Vehicle parking and taking control method and control system

Abstract

The invention discloses a vehicle parking and picking-up control method and system. The method comprises the steps that a vehicle parking or picking-up instruction of a user is read; reading map navigation information, real-time road condition information and vehicle state information of a vehicle position based on the vehicle parking or taking instruction; road condition pre-judgment is carried out on the vehicle parking or taking process based on the map navigation information of the vehicle position, the real-time road condition information and the vehicle state information, and a driving route of the vehicle parking or taking process is determined according to a pre-judgment result; the driving route in the vehicle parking or taking process is converted into a corresponding driving control instruction, and the vehicle is controlled according to the corresponding driving control instruction; in the process of controlling the vehicle through the driving control instruction, the process of reading the map navigation information, the real-time road condition information and the vehicle state information of the vehicle position and the process of processing the read information are still carried out. In the vehicle driving process, the driving route is re-planned along with the environment change, the vehicle parking and taking efficiency is improved, and the parking and taking experience of a user is improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of vehicle parking and taking control, and particularly relates to a vehicle parking and taking control method and a control system. BACKGROUND

[0002] In the prior art, remote parking technology is that a vehicle acquires surrounding parking space types (available parking space, obstacle, drivable area) through a sensor, and transmits information to a mobile phone App. A user specifies a target parking area by dragging a selection box on the App interface, and after confirmation, the vehicle is automatically parked. Although the existing driving assistance system can safely park the vehicle in the idle specified parking space according to the needs of the driver. However, the traditional remote parking system usually does not have the function of autonomously selecting a parking space and parking accordingly. This is because the vehicle ECU cannot identify the optimal parking space in the parking environment through sensors and other devices, and needs to rely on the driver to control the vehicle to select the parking space; which affects the user's parking and taking experience.

[0003] A new vehicle parking and taking control method is needed to solve the above technical problems. SUMMARY

[0004] The purpose of the present application is to provide a vehicle parking and taking control method to solve the technical problems of low parking and taking efficiency and poor customer experience in the prior art.

[0005] The purpose of the present application is also to provide a control system using the vehicle parking and taking control method.

[0006] The technical solution of the present application to solve the technical problems is: A vehicle parking and taking control method, comprising the following steps: reading a vehicle parking or taking instruction of a user; reading map navigation information, real-time traffic information, and vehicle state information of a vehicle location based on the vehicle parking or taking instruction; making a traffic prediction on a vehicle parking or taking process based on the map navigation information, real-time traffic information, and vehicle state information of the vehicle location, determining a driving route of the vehicle parking or taking process according to a prediction result, converting the driving route of the vehicle parking or taking process into a corresponding driving control instruction, and controlling the vehicle with the corresponding driving control instruction; In the process of controlling the vehicle with the corresponding driving control instruction, the map navigation information, real-time traffic information, and vehicle state information of the vehicle location are still being read and processed.

[0007] Preferably, the driving route of the vehicle parking or taking process is determined according to the prediction result; and the driving route of the vehicle parking or taking process is converted into corresponding driving control instructions, including: performing grid processing on the driving environment based on the prediction result, calculating the risk value of each grid unit, and generating the driving route and the driving strategy corresponding to the driving route according to the size of the risk value, and outputting the driving control instructions corresponding to the driving strategy.

[0008] Preferably, the method further comprises: In the vehicle parking and taking control process, the user takeover instruction is read, and after the user takeover instruction is read, the vehicle parking and taking control is exited, and the vehicle is taken over by the user.

[0009] Preferably, the driving route is generated according to the risk value, specifically: the driving route is generated according to the risk value relationship of dynamic obstacle risk value> static obstacle risk value> vehicle control limited risk value> vehicle parking and taking efficiency risk value.

[0010] Preferably, the map navigation information of the vehicle position, the real-time traffic information, and the vehicle state information are specifically: the positions and speeds of surrounding objects and the vehicle self-positioning information are obtained through the vehicle self-sensor; the states of surrounding vehicles, signal light phases, and road side unit sensing information are obtained through the Internet of Vehicles; the lane lines, traffic signs, and curb stone static environment information are obtained through the navigation positioning system; and the traffic congestion, accidents, and weather information are obtained through the cloud.

