CN110751847B - A method and system for autonomous vehicle behavior decision-making - Google Patents
A method and system for autonomous vehicle behavior decision-making Download PDFInfo
- Publication number
- CN110751847B CN110751847B CN201910976369.9A CN201910976369A CN110751847B CN 110751847 B CN110751847 B CN 110751847B CN 201910976369 A CN201910976369 A CN 201910976369A CN 110751847 B CN110751847 B CN 110751847B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- key
- traffic participants
- real
- participants
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 230000003068 static effect Effects 0.000 claims abstract description 19
- 230000008859 change Effects 0.000 claims abstract description 8
- 230000006399 behavior Effects 0.000 claims description 71
- 238000012216 screening Methods 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 16
- 238000013135 deep learning Methods 0.000 claims description 5
- 238000010801 machine learning Methods 0.000 claims description 5
- 230000004888 barrier function Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits
- G08G1/096708—Systems involving transmission of highway information, e.g. weather, speed limits where the received information might be used to generate an automatic action on the vehicle control
- G08G1/096725—Systems involving transmission of highway information, e.g. weather, speed limits where the received information might be used to generate an automatic action on the vehicle control where the received information generates an automatic action on the vehicle control
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Mechanical Engineering (AREA)
- Theoretical Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Human Computer Interaction (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Atmospheric Sciences (AREA)
- Traffic Control Systems (AREA)
Abstract
本发明涉及一种自动驾驶车辆行为决策方法及系统，其特征在于，包括以下内容：1)确定本车辆的参考路径；2)根据确定的参考路径，对本车辆所在的自动驾驶场景范围内的交通参与者进行筛选，得到关键交通参与者，包括关键真实交通参与者和关键虚拟交通参与者；3)针对关键真实交通参与者中的全部动态交通参与者，根据确定的关键交通参与者，对本车辆进行纵向行为决策，得到本车辆的速度调节结果；4)针对关键真实交通参与者中的全部静态交通参与者，对本车辆进行横向行为决策，得到本车辆的路径变化结果，完成本车辆的自动驾驶场景行为决策，本发明可以广泛应用于自动驾驶领域中。
The present invention relates to an automatic driving vehicle behavior decision-making method and system, which is characterized by comprising the following contents: 1) determining the reference path of the vehicle; Participants are screened to obtain key traffic participants, including key real traffic participants and key virtual traffic participants; 3) For all dynamic traffic participants in the key real traffic participants, according to the determined key traffic Make vertical behavior decisions to obtain the speed adjustment results of the vehicle; 4) For all the static traffic participants in the key real traffic participants, make lateral behavior decisions for the vehicle, obtain the path change results of the vehicle, and complete the automatic driving of the vehicle Scenario behavior decision, the present invention can be widely used in the field of automatic driving.
Description
Technical Field
The invention relates to a behavior decision method and a behavior decision system, in particular to a behavior decision method and a behavior decision system for an automatic driving vehicle, and belongs to the field of automatic driving.
Background
With the rapid development of the automatic driving technology, the application field of the automatic driving vehicle is increasingly expanded, and a great amount of scientific research power is put into various colleges and universities, vehicle enterprises and internet enterprises. The automatic driving vehicle is a comprehensive intelligent system integrating multiple functions of navigation, environment perception, decision planning, man-machine interaction and the like, and according to estimation, the market potential of the automatic driving vehicle in the next decades is quite large, and huge economic and social benefits can be generated. Crossroads are the most common and important scenes in urban roads, the road structure is complex, pedestrians and traffic flow are collected and shunted from multiple directions, and meanwhile, the crossroads are influenced by traffic signals and traffic signs, and the crossroads are always the key points and the difficulties of research in the field of automatic driving. Therefore, it is very important to design a safe and intelligent automatic driving behavior decision method for the intersection.
At present, the decision-making methods of the behaviors of the automatic driving vehicles which are applied more can be roughly divided into rule-based and learning-based decision-making methods of the behaviors of the crossroads, wherein the rule-based decision-making method of the behaviors of the crossroads provides suggested behaviors for each traffic participant, and specifically comprises following, parking, yielding, overtaking, avoiding, neglecting and the like; the method comprises the steps of establishing a state model of a vehicle and surrounding traffic participants based on a learned intersection behavior decision method, selecting acceleration, deceleration or uniform speed based on each future time period, and generating an optimal driving strategy according to accumulated income.
