVP+PEPS node cooperates with the other four PEPS nodes to complete the measurement of the smart key position, and sends the ranging information to the automatic driving controller to complete the calculation of the smart key position, thereby determining the subsequent unlocking and starting operations. When entering the AVP mode, the AVP+PEPS node begins to continuously receive UWB messages that comply with the 4z protocol broadcast by the UWB base station deployed in the venue. The information contained in the UWB message is similar to the satellite positioning signal received in the open air. The AVP+PEPS node sends the received UWB message to the industrial computer to calculate the coordinates of the vehicle in the venue, thereby realizing the AVP operation. Since the RF performance of the AVP+PEPS node itself is better than that of the PEPS node and can cope with more complex scenarios, the AVP+PEPS node fully meets the needs of PEPS when working in PEPS mode. The AVP+PEPS node supports signals from the roadside and UWB base stations at least 50 meters away.

[0035] The automatic parking system of this application uses UWB to achieve high-precision positioning and interconnected communication, and can realize low-speed automatic driving, autonomous obstacle avoidance, intelligent parking space search, and autonomous vehicle parking in and out in smart parking lots. After the driver gets off the car at the designated drop-off point in the smart parking lot, he issues a parking instruction through a mobile terminal. After the vehicle receives the parking instruction, it can automatically drive to the parking space without the driver's operation and monitoring; the driver issues a pick-up instruction through a mobile terminal, and after the vehicle receives the pick-up instruction, it can automatically drive from the parking space to the designated pick-up point; if multiple vehicles receive parking instructions at the same time, multiple vehicles can be dynamically and automatically waiting to enter the parking space; during the automatic driving of the vehicle, the on-site traffic rules formulated by the parking lot operator are followed. Complying with the on-site traffic rules formulated by the parking lot operator means that the vehicle understands and judges the traffic rules based on the signs in the venue.

[0036] In order to achieve higher precision automatic parking, the automatic parking system of the present application is also equipped with an advanced driver assistance system (ADAS), which uses UWB technology and ADAS technology to achieve high-precision automatic parking, reduce the driver's operating burden, reduce system costs, and promote the popularization of automatic parking technology on ordinary models.

[0037] Based on the same inventive concept, an embodiment of the present application also provides a smart parking system using the above-mentioned automatic parking system. In an exemplary embodiment, a smart parking system includes: multiple UWB base stations, a mobile charging vehicle, and the aforementioned automatic parking system. Multiple UWB base stations are arranged at different locations in the parking lot; the parking lot is provided with a parking and charging area. Multiple UWB base stations are used to broadcast multiple UWB positioning messages to the first UWB tag module in the automatic parking system; the automatic parking system is used to control the vehicle to automatically run to an empty parking space according to the received multiple UWB positioning messages.

[0038] When the free parking space is a free parking space in the parking and charging area, after the automatic parking system controls the vehicle to automatically move to the free parking space in the parking and charging area, the UWB base station is used to receive the parking coordinates of the vehicle located by the industrial computer through the first UWB tag module, and simultaneously obtain the current coordinates of the mobile charging vehicle. The UWB base station is also used to control the mobile charging vehicle to move to the location of the vehicle according to the current coordinates of the mobile charging vehicle and the parking coordinates of the vehicle. After the mobile charging vehicle moves to the location of the vehicle, the UWB base station is also used to transmit the charging port opening instruction to the industrial computer through the first UWB tag module; the industrial computer is used to open the charging port cover of the vehicle according to the charging port opening instruction, and feedback the information that the charging port cover has been opened to the UWB base station through the first UWB tag module; the UWB base station is also used to control the mobile charging vehicle to perform the gun insertion operation after receiving the information that the charging port cover has been opened, and start charging the vehicle.

[0039] Parking process Figure 3 As shown, multiple UWB base stations in the parking lot constitute Figure 3 The UWB field end in the docking process is as follows Figure 4 shown.

[0040] Existing charging stations require drivers to manually connect charging equipment for charging, which is time-consuming and laborious. This inconvenience is particularly prominent when fast charging is required. This application uses mobile charging vehicles and automatic charging technology to achieve automatic charging of vehicles after parking, solve the time-consuming and inconvenient problems of manual charging, improve charging efficiency, and meet the needs of fast charging.