[0011] Preferably, the specific process of predicting the road condition of the vehicle parking or taking process based on the map navigation information of the vehicle position, the real-time traffic information, and the vehicle state information is as follows: The obtained data information is subjected to data cleaning, time synchronization, and space unification processing, and different source data information is integrated to obtain fused vehicle information, and a dynamic and static combined scene graph is generated by matching the fused vehicle information with the map navigation information; Target recognition and tracking, target map matching and context perception are performed based on the scene graph; The behavior intention of the target is predicted, and a multi-modal trajectory is generated; The multi-modal trajectory result is subjected to conflict detection and risk quantification, a risk map is generated, and a prediction result is output.

[0012] Preferably, the method further comprises: In the vehicle parking and taking control process, the surrounding situation and the vehicle speed during the vehicle driving process are reminded.

[0013] A control system adopting a vehicle parking and taking control method, comprising: A reading unit is configured to read a vehicle parking or taking instruction of a user, and read map navigation information of a vehicle position, real-time traffic information, and vehicle state information based on the vehicle parking or taking instruction; A processing unit is configured to make a traffic prediction on a vehicle parking or taking process based on the map navigation information of the vehicle position, the real-time traffic information, and the vehicle state information, determine a driving route of the vehicle parking or taking process according to a prediction result, and convert the driving route of the vehicle parking or taking process into a corresponding driving control instruction; A control unit is configured to control the vehicle according to the corresponding driving control instruction.

[0014] Preferably, the processing unit comprises: A data preprocessing and fusion module is configured to perform data cleaning, time synchronization, and space unification processing on the obtained data information, integrate data information of different sources, obtain fused vehicle information, match the fused vehicle information with the map navigation information, and generate a dynamic and static combined scene graph; A scene understanding and state estimation module is configured to perform target identification and tracking, map matching, and context perception of the target; A behavior prediction and trajectory generation module is configured to perform behavior intention prediction of the target and generate a multi-modal trajectory; An integrated decision and risk mapping module is configured to perform conflict detection and risk quantification on the multi-modal trajectory result, generate a risk map, and output a prediction result; A decision and execution optimization module is configured to perform grid processing on a driving environment based on the prediction result, calculate a risk value of each grid unit, generate a driving route and a driving strategy corresponding to the driving route according to the size of the risk value, and output a control instruction corresponding to the driving strategy.

[0015] Preferably, the reading unit is further configured to read a user takeover instruction, and the processing unit is further configured to generate a control instruction for exiting parking or stopping.

[0016] The present application has the following advantages: the vehicle parking or taking instruction of the user is read, and the map navigation information of the vehicle position, the real-time traffic information, and the vehicle state information are read based on the vehicle parking or taking instruction; The map navigation information of the vehicle position, the real-time road condition information, and the vehicle state information are used to make a road condition prediction for the vehicle parking or taking process, a driving route for the vehicle parking or taking process is determined according to a prediction result, the driving route for the vehicle parking or taking process is converted into corresponding driving control instructions, and the vehicle is controlled by using the corresponding driving control instructions; in the process of controlling the vehicle by using the driving control instructions, the map navigation information of the vehicle position, the real-time road condition information, and the vehicle state information are read, and the information processing after reading is still in progress. The vehicle can re-plan a driving route according to environmental changes in the driving process, the vehicle parking or taking efficiency is improved, and the user experience of parking or taking is improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is a flowchart of the vehicle parking and taking control method of the present application; Figure 2 is a structural diagram of the control system using the vehicle parking and taking control method of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application.