However, the rule-based intersection behavior decision method needs to integrate and combine the individual decisions for each traffic participant into a final decision, and dependence and conflict exist between each individual decision, so that it is difficult to ensure the consistency of the final behavior decision result. The learning-based intersection behavior decision method needs a large amount of data for model training and testing, and the behavior decision result cannot guarantee absolute safety and is difficult to apply to actual automatic driving.
Disclosure of Invention
In view of the above problems, the present invention provides a method and a system for decision-making of behavior of an autonomous vehicle, which can ensure the safety and consistency of the decision-making result.
In order to achieve the purpose, the invention adopts the following technical scheme: an autonomous vehicle behavior decision method, comprising: 1) determining a reference path of the vehicle; 2) screening traffic participants in the automatic driving scene range where the vehicle is located according to the determined reference path to obtain key traffic participants including key real traffic participants and key virtual traffic participants; 3) aiming at all dynamic traffic participants in the key real traffic participants, making a longitudinal behavior decision on the vehicle according to the determined key traffic participants to obtain a speed regulation result of the vehicle; 4) and (4) performing transverse behavior decision on the vehicle aiming at all static traffic participants in the key real traffic participants to obtain a path change result of the vehicle and finish automatic driving scene behavior decision of the vehicle.
Further, the specific process of the step 1) is as follows: 1.1) judging whether the vehicle is positioned in an automatic driving scene range according to vehicle positioning acquired by a vehicle-mounted sensor and data acquired by a high-precision map; 1.2) when the vehicle enters the automatic driving scene range, acquiring an optimal global path of the vehicle in advance, and selecting a target lane of the vehicle according to the optimal global path; 1.3) extracting or planning a path from the current lane of the vehicle to the target lane in real time from the high-precision map, and taking the path as a reference path.
Further, the specific process of the step 2) is as follows: 2.1) calculating the future track of each real traffic participant for a certain time by adopting a machine learning or deep learning method according to the real driving data of the vehicle; 2.2) screening real traffic participants in the automatic driving scene range according to the determined reference path and the calculated track, and determining key real traffic participants which conflict with the vehicle; 2.3) screening the virtual traffic participants in the automatic driving scene range by adopting a vision or V2X mode, and determining key virtual traffic participants which conflict with the vehicle.
Further, the specific process of step 2.2) is as follows: judging whether the determined reference path is intersected with the calculated track, if so, determining that a real traffic participant corresponding to the calculated track conflicts with the vehicle, wherein the real traffic participant is a key real traffic participant; if the calculated trajectories do not intersect, the real traffic participant corresponding to the calculated trajectory does not conflict with the vehicle, and the real traffic participant is not a key real traffic participant.
Further, the real traffic participants comprise pedestrians and vehicles, and the virtual traffic participants comprise traffic signal lamps, pedestrian crosswalks, parking signboards and yield signboards.
Further, the specific process of step 3) is as follows: 3.1) determining a collision point, wherein the collision point of the vehicle and the key real traffic participant is determined according to the reference path of the vehicle and the calculated track of the key real traffic participant, and the collision point of the vehicle and the key virtual traffic participant is positioned on a stop line within the range of the automatic driving scene; 3.2) if the collision point exists, selecting the collision point closest to the vehicle, and setting a corresponding parking point in front of the collision point; 3.3) determining the position difference value between the current parking point and the original parking point, and if no collision point exists, clearing the existing parking point; 3.4) when the position difference exceeds a preset position threshold value, taking the current parking spot as an updated parking spot; and when the position difference value does not exceed the preset position threshold value, keeping the original parking point, and entering the step 3.1) until the vehicle leaves the automatic driving scene range.