[0041] In another exemplary embodiment of the present application, the smart parking system further includes: a parking host. Multiple UWB base stations are connected to the parking host; the industrial computer in the automatic parking system is used to transmit the real-time coordinates of the vehicle in the venue to the parking host through the first UWB tag module and multiple UWB base stations; the parking host has a digital twin parking virtual model stored inside; the parking host is used to locate the vehicle position in the digital twin parking virtual model in real time according to the real-time coordinates of the vehicle in the venue, and display it in the digital twin parking virtual model in the form of a virtual vehicle model.

[0042] The goal of digital twins is to create a virtual model that can be as close to objects in the real world as possible and describe the situation in real time. To achieve this goal, there are three steps: the first step is to create a virtual model digitally and simulate it. Simulation refers to the dynamic simulation of the working environment of the physical entity and the physical process occurring in the real environment based on modeling. Unlike general three-dimensional modeling, digital twins have the characteristics of multi-physics and multi-scale, and can accurately simulate events according to the changes in time and space levels. This is also one of the main reasons why digital twins are currently used in many industries. A major advantage of digital twin modeling and simulation is the low cost of virtual entity operation and testing: by directly modifying some parameters such as thermal conductivity, hardness, humidity, current behavior, etc., to simulate various situations, there is no need to frequently operate the physical entity. The second step is data perception and fusion. From the establishment of the model to the establishment of the physical virtual system, each stage requires a lot of data support. This is a concept throughout the life cycle of digital twin products. Therefore, the effective extraction, fusion and analysis of data are also important topics. Due to the development of the Internet of Things, more and more devices equipped with Radio Frequency Identification (RFID) or smart sensors are put into use, making it easier for the physical end to capture data throughout the product life cycle. However, the data generated by the product at work is often massive and multi-dimensional. In addition to its own real-time data, there are also environmental data, historical data, etc., some of which are meaningless for maintaining the synchronization of physical-virtual models. Therefore, in order to simplify data, improve channel utilization and data processing efficiency, data filtering is of great significance. More importantly, the physical end usually uses a multi-sensor model to improve the accuracy of the data, so data from different sources must be fused before analysis. The third step is the interaction between virtual and physical space. The physical side notifies the virtual side of the parameter update at an appropriate frequency to maintain the synchronization and effectiveness of the model. At the same time, the virtual side feeds back the analysis results to the physical side in real time to assist in decision-making. This physical-virtual interaction gives the Digital Twin (DT) model a dynamic feature, enabling it to simultaneously match the work progress and working status of the product to evaluate its quality, life and health. Among them, the third step of the interaction process between virtual and physical space is: the parking lot contains monitoring information, such as parking space geomagnetic sensors and video surveillance information. By adding the visualization display of these sensors in the digital twin environment and using the FanLink Internet of Things platform in the background, the sensor information is pushed to the background of the smart parking lot, realizing the fusion of virtual and real physical information.

[0043] The physical virtual system is built on the information exchange mechanism of the real system, its high-level virtual model, and subsequent adjustments according to the established workflow. In the real workspace, the controllers and external sensors of the physical virtual system are used to accumulate the necessary information. This accumulated information is then organized, checked, and compared with the analysis results obtained by the virtual sensor components. Subsequently, the data is updated to achieve synchronous operation of the virtual prototype, which in turn enables the continuous accurate operation of the digital twin and can be used to provide guidance for the future operation of the physical model in the real workspace.

[0044] The method for creating a virtual model of a digital twin parking lot is as follows: (1) Use a handheld device to collect point cloud information, such as XGRIDS's innovative handheld scanner, to collect data and obtain image data from multiple angles containing point cloud information. (2) Use Context Capture software to automatically identify and match feature points in the image to generate an accurate point cloud model. (3) Use 3Ds Max software to perform local refinement on the constructed point cloud model. For traffic elements in the simulation, use 3Ds Max to construct road markings, signs, traffic signs, etc., and use Photoshop to restore the map to obtain a virtual model. (4) Use World Edit software to edit road information so that the virtual model has road traffic rules information.

[0045] In another exemplary embodiment of the present application, the smart parking system further includes: a non-contact payment system. The non-contact payment system is used to receive payment information generated by the parking lot host through the first UWB tag module, and deduct parking fees from the account bound between the user and the parking lot according to the payment information.

