WO2023173678A1

WO2023173678A1 - Internet-of-vehicles-based parking space allocation and parking system for autonomous vehicles in park

Info

Publication number: WO2023173678A1
Application number: PCT/CN2022/113362
Authority: WO
Inventors: 刘明春; 李春; 聂石启; 谭福伦; 张智清; 邵立夫
Original assignee: 金龙联合汽车工业（苏州）有限公司
Priority date: 2022-03-14
Filing date: 2022-08-18
Publication date: 2023-09-21
Also published as: CN114677857A

Abstract

An Internet-of-Vehicles-based parking space allocation and parking system for autonomous vehicles in a park. The system comprises a parking space detection end, a parking space release end, a parking space application end and a communication network architecture, wherein the parking space detection end comprises a detection unit of a parking lot end and a roadside data unit, and is responsible for parking space state detection and releasing; the parking space release end is a data cloud platform and is responsible for the management of vehicles and parking spaces, and optimized recommendation of the parking spaces; the parking space application end is an autonomous vehicle and is responsible for submitting a parking space application; and the communication network architecture comprises a communication base station, and the communication base station realizes the communication and cooperation among the parking space detection end, the parking space release end and the parking space application end by means of a plurality of communication modes. Therefore, the information interaction of "vehicle-road-cloud-parking lot end" is fused, and unified scheduling and efficient management of vehicles can be better realized by means of parking space detection at the parking lot end, route planning at a cloud and autonomous driving at a vehicle end.

Description

Parking space allocation and parking system for park autonomous vehicles based on Internet of Vehicles

Technical field

The invention belongs to the field of automatic driving, and specifically relates to a parking space allocation and parking system for automatic driving vehicles in a park based on the Internet of Vehicles.

Background technique

An autonomous vehicle is an intelligent system that integrates environmental perception, decision-making and planning, control execution and other functions. It is also called driverless, driverless vehicle, etc. It is combined with the Internet of Vehicles to be intelligent and connected. Characteristics are the inevitable trend of the current development of the automobile industry and the main area of technological innovation. Park traffic scenarios include scenic sightseeing, park site ferrying, park cleaning, fixed route patrols and other operating scenarios. It has the characteristics of low speed, relatively fixed routes, fewer traffic participants, and controllable safety. It is considered to be the first to realize autonomous driving. One of the key scenes of industrialization. Autonomous driving vehicles in the park need to perform repetitive operations for a long time, and the operation routes are relatively fixed. Under the instructions of the vehicle operation status and the background dispatch system, they can "come and leave as soon as they are called", and need to drive frequently in the park. Between the parking lot and the work site. How to rationally allocate parking spaces to autonomous vehicles in real time and complete efficient parking is of great significance to improving the operating efficiency and management dispatch of autonomous vehicles in the park.

The Chinese invention patent "An automatic parking method based on driverless driving and Internet of Vehicles" (Patent No. ZL201910892986.0) invented an automatic parking method for driverless vehicles, through the combination of Internet of Vehicles and driverless technology. , the car owner can realize the automatic parking action of the vehicle by inputting parking parameter information and selecting the parking mode on the mobile terminal; this patent is mainly applicable to manned automatic driving vehicles, targeting specific parking lots and vehicles with automatic parking functions. Solve customers’ pain points of parking difficulties.

The Chinese invention patent application "Unmanned vehicle parking method, device and electronic equipment based on temporary events" (application number: CN 202110810680.3, application publication date: 2021.08.17) discloses an unmanned vehicle parking method based on temporary events, which can Realize the parking function of unmanned vehicles triggered based on temporary events, comprehensively detect the road environment, traffic rules, obstacles and other constraints through on-board sensors and controllers, dynamically adjust the parking position of unmanned vehicles, and improve the interaction between unmanned vehicles and external users. capabilities and environmental response capabilities; this patent application mainly solves the parking decision-making problem of unmanned vehicles in the case of temporary incident intervention, to determine whether to respond to temporary incidents, whether to park, how to park and where to park, etc.