Further, the specific process of the step 4) is as follows: at the same time, when the key real traffic participant which is closest to the vehicle and conflicts with the vehicle is a static traffic participant, making a transverse behavior decision on the vehicle, selecting one side with larger passable space to drive, and avoiding the key real traffic participant; and when the key real traffic participant which is closest to the vehicle and conflicts with the vehicle is a dynamic traffic participant, the lateral behavior decision of the vehicle is not carried out.
Further, the reference path in the step 1) is generated in real time by adopting a lane line identification method based on vision and laser.
Further, the dynamic traffic participants are key real traffic participants moving within the scope of the automatic driving scene; the static traffic participants are key real traffic participants with a speed of zero or below 0.5m/s within the scope of the automatic driving scenario.
An autonomous vehicle behavior decision system, comprising: a reference path determination module for determining a reference path of the vehicle; the traffic participant screening module is used for screening traffic participants in the automatic driving scene range where the vehicle is located according to the determined reference path to obtain key traffic participants, including key real traffic participants and key virtual traffic participants; the longitudinal behavior decision module is used for carrying out longitudinal behavior decision on the vehicle according to the determined key traffic participants aiming at all dynamic traffic participants in the key real traffic participants to obtain a speed regulation result of the vehicle; and the transverse behavior decision module is used for carrying out transverse behavior decision on the vehicle aiming at all static traffic participants in the key real traffic participants to obtain a path change result of the vehicle and finish the automatic driving scene behavior decision of the vehicle.
Due to the adoption of the technical scheme, the invention has the following advantages: the invention represents the behavior decision of the automatic driving vehicle as setting a parking spot or avoiding a certain static barrier, realizes the longitudinal behavior decision by adopting a mode of updating the parking spot, realizes the transverse behavior decision by avoiding a certain static barrier, avoids the dependence and conflict caused by the behavior decision of each traffic participant, has simple and clear decision process, does not have the dependence and conflict between each independent decision, can ensure the safety and consistency of the behavior decision result, and can be widely applied to the field of automatic driving.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
fig. 2 is a flow chart of the parking spot setting in the method of the present invention.
Detailed Description
The present invention is described in detail below with reference to the attached drawings. It is to be understood, however, that the drawings are provided solely for the purposes of promoting an understanding of the invention and that they are not to be construed as limiting the invention.
As shown in fig. 1, the method for deciding the behavior of the autonomous vehicle provided by the present invention comprises the following steps:
1) according to the vehicle positioning acquired by the vehicle-mounted sensor and the data acquired by a high-precision map (the precision reaches centimeter level), the reference path of the vehicle is determined, and the method specifically comprises the following steps:
1.1) judging whether the vehicle is positioned in an automatic driving scene range according to the vehicle positioning acquired by the vehicle-mounted sensor and the data acquired by the high-precision map, wherein the automatic driving scene comprises a crossroad, a common urban road, a low-speed park and the like, and the automatic driving scene is the crossroad.
1.2) when the vehicle enters the range of the crossroad, acquiring the optimal global path of the vehicle in advance, and selecting the target lane of the vehicle according to the optimal global path, wherein the range of the crossroad comprises the crossroad and a range which is dozens of meters away from the crossroad, the specific range can be set according to the actual situation, the method for acquiring the optimal global path of the vehicle is disclosed in the prior art, and the specific process is not repeated herein.
1.3) extracting or planning a smooth path from the current lane of the vehicle to the target lane in real time from the high-precision map, and taking the smooth path as a reference path.
2) Screening the traffic participants in the range of the crossroad according to the determined reference path to obtain key traffic participants, which specifically comprises the following steps:
a large number of traffic participants are contained in the range of the crossroad, and the decision complexity and the calculation resources can be reduced by pre-screening key traffic participants. Traffic participants in the range of the crossroad can be divided into real traffic participants and virtual traffic participants, wherein the real traffic participants comprise pedestrians, vehicles and the like, the virtual traffic participants comprise traffic signal lamps, pedestrian crosswalks, parking signboards, yield signboards and the like, and the conflict relationship between the virtual traffic participants and the vehicles needs to be judged respectively.