[0046] Most of the existing parking lot and charging station payment systems require on-site payment or manual payment using mobile phone applications, which is cumbersome and inconvenient. It not only increases the waiting time of users, but also reduces the operational efficiency of parking lots. This application integrates contactless payment technology to automatically complete the settlement of parking fees without manual operation by users, simplifies the payment process, and improves user experience and the operational efficiency of parking lots.

[0047] Mobile charging is a process to achieve automatic charging after parking new energy vehicles. In this process, the parking lot, mobile charging vehicles and vehicles are integrated and interacted to achieve motion control, mutual positioning, extension of the robotic arm, etc. of the mobile charging vehicles, and the action of inserting the charging gun into the vehicle charging port. Mobile charging is a key link in liberating car owners to fully realize automatic parking. At the same time, the cloud can be used to optimize the parking situation of vehicles in the smart charging area, realize "car looking for charging piles" and "charging piles looking for cars", and improve the utilization rate of charging piles.

[0048] Drivers can request the charging function through mobile terminals, and the mobile charging vehicle will go to the designated parking space by itself to provide charging services, achieving 100% coverage of parking space charging functions, and relying on UWB communication functions to realize functions such as contactless payment, vehicle interconnection, and remote takeover.

[0049] In order to cooperate with the realization of automatic parking and automatic charging, improve the functions of the smart parking system, and further carry out hierarchical and zoning planning of parking lots. In view of the large number of intelligent entities in the field, the diversity and flexibility of user needs, and considering factors such as economy, safety, and efficiency of use, the parking lot is divided into regions based on multi-modular construction planning and intelligent iterative development methods to realize the intelligent planning and design of each area of ​​the parking lot, providing users with efficient, intelligent, and safe parking and charging services. In view of the characteristics of user parking and charging needs, stay time, diversity and personalization of vehicle models, based on clustering analysis methods such as K-means clustering, the parking and charging area is divided into ordinary parking areas, ordinary charging areas, and emergency charging areas, and charging areas are reserved for mobile charging robots and other equipment. In view of the many emergencies, congestion and chaos in the parking lot, special traffic area designs are carried out based on road grid design and location optimization methods, such as emergency passages and disabled passages, to ensure that vehicles to be parked and charged can quickly enter specific areas or evacuate vehicles in time under special circumstances. In view of the large disparity in utilization of different areas and insufficient dynamic allocation, based on intelligent decision-making algorithms such as Markov Decision Processes (MDP), a method of intelligent dynamic division of regions is carried out. According to vehicle flow, vehicle parking and charging demand, and the demand for mobile power equipment in the parking lot, the space allocation of each area in the parking lot is automatically adjusted to improve the utilization rate of parking spaces and reduce vacancy. In view of the safety issues such as collision safety, electrical safety and equipment safety in the integrated parking and charging parking lot, based on risk assessment and safety risk methods, the interaction between vehicles and parking lots and the potential safety hazards of equipment in the parking lot are considered to set standards for the safe distance between parking spaces and arrange emergency equipment. In view of the problems of parking space reservation, management, and guidance in the parking lot, based on questionnaires and on-site surveys, parking space marking schemes are involved, including digital indications, LED display screens, and ground markings. In view of the problems of large parking demand and many vehicle driving conflicts during the peak parking period, based on the traffic flow modeling method combined with the Internet of Things, considering the road network and geographical location around the parking lot, a simulation model of vehicle entry and exit and parking demand is established, and the traffic flow planning and optimization methods of the parking lot are studied to improve the operation efficiency of the parking lot.

[0050] In order to improve the flexibility, scalability and coordination of the parking system, based on the modular principle and software architecture theory, the modular design architecture of the parking system is studied, and different modules are divided according to functions, such as automatic parking module, parking space management module, charging pile management module, parking lot monitoring module and user management module, and interfaces are reserved for communication and integration between different modules. In order to ensure the reliability and stability of the parking system and achieve load balancing and fault tolerance, a distributed server architecture based on distributed consistency analysis methods is studied, and different servers are arranged according to the categories of functional modules. For the problem of numerous intelligent entities and complex communication networks in the field, an efficient and reliable communication architecture is constructed based on MQTT, CoAP, and RESTful Internet of Things communication protocols to achieve data transmission between devices and interaction between the system and users. The layout scheme of wireless / wired communication equipment, communication gateway equipment, and data transmission equipment based on graph theory, network topology analysis and other methods is designed to efficiently arrange the communication network in the parking lot. The coverage of different layout schemes is evaluated by using technologies such as coverage analysis and signal strength simulation, and the layout is optimized based on particle swarm algorithms. In response to problems such as untimely vehicle information feedback, difficulty in parking space reservation, and inefficient parking lot management, a safe and intelligent parking-vehicle-person interaction software architecture was developed based on user demand analysis and user interface design principles.