The Chinese invention patent "Method for accurately selecting parking locations, intelligent control equipment and autonomous vehicles" (application number: CN 202110141603.3, application publication date: 2021.05.11) discloses a method for autonomous vehicles to accurately select parking locations. By constructing Autonomous vehicles associated with high-precision maps can directly identify the location of the map and accurately confirm the user's location; when the distance between the vehicle's current location and the parking location is less than the preset distance, obtain the user's confirmation information about the parking location. ; Provide users with a more accurate map of the parking location to ensure that users get on and off the car at the exact location; This patent application mainly solves the problems of rational decision-making and parking accuracy for autonomous vehicle parking.

The Chinese invention patent application "A parking space allocation method and device" (application number: CN202011329692.6, application publication date: 2021.02.19) discloses a parking space allocation method suitable for automatic valet parking. This method uses preset The "car-park" matching table obtains the prioritized set of nearby available parking lots and parking spaces and sends them to the user for user selection, which can provide appropriate automatic valet parking for vehicles with different autonomous driving functions. Parking lots and parking spaces; this patent application is applicable to manned autonomous vehicles and mainly depends on conditions such as parking lot type setting, vehicle function division, vehicle-park pre-matching and other conditions.

China’s invention patent application “An automatic driving parking system for new energy vehicles based on artificial intelligence” (application number: 202011368423.0, application publication number: 2021.02.19) discloses an automatic driving parking system for new energy vehicles based on artificial intelligence. The remaining power detection module detects the remaining power of the new energy vehicle after it arrives at the parking lot, and determines whether the car can reach its destination on the remaining power based on historical data. It selects charging and non-charging empty parking spaces based on the judgment results; after finding a suitable empty parking space, it passes The vacancy scanning module scans to confirm whether there are obstacles in the empty parking space and further decides whether to park. This patent application is mainly based on on-board sensors and controllers and is used to solve the problem of whether the vehicle is charging and the parking space is actively selected.

The invention patent application "A parking system and method based on automatic driving" (application number: CN202011473694.2, application publication date: 2021.04.06) discloses a parking system and method based on automatic driving, using three or more The field-end ultrasonic radar determines the positioning information of the vehicle, plans the driving path to the empty parking space based on the electronic map of the parking lot, and sends the driving path to the autonomous vehicle, so that the autonomous vehicle automatically drives to the empty parking space based on the driving path; This patent application is mainly aimed at indoor parking lots, using multiple ultrasonic radars to achieve vehicle positioning.

Regarding the above-mentioned patents, most of the existing parking systems for self-driving vehicles are used in parking scenarios, and the driving routes targeted are relatively changeable, and the path between the self-driving vehicle from the location where the parking application is issued to the parking space is not considered. Navigation; Most of the existing parking systems for self-driving vehicles are based on single-vehicle intelligence, which relies on the vehicle's own sensors to locate and identify parking spaces. It does not integrate the information interaction between "vehicle-road-cloud-parking lot" and cannot Carry out unified parking space scheduling and path planning well.

Based on the need for automatic driving vehicles in the park to frequently travel between the parking lot and the operation point, and the relatively fixed characteristics of the operation route, this invention integrates the information interaction between "vehicle-road-cloud-parking lot end" to complete automatic driving Automatic vehicle application, optimal parking space recommendation, and automatic navigation to parking spaces realize efficient management and dispatch of autonomous vehicles in the park.

Contents of the invention

The purpose of this invention is to propose a parking space allocation and parking system for autonomous vehicles in the park based on the Internet of Vehicles in view of the frequent driving needs of autonomous vehicles in the park between the parking lot and the operation point, and the relatively fixed operating routes, so as to improve the automation of the park. Driving vehicle management and dispatch efficiency.

The technical solution of the present invention is:

The parking space allocation and parking system for park autonomous vehicles based on the Internet of Vehicles includes a parking space detection end, a parking space release end, a parking space application end and a communication network architecture, including:

The parking space detection terminal is responsible for detecting and releasing parking space status;

The parking space release end is responsible for the management of vehicles and parking spaces and the optimization and recommendation of parking spaces;

The parking space application end is an autonomous vehicle and is responsible for applying for parking spaces;

The communication network architecture includes communication base stations. The communication base stations realize communication and collaboration between the parking space detection end, the parking space release end, and the parking space application end through multiple communication methods.

Preferably, the parking space detection end includes a field-side parking space detection unit and a roadside data unit;

The parking space detection unit at the yard detects the idle status and parking space number of the parking space through the parking space detection sensor, and preprocesses the parking space signal through the parking space detection signal processor;

The roadside data unit is responsible for receiving parking space application commands sent by autonomous vehicles and parking space information released by the broadcast cloud platform.

Preferably, the roadside data unit includes an edge calculator, a roadside data processing module and a roadside communication module;

The edge calculator converts the unstructured data of the roadside data unit into structured data, and the roadside data processing module performs preliminary processing of the data; the unstructured data includes the image data of the camera, and the structured data includes the parking space number and parking space location. and lane lines.

Preferably, the parking space publishing end is deployed in the cloud platform, including a parking space identification and positioning unit, a parking space recommendation unit, and a parking management unit;

The parking space identification and positioning unit uses artificial intelligence technology to identify parking spaces and obtain the number and location of the free parking space. After the free parking space is identified and located, the cloud platform stores the parking space information in the parking space database and waits for the parking space application command;

When the self-driving vehicle issues a parking application command, the parking space recommendation unit uses a sorting algorithm that approximates the ideal solution to obtain the optimal ranking of available parking spaces, and recommends the optimal parking spaces to the self-driving vehicle. The cloud platform will update the parking space database and Then publish the recommended parking spaces to other vehicles; based on the positioning coordinates of the parking space application vehicle uploaded to the cloud platform, the cloud platform will obtain the navigation route from the parking space application vehicle to the recommended parking space by accessing the pre-stored map, and ultimately guide the vehicle to drive automatically. Go to the parking space and complete the parking.

Preferably, the parking space application end includes a vehicle-mounted unit, a human-computer interaction interface, a camera/inertial integrated navigation and a microcontroller; the vehicle-mounted unit is responsible for sending and receiving signals, and the human-computer interaction interface is mainly responsible for the interaction between personnel and the parking system. The parking space and navigation path are displayed. The camera/inertial integrated navigation is responsible for collecting image data on both sides of the driving road and the vehicle's own positioning data. The microcontroller is responsible for processing, sending and receiving signals from each module.

Preferably, in the communication network architecture:

The vehicle-mounted unit uses the Uu port or PC5 port to communicate with the roadside unit and the communication base station. The data sent includes vehicle positioning data, vehicle movement data, and parking space application commands. The data received includes the number and location data of the free parking space, and the navigation route. , Recommended free parking space information;

The roadside unit communicates with the vehicle-mounted unit and the communication base station, sends the status information of the autonomous vehicle and the parking space application command to the cloud platform, and sends the recommended parking space information and navigation route information from the cloud platform to the vehicle-mounted unit;

The cloud platform communicates with the communication base station and receives and sends data through the communication base station, including receiving vehicle positioning data, vehicle status data, and parking space application commands uploaded by the roadside unit and initially processed by the edge computer, and uploading parking space application commands through the Uu communication interface. , and publish recommended parking space information and navigation route information.

Preferably, when the self-driving vehicle returns to the parking lot, the vehicle controller issues a command to apply for a parking space to the vehicle-mounted unit, and the vehicle-mounted unit uploads the parking space application command, current body positioning data, and vehicle status data to the cloud platform;

The cloud platform will query whether there is an available parking lot near the vehicle and whether there are free parking spaces in the parking lot. If there are no free parking spaces, the distance range of the search space will be increased until available parking lots and free parking spaces are found;

After completing the parking lot and free parking space search, the cloud platform automatically confirms the parking lot location, number and number of free parking spaces; the cloud platform sorts the free parking spaces, selects the optimal parking space, and recommends the parking space to the autonomous vehicle;

After the self-driving vehicle confirms the free parking space, the cloud platform will update the free parking space library, change the status of the parking space to occupied, and no longer recommend the parking space to other self-driving vehicles; the cloud platform will locate the free parking space based on the vehicle's own positioning data at this time. data, call the pre-stored high-precision map, plan the path for the self-driving vehicle to drive to the recommended parking space, and send the navigation path to the self-driving vehicle;

After receiving the navigation route published by the cloud platform, the vehicle-mounted unit confirms the navigation route, activates the automatic driving function, and displays the navigation route and vehicle status information on the human-computer interaction interface of the vehicle and the cloud platform to facilitate real-time management and control by relevant personnel. Vehicle driving status;

After the autonomous vehicle arrives at the recommended parking space and completes the parking operation, it uploads parking completion, vehicle number, and parking start time information to the cloud platform, and the cloud platform continuously monitors the status of this parking space.

Preferably, if the driving path recommended by the cloud platform is unobstructed, the on-board sensor of the autonomous vehicle will detect the cause of the obstruction in the driving path and send the information about the blocked road to the cloud platform through the vehicle-mounted unit. The cloud platform will determine the location of the vehicle according to its location. location, and re-recommend optimal parking spaces and navigation routes.

Preferably, the parking space recommendation unit uses a sorting algorithm that approximates the ideal solution to obtain the optimal sorting of free parking spaces, including:

S1. Construct initial matrix

Suppose there are m plans and n evaluation indicators. The plan set is D={d ₁ , d ₂ ,...d _m }, and the attribute variables to measure the quality of the plan are x ₁ ,..., x _n . At this time, the plan set D is The initial matrix constructed by n indicators for each plan d _i (i=1,...,m) is A=[ai1,...,ain]. As a point in the n-dimensional space, the matrix A can uniquely characterize the plan d _i quality;

S2, standardization/standardization

Use the matrix programming method to obtain the standardized decision matrix. Assume the decision matrix A=(a _ij ) _m×n for the multi-attribute decision-making problem, and calculate the standardized decision matrix B=(b _ij ) _m×n , where,

S3. Construct a weighted norm matrix C = (c _ij ) _m×n

Assume that the weight vector of each attribute/index given by the decision-maker is w=[w ₁ ,w ₂ ,…,w _n ] ^T , then

c _ij =w _j ·b _ij , where i=1,2,…,m; j=1,2,…,n;

S4. Determine the positive ideal solution and negative ideal solution

The ideal solution C ⁺ consists of the maximum value in each column in C:

C ⁺ =max(c _i1 ,c _i2 ,…,c _im )

The negative ideal solution C ^- consists of the minimum value in each column in C:

C ^- =min(c _i1 ,c _i2 ,...,c _im )

S5. Calculate the Euclidean distance between each target and the ideal value

That is: calculate the distance of each solution to the positive ideal solution and the negative ideal solution

The distance between alternative d _i and the ideal solution is:

The distance between alternative d _i and the negative ideal solution is:

S6. Calculate the queuing index value of each plan

That is: calculate the closeness of the evaluation object to the optimal solution

Arrange the advantages and disadvantages of each plan in descending order according to the proximity f _i , select the optimal parking space from the sorted list, and publish it to the self-driving vehicle that applies for the parking space.

Compared with the existing technology, the present invention has the following advantages:

(1) The present invention integrates the information interaction of "vehicle-road-cloud-parking lot", and can better realize unified dispatching and efficient management of vehicles through parking space detection on the field, route planning on the cloud, and automatic driving on the vehicle. ;

(2) This invention takes into account the need for automatic driving vehicles in the park to frequently travel between the parking lot and the operation point, as well as the relatively fixed characteristics of the operation route. The route planning is carried out in the cloud and sent to the vehicle as a navigation route, which can effectively improve the vehicle's safety. Driving efficiency and safety;

(3) The present invention adopts a sorting algorithm that approximates the ideal solution, takes into account the parking space information, vehicle status information, driving distance and other information of many vacant parking spaces to sort the parking spaces, calculates the optimal parking space recommendation car terminal, and improves the parking of autonomous vehicles in the park. efficiency.

Description of the drawings

Figure 1 is the architecture diagram of the parking space allocation and parking system for autonomous vehicles in the park based on the Internet of Vehicles;

Figure 2 is a scene architecture diagram of the parking space allocation and parking system for autonomous vehicles in the park based on the Internet of Vehicles;

Figure 3 is a flow chart of parking space allocation and parking system implementation for park autonomous vehicles based on the Internet of Vehicles;

Figure 4 is a flow chart of optimal parking space recommendation.

Detailed ways

The technical solutions in the embodiments of the present invention will be described clearly and in detail below with reference to the accompanying drawings in the embodiments of the present invention. The described embodiments are only some of the embodiments of the invention.

As shown in Figure 1, the parking space allocation and parking system for park autonomous vehicles based on the Internet of Vehicles of the present invention includes a parking space detection end, a parking space release end, a parking space application end, and a communication network architecture.

The parking space detection terminal is mainly responsible for parking space status detection and release, including the parking space detection unit at the field end and the roadside data unit. The parking space detection unit at the yard detects the idle status and parking space number of the parking space through parking space detection sensors (such as cameras, ultrasonic radars, infrared detectors, etc.), and preprocesses the parking space signals through the parking space detection signal processor. The roadside data unit is responsible for receiving parking space application commands sent by autonomous vehicles and parking space information released by the broadcast cloud platform. It consists of an edge calculator, a roadside data processing module and a roadside communication module. Due to the large amount of data transmitted in real time , using the existing communication network to upload all data to the cloud platform will cause greater communication pressure, so the roadside data processing module performs preliminary data processing and deploys edge calculators to convert unstructured data into structures Unstructured data includes camera image data, etc., and structured data includes parking space numbers, parking space locations, lane lines, etc.

The parking space publishing end is deployed in the cloud control platform and is mainly responsible for the management of vehicles and parking spaces and the optimization and recommendation of parking spaces. It consists of three parts: a parking space identification/positioning unit, a parking space recommendation unit, and a parking management unit. The parking space identification/positioning unit uses artificial intelligence technology to identify parking spaces and obtain the number and location of the free parking space. After the free parking space is identified and located, the cloud platform stores the parking space information in the service and waits for the parking space application command; when it automatically After the driving vehicle issues a parking application command, the parking space recommendation unit uses a sorting algorithm that approximates the ideal solution to obtain the optimal ranking of available parking spaces, and recommends the optimal parking spaces to autonomous vehicles. The cloud platform will update the parking space database and will no longer publish it. Recommended parking spaces to other vehicles; according to the positioning coordinates of the parking space application vehicle uploaded to the cloud platform, the cloud platform will obtain the navigation route from the parking space application vehicle to the recommended parking space by accessing the pre-stored high-precision map, and ultimately guide the vehicle to drive automatically Go to the parking space and complete the parking.

The parking space application end refers to the autonomous vehicle, including the vehicle-mounted unit, human-computer interaction interface, camera/inertial integrated navigation and microcontroller (MCU), etc. The vehicle-mounted unit is mainly responsible for sending and receiving signals, and the human-computer interaction interface is mainly responsible for the personnel and The interaction of the parking system displays parking spaces and navigation paths. The camera/inertial integrated navigation is mainly responsible for collecting image data on both sides of the driving road and the vehicle's own positioning data. The microcontroller (MCU) is mainly responsible for processing, sending and receiving signals from each module.

Figure 2 shows one of the scenarios of parking space allocation and parking system implementation for park autonomous vehicles based on the Internet of Vehicles. After receiving the parking instruction, the autonomous vehicle automatically drives the vehicle to the parking space recommended by the cloud platform to complete the parking process. . As shown in the figure, there are side parking spaces on both sides of the road. There are other types of parking spaces in the park scene. As long as they can be effectively identified and located by the parking space detection device, the implementation method is consistent with this embodiment.

In the embodiment shown in Figure 2, the vehicle-mounted unit uses different communication methods (Uu port or PC5 port) to communicate with the roadside unit and communication base station. The data sent includes self-vehicle positioning data, vehicle movement data, parking space application Commands, etc. The data received includes the number and location data of the free parking space, navigation route, recommended free parking space information, etc. The roadside unit communicates with the vehicle-mounted unit and the communication base station, sends the status information of the autonomous vehicle and the parking space application command to the cloud platform, and sends the recommended parking space information and navigation route information from the cloud platform to the vehicle-mounted unit. The cloud platform mainly communicates with the communication base station, receiving and sending data through the communication base station, including receiving vehicle positioning data, vehicle status data, parking space application commands, etc. uploaded by the roadside unit and initially processed by the edge computer, and uploading parking spaces through the Uu communication interface. Apply for an order and publish recommended parking space information and navigation route information.