2.1) according to the real driving data of the vehicle, adopting a machine learning or deep learning method to calculate the intention (behavior) and the track of each real traffic participant for a certain time in the future (specific time can be determined according to actual conditions), wherein the real driving data is a data set comprising a large number of speeds, moving directions, behaviors and tracks of vehicles, the machine learning or deep learning method is a method disclosed by the prior art, and the specific process is not repeated herein.
2.2) screening real traffic participants in the range of the crossroad according to the determined reference path and the calculated intention and track, and determining key real traffic participants which conflict with the vehicle, namely judging whether the determined reference path intersects with the calculated track, if so, determining that the real traffic participants corresponding to the calculated track conflict with the vehicle, and the real traffic participants are the key real traffic participants; if the calculated trajectories do not intersect, the real traffic participant corresponding to the calculated trajectory does not conflict with the vehicle, and the real traffic participant is not a key real traffic participant.
2.3) screening virtual traffic participants in the range of the intersection by adopting a vision or V2X (vehicle to outside information exchange) mode, and determining key virtual traffic participants which conflict with the vehicle, for example, according to traffic lights corresponding to the current lane of the vehicle, namely if the traffic lights are yellow or red, the virtual traffic participants conflict with the vehicle, wherein the virtual traffic participants are key virtual traffic participants; if the traffic signal light is green, the virtual traffic participant does not conflict with the vehicle, and the virtual traffic participant is not a key virtual traffic participant. Therefore, the real traffic participants and the virtual traffic participants which conflict with the vehicle are the screened key traffic participants.
3) Because the structure of the crossroad is complex and the number of traffic participants is large, and the principle of 'speed giving no way to give way' is required to be followed as much as possible, therefore, aiming at all the dynamic traffic participants in the key real traffic participants, the longitudinal behavior decision (i.e. the speed regulation decision) is carried out on the vehicle according to the determined key traffic participants to obtain the speed regulation result of the vehicle, namely the updated parking point is obtained, in order to ensure the safety and the stability of the behavior decision of the automatic driving vehicle, the longitudinal behavior decision adopts the mode of setting the parking point, as shown in figure 2, a parking point is arranged in front of the position of each key traffic participant which conflicts with the vehicle, the position of the parking point is updated or cleared according to the positions of the key traffic participants, wherein the dynamic traffic participants are the key real traffic participants moving in the range of the crossroad, the method specifically comprises the following steps:
3.1) determining a collision point, wherein the collision point of the vehicle and the key real traffic participants is determined according to the reference path of the vehicle and the track of the key real traffic participants, and the collision point of the vehicle and the key virtual traffic participants is positioned on a stop line of the intersection.
And 3.2) if the collision point exists, selecting the collision point closest to the vehicle, and setting a corresponding parking point in front of the collision point based on the safe distance.
3.3) determining the position difference value between the current parking point and the original parking point, and clearing the existing parking point if no collision point exists, wherein the safe distance can be set according to the actual situation, and can be 4 meters for example.
3.4) in order to ensure the consistency of the decision result, when the position difference value exceeds a preset position threshold value, taking the current parking spot as an updated parking spot; and when the position difference value does not exceed the preset position threshold value, keeping the original parking point, and entering the step 3.1) until the vehicle leaves the range of the intersection, wherein the preset position threshold value can be set according to the actual situation, and can be 3 meters for example.
4) Considering that a conflict exists between a static traffic participant and a vehicle, in order to ensure that an automatically driven vehicle can smoothly pass through a crossroad range, it is necessary to make a transverse behavior decision (i.e., a path change) on the vehicle for all static traffic participants in the critical real traffic participants, and avoid the static traffic participants having a conflict on the vehicle, so as to obtain a path change result of the vehicle, i.e., obtain a final reference path, where the static traffic participants are the critical real traffic participants with zero or extremely low speed (lower than 0.5m/s) in the crossroad range, and specifically:
because the determined reference path is changed when a transverse behavior decision is executed for a certain static traffic participant, and other transverse and longitudinal behavior decision results are influenced, when a key real traffic participant which is closest to the vehicle and conflicts with the vehicle is a static traffic participant at the same time, the transverse behavior decision is carried out on the vehicle, namely the vehicle takes an avoidance behavior decision, one side with larger passable space is selected to run, and the key real traffic participant is avoided; and when the key real traffic participant which is closest to the vehicle and conflicts with the vehicle is a dynamic traffic participant, the lateral behavior decision of the vehicle is not carried out. And the combination of the longitudinal behavior decision result and the transverse behavior decision result is the intersection behavior decision result of the vehicle.