[0051] Based on the above parking lot zoning planning and the modular design architecture of the parking lot system, in order to further optimize the global perception information fusion of the parking lot for the parking-space matching needs, the distribution layout of sensors in the parking lot is optimized: for the limited coverage of sensors, based on genetic algorithms, simulated annealing and other intelligent optimization algorithms, according to the size, layout and parking density of the parking lot, the number and range of parking spaces that each sensor can cover are determined. For the problem of weak perception breadth and poor robustness of the parking lot, considering the layout and complex electromagnetic environment in the parking lot, based on ultra-wideband carrier communication technology, the location and number of anchor points and sensors are optimized to improve the layout efficiency and coverage quality. Based on the structure and terrain of the parking lot, the hybrid sensor layout scheme is studied to determine the installation location of the sensor, including ground installation, wall installation, ceiling installation, etc. Based on heuristic algorithms, ant colony algorithms and other methods, the location and layout of the sensor are optimized by searching and adjusting the sensor location, maximizing the parking space coverage of the parking lot and minimizing the overlapping area of ​​the sensors, thereby reducing data conflicts and improving the accuracy of parking space matching.

[0052] In order to ensure the normal charging of vehicles in the parking lot, the parking lot electricity can be optimized and the charging equipment can be planned. In view of the characteristics of large energy demand and power fluctuations of electrical equipment in parking lots, based on time series analysis and regression analysis, the design of the power supply system is studied, including the access method of power supply (such as mains), distribution system design (such as distribution cabinets, line layout, etc.) and backup power supply (such as uninterruptible power supply system) configuration design. In view of the complex electric field environment of parking lots and the difficulty in obtaining the power status of electrical equipment, based on sensor networks and data acquisition technology, real-time monitoring and data acquisition methods of the power system in parking lots are studied to achieve accurate acquisition of the power status of each electrical equipment. Based on distributed control and energy flow coordination methods, the optimal power distribution management decision framework is constructed, and the real-time, reliable and efficient power distribution architecture is studied.

[0053] In view of the large number of electrical equipment in the field, high time-varying power consumption and strong uncertainty, based on the energy demand prediction and analysis method combined with data mining technology, the division method of the system energy demand prediction module and the energy analysis module is studied. The energy demand prediction module predicts the power demand of each module in the field based on the historical data of the electrical equipment in the field and the power consumption behavior pattern. The energy analysis module determines the energy supply strategy in the field based on the prediction results, providing a reference for the decision-making of power distribution management. Considering the prediction and analysis results of power usage, based on the optimization scheduling algorithm, the overall optimization configuration method of the parking lot power is studied to achieve the rational utilization and management of power resources.

[0054] In view of the limited overall power of the parking lot and the large fluctuations in charging power demand, the total number of fast and slow charging piles and mobile charging vehicles in the parking lot is determined based on game theory and time series analysis methods. In view of the various types of vehicles, different charging rates, and differences in vehicle and pile types, the charging configuration plan is optimized based on the combination of subjective and objective weighting methods to meet the charging needs of car owners and optimize energy utilization. Considering the layout of charging equipment and the relative position of parking spaces, the existing power grid capacity and power supply stability, the safety and protection measures of power facilities, reserved space and power supply capacity, the layout plan of charging facilities is studied based on the hierarchical analysis method and principal component analysis method, including the spatial layout of fast and slow piles and the reserved position of mobile charging vehicles, to balance the overall charging demand of vehicles and energy utilization efficiency.

