A roadside perception method and system for vehicle-road collaboration

By adopting dual monitoring route and coordinate conversion models in vehicle-road collaboration, the blind spot problem of line of sight in traditional road-side perception technology is solved, real-time monitoring and data collection of target vehicles are achieved, and the comprehensiveness of road-side perception is ensured. reliability.

CN115810283BActive Publication Date: 2024-11-22CHENGDU INTELLIGENT CONNECTED VEHICLE TECH DEV CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202211439612.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-11-17
Publication Date
2024-11-22
Estimated Expiration
2042-11-17

AI Technical Summary

Technical Problem

In vehicle-road collaboration, traditional roadside perception technology cannot monitor all vehicle information on the road in real time due to visual occlusion, resulting in incomplete collection of road section information.

Method used

Using the dual monitoring route method, by establishing a connection between the perceived section plane model and the cloud platform, using GPS positioning and image data to track the target vehicle in real time, establishing a coordinate conversion model, converting image coordinate information into map coordinates, and through the communication module Transmission to the moving vehicle ensures that the target vehicle can be continuously monitored even if its location is blocked.

Benefits of technology

It realizes tracking of the target vehicle through GPS monitoring routes when image monitoring data is temporarily difficult to capture, solves the problem of blind spots of the roadside perception of the line of sight, and ensures real-time monitoring of the target vehicle and comprehensive data collection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115810283B_ABST
    Figure CN115810283B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of traffic technology, and more specifically, to a roadside sensing method in vehicle-road collaboration, including a roadside sensing unit and a vehicle positioning system. The roadside sensing unit collects vehicle image information, including position information, and the vehicle positioning system collects vehicle real-time positioning information, and synchronizes information in real time through dual monitoring routes to achieve vehicle road section driving beyond the field of view, avoiding the realization of blind spots caused by large-volume vehicle occlusion.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of traffic technology, and in particular to a roadside perception method and system thereof in vehicle-road collaboration. Background Art

[0002] At present, intelligent transportation and autonomous driving technologies have begun to flourish. In autonomous driving, it is difficult and unreliable to complete autonomous driving by relying solely on "smart cars" to collect road information on traffic roads. The emphasis on "smart roads" is an important direction for the advancement of autonomous driving technology. The vehicle-road collaborative approach can make road information collection more comprehensive, safer and more reliable.

[0003] Roadside perception technology is an important component of vehicle-road collaboration. In roadside perception technology, sensors are placed on the road to collect real-time information about vehicles and pedestrians on the road. The vehicles and pedestrians on the road are of different sizes, and the monitoring direction of the sensor is generally directional monitoring. There will be visual occlusion in three-dimensional space. Large vehicles will block small vehicles, and there will be blind spots in the system's line of sight. It is impossible to monitor all vehicle information on the monitored section in real time, and the section information collection is incomplete. Summary of the invention

[0004] The purpose of the present invention is to provide a roadside perception method and system in vehicle-road collaboration, which can solve the problem of blind spots in roadside perception image data by means of dual monitoring routes, and can continue to track the target vehicle through the GPS monitoring route after the image monitoring data is temporarily difficult to capture, thereby solving the problem of roadside perception visual blind spots.

[0005] The embodiments of the present invention are implemented by the following technical solutions:

[0006] A roadside perception method in vehicle-road collaboration includes:

[0007] Establish a plane model of the perceived road section;

[0008] The cloud platform is connected to the target vehicle positioning system to accurately locate the current GPS geographic location of the target vehicle and project it into the plane model of the road section to form a second map coordinate;

[0009] The roadside sensing unit obtains the speed and image information of the target vehicle on the current road section, establishes an image coordinate system, transmits the image coordinate and speed information of the target vehicle to the cloud platform, tracks the target vehicle on the current monitored road section and continuously obtains the image coordinate information;

[0010] Establish a coordinate conversion model, convert the acquired image coordinate information of the target vehicle into first map coordinates and overlap them with the map coordinate system in the road section plane model, transmit the real-time first map coordinate information of the target vehicle to the moving vehicle through the communication module, and construct the first driving route information of the vehicle according to the monitored first map coordinates;

[0011] Constructing second driving route information of the target vehicle according to the second map coordinate information obtained by the target vehicle positioning system, and updating the second driving route information of the vehicle in the plane model in real time according to the vehicle positioning system;

[0012] The cloud platform continuously synchronizes and compares the first driving route with the second driving route. When the roadside sensing unit monitors that the target vehicle is blocked and disappears from the field of vision, if the first driving route and the second driving route monitored by the target vehicle in the same period are consistent, the latest GPS second map coordinates obtained by the target vehicle positioning system are used as the real-time location information of the target vehicle.