In the step 1), the reference path may be generated in real time by using a lane line recognition method based on vision and laser.
Based on the above method for deciding the behavior of the automatically driven vehicle, the present invention further provides a system for deciding the behavior of the automatically driven vehicle, comprising:
a reference path determination module for determining a reference path of the vehicle; the traffic participant screening module is used for screening traffic participants in the automatic driving scene range where the vehicle is located according to the determined reference path to obtain key traffic participants, including key real traffic participants and key virtual traffic participants; the longitudinal behavior decision module is used for carrying out longitudinal behavior decision on the vehicle according to the determined key traffic participants aiming at all dynamic traffic participants in the key real traffic participants to obtain a speed regulation result of the vehicle; and the transverse behavior decision module is used for carrying out transverse behavior decision on the vehicle aiming at all static traffic participants in the key real traffic participants to obtain a path change result of the vehicle and finish the automatic driving scene behavior decision of the vehicle.
The above embodiments are only used for illustrating the present invention, and the structure, connection mode, manufacturing process, etc. of the components may be changed, and all equivalent changes and modifications performed on the basis of the technical solution of the present invention should not be excluded from the protection scope of the present invention.
Claims (6)
1. An autonomous vehicle behavior decision method, comprising:
1) determining a reference path of the vehicle;
2) according to the determined reference path, screening traffic participants in the automatic driving scene range where the vehicle is located to obtain key traffic participants including key real traffic participants and key virtual traffic participants, wherein the real traffic participants include pedestrians and vehicles, the virtual traffic participants include traffic signal lamps, pedestrian crossings, parking signboards and yield signboards, and the specific process is as follows:
2.1) calculating the future track of each real traffic participant for a certain time by adopting a machine learning or deep learning method according to the real driving data of the vehicle;
2.2) screening real traffic participants in the automatic driving scene range according to the determined reference path and the calculated track, and determining key real traffic participants which conflict with the vehicle;
2.3) screening the virtual traffic participants in the automatic driving scene range by adopting a vision or V2X mode, and determining key virtual traffic participants which conflict with the vehicle;
3) aiming at all dynamic traffic participants in the key real traffic participants, making a longitudinal behavior decision on the vehicle according to the determined key traffic participants to obtain a speed regulation result of the vehicle, wherein the specific process is as follows:
3.1) determining a collision point, wherein the collision point of the vehicle and the key real traffic participant is determined according to the reference path of the vehicle and the calculated track of the key real traffic participant, and the collision point of the vehicle and the key virtual traffic participant is positioned on a stop line within the range of the automatic driving scene;
3.2) if the collision point exists, selecting the collision point closest to the vehicle, and setting a corresponding parking point in front of the collision point;
3.3) determining the position difference value between the current parking point and the original parking point, and if no collision point exists, clearing the existing parking point;
3.4) when the position difference exceeds a preset position threshold value, taking the current parking spot as an updated parking spot; when the position difference value does not exceed the preset position threshold value, the original parking point is reserved, and the step 3.1) is carried out until the vehicle leaves the automatic driving scene range;
4) aiming at all static traffic participants in the key real traffic participants, transverse behavior decision is carried out on the vehicle to obtain a path change result of the vehicle, and automatic driving scene behavior decision of the vehicle is completed, wherein the specific process is as follows:
at the same time, when the key real traffic participant which is closest to the vehicle and conflicts with the vehicle is a static traffic participant, making a transverse behavior decision on the vehicle, selecting one side with larger passable space to drive, and avoiding the key real traffic participant; and when the key real traffic participant which is closest to the vehicle and conflicts with the vehicle is a dynamic traffic participant, the lateral behavior decision of the vehicle is not carried out.