[0055] Aiming at the demand for integrated control of parking and charging multi-tasks, we design a method for the collaborative management and control of parking guidance and charging planning, and build a highly reliable, low-latency, and multi-cooperative central control platform for integrated parking and charging. Based on distributed algorithm and protocol design, we study the multi-agent multi-task distributed architecture, including the core control center, parking space management system, charging management system, and agent management subsystem, to achieve comprehensive intelligent management of integrated parking and charging parking lots. Based on domain-driven design and containerization and orchestration technology, we study the high-cohesion and low-coupling microservice platform architecture, including authentication services, parking services, charging services, task scheduling services, etc., to provide an efficient, reliable, and secure microservice operating environment to support the construction and operation of complex business systems. Based on data mining and machine learning methods, we study real-time monitoring and feedback mechanisms to solve problems such as abnormal situation monitoring, resource scheduling optimization, service quality assurance, system operation status evaluation, and decision support of the central control platform. Aiming at problems such as parking lot performance optimization, function improvement, and business expansion, we study the continuous optimization and improvement strategy of the system based on automated operation and maintenance and deployment technology, and continuously improve the system's performance, stability, and user experience based on actual operation conditions and user feedback. In order to solve the problems of high concurrency of decision-making tasks, we study multi-task decision-making methods based on collaborative learning and knowledge transfer mechanisms, design multi-objective optimization algorithms, and build economical and efficient multi-agent distributed architecture and collaborative decision-making systems. In order to solve the problems of slow task scheduling response and poor resource distribution dynamics, we study dynamic switching strategies based on task criticality based on dynamic scheduling algorithms based on dynamic demand boundary theory to improve task execution efficiency, and design real-time quantitative evaluation methods for computing power and application indicator design methods.

[0056] In order to solve the problem of unsmooth multi-modal information flow of multi-agents in the field, a model development environment integrating statistical computing and machine learning algorithms, compatible with privacy computing protocols such as differential privacy mechanisms and homomorphic encryption is established based on a high-performance, distributed, loosely coupled cloud computing environment, to build a data interaction system serving multi-party computing. In view of the high communication delay in the field and the inconsistency of communication protocols among various agents, an information interaction architecture with high reliability, good real-time performance and strong security is studied based on point-to-point communication and centralized coordination methods. In view of the multi-modality of multi-agent data information in the parking lot, a data fusion method based on statistical methods and multi-cluster analysis is considered to construct a feature matching and feature complementary fusion strategy for multi-party and multi-agent data of vehicle-field-electricity, and a feature database that integrates multi-source data. In view of the problem of inconsistent multi-party data standards, the differences in the representation methods of vehicle-field-electricity multi-agent data items under different tasks are studied, and a data processing model based on unified semantics and unified precision is established.

[0057] In response to problems such as slow task scheduling response and poor resource distribution dynamics, a dynamic scheduling algorithm based on dynamic demand boundary theory is used to study dynamic task switching strategies, dynamically schedule and allocate tasks according to the system load and task priority, and allocate subtasks to different agents to execute, thereby improving task execution efficiency. In response to problems such as system computing real-time requirements and fluctuations in computing power demand, based on real-time quantitative evaluation methods of computing power and application indicator measurement technology, the system's performance indicators and resource utilization are regularly collected, and the system's load, performance bottlenecks, and resource utilization analysis methods are studied to provide data support for computing power resource management and optimization. Based on time series analysis methods (such as autoregressive moving average models, seasonal difference autoregressive sliding average models) and regression analysis methods, the system's load fluctuations are studied to achieve load balancing and resource optimization; based on the optimization method of genetic algorithms, the elastic computing power allocation architecture for computing power scheduling and allocation between agents is studied to ensure that each agent obtains sufficient computing resources to execute tasks.

[0058] Based on the same inventive concept, the present application embodiment also provides an automatic parking method, such as Figure 5 As shown, the following steps 101 to 110 are included.

[0059] Step 101: After the vehicle receives the automatic parking instruction, a plurality of UWB positioning messages currently broadcasted by the site where the vehicle is located are received through a first UWB tag module.

[0060] Step 102: Obtain point cloud data of the current site.

[0061] Step 103: Locate the vehicle position according to the multiple UWB positioning messages currently broadcast, and determine the current coordinates of the vehicle in the venue.

[0062] Step 104: construct a point cloud map of the current venue, and determine the Lanelet2 map of the current venue according to the point cloud map of the current venue.

[0063] Step 105: The Lanelet2 map of the current site, the point cloud map of the current site and the coordinates of the vehicle currently in the site are integrated to obtain a grid occupancy map of the current site.

[0064] Step 106: Identify an empty parking space from the grid occupancy map of the current site, and plan an automatic parking trajectory based on the grid occupancy map of the current site, with the current coordinates of the vehicle in the site as the starting point and the empty parking space as the end point, and then control the vehicle to start according to the planned automatic parking trajectory.

[0065] Step 107: During the automatic parking process of the vehicle, the position of the vehicle is located according to the multiple UWB positioning messages received in real time by the first UWB tag module to determine the real-time coordinates of the vehicle in the venue.