[0019] As shown in Figure 1 , the present application discloses a vehicle parking and taking control method, which comprises the following steps: reading a vehicle parking or taking instruction of a user; Based on the vehicle parking or taking vehicle instruction, the map navigation information of the vehicle position, the real-time traffic information, and the vehicle state information are read; the map navigation information of the vehicle position, the real-time traffic information, and the vehicle state information are specifically: the position and speed of the surrounding objects and the vehicle self-positioning information are obtained through the vehicle self-sensor, and the specific sensor is a camera, a laser radar, a millimeter wave radar, a GPS / IMU, etc., which can be selected according to specific needs; the state of the surrounding vehicles, the signal lamp phase, and the road side unit sensing information are obtained through the Internet of Vehicles, which can be V2X, i.e., vehicle-to-vehicle communication, vehicle-to-road communication, and vehicle-to-cloud communication; the lane line, traffic sign, and curb stone static environment information are obtained through the navigation positioning system; and the traffic congestion situation, accident, and weather information are obtained through the cloud. In the actual application process, the information collection utilizes the 5G communication technology, and the ultra-low delay characteristic (theoretically, it can reach 1 millisecond) of the 5G network enables the vehicle to exchange information with other vehicles, infrastructure, or cloud servers in almost real time. The navigation positioning system can adopt the Beidou navigation positioning system, which has a positioning accuracy of centimeter level, providing extremely accurate position and environment information for the vehicle, which is helpful to realize accurate map matching and path planning, and the Beidou system supports multiple frequency working modes, improving the anti-interference ability and positioning accuracy, especially in complex environments. When the 5G communication Bluetooth technology is combined with the Beidou navigation, they can provide a more comprehensive and reliable intelligent parking and taking vehicle solution. With the precise position information provided by Beidou and the fast response capability of 5G, the vehicle can detect potential dangers and take measures in a very short time, greatly reducing the probability of traffic accidents. Whether in urban or rural areas, users can enjoy continuous service experience, because even in places without 4G / 5G coverage, Beidou can guarantee basic navigation services; and in places with 5G coverage, more rich service experience can be obtained.

[0020] Based on the map navigation information of the vehicle position, the real-time traffic information, and the vehicle state information, the road condition of the vehicle parking or taking vehicle process is predicted, and the driving route of the vehicle parking or taking vehicle process is determined according to the prediction result; the driving route of the vehicle parking or taking vehicle process is converted into corresponding driving control instructions, and the vehicle is controlled by the corresponding driving control instructions; The specific process of predicting the road condition of the vehicle parking or taking vehicle process based on the map navigation information of the vehicle position, the real-time traffic information, and the vehicle state information is as follows: The obtained data information is subjected to data cleaning, time synchronization, and space unification processing, different source data information is integrated, and fused vehicle information is obtained; by matching the fused vehicle information with the map navigation information, a dynamic and static combined scene graph is generated; Target recognition and tracking, map matching, and context perception of the target are performed based on the scene graph; Predicting the behavior intention of the target and generating a multi-modal trajectory; Performing conflict detection and risk quantification on the multi-modal trajectory result, generating a risk map, and outputting a prediction result.

[0021] The vehicle parking or taking process route is determined according to the prediction result; and the vehicle parking or taking process route is converted into a corresponding driving control instruction, which includes: performing grid processing on the driving environment based on the prediction result, calculating the risk value of each grid unit, and generating a driving route and a driving strategy corresponding to the driving route according to the size of the risk value, and outputting a driving control instruction corresponding to the driving strategy. The driving route is generated according to the size of the risk value, which is: the driving route is generated according to the risk value relationship of dynamic obstacle risk value> static obstacle risk value> vehicle control limited risk value> vehicle parking and taking efficiency. That is, in the vehicle parking and taking process, when a dynamic obstacle is encountered, the driving route should be stopped immediately or changed. When a static obstacle is encountered, it is ensured that no collision occurs during the formation process. When the vehicle control is limited, the user is reminded to take over and the driving route is changed. In the vehicle parking and taking process, efficiency is placed last, and in the above premise, the road with the shortest time is selected, and driving safety is further placed first.

[0022] In the process of controlling the vehicle with the corresponding driving control instruction, the map navigation information of the vehicle position, the real-time traffic information, the vehicle state information, and the information processing process after reading are still being performed.