[0013] The cloud platform transmits the real-time second map coordinate information to the moving vehicle through the communication module.

[0014] Furthermore, the plane model is a plane model of a road section from a bird's-eye view, and the plane model is proportional to the size of the road section within the monitoring range of the perception unit.

[0015] Furthermore, the vehicle positioning system transmits the received vehicle positioning GPS geographic location information to the cloud platform through the communication module in the vehicle.

[0016] Furthermore, reference points are selected on the sensing road section of the roadside unit to establish the image coordinate system.

[0017] Furthermore, the coordinate conversion model includes establishing height data, inclination angle data, image coordinate data of at least four reference points of the perception section, image coordinates of the origin of the target vehicle, map coordinates of the origin of the target vehicle, etc., which need to be collected by the roadside perception unit, and establishing a mathematical coordinate conversion model through various data; and then converting the image coordinate information of the captured target vehicle into map coordinates corresponding to the first image coordinates of the target vehicle according to the coordinate conversion model.

[0018] Furthermore, the first driving route information includes first map coordinate data of the target vehicle and time data corresponding to the first map coordinate data.

[0019] Furthermore, the second map coordinates are projected into a map coordinate system within the road section plane model, and the second driving route information includes second coordinate data and time data corresponding to the second map coordinate data.

[0020] The present invention also provides a roadside perception system in vehicle-road collaboration that applies the above method, including a roadside perception unit located above one side of a monitored section for monitoring the direction of traffic flow, wherein the monitored section includes at least three groups of connected sections, and the roadside perception unit includes three groups of perception units; it also includes a computing cloud platform that is communicatively connected to several of the roadside perception units, and it also includes a vehicle-mounted unit that is communicatively connected to the computing cloud platform.

[0021] Furthermore, the roadside perception unit includes a camera and a radar.

[0022] The technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects:

[0023] The present invention has a reasonable design and a simple structure. A perception road section model is established and simultaneously projected into two monitoring routes. The two monitoring routes are image data monitoring and collection of the road section by the perception unit and GPS positioning data information of the vehicle positioning system. When it is difficult for the roadside perception unit to capture image information and location information, it can be switched to GPS positioning information for monitoring in a timely manner. This enables close real-time monitoring of target vehicles on the perception road, as the perception road has incomplete monitoring data of the perception road section due to the existence of a monitoring blind spot. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic diagram of a sensing unit monitoring a road surface provided by an embodiment of the present invention;

[0025] Figure 2 A flow chart of a roadside sensing method provided by an embodiment of the present invention;

[0026] Figure 3 A schematic diagram of a road section monitoring unit provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0028] The present embodiment provides a roadside perception method in vehicle-road collaboration, including: establishing a plane model of a perceived road section; the plane model is a plane model of the road section from a bird's-eye view, and the plane model is proportional to the size of the road section within the monitoring range of the perception unit. According to the road type of the perceived road section, a plane model with the same proportion is established in the cloud platform to associate the perception data with the road section in a precise positional relationship and to construct a route for the target vehicle.

[0029] The cloud platform is communicatively connected with the target vehicle positioning system to accurately locate the current GPS geographic location of the target vehicle and project it into the plane model of the road section to form a second map coordinate; the vehicle positioning system includes a satellite positioning antenna, which receives vehicle positioning information transmitted by the satellite and transmits the received vehicle positioning information to the cloud platform through a communication module in the vehicle.

[0030] The roadside perception unit obtains the speed and image information of the target vehicle on the current road section, and establishes an image coordinate system. The middle of the lowest position of the captured target vehicle is used as the coordinate point, and the image coordinates and speed information of the target vehicle are transmitted to the cloud platform. The target vehicle on the current monitored section is tracked and the image coordinate information is continuously obtained; a reference point is selected on the perception section of the roadside unit to establish the image coordinate system.

[0031] Establish a coordinate conversion model, convert the acquired image coordinate information of the target vehicle into the first map coordinate and overlap it with the map coordinate system in the road section plane model, transmit the real-time first map coordinate information of the target vehicle to the driving vehicle through the communication module, and construct the first driving route information of the vehicle according to the monitored first map coordinate; the coordinate conversion model includes the establishment of the height data, inclination angle data, image coordinate data of at least four reference points of the sensing section, the image coordinates of the origin of the target vehicle, the map coordinates of the origin of the target vehicle, etc., and establish a mathematical coordinate conversion model through each data, the coordinate conversion model includes the image coordinates of the four reference points selected according to the known one-to-one correspondence and affine conversion, thereby or the conversion parameters directly between the image coordinates and the map coordinates; then, according to the coordinate conversion model, the image coordinate information of the captured target vehicle is converted into the map coordinates corresponding to the first image coordinates of the target vehicle through the conversion parameters. The first driving route information includes the first map coordinate data of the target vehicle and the time data corresponding to the first map coordinate data.

[0032] The second driving route information of the target vehicle is constructed according to the second map coordinate information obtained by the target vehicle positioning system, and the second driving route information of the vehicle in the plane model is updated in real time according to the vehicle positioning system; the second map coordinate is projected into the map coordinate system in the road section plane model, and the second driving route information includes second coordinate data and time data corresponding to the second map coordinate data.

[0033] The cloud platform continuously synchronizes and compares the first driving route with the second driving route, and continuously synchronizes the target vehicle coordinate information of the dual monitoring routes. When the roadside sensing unit monitors that the target vehicle is blocked and disappears from the field of vision, if the first driving route and the second driving route monitored by the target vehicle in the same period are consistent, that is, whether the map coordinates of the first driving route and the second driving route at the same time are consistent is obtained. If they are consistent, the latest GPS second map coordinates obtained by the target vehicle positioning system are used as the real-time location information of the target vehicle. The dual monitoring route enables the route and direction of the vehicle to be predicted even if it disappears from the monitoring line of sight, solving the problem of monitoring blind spots.

[0034] This embodiment also involves a roadside perception system in vehicle-road collaboration. The system uses the above-mentioned method for monitoring, including a roadside perception unit located above one side of the monitored road section to monitor the direction of traffic flow. The roadside perception unit is mainly composed of various sensors for obtaining data information such as images, speeds, and driving status of vehicles on the road section.

[0035] The monitoring section includes at least three groups of connected sections, and the roadside sensing unit includes three groups of monitoring units. Several monitoring sensors of a section form a monitoring unit. Since the monitoring range and distance of a group of monitoring units are limited, when performing roadside sensing, in addition to monitoring the target vehicle status of the current section, the system will also monitor the two sections before and after the current monitoring section through segmented monitoring, that is, the two sections before and after the current section, and transmit the sensing information of the three sensing sections to the driving vehicle as an information module. When the driving vehicle enters the next sensing section, the sensing information of the current section and the two sections before and after it is also transmitted to the driving vehicle as an information module.

[0036] The system also includes a computing cloud platform that is communicatively connected to the plurality of roadside sensing units, and also includes an on-board unit that is communicatively connected to the computing cloud platform. When a vehicle enters a second monitoring section, and assuming that there are adjacent first and third monitoring sections before and after the monitoring section, the cloud platform transmits all vehicle information in the second monitoring section to the on-board unit, and transmits the main vehicle information in the third monitoring section in the vehicle's forward direction to the on-board unit, such as vehicle abnormal driving data, vehicle collision information, and road section abnormal data, etc.; transmits vehicle abnormal driving data in the first monitoring section behind the vehicle's forward direction to the on-board unit, such as vehicle abnormal speeding data, etc.

[0037] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A roadside perception method in vehicle-road collaboration, characterized by: include Establish a plane model of the perceived road section; The cloud platform is connected to the target vehicle positioning system to accurately locate the current GPS geographic location of the target vehicle and project it into the plane model of the road section to form a second map coordinate; The roadside sensing unit obtains the speed and image information of the target vehicle on the current road section, establishes an image coordinate system, transmits the image coordinate and speed information of the target vehicle to the cloud platform, tracks the target vehicle on the current monitored road section and continuously obtains the image coordinate information; Establish a coordinate conversion model, convert the acquired image coordinate information of the target vehicle into first map coordinates and overlap them with the map coordinate system in the road section plane model, transmit the real-time first map coordinate information of the target vehicle to the moving vehicle through the communication module, and construct the first driving route information of the vehicle according to the monitored first map coordinates; Constructing second driving route information of the target vehicle according to the second map coordinate information obtained by the target vehicle positioning system, and updating the second driving route information of the vehicle in the plane model in real time according to the vehicle positioning system; The cloud platform continuously synchronizes and compares the first driving route with the second driving route. When the roadside sensing unit monitors that the target vehicle is blocked and disappears from the field of vision, if the first driving route and the second driving route monitored by the target vehicle in the same period are consistent, the latest GPS second map coordinates obtained by the target vehicle positioning system are used as the real-time location information of the target vehicle. The cloud platform transmits the real-time second map coordinate information to the moving vehicle through the communication module.

2. The roadside perception method in vehicle-road collaboration according to claim 1 is characterized in that: The plane model is a plane model of a road section from a bird's-eye view, and the plane model is proportional to the size of the road section within the monitoring range of the perception unit.

3. The roadside perception method in vehicle-road collaboration according to claim 1 is characterized in that: The vehicle positioning system transmits the received vehicle positioning GPS geographic location information to the cloud platform through the communication module in the vehicle.

4. The roadside perception method in vehicle-road collaboration according to claim 1 is characterized in that: At least four reference points are selected on the perception section of the roadside perception unit to establish the image coordinate system.

5. The roadside perception method in vehicle-road collaboration according to claim 1 is characterized in that: The coordinate conversion model includes establishing the height data, inclination angle data, image coordinate data of at least four reference points of the perception section, the image coordinates of the origin of the target vehicle, and the map coordinates of the origin of the target vehicle that need to be collected by the roadside perception unit, and establishing a mathematical coordinate conversion model through each data; and then converting the image coordinate information of the captured target vehicle into map coordinates corresponding to the first image coordinates of the target vehicle according to the coordinate conversion model.

6. The roadside perception method in vehicle-road collaboration according to claim 1 is characterized in that: The first driving route information includes first map coordinate data of the target vehicle and time data corresponding to the first map coordinate data.

7. The roadside perception method in vehicle-road collaboration according to claim 1 is characterized by: The second map coordinates are projected into a map coordinate system within the road section plane model, and the second driving route information includes second coordinate data and time data corresponding to the second map coordinate data.

8. A roadside perception system in vehicle-road collaboration, characterized by: It includes a roadside sensing unit located above one side of a monitored road section and monitoring the direction of vehicle flow, wherein the monitored road section includes at least three groups of connected road sections, and the roadside sensing unit includes three groups of sensing units; It also includes a computing cloud platform that is communicatively connected to the plurality of roadside sensing units, and an on-board unit that is communicatively connected to the computing cloud platform; The working method of the perception system is as follows: Establish a plane model of the perceived road section; The cloud platform is connected to the target vehicle positioning system to accurately locate the current GPS geographic location of the target vehicle and project it into the plane model of the road section to form a second map coordinate; The roadside sensing unit obtains the speed and image information of the target vehicle on the current road section, establishes an image coordinate system, transmits the image coordinate and speed information of the target vehicle to the cloud platform, tracks the target vehicle on the current monitored road section and continuously obtains the image coordinate information; Establish a coordinate conversion model, convert the acquired image coordinate information of the target vehicle into first map coordinates and overlap them with the map coordinate system in the road section plane model, transmit the real-time first map coordinate information of the target vehicle to the moving vehicle through the communication module, and construct the first driving route information of the vehicle according to the monitored first map coordinates; Constructing second driving route information of the target vehicle according to the second map coordinate information obtained by the target vehicle positioning system, and updating the second driving route information of the vehicle in the plane model in real time according to the vehicle positioning system; The cloud platform continuously synchronizes and compares the first driving route with the second driving route. When the roadside sensing unit monitors that the target vehicle is blocked and disappears from the field of vision, if the first driving route and the second driving route monitored by the target vehicle in the same period are consistent, the latest GPS second map coordinates obtained by the target vehicle positioning system are used as the real-time location information of the target vehicle. The cloud platform transmits the real-time second map coordinate information to the moving vehicle through the communication module.

9. The roadside perception system in vehicle-road collaboration according to claim 8 is characterized by: The roadside perception unit includes a camera and a radar.

Citation Information

Patent Citations

  • Sensing method, device and system

    CN110491156A

  • Urban road road-vehicle fusion global sensing method based on 5G

    CN111768621A

  • Vehicle infrastructure collaborative visual perception processing method based on gridding management

    CN114937249A

  • Vehicle traveling locus observation system, vehicle traveling locus observation method and program

    JP2010020729A

  • Method, apparatus, and system for detecting road incidents

    US20220084398A1