[0066] Step 108: Acquire the real-time point cloud data of the venue to construct a real-time point cloud map of the venue, and determine the real-time Lanelet2 map of the venue according to the real-time point cloud map of the venue.

[0067] Step 109: The real-time Lanelet2 map, the real-time point cloud map and the real-time coordinates of the vehicle in the venue are integrated to obtain a real-time grid occupancy map of the venue.

[0068] Step 110: Based on the real-time grid occupancy map, the automatic parking trajectory is adjusted in real time so that the vehicle automatically moves to an empty parking space.

[0069] By implementing the above steps 101 to 110 , the vehicle can accurately navigate to an empty parking space after receiving a parking instruction.

[0070] In another exemplary embodiment of the present application, if multiple vehicles receive automatic parking instructions at the same time, and the multiple vehicles dynamically and automatically wait to enter the parking space, the real-time adjustment of the automatic parking trajectory in the above step 110 can be replaced by the following steps 201 to 206.

[0071] Step 201: define objects whose moving speed is greater than a speed threshold as dynamic interference objects; the objects include pedestrians and vehicles.

[0072] Step 202: predict the trajectory of the dynamic interference object, and project the dynamic interference object in the form of a rectangular outline at each predicted trajectory point into a Frenet coordinate system established based on a reference path; the reference path is the automatic parking trajectory planned by the vehicle; the horizontal axis of the Frenet coordinate system represents the reference path, and the vertical axis of the Frenet coordinate system represents the distance.

[0073] Step 203: Determine the shortest distance between the rectangular contour points on the predicted trajectory and the horizontal axis.

[0074] Step 204: If the shortest distance is less than the automatic parking path safety distance, it is determined that the dynamic interference object intrudes into the automatic parking trajectory planned by the vehicle.

[0075] Step 205: Project the dynamic interferer at the trajectory point where the shortest distance is located into the ST graph, and use a dynamic programming algorithm to plan a speed curve that avoids the dynamic interferer in the ST graph; the horizontal axis of the ST graph represents time, and the vertical axis of the ST graph represents the reference path.

[0076] Step 206: Perform secondary smoothing on the speed curve to obtain an optimal speed curve that avoids dynamic interferences.

[0077] Compared with the prior art, the present invention has the following obvious advantages and beneficial effects.

[0078] (1) Improved positioning accuracy and safety: This application uses ultra-wideband technology to enable vehicles to achieve higher positioning accuracy in complex parking environments, such as multi-story or underground parking lots. UWB technology can provide centimeter-level positioning accuracy, which is significantly higher than traditional ultrasonic or camera systems. This high-precision positioning directly improves the safety and reliability of parking and reduces collisions and damage caused by positioning errors.

[0079] (2) Optimizing parking efficiency: Through the holographic ubiquitous perception information system, this application can process and optimize the flow of vehicles in the parking lot in real time. The system can not only identify vacant parking spaces, but also dynamically adjust parking strategies according to current traffic conditions and vehicle flow, thereby reducing the waiting and driving time of vehicles when looking for parking spaces.

[0080] (3) Enhanced intelligent collaborative operation of parking lots: The collaborative perception, decision-making and control methods proposed in this application enable a single vehicle to not only park autonomously, but also exchange information with other vehicles and infrastructure in the parking lot. This collaborative operation mode can effectively manage the parking needs of multiple vehicles during peak hours, significantly improving the overall operational efficiency and user satisfaction of the parking lot.

[0081] (4) Integration of contactless payment and value-added services: By integrating the contactless payment system, users can automatically complete the payment process without any physical operation, which greatly improves customer convenience. In addition, combined with the automatic charging function, this application provides a one-stop service for electric vehicle users, which not only enhances the attractiveness of parking lots, but also helps promote the popularization of electric vehicles.

[0082] (5) Solve the operation problem under multi-source random interference: In the environment of multi-source random interference, traditional parking technology is often limited by the perception ability of the equipment. This application can effectively analyze the signal interference in complex environments through advanced algorithms and data processing capabilities to ensure the accuracy and stability of parking operations. This robustness is particularly important in complex environments such as large urban shopping malls and underground parking lots in commercial buildings.