[0023] It also includes: In the vehicle parking and taking control process, the user takeover instruction is read, and after the user takeover instruction is read, the vehicle parking and taking control is exited, and the vehicle is taken over by the user. In the driving process, if the user actively intervenes (steering wheel, brake, cancel button), the system must immediately and unconditionally exit the automatic control and return the authority to the driver. This reflects the respect for the user's sovereignty and the adherence to the safety bottom line.

[0024] In the vehicle parking and taking control process, the surrounding situation and the vehicle speed during the vehicle driving process are reminded to further ensure the safety of the driving process.

[0025] The method reads the vehicle parking or taking instruction of the user, reads the map navigation information, real-time traffic information and vehicle state information of the vehicle position based on the vehicle parking or taking instruction, makes a traffic prediction on the vehicle parking or taking process based on the map navigation information, real-time traffic information and vehicle state information of the vehicle position, determines the driving route of the vehicle parking or taking process according to the prediction result, converts the driving route of the vehicle parking or taking process into corresponding driving control instructions, and controls the vehicle with the corresponding driving control instructions. In the process of controlling the vehicle with the driving control instructions, the map navigation information, real-time traffic information and vehicle state information of the vehicle position are read, and the information processing after reading is still in progress. The vehicle can re-plan the driving route according to the environmental changes during driving, improve the vehicle parking and taking efficiency, and improve the user's parking and taking experience.

[0026] As shown in Figure 2 A control system adopting a vehicle parking and taking control method, comprising: A reading unit for reading the vehicle parking or taking instruction of the user, and reading the map navigation information, real-time traffic information and vehicle state information of the vehicle position based on the vehicle parking or taking instruction, and reading the user's takeover instruction; A processing unit for making a traffic prediction on the vehicle parking or taking process based on the map navigation information, real-time traffic information and vehicle state information of the vehicle position, determining the driving route of the vehicle parking or taking process according to the prediction result, converting the driving route of the vehicle parking or taking process into corresponding driving control instructions, and generating the control instructions for exiting parking and stopping; wherein the processing unit comprises: A data preprocessing and fusion module for performing data cleaning, time synchronization, space unification processing on the obtained data information, integrating data information of different sources, obtaining fused vehicle information, and generating a dynamic and static combined scene graph by matching the fused vehicle information with the map navigation information; A scene understanding and state estimation module for target recognition and tracking, map matching and context perception of the target; A behavior prediction and trajectory generation module for target behavior intention prediction and multi-modal trajectory generation; A comprehensive decision and risk mapping module for conflict detection and risk quantification of multi-modal trajectory results, generation of a risk map, and output of a prediction result; A decision and execution optimization module for grid processing of the driving environment based on the prediction result, calculation of the risk value of each grid unit, generation of the driving route and the driving strategy corresponding to the driving route according to the size of the risk value, and output of the control instructions corresponding to the driving strategy.

[0027] A control unit for controlling the vehicle with the corresponding driving control instructions.

[0028] All other embodiments that would be obvious to those of ordinary skill in the art based on the disclosure herein, in the absence of doing creative work, are within the scope of the present invention.

Claims

1. A vehicle parking and taking control method, characterized by, The method comprises the following steps: reading a vehicle parking or taking instruction of a user; reading map navigation information, real-time traffic information, and vehicle state information of a vehicle location based on the vehicle parking or taking instruction; making a traffic prediction for a vehicle parking or taking process based on the map navigation information, real-time traffic information, and vehicle state information of the vehicle location, determining a driving route of the vehicle parking or taking process according to a prediction result, converting the driving route of the vehicle parking or taking process into corresponding driving control instructions, and controlling the vehicle with the corresponding driving control instructions; wherein, during the process of controlling the vehicle with the corresponding driving control instructions, the reading of the map navigation information, real-time traffic information, and vehicle state information of the vehicle location and the reading information processing are still in progress.

2. The vehicle parking and taking control method according to claim 1, wherein: the driving route of the vehicle parking or taking process is determined according to the prediction result; the conversion of the driving route of the vehicle parking or taking process into corresponding driving control instructions comprises: performing a grid processing on a driving environment based on the prediction result, calculating a risk value of each grid unit, generating a driving route and a driving strategy corresponding to the driving route according to the size of the risk value, and outputting a driving control instruction corresponding to the driving strategy.

3. The vehicle parking and pickup control method according to claim 2, characterized by, further comprising: during the vehicle parking and taking control process, reading a user takeover instruction, and after reading the user takeover instruction, exiting the vehicle parking and taking control, and the vehicle is taken over by the user.

4. The vehicle parking and pickup control method according to claim 3, characterized by: the driving route is generated according to a risk value relationship of dynamic obstacle risk value > static obstacle risk value > vehicle control limited risk value > vehicle parking and taking efficiency risk value.

5. The vehicle parking and pickup control method according to claim 1, characterized by, the map navigation information, real-time traffic information, and vehicle state information of the vehicle location are specifically: obtaining the position and speed of surrounding objects and the vehicle self-positioning information through the vehicle's own sensors; obtaining the state of surrounding vehicles, signal light phases, and road side unit sensing information through the Internet of Vehicles; obtaining lane lines, traffic signs, and curb stone static environment information through a navigation positioning system; and obtaining traffic congestion, accidents, and weather information through the cloud.

6. The vehicle parking and pickup control method according to claim 1, characterized by, the specific process of making a traffic prediction for the vehicle parking or taking process based on the map navigation information, real-time traffic information, and vehicle state information of the vehicle location is as follows: performing data cleaning, time synchronization, and space unification processing on the obtained data information, integrating data information of different sources, obtaining fused vehicle information, matching the fused vehicle information with map navigation information, and generating a dynamic and static combined scene graph; performing target recognition and tracking, map matching, and context perception on the target based on the scene graph; predicting the behavior intention of the target and generating a multi-modal trajectory; performing conflict detection and risk quantification on the multi-modal trajectory result, generating a risk map, and outputting a prediction result.

7. The vehicle parking and pickup control method according to claim 1, characterized by, further comprising: during the vehicle parking and taking control process, reminding the user of the surrounding situation and vehicle speed during the vehicle driving process.

8. A control system employing the vehicle parking and pickup control method according to any one of claims 1 to 7, characterized by comprising: a reading unit configured to read a vehicle parking or taking instruction of a user; read map navigation information of the vehicle position, real-time traffic information, and vehicle state information based on the vehicle parking or taking vehicle instruction; a processing unit configured to make a traffic prediction on the vehicle parking or taking vehicle process based on the map navigation information of the vehicle position, the real-time traffic information, and the vehicle state information, determine a driving route of the vehicle parking or taking vehicle process according to a prediction result, and convert the driving route of the vehicle parking or taking vehicle process into a corresponding driving control instruction; a control unit configured to control the vehicle according to the corresponding driving control instruction.

9. The control system employing the vehicle parking and taking control method according to claim 8, characterized by, The processing unit comprises: a data preprocessing and fusion module configured to perform data cleaning, time synchronization, and space unification processing on the obtained data information, integrate data information of different sources, obtain fused vehicle information, match the fused vehicle information with the map navigation information, and generate a dynamic and static combined scene graph; a scene understanding and state estimation module configured to perform target recognition and tracking, map matching, and context perception; a behavior prediction and trajectory generation module configured to perform behavior intention prediction of a target and generate a multi-modal trajectory; a comprehensive decision and risk mapping module configured to perform conflict detection and risk quantification on the multi-modal trajectory result, generate a risk map, and output a prediction result; a decision and execution optimization module configured to perform grid processing on a driving environment based on the prediction result, calculate a risk value of each grid unit, generate a driving route and a driving strategy corresponding to the driving route according to the size of the risk value, and output a control instruction corresponding to the driving strategy.

10. The control system employing the vehicle parking and taking control method according to claim 8, characterized by, The reading unit is further configured to read a user takeover instruction, and the processing unit is further configured to generate a control instruction for exiting parking or stopping.

Images