2. The automated driving vehicle behavior decision method according to claim 1, wherein the specific process of step 1) is as follows:
1.1) judging whether the vehicle is positioned in an automatic driving scene range according to vehicle positioning acquired by a vehicle-mounted sensor and data acquired by a high-precision map;
1.2) when the vehicle enters the automatic driving scene range, acquiring an optimal global path of the vehicle in advance, and selecting a target lane of the vehicle according to the optimal global path;
1.3) extracting or planning a path from the current lane of the vehicle to the target lane in real time from the high-precision map, and taking the path as a reference path.
3. An automated driving vehicle behavior decision method as claimed in claim 1, characterized in that the specific process of step 2.2) is:
judging whether the determined reference path is intersected with the calculated track, if so, determining that a real traffic participant corresponding to the calculated track conflicts with the vehicle, wherein the real traffic participant is a key real traffic participant;
if the calculated trajectories do not intersect, the real traffic participant corresponding to the calculated trajectory does not conflict with the vehicle, and the real traffic participant is not a key real traffic participant.
4. A method as claimed in any one of claims 1 to 3, wherein the reference path in step 1) is generated in real time using a lane line recognition method based on vision and laser.
5. A method as claimed in any one of claims 1 to 3, wherein the dynamic traffic participants are key real traffic participants moving within the context of the autonomous driving scenario; the static traffic participants are key real traffic participants with a speed of zero or below 0.5m/s within the scope of the automatic driving scenario.
6. An autonomous vehicle behavior decision system, comprising:
a reference path determination module for determining a reference path of the vehicle;
the traffic participant screening module is used for screening traffic participants in an automatic driving scene range where the vehicle is located according to the determined reference path to obtain key traffic participants, wherein the key traffic participants comprise key real traffic participants and key virtual traffic participants, the real traffic participants comprise pedestrians and vehicles, the virtual traffic participants comprise traffic signal lamps, pedestrian crossings, parking signboards and yield signboards, and the specific process is as follows:
2.1) calculating the future track of each real traffic participant for a certain time by adopting a machine learning or deep learning method according to the real driving data of the vehicle;
2.2) screening real traffic participants in the automatic driving scene range according to the determined reference path and the calculated track, and determining key real traffic participants which conflict with the vehicle;
2.3) screening the virtual traffic participants in the automatic driving scene range by adopting a vision or V2X mode, and determining key virtual traffic participants which conflict with the vehicle;
the longitudinal behavior decision module is used for carrying out longitudinal behavior decision on the vehicle according to the determined key traffic participants aiming at all dynamic traffic participants in the key real traffic participants to obtain a speed regulation result of the vehicle, and the specific process is as follows:
3.1) determining a collision point, wherein the collision point of the vehicle and the key real traffic participant is determined according to the reference path of the vehicle and the calculated track of the key real traffic participant, and the collision point of the vehicle and the key virtual traffic participant is positioned on a stop line within the range of the automatic driving scene;
3.2) if the collision point exists, selecting the collision point closest to the vehicle, and setting a corresponding parking point in front of the collision point;
3.3) determining the position difference value between the current parking point and the original parking point, and if no collision point exists, clearing the existing parking point;
3.4) when the position difference exceeds a preset position threshold value, taking the current parking spot as an updated parking spot; when the position difference value does not exceed the preset position threshold value, the original parking point is reserved, and the step 3.1) is carried out until the vehicle leaves the automatic driving scene range;