[0083] (6) Improve user experience and operational efficiency: This application comprehensively improves the user's parking experience through automated and intelligent technical means. The touchless payment and automatic charging functions greatly simplify the user's operation steps and improve the service level of the parking lot. At the same time, intelligent collaborative operation and efficient parking strategy optimization have greatly improved the operational efficiency of the parking lot, reduced resource waste, and optimized the space and energy utilization of the parking lot.

[0084] To sum up, this application not only solves many problems in the existing technology by integrating UWB positioning technology, automatic parking technology, contactless payment system and automatic charging technology, but also significantly improves the intelligence and automation level of parking lots, bringing significant technological progress and practical value.

[0085] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0086] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. An automatic parking system, characterized in that: The automatic parking system includes: a first UWB tag module, a laser radar, an industrial computer and an automatic driving controller; The first UWB tag module is arranged on the roof of the vehicle; The first UWB tag module and the laser radar are connected to the industrial computer respectively, and the industrial computer is connected to the automatic driving controller; After the vehicle receives the automatic parking instruction, the first UWB tag module is used to receive multiple UWB positioning messages currently broadcast by the site where the vehicle is located; the laser radar is used to scan the site environment and obtain the point cloud data of the current site; the industrial computer is used to locate the vehicle according to the multiple UWB positioning messages currently broadcast, and determine the current coordinates of the vehicle in the site; construct a point cloud map of the current site according to the point cloud data of the current site, and determine the Lanelet2 map of the current site according to the point cloud map of the current site, and fuse the Lanelet2 map of the current site, the point cloud map of the current site and the current coordinates of the vehicle in the site to obtain the grid occupancy map of the current site; the industrial computer is also used to identify vacant parking spaces from the grid occupancy map of the current site, and based on the grid occupancy map of the current site, take the current coordinates of the vehicle in the site as the starting point and the vacant parking space as the end point, plan the automatic parking trajectory, and then control the vehicle to start according to the planned automatic parking trajectory; During the automatic parking process of the vehicle, the laser radar is used to scan the site environment in real time to obtain real-time point cloud data of the site; the industrial computer is also used to locate the vehicle position according to multiple UWB positioning messages received in real time by the first UWB tag module, and determine the real-time coordinates of the vehicle in the site; a real-time point cloud map of the site is constructed according to the real-time point cloud data of the site obtained by the laser radar, and a real-time Lanelet2 map of the site is determined according to the real-time point cloud map of the site, and the real-time Lanelet2 map, the real-time point cloud map and the real-time coordinates of the vehicle in the site are integrated to obtain a real-time grid occupancy map of the site; the industrial computer is also used to plan or adjust the automatic parking trajectory in real time based on the real-time grid occupancy map, so that the vehicle automatically runs to an empty parking space.

2. The automatic parking system according to claim 1, characterized in that: The automatic parking system further includes: a second UWB tag module, a third UWB tag module, a fourth UWB tag module and a fifth UWB tag module; The second UWB tag module, the third UWB tag module, the fourth UWB tag module and the fifth UWB tag module are respectively arranged at the two headlights and the two taillights of the vehicle in a one-to-one correspondence; The second UWB tag module, the third UWB tag module, the fourth UWB tag module and the fifth UWB tag module are all connected to the automatic driving controller; When the vehicle operates in keyless entry and start mode, the first UWB tag module, the second UWB tag module, the third UWB tag module, the fourth UWB tag module and the fifth UWB tag module all receive the ranging message from the key end, and send the ranging message to the automatic driving controller; the automatic driving controller is used to calculate the position of the key end, and unlock the vehicle according to the unlocking instruction sent by the key end at the position, or start the vehicle according to the start instruction sent by the key end at the position; the key end is a mobile phone or a car key.

3. A smart parking system, characterized in that: The smart parking system comprises: a plurality of UWB base stations, a mobile charging vehicle and the automatic parking system according to any one of claims 1-2; A plurality of UWB base stations are arranged at different locations of the parking lot; the parking lot is provided with a parking area; The plurality of UWB base stations are used to broadcast a plurality of UWB positioning messages to the first UWB tag module in the automatic parking system; the automatic parking system is used to control the vehicle to automatically run to an empty parking space according to the received plurality of UWB positioning messages; When the free parking space is a free parking space in the parking and charging area, after the automatic parking system controls the vehicle to automatically move to the free parking space in the parking and charging area, the UWB base station is used to receive the parking coordinates of the vehicle located by the industrial computer through the first UWB tag module, and simultaneously obtain the current coordinates of the mobile charging vehicle; the UWB base station is also used to control the mobile charging vehicle to move to the location of the vehicle according to the current coordinates of the mobile charging vehicle and the parking coordinates of the vehicle; After the mobile charging vehicle moves to the location of the vehicle, the UWB base station is also used to transmit the charging port opening instruction to the industrial computer through the first UWB tag module; the industrial computer is used to open the charging port cover of the vehicle according to the charging port opening instruction, and to feedback the information that the charging port cover has been opened to the UWB base station through the first UWB tag module; the UWB base station is also used to control the mobile charging vehicle to perform the gun insertion operation and start charging the vehicle after receiving the information that the charging port cover has been opened.

4. The smart parking system according to claim 3 is characterized in that: The smart parking system also includes: a parking host; Multiple UWB base stations are connected to the parking lot host; The industrial computer in the automatic parking system is used to transmit the real-time coordinates of the vehicle in the venue to the parking lot host through the first UWB tag module and multiple UWB base stations; The parking lot host internally stores a digital twin parking lot virtual model; the parking lot host is used to locate the vehicle position in real time in the digital twin parking lot virtual model according to the real-time coordinates of the vehicle in the site, and display it in the digital twin parking lot virtual model in the form of a virtual vehicle model.

5. The smart parking system according to claim 3, characterized in that: The smart parking system also includes: a contactless payment system; The contactless payment system is used to receive payment information generated by the parking lot host through the first UWB tag module, and deduct parking fees from the account bound between the user and the parking lot according to the payment information.

6. An automatic parking method, characterized in that: The automatic parking method comprises: After the vehicle receives the automatic parking instruction, multiple UWB positioning messages currently broadcasted by the site where the vehicle is located are received through the first UWB tag module; Get the point cloud data of the current site; Locate the vehicle's position based on the multiple UWB positioning messages currently broadcast, and determine the vehicle's current coordinates within the venue; Construct a point cloud map of the current site, and determine the Lanelet2 map of the current site based on the point cloud map of the current site; The Lanelet2 map of the current site, the point cloud map of the current site and the coordinates of the vehicle in the site are integrated to obtain the grid occupancy map of the current site; Identify an empty parking space from a grid occupancy map of the current site, and plan an automatic parking trajectory based on the grid occupancy map of the current site, with the current coordinates of the vehicle in the site as a starting point and the empty parking space as an end point, and then control the vehicle to start according to the planned automatic parking trajectory; During the automatic parking process of the vehicle, the vehicle position is located according to the multiple UWB positioning messages received in real time by the first UWB tag module to determine the real-time coordinates of the vehicle in the venue; Acquire the real-time point cloud data of the site to construct a real-time point cloud map of the site, and determine the real-time Lanelet2 map of the site based on the real-time point cloud map of the site; The real-time Lanelet2 map, the real-time point cloud map and the real-time coordinates of the vehicle in the venue are integrated to obtain a real-time grid occupancy map of the venue; Based on the real-time grid occupancy map, the automatic parking trajectory is adjusted in real time so that the vehicle can automatically move to an empty parking space.

7. The automatic parking method according to claim 6, characterized in that: If multiple vehicles receive automatic parking instructions at the same time, the automatic parking trajectory will be adjusted in real time, including: An object whose moving speed is greater than a speed threshold is defined as a dynamic interference object; the object includes pedestrians and vehicles; Predicting the trajectory of the dynamic interference object, and projecting the dynamic interference object in the form of a rectangular outline at each predicted trajectory point into a Frenet coordinate system established based on a reference path; the reference path is an automatic parking trajectory planned by the vehicle; the horizontal axis of the Frenet coordinate system represents the reference path, and the vertical axis of the Frenet coordinate system represents the distance; Determine the shortest distance between the rectangular contour points on the predicted trajectory and the horizontal axis; If the shortest distance is less than the automatic parking path safety distance, it is determined that the dynamic interference object intrudes into the automatic parking trajectory planned by the vehicle; The dynamic interference object of the trajectory point where the shortest distance is located is projected into the ST graph, and a speed curve that avoids the dynamic interference object is planned in the ST graph using a dynamic programming algorithm; the horizontal axis of the ST graph represents time, and the vertical axis of the ST graph represents a reference path; The speed curve is subjected to secondary smoothing processing to obtain an optimal speed curve that avoids dynamic interferences.

Images