Figure 3 shows the implementation process of parking space allocation and parking system for park autonomous vehicles based on the Internet of Vehicles. First, the system completes initialization. When the self-driving vehicle encounters working conditions that require it to return to the parking lot, the vehicle controller (MCU) issues a parking "application for parking space" command to the on-board unit (OBU) (which prompts the self-driving vehicle to issue the "application for parking space" command). There are many factors, such as the completion of the automatic driving operation in the park, insufficient vehicle power, background management and scheduling, etc.), the OBU communicates with the roadside unit and communication base station, and uploads the parking space application command, current body positioning data and vehicle status data to the cloud platform .

The parking management server in the cloud platform will query whether there are available parking lots near the vehicle and whether there are free parking spaces in the parking lot. If there are no free parking spaces, the distance range of the search space will be increased until available parking lots and free parking spaces are found. ;

Further, after completing the parking lot and free parking space search, the cloud platform system automatically confirms the parking lot location, number of free parking spaces, location and number;

Furthermore, the optimal parking space recommendation server in the cloud platform will use a sorting algorithm that approximates the ideal solution to sort the available parking spaces. This algorithm considers parking space information, vehicle status, driving distance and other factors to select the optimal parking space and place it The parking space is recommended to self-driving vehicles, and basic information about the parking space is attached, such as parking space type, distance, location, etc.;

Further, after the self-driving vehicle confirms the free parking space, the cloud platform will update the free parking space library, change the status of the parking space to occupied, and no longer recommend the parking space to other self-driving vehicles;

Further, the cloud platform calls the pre-stored high-precision map based on the vehicle's own positioning data and the free parking space positioning data at this time, plans the path for the self-driving vehicle to travel to the recommended parking space, and sends the navigation path to the self-driving vehicle;

Further, after receiving the navigation route published by the cloud platform, the OBU confirms the navigation route, activates the automatic driving function (including but not limited to tracking driving, parking when encountering obstacles, obstacle avoidance driving, etc.), and transmits the navigation route and vehicle status information Displayed on the human-computer interaction interface of the vehicle and cloud platform to facilitate relevant personnel to control the vehicle driving status in real time;

Furthermore, after the autonomous vehicle arrives at the recommended parking space and completes the parking operation, it uploads parking completion, vehicle number, parking start time and other information to the cloud platform, and the cloud platform continuously monitors the status of this parking space.

In particular, if the recommended driving path is not smooth (such as traffic lane blockage, traffic control, road construction, etc.), the on-board sensors (such as cameras, lidar, etc.) of the autonomous vehicle will detect the reasons for the obstruction in the driving path. The "road is blocked" information is sent to the cloud platform through OBU, and the cloud platform will use the TOPSIS algorithm to re-recommend the optimal parking space and navigation route based on the vehicle's location.

Figure 4 shows the flow chart of the optimal parking space recommendation. It mainly uses a sorting algorithm that approximates the ideal solution and takes into account the parking space information, vehicle status information, driving distance and other information of many vacant parking spaces to sort the parking spaces. The implementation steps of the sorting algorithm that approximates the ideal solution include:

S1. Construct initial matrix

Suppose there are m plans and n evaluation indicators. The plan set is D={d ₁ , d ₂ ,...d _m }, and the attribute variables to measure the quality of the plan are x ₁ ,..., x _n . At this time, the plan set D is The initial matrix constructed by the n indicators (attribute values) of each scheme di ( _i =1,...,m) is A=[ai1,...,ain]. As a point in the n-dimensional space, the matrix A can be unique Characterizes the quality of solution _di .

S2, standardization/standardization

Use matrix programming method to obtain the standardized decision matrix. Assume the decision matrix A=(a _ij ) _m×n for multi-attribute decision-making problem, and calculate the normalized decision matrix B=(b _ij ) _m×n , where

S3. Construct a weighted norm matrix C = (c _ij ) _m×n

c _ij =w _j ·b _ij , where i=1,2,…,m; j=1,2,…,n.

S4. Determine the positive ideal solution and negative ideal solution

The ideal solution C ⁺ consists of the maximum value in each column in C:

C ⁺ =max(c _i1 ,c _i2 ,…,c _im )

C ^- =min(c _i1 ,c _i2 ,...,c _im )

S5. Calculate the Euclidean distance between each target (scheme) and the ideal value

The distance between alternative d _i and the ideal solution is:

The distance between alternative d _i and the negative ideal solution is:

S6. Calculate the queuing index value of each plan (i.e. comprehensive evaluation index)

The above embodiments are only for illustrating the technical concepts and characteristics of the present invention. Their purpose is to enable those familiar with the technology to understand the content of the present invention and implement it accordingly. They cannot limit the scope of protection of the present invention. All modifications made based on the spirit of the main technical solution of the present invention should be included in the protection scope of the present invention.

Claims

The parking space allocation and parking system for park autonomous vehicles based on the Internet of Vehicles is characterized by including a parking space detection end, a parking space release end, a parking space application end and a communication network architecture, including:

The parking space detection terminal is responsible for detecting and releasing parking space status;

The parking space release end is responsible for the management of vehicles and parking spaces and the optimization and recommendation of parking spaces;

The parking space application end is an autonomous vehicle and is responsible for applying for parking spaces;

The communication network architecture includes communication base stations. The communication base stations realize communication and collaboration between the parking space detection end, the parking space release end, and the parking space application end through multiple communication methods.
The parking space allocation and parking system for park autonomous vehicles based on the Internet of Vehicles according to claim 1, characterized in that the parking space detection end includes a field-end parking space detection unit and a roadside data unit;

The parking space detection unit at the yard detects the idle status and parking space number of the parking space through the parking space detection sensor, and preprocesses the parking space signal through the parking space detection signal processor;

The roadside data unit is responsible for receiving parking space application commands sent by autonomous vehicles and parking space information released by the broadcast cloud platform.
The parking space allocation and parking system for park autonomous vehicles based on the Internet of Vehicles according to claim 2, wherein the roadside data unit includes an edge calculator, a roadside data processing module and a roadside communication module;

The edge calculator converts the unstructured data of the roadside data unit into structured data, and the roadside data processing module performs preliminary processing of the data; the unstructured data includes the image data of the camera, and the structured data includes the parking space number and parking space location. and lane lines.
The parking space allocation and parking system for park autonomous vehicles based on the Internet of Vehicles according to claim 3, characterized in that the parking space issuing end is deployed in a cloud platform and includes a parking space identification and positioning unit, a parking space recommendation unit, and a parking management unit. ;

The parking space identification and positioning unit uses artificial intelligence technology to identify parking spaces and obtain the number and location of the free parking space. After the free parking space is identified and located, the cloud platform stores the parking space information in the parking space database and waits for the parking space application command;

When the self-driving vehicle issues a parking application command, the parking space recommendation unit uses a sorting algorithm that approximates the ideal solution to obtain the optimal ranking of available parking spaces, and recommends the optimal parking spaces to the self-driving vehicle. The cloud platform will update the parking space database and Then publish the recommended parking spaces to other vehicles; based on the positioning coordinates of the parking space application vehicle uploaded to the cloud platform, the cloud platform will obtain the navigation route from the parking space application vehicle to the recommended parking space by accessing the pre-stored map, and ultimately guide the vehicle to drive automatically. Go to the parking space and complete the parking.
The parking space allocation and parking system for park autonomous vehicles based on the Internet of Vehicles according to claim 4, characterized in that the parking space application terminal includes a vehicle-mounted unit, a human-computer interaction interface, a camera/inertial combined navigation and a microcontroller ; The vehicle-mounted unit is responsible for sending and receiving signals. The human-computer interaction interface is mainly responsible for the interaction between personnel and the parking system, displaying parking spaces and navigation paths. The camera/inertial combined navigation is responsible for collecting image data on both sides of the driving road and the vehicle's own positioning data. The controller is responsible for processing, sending and receiving signals from each module.
The parking space allocation and parking system for park autonomous vehicles based on the Internet of Vehicles according to claim 5, characterized in that in the communication network architecture:

The vehicle-mounted unit uses the Uu port or PC5 port to communicate with the roadside unit and the communication base station. The data sent includes vehicle positioning data, vehicle movement data, and parking space application commands. The data received includes the number and location data of the free parking space, and the navigation route. , Recommended free parking space information;

The roadside unit communicates with the vehicle-mounted unit and the communication base station, sends the status information of the autonomous vehicle and the parking space application command to the cloud platform, and sends the recommended parking space information and navigation route information from the cloud platform to the vehicle-mounted unit;

The cloud platform communicates with the communication base station and receives and sends data through the communication base station, including receiving vehicle positioning data, vehicle status data, and parking space application commands uploaded by the roadside unit and initially processed by the edge computer, and uploading parking space application commands through the Uu communication interface. , and publish recommended parking space information and navigation route information.
The parking space allocation and parking system for park autonomous vehicles based on Internet of Vehicles according to claim 6, characterized by:

When the self-driving vehicle returns to the parking lot, the vehicle controller issues a command to apply for a parking space to the on-board unit, and the on-board unit uploads the parking space application command, current body positioning data, and vehicle status data to the cloud platform;

The cloud platform will query whether there is an available parking lot near the vehicle and whether there are free parking spaces in the parking lot. If there are no free parking spaces, the distance range of the search space will be increased until available parking lots and free parking spaces are found;

After completing the parking lot and free parking space search, the cloud platform automatically confirms the parking lot location, number and number of free parking spaces; the cloud platform sorts the free parking spaces, selects the optimal parking space, and recommends the parking space to the autonomous vehicle;

After the self-driving vehicle confirms the free parking space, the cloud platform will update the free parking space library, change the status of the parking space to occupied, and no longer recommend the parking space to other self-driving vehicles; the cloud platform will locate the free parking space based on the vehicle's own positioning data at this time. data, call the pre-stored high-precision map, plan the path for the self-driving vehicle to drive to the recommended parking space, and send the navigation path to the self-driving vehicle;

After receiving the navigation route published by the cloud platform, the vehicle-mounted unit confirms the navigation route, activates the automatic driving function, and displays the navigation route and vehicle status information on the human-computer interaction interface of the vehicle and the cloud platform to facilitate real-time management and control by relevant personnel. Vehicle driving status;

After the autonomous vehicle arrives at the recommended parking space and completes the parking operation, it uploads parking completion, vehicle number, and parking start time information to the cloud platform, and the cloud platform continuously monitors the status of this parking space.
The parking space allocation and parking system for park autonomous vehicles based on Internet of Vehicles according to claim 7, characterized by:

If the driving path recommended by the cloud platform is not smooth, the on-board sensor of the autonomous vehicle will detect the cause of the blocking and send the information about the blocked road to the cloud platform through the vehicle-mounted unit. The cloud platform will re-establish the route based on the location of the vehicle. Recommend optimal parking spaces and navigation routes.
The parking space allocation and parking system for park autonomous vehicles based on the Internet of Vehicles according to claim 4, characterized in that the parking space recommendation unit uses a sorting algorithm that approximates an ideal solution to obtain the optimal sorting of free parking spaces, including:

S1. Construct initial matrix

Suppose there are m plans and n evaluation indicators. The plan set is D={d 1 , d 2 ,...d m }, and the attribute variables to measure the quality of the plan are x 1 ,..., x n . At this time, the plan set D is The initial matrix constructed by n indicators for each plan d i (i=1,...,m) is A=[ai1,...,ain]. As a point in the n-dimensional space, the matrix A can uniquely characterize the plan d i quality;

S2, standardization/standardization

Use the matrix programming method to obtain the standardized decision matrix. Assume the decision matrix A=(a ij ) m×n for the multi-attribute decision-making problem, and calculate the standardized decision matrix B=(b ij ) m×n , where,

S3. Construct a weighted norm matrix C = (c ij ) m×n

Assume that the weight vector of each attribute/index given by the decision-maker is w=[w 1 ,w 2 ,…,w n ] T , then

c ij =w j ·b ij , where i=1,2,…,m; j=1,2,…,n;

S4. Determine the positive ideal solution and negative ideal solution

The ideal solution C + consists of the maximum value in each column in C:

C + =max(c i1 ,c i2 ,…,c im )

The negative ideal solution C - consists of the minimum value in each column in C:

C - =min(c i1 ,c i2 ,...,c im )

S5. Calculate the Euclidean distance between each target and the ideal value

That is: calculate the distance of each solution to the positive ideal solution and the negative ideal solution

The distance between alternative d i and the ideal solution is:

The distance between alternative d i and the negative ideal solution is:

S6. Calculate the queuing index value of each plan

That is: calculate the closeness of the evaluation object to the optimal solution

Arrange the advantages and disadvantages of each plan in descending order according to the proximity f i , select the optimal parking space from the sorted list, and publish it to the self-driving vehicle that applies for the parking space.