the transverse behavior decision module is used for carrying out transverse behavior decision on the vehicle aiming at all static traffic participants in the key real traffic participants to obtain a path change result of the vehicle and finish the automatic driving scene behavior decision of the vehicle, and the specific process is as follows:
at the same time, when the key real traffic participant which is closest to the vehicle and conflicts with the vehicle is a static traffic participant, making a transverse behavior decision on the vehicle, selecting one side with larger passable space to drive, and avoiding the key real traffic participant; and when the key real traffic participant which is closest to the vehicle and conflicts with the vehicle is a dynamic traffic participant, the lateral behavior decision of the vehicle is not carried out.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910976369.9A CN110751847B (en) | 2019-10-15 | 2019-10-15 | A method and system for autonomous vehicle behavior decision-making |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910976369.9A CN110751847B (en) | 2019-10-15 | 2019-10-15 | A method and system for autonomous vehicle behavior decision-making |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110751847A CN110751847A (en) | 2020-02-04 |
| CN110751847B true CN110751847B (en) | 2021-03-30 |
Family
ID=69278316
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910976369.9A Active CN110751847B (en) | 2019-10-15 | 2019-10-15 | A method and system for autonomous vehicle behavior decision-making |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110751847B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111832154B (en) * | 2020-06-11 | 2021-05-18 | 北京华如科技股份有限公司 | Identification test method for space decision consistency under large-scale data drive |
| CN111915888B (en) * | 2020-07-14 | 2021-09-03 | 同济大学 | Method for calculating complexity of traffic participants in automatic driving test scene |
| CN112364439B (en) * | 2020-12-02 | 2021-11-02 | 腾讯科技（深圳）有限公司 | Simulation test method and device for automatic driving system and storage medium |
| CN113071487B (en) * | 2021-02-24 | 2022-02-18 | 中移智行网络科技有限公司 | Automatic driving vehicle control method and device and cloud equipment |
| CN112799412A (en) * | 2021-04-12 | 2021-05-14 | 北京三快在线科技有限公司 | Control method and control device of unmanned equipment |
| CN113569378B (en) * | 2021-06-16 | 2024-01-05 | 阿波罗智联(北京)科技有限公司 | Simulation scene generation method and device, electronic equipment and storage medium |
| CN113911136B (en) * | 2021-09-29 | 2023-03-21 | 广州文远知行科技有限公司 | Driverless vehicle lane change decision-making method, driverless vehicle lane change decision-making system, driverless vehicle lane change decision-making equipment and driverless vehicle lane change decision-making medium |
| CN114084155B (en) * | 2021-11-15 | 2023-10-20 | 清华大学 | Predictive intelligent automobile decision control method and device, automobile and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2299969A1 (en) * | 2000-02-22 | 2001-08-22 | Peter Johann Kielland | Vehicle collision avoidance system using cell phones devices communicating through the internet |
| SE515655C2 (en) * | 1999-12-22 | 2001-09-17 | Saab Ab | "System and method of collision avoidance between vehicles |
| CN102496285A (en) * | 2011-12-23 | 2012-06-13 | 上海海事大学 | Method for determining red-light running of vehicles at intersection based on video detection and signal control system |
| CN102881180A (en) * | 2012-09-25 | 2013-01-16 | 浙江吉利汽车研究院有限公司杭州分公司 | Control device and method for preventing vehicles from running red light |
| CN104106103A (en) * | 2012-02-10 | 2014-10-15 | 丰田自动车株式会社 | Driving assistance apparatus |
| CN106530789A (en) * | 2016-10-31 | 2017-03-22 | 成都路行通信息技术有限公司 | Automobile parking point static drift detection method and system |
| CN109979238A (en) * | 2017-12-28 | 2019-07-05 | 北京百度网讯科技有限公司 | Barrier based reminding method, device and equipment in lane |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2726388A1 (en) * | 1994-10-26 | 1996-05-03 | Nelrow Sa | Computation of anti-collision trajectories for motor vehicles |
| US7639841B2 (en) * | 2004-12-20 | 2009-12-29 | Siemens Corporation | System and method for on-road detection of a vehicle using knowledge fusion |
| CN107274699A (en) * | 2017-06-30 | 2017-10-20 | 重庆交通大学 | Based on what position was shared passing method is cooperateed with without signal-control crossing vehicle certainly |
| WO2019006743A1 (en) * | 2017-07-07 | 2019-01-10 | 驭势科技（北京）有限公司 | Method and device for controlling travel of vehicle |
| CN107738644B (en) * | 2017-09-30 | 2019-06-21 | 长安大学 | A vehicle collision avoidance control method |
| CN109829351B (en) * | 2017-11-23 | 2021-06-01 | 华为技术有限公司 | Lane information detection method, device and computer readable storage medium |
| CN109017786B (en) * | 2018-08-09 | 2020-09-22 | 北京智行者科技有限公司 | Vehicle obstacle avoidance method |
| WO2020133208A1 (en) * | 2018-12-28 | 2020-07-02 | 驭势科技（北京）有限公司 | Control method for self-driving vehicle, and self-driving system |
| CN110209485B (en) * | 2019-06-05 | 2020-06-02 | 青岛海通胜行智能科技有限公司 | Dynamic avoidance method for multiple robots during cooperative operation |
| CN110288847B (en) * | 2019-06-28 | 2021-01-19 | 浙江吉利控股集团有限公司 | Automatic driving decision method, device and system, storage medium and terminal |
-
2019
- 2019-10-15 CN CN201910976369.9A patent/CN110751847B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SE515655C2 (en) * | 1999-12-22 | 2001-09-17 | Saab Ab | "System and method of collision avoidance between vehicles |
| CA2299969A1 (en) * | 2000-02-22 | 2001-08-22 | Peter Johann Kielland | Vehicle collision avoidance system using cell phones devices communicating through the internet |
| CN102496285A (en) * | 2011-12-23 | 2012-06-13 | 上海海事大学 | Method for determining red-light running of vehicles at intersection based on video detection and signal control system |
| CN104106103A (en) * | 2012-02-10 | 2014-10-15 | 丰田自动车株式会社 | Driving assistance apparatus |
| CN102881180A (en) * | 2012-09-25 | 2013-01-16 | 浙江吉利汽车研究院有限公司杭州分公司 | Control device and method for preventing vehicles from running red light |
| CN106530789A (en) * | 2016-10-31 | 2017-03-22 | 成都路行通信息技术有限公司 | Automobile parking point static drift detection method and system |
| CN109979238A (en) * | 2017-12-28 | 2019-07-05 | 北京百度网讯科技有限公司 | Barrier based reminding method, device and equipment in lane |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110751847A (en) | 2020-02-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110751847B (en) | A method and system for autonomous vehicle behavior decision-making | |
| CN111768637B (en) | Signal intersection traffic signal lamp and vehicle track control method | |
| WO2020135742A1 (en) | Autonomous driving vehicle horizontal decision system and horizontal decision-making method | |
| GB2592461A (en) | Traffic light detection system for vehicle | |
| US20190143992A1 (en) | Self-driving learning apparatus and method using driving experience information | |
| CN110634284A (en) | Association of traffic lights with lanes to which they belong | |
| CN106080590A (en) | Control method for vehicle and device and the acquisition methods of decision model and device | |
| CN106092114A (en) | The automobile real scene navigation apparatus of a kind of image recognition and method | |
| CN108622103A (en) | The scaling method and system of driving Risk Identification model | |
| CN112437501B (en) | Multi-sensor beyond-the-horizon ad hoc network method based on traffic semantics and game theory | |
| CN113112830A (en) | Signal control intersection emptying method and system based on laser radar and track prediction | |
| CN115523934A (en) | A vehicle trajectory prediction method and system based on deep learning | |
| DE102022105579A1 (en) | GRAPH EXPLORATION FOR RULEBOOK TRAJECTORY GENERATION | |
| CN116597690B (en) | Highway test scene generation method, equipment and medium for intelligent network-connected automobile | |
| CN116266380A (en) | Environment data reconstruction method, device, system and storage medium | |
| CN110765224A (en) | Processing method of electronic map, vehicle vision repositioning method and vehicle-mounted equipment | |
| CN115272994A (en) | Automatic driving prediction model training method, device, terminal and medium | |
| CN116842688A (en) | Online compliance verification system oriented to automatic driving decision algorithm | |
| CN113306558A (en) | Lane changing decision method and system based on lane changing interaction intention | |
| CN111103882A (en) | Autonomous following control method for unmanned electric vehicle | |
| CN113156923A (en) | Road data generation, navigation, and driving control method, device, equipment, and medium | |
| CN108447306A (en) | Real time position shares the analogy method of intelligent collision warning between No-shell culture conflict vehicle | |
| CN113176778A (en) | Control method and control device for unmanned vehicle and unmanned vehicle | |
| CN118072553A (en) | Intelligent traffic safety management and control system | |
| CN112810604B (en) | Intelligent vehicle behavior decision method and system based on parking lot scene |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |