WO2021196145A1 - 一种车辆盲区识别方法、自动驾驶辅助系统以及包括该系统的智能驾驶车辆 - Google Patents

一种车辆盲区识别方法、自动驾驶辅助系统以及包括该系统的智能驾驶车辆 Download PDF

Info

Publication number
WO2021196145A1
WO2021196145A1 PCT/CN2020/083075 CN2020083075W WO2021196145A1 WO 2021196145 A1 WO2021196145 A1 WO 2021196145A1 CN 2020083075 W CN2020083075 W CN 2020083075W WO 2021196145 A1 WO2021196145 A1 WO 2021196145A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
driving
obstacle
blind
blind zone
Prior art date
Application number
PCT/CN2020/083075
Other languages
English (en)
French (fr)
Inventor
陈亮亮
刘天放
陈曦
湛鹤峰
Original Assignee
华为技术有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2020/083075 priority Critical patent/WO2021196145A1/zh
Priority to CN202080004420.0A priority patent/CN113348119A/zh
Publication of WO2021196145A1 publication Critical patent/WO2021196145A1/zh

Links

Images

Classifications

    • 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
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/095—Predicting travel path or likelihood of collision
    • 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
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • 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
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/09—Taking automatic action to avoid collision, e.g. braking and steering
    • 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
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/095—Predicting travel path or likelihood of collision
    • B60W30/0956—Predicting travel path or likelihood of collision the prediction being responsive to traffic or environmental parameters
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes

Definitions

  • This application relates to the field of automatic driving, and in particular, to a vehicle blind spot recognition method, an automatic driving assistance system, and an intelligent driving vehicle including the system.
  • assisted driving and autonomous driving technologies are gradually entering the market. More and more advanced assisted driving and autonomous driving technologies also require vehicles to be able to cope with more complex traffic scenarios and dangerous situations, especially in In the complex urban traffic environment, due to dynamic and static traffic vehicles, intersection green belts, walls and other obstacles form obstructions, resulting in the limitation of the detection area of the autonomous driving perception sensor, and for the traffic participants in the blind area to suddenly appear in their own vehicles. Collision accidents are prone to occur in front of the trajectory, especially pedestrians and cyclists who do not follow traffic rules and the trajectory changes, which greatly increases the risk of collision.
  • target tracking and trajectory prediction are carried out for traffic participants. Once a collision risk is found, avoidance control will be adopted. The hidden collision risk is not identified for the current driving scene, and the blind area is blocked. Sudden risks are difficult to solve well, and the automatic emergency braking obstacle avoidance system can only be triggered when traffic participants enter the sensing range for this kind of emergency. It is difficult to avoid collisions when the reaction distance is small.
  • V2X technology uses V2X technology to obtain information from traffic participants to the autopilot system.
  • the autopilot system obtains rich data from the information collected by the sensors of its own vehicle and the information obtained from the V2X technology, which can provide information on the behavior of traffic participants in the current traffic scene. Effective predictions can be made to avoid collisions between vehicles and traffic participants.
  • the realization of current automatic driving technology is difficult to rely on V2X technology. It requires complete road infrastructure and intelligent equipment for vehicles and pedestrians to achieve commercial applications in the short term.
  • various embodiments of the present application provide a vehicle blind spot recognition method, a driving assistance system and a vehicle adopting the system, so as to accurately identify the blind spot of the vehicle driving and make effective safe driving on this basis.
  • various embodiments of the present application provide a vehicle blind spot recognition method, which includes: first obtaining vehicle surrounding environment information, such as road images, dynamic/static target information, traffic light information, etc.; and then judging the vehicle based on the vehicle surrounding environment information
  • Driving scenes such as lane line detection, traffic sign detection, self-car positioning, static/dynamic object detection, etc.; to determine whether the vehicle is currently in an obstacle-free scene or an obstacle scene, in an obstacle scene, at least based on obstacles and obstacles
  • the speed determines the dangerous driving blind zone.
  • the technical solution of the first aspect can confirm whether the scene where the vehicle is located has obstacles or no obstacles. In the case of obstacles, further comprehensively consider the influence of obstacles and obstacle speeds on the blind area.
  • the speed of the obstacle when there is an obstacle, it is further judged whether the speed of the obstacle is zero. If the speed of the obstacle is zero, the area covered by the obstacle is determined to be a dangerous driving blind zone; if the speed of the obstacle is not At zero hour, the area covered by obstacles is the total blind zone, and the length or area swept by the obstacle in the total blind zone with maximum deceleration braking is the safe driving blind zone.
  • the dangerous driving blind zone is determined based on the total blind zone and the safe driving blind zone.
  • the dangerous driving blind area when the vehicle and the obstacle are driving in the same direction, the dangerous driving blind area is the total blind area minus the safe driving blind area; when the vehicle and the obstacle are driving in opposite directions, the dangerous driving blind area is the total blind area plus the safe driving blind area . This distinguishes whether the speed of the vehicle and the obstacle are in the same direction or facing each other, and the dangerous driving blind spot is confirmed accordingly.
  • the risk factor of the scene in which the vehicle is located is determined according to the driving scene of the vehicle.
  • the risk of the blind zone of the vehicle in different scenes is different. This is distinguished in the technical solution; and then according to the said The dangerous driving blind zone and the risk coefficient of the scene in which the vehicle is located determine the risk level of the vehicle blind zone, so that the risk level of the vehicle blind zone can be quantitatively evaluated.
  • the risk level of the vehicle blind area is determined according to parameters such as the dangerous driving blind area, the lateral distance from the vehicle to the blind area, the current speed of the vehicle, the time when the vehicle reaches the conflict area, and the risk factor of the scene in which the vehicle is located. Makes the assessment of the risk level of the blind area of the vehicle more accurate.
  • the corresponding control strategy can be confirmed.
  • the vehicle will give priority to lateral avoidance, which means that the vehicle enters the lane away from the blind area in the lateral direction; and if the risk level of the blind area of the vehicle is lower than the set threshold, the vehicle continues to drive according to the Assist system planning driving.
  • the risk level of the blind zone is high, the technical solution will preferably be selected to make the vehicle away from the blind zone in the lateral direction, so as to avoid the risk of collision to the greatest possible extent.
  • In-track avoidance means that the vehicle moves away from the blind area in the lateral direction in the current lane, so as to avoid the blind area. It is possible to avoid collisions.
  • the vehicle will drive at the minimum speed in the speed planning of the automatic driving assistance system and the safe driving speed planning. .
  • an automatic driving assistance system including: a sensor unit composed of a camera, GPS, radar, etc., the sensor unit is used to obtain information about the surrounding environment of the vehicle; a sensing unit communicatively connected with the sensor unit, and the sensing unit receives the sensor unit
  • the acquired information determines the vehicle driving scene based on the acquired information
  • the decision-making department communicatively connected with the perception department, which determines the dangerous driving blind zone and the risk level of the vehicle blind zone according to the vehicle driving scene
  • the control department communicating with the decision-making department , The control unit determines the driving strategy at least according to the risk level of the blind spot of the vehicle.
  • an intelligent vehicle which includes the automatic driving assistance system of the aforementioned second aspect.
  • This application provides a vehicle blind spot recognition method, a driving assistance system, and a vehicle including the vehicle.
  • the technical solution of this application first confirms whether the scene where the vehicle is located has obstacles or obstacles. In the case of obstacles, it is further integrated Considering the influence of obstacle speed on the blind zone, by introducing the concept of safe driving blind zone and evaluating the risk level of vehicle blind zone, the technical solution can accurately identify the blind zone of vehicle driving and make effective safety on this basis. Driving strategy.
  • the technical solution of the present application can be completed based on the existing driving assistance system without expensive road infrastructure. Therefore, the technical solution of the present application is highly economical, facilitates promotion and popularization, and can improve the safety of automatic driving.
  • FIG. 1 is a schematic diagram of the blind spot recognition and driving control process provided by an embodiment of the present application
  • Fig. 2 is a schematic diagram of providing blind spot recognition according to an embodiment of the present application
  • Fig. 3 is a schematic diagram of blind zone separation and calculation provided by an embodiment of the present application.
  • FIG. 4 is a schematic diagram of a safe driving strategy based on blind spot recognition provided by an embodiment of the present application
  • Fig. 5 is a schematic diagram of an automatic driving assistance system provided by an embodiment of the present application.
  • FIG. 1 shows a schematic flowchart of a method 100 for identifying a vehicle blind spot based on an embodiment of the present application, including:
  • the on-board equipment can include mainstream devices for autonomous vehicles, namely cameras, GPS, and radar.
  • the cameras can include monocular and/or binocular cameras, and the radar can include millimeter waves.
  • Radar and/or Lidar. Cameras are generally used to collect road scene images, lidar and/or millimeter-wave radar are responsible for collecting dynamic and static target data, and GPS is used to locate vehicle positions.
  • the information obtained by the camera and radar is sent to the on-board image processing device or the cloud communicating with the vehicle to determine the driving scene of the vehicle.
  • the driving scene of the vehicle may include, for example, road structure information, vehicle/pedestrian recognition, dynamic and static target recognition, vehicle position, Speed, acceleration, heading; and state observation information of other vehicles and pedestrians around the vehicle.
  • a certain range may be a range of 0-50 meters in front of the vehicle traveling direction. It should be understood that the range may be adjusted according to actual conditions, for example, it may also be 0-75 meters, 10 meters in front of the vehicle traveling direction. The range of -80 meters is not limited in this application.
  • 103 Blind zone classification and calculation. Based on the result of 101, if it is judged that the vehicle is in an obstacle-free scene, it is considered that there is no blind zone for the current vehicle; if it is judged that the vehicle is in an obstacle scene, the blind zone is classified according to the speed of the obstacle . Specifically, if the speed of the obstacle is zero, the blind zone caused by the occlusion of the stationary obstacle is a dangerous driving blind zone; if the speed of the obstacle is not zero, it is caused by the obstacle at the current moment. The blind zone is the total blind zone. The distance (length, area) swept by the obstacle at the maximum deceleration braking is the safe driving blind zone, and the total blind zone minus the safe driving blind zone is the dangerous driving blind zone. It should be pointed out that the speed of obstacles is based on the geodetic coordinate system.
  • the vehicle 21 is traveling along the lane at a speed V0, and there is an obstacle 22 in the front left of the vehicle, that is, the vehicle 21 is currently in an obstacle scene.
  • the obstacle 22 is a stationary obstacle.
  • the blind zone caused by the obstacle 22 is for the vehicle 22. It is a dangerous driving blind zone.
  • the blind area caused by the sight line of the vehicle 21 being blocked by the obstacle 22 is the area ABC from the two-dimensional area, and the section AB (the length is S1) from the one-dimensional length.
  • the area ABC may be defined as a dangerous driving blind spot; in other embodiments, the section AB may be positioned as a dangerous driving blind spot. It should be understood that there is no essential difference in selecting a one-dimensional section or a two-dimensional area as the blind zone, and it only needs to be consistent in one solution.
  • the line of sight of the aforementioned vehicle 21 refers to the line of sight of the on-board camera.
  • the speed of the obstacle 22 is zero, the above-mentioned area ABC or section AB is "invisible" to the vehicle 21. If an object 23 with a lateral speed V3 appears within the range of the area ABC or section AB ( For example, it may be a person or a vehicle). At this time, the object 23 is invisible to the vehicle 21 and has a potential collision risk. Therefore, the area ABC or the section AB is a dangerous driving blind area for the vehicle 21.
  • S1 (L0-L1)*tan ⁇ .
  • the speed V1 of the obstacle 22 is not zero, refer to FIG. 2. If the obstacle 22 is also a vehicle and its speed is V1, the angle between V1 and the speed V0 of the vehicle 21 is ⁇ . In this case, consider the distance (area) over which the obstacle 22 brakes at the maximum deceleration. Assuming that the deceleration of the obstacle 22 is a, and the direction of a is consistent with the speed V1, then a in the V0 direction is a*cos ⁇ . Based on the kinematics principle, it can be known that the obstacle 22 passes in the V0 direction from moving to stopping. The distance of is (V1*cos ⁇ ) 2 /(2a*cos ⁇ ). In Fig. 2 the distance is identified by S3, and if expressed by area, it is the area DECB.
  • This application distinguishes the safe driving blind zone and the dangerous driving blind zone in this case, and defines the section AB (S1) or the area DECB as the total blind area, and the section DB (S3) or the area DECB as the safe driving blind area.
  • the meaning is: if there is an object 24 (vehicle or pedestrian) in the DB (S3) or area DECB, which has a lateral speed of V4, then the object 24 may collide with the obstacle (vehicle) 22 first, and the vehicle 22 is the largest
  • the distance/area swept by the deceleration brake is therefore a “safe area” for the vehicle 21, because within this distance/area, if the object 24 collides with the vehicle 22, it should first occur.
  • the section S4 or the area ADE is a dangerous driving blind zone for the vehicle 21.
  • the maximum deceleration of the vehicle can take a value in the range of 8-12 m/S 2.
  • the system delay time T delay needs to be considered when calculating the obstacle 22 from moving to stationary.
  • the distance (S3) that the obstacle passes in the V0 direction from moving to stopping is:
  • the system delay time T delay may take a value in the range of 0.1-1 second.
  • the aforementioned driving in the same direction or opposite direction does not mean that the speed directions of the vehicle 21 and the obstacle 22 are completely the same and parallel, or completely opposite and parallel; in general, if the vehicle 21 If the cosine value cos ⁇ of the angle ⁇ between the vehicle speed of the obstacle and the obstacle is greater than zero, the vehicle 21 and the obstacle 22 are considered to be traveling in the same direction, and if the cosine value cos ⁇ is less than zero, the vehicle 21 and the obstacle 22 are considered to be traveling in opposite directions.
  • the total blind zone is smaller than the safe driving blind zone, that is It can be considered that the total blind zone is a safe driving blind zone for the vehicle 21 at this time.
  • step 34 Determine whether the obstacle is stationary, if it is determined that the obstacle is stationary, go to step 34, directly determine that the blind area caused by the obstacle is a dangerous driving blind area; if it is determined that the obstacle is in motion, go to step 33;
  • step 33 Calculate the blind area caused by obstacles and record it as the total blind area, and then go to step 35;
  • step 35 Calculate the safe driving blind zone, that is, the length (area) swept by the obstacle at the maximum deceleration braking, and then go to step 36;
  • step 36 Determine whether the safe driving blind zone is greater than the total blind zone; if the safe driving blind zone is greater than the total blind zone, directly determine the safe driving blind zone in step 38; if the safe driving blind zone is less than the total blind zone, go to step 37;
  • the current scene of the vehicle can be determined according to the driving scene of the vehicle, and the risk level r of the scene can be given based on the big data analysis according to the scene of the vehicle.
  • the current scene of the vehicle can be divided into: urban roads, highways, and parking lot roads.
  • different risk factors r can be given, for example:
  • the risk coefficient r of urban roads is 0.4-0.6;
  • the high-speed road risk coefficient r is 0.2-0.8;
  • the risk coefficient r of the parking lot road is 0.1-0.6;
  • Different subdivided scenes correspond to different risk factors r.
  • the risk coefficient r in the scenario can be defined as 0.6; and if the urban road scene is an urban elevated, the probability of accidents will be relatively low, and the risk coefficient r in this scenario can be defined as 0.4.
  • the risk factor can be defined as 0.2 in normal expressway scenes, and if it is when the expressway merges and meets, due to the intersection of vehicles There will be more lane changing actions, and the speed of the vehicle on the highway is faster, so the risk coefficient r can be defined as 0.8 in this scenario.
  • scenario and risk coefficient r are only exemplary rather than limiting, and those skilled in the art can determine the scenario and the risk coefficient r corresponding to the scenario based on the actual situation.
  • the risk level of the blind area of the vehicle can be determined. Specifically, the risk level of the blind area of the vehicle can be determined according to the following parameters:
  • Dangerous driving blind zone B can be the dangerous driving blind zone determined in step 103, and B can take a one-dimensional value or a two-dimensional value;
  • the lateral distance D from the vehicle to the blind zone; D can be the value from the center of the vehicle to the edge of the obstacle.
  • D can be L1;
  • TTC Time to Collision
  • TTC can be the time when the vehicle 21 reaches the lower edge of the dangerous driving blind zone, that is, the time required for the vehicle to travel at the current speed V to DF .
  • the risk factor r of the scene where the vehicle is currently located
  • a function with parameters can be expressed as:
  • Vehicle blind zone risk level R f (B, D, TTC, V, r)
  • the expression of the risk level of the vehicle blind zone may be:
  • e B * can be used.
  • r is used to calculate the risk level of the vehicle blind zone, where e is the base of the natural logarithm.
  • the driving strategy can be determined according to the risk R level of the blind spot of the vehicle.
  • ADAS automatic driving assistance system
  • Fig. 4 shows the process of determining the driving strategy based on the risk level of the blind area of the vehicle, including:
  • the set threshold can be set to 0.5. It should be understood that the threshold can be adjusted up or down according to actual needs; if R is greater than the set threshold, go to step 43;
  • Step 43 Determine whether the vehicle can change lanes and avoid the lane.
  • the so-called lane-change avoidance here refers to the vehicle moving away from the blind zone in the lateral direction to avoid possible collision risks.
  • lane-change avoidance refers to the vehicle moving away from the blind zone in the lateral direction to avoid possible collision risks.
  • Step 44 It should be understood that ADAS is not allowed to change lanes and avoidance is determined by ADAS based on the actual situation of the vehicle. For example, when there are too many vehicles in the lane next to the current vehicle that cannot change lanes, ADAS will not allow it. Judgment of changing lanes and avoiding;
  • Step 44 Determine whether to allow in-street avoidance, that is, consider whether possible collisions can be avoided without changing lanes.
  • the dangerous driving blind zone such as ADE or S4
  • the vehicle 21 can only avoid the current lane without changing lanes, and the vehicle 21 can move laterally away from the blind area in the current lane to realize the avoidance within the lane; After success, continue to return to step 41 to obtain the vehicle blind zone risk level R at the next moment; if it is not possible to avoid in-street, proceed to step 46;
  • step 46 Carry out safe driving speed planning to make the risk factor within the threshold. Generally speaking, it means to reduce the speed of the vehicle below a certain value.
  • step 45 namely the speed planning of the automatic driving assistance system, takes the minimum safe speed (step 48) As the planning speed, it is given to the execution system for execution.
  • FIG. 5 exemplarily illustrates an automatic driving assistance system 50, including:
  • the sensor part 51 may include commonly used vehicle sensors, such as cameras, GPS, radar, etc.
  • the cameras may include monocular and/or binocular cameras, and the radar may include Lidar or millimeter wave radar; the camera is responsible for collecting roads Scene images, lidar and millimeter-wave radar are responsible for collecting dynamic and static target data, and GPS is used to obtain the current position of the vehicle;
  • the sensing unit 52 which receives the information collected by the sensor unit and determines the driving scene of the vehicle based on the information.
  • the driving scene of the vehicle may include, for example (but not limited to) lane line detection, traffic sign detection, self-vehicle positioning, static/dynamic object detection Wait;
  • the decision-making unit 53 receives the vehicle driving scene information from the perception unit and makes corresponding decisions, such as (but not limited to) confirming the dangerous driving blind zone and the risk level of the vehicle blind zone.
  • corresponding decisions such as (but not limited to) confirming the dangerous driving blind zone and the risk level of the vehicle blind zone.
  • For the specific confirmation method please refer to the description of the above embodiments. Make corresponding control strategies for the risk level of vehicle blind spots;
  • the control unit 54 controls the vehicle according to the control strategy of the decision unit 53, including (but not limited to) the various lateral control and longitudinal control described in the foregoing embodiments.
  • Some embodiments also provide an intelligent driving vehicle, which includes the above-mentioned automatic driving assistance system 50, so that the intelligent driving vehicle can implement various solutions of the above-mentioned embodiments.
  • Various embodiments of the present application provide a method for identifying a blind spot of a vehicle, a driving assistance system, and a vehicle including the vehicle.
  • the technical solution of the present application first confirms whether the scene where the vehicle is located has obstacles or obstacles. In the case of obstacles, Next, further comprehensively consider the impact of obstacle speed on blind spots, introduce the concept of safe driving blind spots and evaluate the risk level of vehicle blind spots, so that the technical solution can accurately identify the blind spots of vehicle driving, and do it on this basis. Develop effective safe driving strategies.
  • the technical solution of the present application can be completed based on the existing driving assistance system, disorderly and expensive road infrastructure, so the technical solution of the present application is highly economical, facilitates promotion and popularization, and can improve the safety of autonomous driving .
  • the disclosed system, device, and method can be implemented in other ways.
  • the device embodiments described above are merely illustrative, for example, the division of units is only a logical business division, and there may be other divisions in actual implementation, for example, multiple units or components can be combined or integrated. To another system, or some features can be ignored, or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • business units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit can be realized in the form of hardware or software business unit.
  • the integrated unit is implemented in the form of a software business unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of the present application essentially or the part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , Including several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods in the various embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks and other media that can store program codes. .
  • the services described in this application can be implemented by hardware, software, firmware, or any combination thereof.
  • these services can be stored in a computer-readable medium or transmitted as one or more instructions or codes on the computer-readable medium.
  • the computer-readable medium includes a computer storage medium and a communication medium, where the communication medium includes any medium that facilitates the transfer of a computer program from one place to another.
  • the storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer.

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种车辆盲区识别方法（100）、自动驾驶辅助系统（50）以及包括该自动驾驶辅助系统（50）的智能驾驶车辆（21），该车辆盲区识别方法（100）包括：首先获取车辆（21）周围环境信息，然后根据车辆（21）周围环境信息判断车辆（21）行驶场景，判断车辆（21）当前是处于无障碍物场景还是有障碍物场景，在有障碍物场景下，至少基于障碍物（22）和障碍物（22）的速度确定危险驾驶盲区。

Description

一种车辆盲区识别方法、自动驾驶辅助系统以及包括该系统的智能驾驶车辆 技术领域
本申请涉及自动驾驶领域，特别地，涉及一种车辆盲区识别方法、自动驾驶辅助系统以及包括该系统的智能驾驶车辆。
背景技术
随着人工智能技术结合汽车应用的快速发展，辅助驾驶和自动驾驶技术渐渐走向市场，越来越先进的辅助驾驶和自动驾驶技术也要求车辆能够应对更复杂的交通场景和危险情况，尤其是在复杂的城市交通环境中，由于动态、静态交通车辆，路口绿化带、墙壁等障碍物形成遮挡，导致自动驾驶感知传感器的探测区域受到限制，对于感知盲区内的交通参与者突然出现在自车行驶轨迹前方极易出现碰撞事故，特别是行人和骑行者易发生不遵守交通规则的行为并且轨迹变化不定，碰撞风险大大增加。
现有技术中应对碰撞风险主要有以下的方案：
1.基于感知传感器探测到的已知信息，对交通参与者进行目标跟踪和轨迹预测，一旦发现有碰撞风险会采取避让控制，并没有针对当前行驶场景识别出隐藏的碰撞风险，对于盲区遮挡区域突发的风险难以很好解决，只能在交通参与者进入感知范围时对于这种紧急情况触发自动紧急制动避障系统，在反应距离较小时难以避免碰撞。
2.利用V2X技术获取交通参与者的信息给到自动驾驶系统，自动驾驶系统通过自车的传感器采集的信息与来自V2X技术获取的信息得到丰富的数据可以对当前交通场景的交通参与者的行为进行有效预测，从而避免车辆与交通参与者的碰撞，但当前自动驾驶技术的实现难以依赖V2X技术，要求完备的道路基础设施及车辆、行人配备智能设备难以在短期内实现商业化应用。
基于以上，需要一种方案，其可以准确地识别车辆在行驶过程中的盲区，并基于盲区做出自动驾驶策略以有效规避可能的碰撞风险；并且，该方案应当可以基于车辆自身就可以实现。
发明内容
为了解决上述问题，本申请各种实施例提供了一种车辆盲区识别方法、驾驶辅助系统和采用该系统的车辆，以实现精确地识别车辆驾驶盲区，并在此基础上做出有效的安全驾驶策略。
第一方面，本申请各种实施例提供一种车辆盲区识别方法，包括：首先获取车辆周围环境信息，例如道路图像、动/静态目标信息、交通灯信息等；然后根据车辆周围环境信息判断车辆行驶场景，例如车道线检测、交通标识检测、自车定位、静态/动态对象检测等；判断车辆当前是处于无障碍场景还是有障碍场景，在有障碍物场景下，至少基于障碍物和障碍物的 速度确定危险驾驶盲区。
第一方面的技术方案可以确认车辆所处的场景是有障碍物还是无障碍物，在有障碍物的情况下，进一步综合地考虑障碍物和障碍物的速度对盲区所产生的影响。
在一个可能的设计中，在有障碍物的时候，进一步判断障碍物的速度是否为零，如果障碍物速度为零时，确定障碍物产生遮挡的区域为危险驾驶盲区；如果障碍物速度不为零时，障碍物产生遮挡的区域为总盲区，障碍物以最大减速度制动在总盲区中扫过的长度或面积为安全驾驶盲区，依据总盲区和安全驾驶盲区确定危险驾驶盲区。通过引入安全驾驶盲区，从而使得对方案对盲区的确认更加精确。
在一个可能的设计中，当车辆和障碍物同向行驶时，危险驾驶盲区为总盲区减去安全驾驶盲区；当车辆和障碍物对向行驶时，危险驾驶盲区为总盲区加上安全驾驶盲区。这里区分了车辆在和障碍物的速度是同向还是对向，并依此确认危险驾驶盲区。
在一个可能的设计中，依据所述车辆行驶场景确定车辆所处场景的风险系数，车辆在不同的场景下出现盲区所带来的风险不同，技术方案中对此做了区分；然后依据所述危险驾驶盲区和所述车辆所处场景的风险系数确定车辆盲区风险等级，从而可以定量化地评估车辆盲区风险等级。
在一个可能的设计中，综合考虑各种因素，依据危险驾驶盲区、车辆距离盲区横向距离、车辆当前车速、车辆到达冲突区域的时间、车辆所处场景的风险系数等参数确定车辆盲区风险等级，使得对于车辆盲区风险等级的评估更加准确。
在一个可能的设计中，在确定了车辆盲区风险等级后，即可确认对应的控制策略。当车辆盲区风险等级高于设定阈值时，车辆优先进行横向避让，横向避让指车辆在横向方向上进入远离盲区的车道行驶；而如果车辆盲区风险等级低于设定阈值时，车辆继续按照驾驶辅助系统的规划行驶。如果盲区风险等级高，技术方案将优选选择使得车辆在横向上远离盲区，从而最大可能地避免碰撞风险。
在一个可能的设计中，如果所述车辆盲区风险等级高于设定阈值，且车辆无法进行横向避让时，车辆进行道内避让，道内避让指车辆在当前车道沿横向方向上远离盲区移动，从而尽可能地避免发生碰撞。
在一个可能的设计中，如果所述车辆盲区风险等级高于设定阈值，且车辆无法进行横向避让和道内避让时，车辆以自动驾驶辅助系统速度规划和安全行驶速度规划中的速度最小值行驶。
第二方面，提供一种自动驾驶辅助系统，包括：由相机、GPS、雷达等构成的传感器部，传感器部用于获取车辆周围环境信息；和传感器部通信连接的感知部，感知部接收传感器部获取的信息并依据所获取的信息确定车辆行驶场景；和感知部通信连接的决策部，决策部依据所述车辆行驶场景确定危险驾驶盲区和车辆盲区风险等级；以及与决策部通信连接的控制部，所述控制部至少依据所述车辆盲区风险等级确定驾驶策略。
可以理解的是，第二方面提供的系统对应于前述第一方面提供的方法，故第二方面各实现方式以及达到的技术效果可参见前述第一方面各实现方式的相关描述。
第三方面，提供一种智能车辆，其包括前述第二方面的自动驾驶辅助系统。
本申请提供了一种车辆盲区识别方法、驾驶辅助系统以及包括该系辆，本申请技术方案首先确认车辆所处的场景是有障碍物还是无障碍物，在有障碍物的情况下，进一步综合地考 虑障碍物的速度对盲区所产生的影响，通过引入安全驾驶盲区的概念以及对车辆盲区风险等级进行评估，使得技术方案可以精确地识别车辆驾驶盲区，并在此基础上做出有效的安全驾驶策略。另一方面，本申请的技术方案基于现有的驾驶辅助系统可完成，无需昂贵的道路基础设施，因此本申请的技术方案经济性高，利于推广和普及，并可以提升自动驾驶的安全性。
附图说明
图1是本申请实施例提供的盲区识别和驾驶控制流程示意图；
图2是本申请实施例提供盲区识别的示意图；
图3是本申请实施例提供的盲区分离及计算示意图；
图4是本申请实施例提供的基于盲区识别的安全驾驶策略示意图；
图5是本申请实施例提供的一种自动驾驶辅助系统示意图。
具体实施方式
参见图1，其示出了基于本申请实施例给出的一种车辆盲区识别方法100的流程示意图，包括：
101：开始；
102：确定车辆行驶场景，此步骤可由车载设备所实现，车载设备可以包括自动驾驶车辆的主流设备，即相机、GPS、雷达，相机可以包括单目和/或双目相机，雷达可以包括毫米波雷达和/或激光雷达(Lidar)。相机一般用于采集道路场景图像，激光雷达和/或毫米波雷达负责采集动、静态目标数据，GPS用于定位车辆位置。相机和雷达获取的信息被输送给车载图像处理装置或者和车辆通信的云端以确定车辆行驶场景，车辆行驶场景可以包括例如：道路结构信息、车辆/行人识别、动静态目标识别、自车位置、速度、加速度、航向；以及自车周围其他车辆及行人的状态观测信息等。
在一些实施例中，基于车辆行驶场景可以判断车辆当前处于有障碍物场景还是无障碍物场景。通过车辆行驶场景信息判断当前车辆在行驶方向的一定范围内如果不存在障碍物则认为车辆当前处于无障碍物场景，否则车辆当前处于有障碍物场景。在一些实施例中，一定范围可以是车辆行驶方向前方0-50米的范围，应当理解的是，该范围可以依据实际情况而进行调整，例如也可以是车辆行驶方向前方0-75米、10-80米的范围，本申请对此不做限定。
103：盲区分类和计算，基于101的结果，如果判断车辆处于无障碍物场景，则认为对当前车辆而言不存在盲区；如果判断车辆处于有障碍物场景，则依据障碍物的速度进行盲区分类。具体而言，如果障碍物的速度为零，则由静止的障碍物产生遮挡所造成的盲区即属于为危险驾驶盲区；如果障碍物的速度不为零，则由障碍物当前时刻遮挡所造成的盲区为总盲区，障碍物以最大减速度制动所掠过的距离(长度、面积)为安全驾驶盲区，总盲区减去安全驾驶盲区为危险驾驶盲区。需要指出的是：障碍物的速度是基于大地坐标系的。
参见图2以示例性地说明盲区分类和计算，在图2中，车辆21沿车道以速度V0行驶，在自车的左前方有障碍物22，即车辆21当前处于有障碍物场景。
如果障碍物22的速度V1为0，即障碍物22相对于大地坐标系静止，则障碍物22为静止障碍物，在此情况下，由障碍物22产生遮拦所造成的盲区对于车辆22而言属于危险驾驶盲区。参见图2，此情况下车辆21的视线被障碍物22遮拦所造成的盲区从二维面积上而言 为区域ABC，从一维长度上而言为区段AB(长度为S1)。在一些实施例中，可以将区域ABC定义为危险驾驶盲区；在另外一些实施例中，可以将区段AB定位为危险驾驶盲区。应当理解的是，选择一维的区段或者二维的区域作为盲区并无本质的区别，只需在一种方案中保持一致即可。
还应当理解的是，上述车辆21的视线指车载相机的视线。当障碍物22的速度为零的时候，上述区域ABC或者区段AB对于车辆21而言是“不可见”的，如果在区域ABC或者区段AB范围内出现了具有横向速度V3的对象23(例如可以是人或者车辆)，那么此时对象23对于车辆21而言是不可见的并具有潜在的碰撞危险，因此区域ABC或区段AB对于车辆21为危险驾驶盲区。
可以使用多种方法来计算S1的值，例如，可以通过激光雷达测量车辆21和障碍物22以及道路边缘的距离(L1和L0)以及车辆视线和道路边缘的夹角α来确定S1，此种情况下，S1＝(L0-L1)*tanα。
应当指出的是，如果车辆21在障碍物22处于静止，对于车辆21而言，危险驾驶盲区的范围是随着时间而变的，因为车辆21在运动状态，其视线也是在变化的，因此区域ABC或者区段AB也是随时间而变的。
如果障碍物22的速度V1不为零，参见图2，如果障碍物22也为车辆，并且其速度为V1，V1和车辆21的速度V0之间的夹角为θ。在此情况下，考虑障碍物22以最大减速度刹车所掠过的距离(面积)。假设障碍物22的减速度为a，a的方向和速度V1相一致，则在V0方向上的a为a*cosθ,基于运动学原理可知，障碍物22从运动到停止在V0方向上所通过的距离为(V1*cosθ) 2/(2a*cosθ)。在图2中该距离以S3标识，而如果以面积表示的话为区域DECB。
本申请区分了此种情况下的安全驾驶盲区和危险驾驶盲区，定义了区段AB(S1)或区域DECB为总盲区，而区段DB(S3)或区域DECB为安全驾驶盲区。其含义在于：如果在DB(S3)或区域DECB内有一个对象24(车辆或行人)，其具有横向速度V4，则对象24最先可能与障碍物(车辆)22发生碰撞，车辆22以最大减速度刹车所掠过的距离/面积因此对于车辆21而言是“安全区域”，因为在此距离/面积内如果对象24发生了碰撞则应当是首先和车辆22所发生的。进一步地，区段S4或者区域ADE对于车辆21而言是危险驾驶盲区。
在一些实施例中，车辆的最大减速度可以取8-12m/S 2中的值。
在一些实施例中，在计算障碍物22从运动到静止的时候还需要考虑系统延迟时间T delay，此种情况下，障碍物从运动到停止在V0方向上所通过的距离(S3)为：
V1*cosθ*T delay+(V1*cosθ) 2/(2a*cosθ)。
在一些实施例中，系统延迟时间T delay可以取0.1-1秒范围内的值。
当车辆21和车辆22同向行驶时，可以用以下公式表示一维情况下的危险驾驶盲区：
危险驾驶盲区＝S1-S3；
而当车辆21和车辆22对向行驶时，可以用以下公式表示一维情况下的危险驾驶盲区：
危险驾驶盲区＝S1+S3。
应当理解的是，以图2为例，上述的同向或对向行驶并不意味着车辆21和障碍物22的速度方向完全相同和平行，或者完全相对和平行；一般而言，如果车辆21和障碍物的车辆速度的夹角θ的余弦值cosθ大于零则认为车辆21和障碍物22同向行驶，而如果余弦值cosθ 小于零则认为车辆21和障碍物22对向行驶。
在一些特殊的情况下，参见图2，如果障碍物(车辆)22以最大加速度刹车时所经过的距离S5大于S1，则此种情况下对于车辆21而言，总盲区小于安全驾驶盲区，即可认为总盲区此时对于车辆21而言均为安全驾驶盲区。
参见图3，其示出了盲区分类和计算的流程，包括：
31：确认当前为有障碍物场景；
32：判断障碍物是否静止，如果判断障碍物静止，则进入步骤34，直接确定由障碍物遮拦所造成的盲区为危险驾驶盲区；如果判断障碍物处于运动状态，进入步骤33；
33：计算由障碍物遮拦所造成的盲区，记为总盲区，然后进入步骤35；
35：计算安全驾驶盲区，即障碍物以最大减速度刹车所掠过的长度(面积)，然后进入步骤36；
36：判断安全驾驶盲区是否大于总盲区；如果安全驾驶盲区大于总盲区，则直接在步骤38确定安全驾驶盲区；如果安全驾驶盲区小于总盲区，进入步骤37；
37：用总盲区减去安全驾驶盲区，获得危险驾驶盲区，如果车辆和障碍物对向行驶，则以总盲区加上安全驾驶盲区获得危险驾驶盲区。
104：确定车辆盲区风险等级；根据车辆行驶场景可以确定车辆当前所处的场景，依据车辆所处的场景，基于大数据分析可以给出场景的风险等级r。
在一些实施例中，可以将车辆当前所处的场景分为：城市道路、高速道路、停车场道路。对于上述三种场景，可以给出不同的风险系数r，例如：
城市道路    风险系数r为0.4-0.6；
高速道路    风险系数r为0.2-0.8；
停车场道路  风险系数r为0.1-0.6；
对于上述三种场景，还可以继续细分场景，不同的细分场景对应于不同的风险系数r，例如：对于城市道路，可以包括路口场景，由于路口场景下车辆交汇，容易发生事故，所以此场景下的风险系数r可以定义为0.6；而如果城市道路的场景是城市高架，那么发生事故的几率会比较低，此场景下的风险系数r可以定义为0.4。在高速道路中，由于高速道路上一般不会发生有行人横向通过道路的事件，所以正常的高速道路场景可以将风险系数定义为0.2，而如果是在高速道路发生合流交汇的时候，由于车辆交汇会有较多的变道动作发生，加上车辆在高速道路上车速较快，所以此种场景下可以将风险系数r定义为0.8。
应当理解的是，上述的场景和风险系数r仅是示例性的而非限定，本领域技术人员可以基于实际情况来确定场景和与场景对应的风险系数r。
在确定了场景的风险等级后，可以确定车辆盲区风险等级，具体而言，可以依据如下参数来确定车辆盲区风险等级：
危险驾驶盲区B；B可以为步骤103所确定的危险驾驶盲区，B可以取一维的值，也可以取二维的值；
车辆距离盲区横向距离D；D可以取车辆中心至障碍物边缘的值，在图2的示例中，D可以取L1；
车辆当前行驶车速V；
车辆到达可能发生碰撞区域的时间TTC(Time to Collision)；在图2的示例中，TTC可 以取车辆21到达危险驾驶盲区下缘的时间，即车辆以当前行驶车速V行驶到DF所需的时间。
车辆当前所处场景的风险系数r；
以带参数的函数可以表示为：
车辆盲区风险等级R＝f(B,D,TTC,V,r)
在一些实施例中，车辆盲区风险等级的表达式可以为：
Figure PCTCN2020083075-appb-000001
在一些实施例中，可以采用相对简略的方式来计算车辆盲区风险等级，即仅利用危险驾驶盲区和所述车辆所处场景的风险系数确定车辆盲区风险等级，此时车辆盲区风险等级＝B*r。
应当理解的是，上述的车辆盲区风险等级的函数表达仅仅是示例性的，本领域技术人员可以依据实际需求而采用任何合适的函数表达而不背离本申请的精神，例如，可以使用e B*r来计算车辆盲区风险等级，其中，e为自然对数的底。
105：确定驾驶策略；在104确定了车辆盲区风险等级后，即可依据车辆盲区风险R等级确定驾驶策略。
首先判断当前车辆盲区风险等级R是否小于安全阈值，若满足则仅需根据自动驾驶辅助系统(Advanced Driving Assist System:简称ADAS)的规划行驶，即步骤47；若不满足，首先进行横向规划避让，判断当前车辆能否进行换道避让，换道避让成功后继续回到风险系数计算步；若不能进行换道避让判断车辆是否能够进行道内避让，若道内避让成功继续回到风险系数计算步；若不能进行道内避障则继续进行安全行驶速度规划使得风险系数达到安全阈值以内，结合安全行驶速度规划和自动驾驶辅助系统速度规划，取速度最小值作为规划速度给到执行系统执行。
具体地，参见图4，其示出了基于车辆盲区风险等级来确定驾驶策略的流程，包括：
41：获取当前车辆盲区风险等级R；
42：判断车辆盲区风险等级R是否小于设定的阈值，如果小于设定的阈值，则表明当前的车辆盲区风险等级较小，因此此时可以按照自动驾驶辅助系统(Advanced Driving Assist System:简称ADAS)的规划进行行驶；在一些实施例中，可以将设定的阈值定为0.5，应当理解的是，可以依据实际需求调高或降低阈值；如果R大于设定的阈值，则进入步骤43；
43：判断车辆是否可以进行换道避让，这里所谓的换道避让指的是车辆在横向方向上远离盲区，以避免可能的碰撞风险，以图2为例，如果车辆21进行换道避让，则车辆将向其右侧的车道切入以实现在横向方向上远离盲区；当换道避让成功后继续返回到步骤41以获取下一时刻的车辆盲区风险等级R；如果不允许换道避让，则进入步骤44；应当理解的是，不允许进行换道避让是由ADAS基于车辆实际情况所判断的，例如在当前车辆旁边的车道上的车辆较多无法进行换道的时候，ADAS会做出不允许换道避让的判断；
44：判断是否允许进行道内避让，即考虑在不进行换道避让的情况下是否可以避免可能发生的碰撞，参见图2，如果危险驾驶盲区(例如ADE或者S4)的范围较小，则代表盲区内有车辆或者行人的几率也较小，此情况下车辆21可以不进行换道避让而仅在当前车道内进行避让，车辆21可以在当前车道内横向地远离盲区以实现道内避让；当道内避让成功后继续返回到步骤41以获取下一时刻的车辆盲区风险等级R；如果不能进行道内避让，则进入步骤46；
46：进行安全行驶速度规划使得风险系数达到阈值以内，通常而言指将车速降低至某一 值以下，同时还考虑引入步骤45，即自动驾驶辅助系统速度规划，取最小安全速度(步骤48)作为规划速度给到执行系统执行。
参见图5，其示例性地示意出了一种自动驾驶辅助系统50，包括：
传感器部51，传感器部可以包括常用的车载传感器，例如相机、GPS、雷达等，相机可以包括单目和/或双目相机，雷达可以包括激光雷达(Lidar)或毫米波雷达；相机负责采集道路场景图像，激光雷达和毫米波雷达负责采集动、静态目标数据，GPS用于获取车辆当前的位置；
感知部52，感知部接收传感器部收集的信息并依据这些信息来确定车辆行驶场景，车辆行驶场景可以包括例如(但不限于)车道线检测、交通标识检测、自车定位、静态/动态对象检测等；
决策部53，决策部接收感知部的车辆行驶场景信息并作出相应决策，例如(但不限于)确认危险驾驶盲区和车辆盲区风险等级，具体的确认方法可以参见上述各实施例的描述，拱依据车辆盲区风险等级做出相应的控制策略；
控制部54，依据决策部53的控制策略对车辆进行控制，包括(但不限于)上述实施例所描述的各种横向控制和纵向控制。
一些实施例还提供了一种智能驾驶车辆，其包括有上述的自动驾驶辅助系统50，从而使得智能驾驶车辆可以实施上述实施例的各种方案。
本申请各种实施例提供了一种车辆盲区识别方法、驾驶辅助系统以及包括该系辆，本申请技术方案首先确认车辆所处的场景是有障碍物还是无障碍物，在有障碍物的情况下，进一步综合地考虑障碍物的速度对盲区所产生的影响，通过引入安全驾驶盲区的概念以及对车辆盲区风险等级进行评估，使得技术方案可以精确地识别车辆驾驶盲区，并在此基础上做出有效的安全驾驶策略。另一方面，本申请的技术方案基于现有的驾驶辅助系统可完成，无序昂贵的道路基础设施，因此本申请的技术方案经济性高，利于推广和普及，并可以提升自动驾驶的安全性。
本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”、“第三”、“第四”等(如果存在)是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的实施例能够以除了在这里图示或描述的内容以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统，装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统，装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，单元的划分，仅仅为一种逻辑业务划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各业务单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件业务单元的形式实现。
集成的单元如果以软件业务单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
本领域技术人员应该可以意识到，在上述一个或多个示例中，本申请所描述的业务可以用硬件、软件、固件或它们的任意组合来实现。当使用软件实现时，可以将这些业务存储在计算机可读介质中或者作为计算机可读介质上的一个或多个指令或代码进行传输。计算机可读介质包括计算机存储介质和通信介质，其中通信介质包括便于从一个地方向另一个地方传送计算机程序的任何介质。存储介质可以是通用或专用计算机能够存取的任何可用介质。
以上的具体实施方式，对本申请的目的、技术方案和有益效果进行了进一步详细说明，所应理解的是，以上仅为本申请的具体实施方式而已。
以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的范围。

Claims (18)

  1. 一种车辆盲区识别方法，包括：
    获取车辆周围环境信息；
    至少根据所述车辆周围环境信息判断车辆行驶场景，所述车辆行驶场景包括无障碍物场景和有障碍物场景；
    在有障碍物场景下，至少基于障碍物和障碍物的速度确定危险驾驶盲区。
  2. 根据权利要求1所述方法，所述在有障碍物场景下，至少基于障碍物和障碍物的速度确定危险驾驶盲区，包括：
    当障碍物速度为零时，确定障碍物产生遮挡的区域为危险驾驶盲区；
    当障碍物速度不为零时，确定障碍物产生遮挡的区域为总盲区，确定障碍物以最大减速度制动在总盲区中扫过的长度或面积为安全驾驶盲区，依据总盲区和安全驾驶盲区确定所述危险驾驶盲区。
  3. 根据权利要求2所述方法，所述依据总盲区和安全驾驶盲区确定所述危险驾驶盲区，包括：
    当车辆和障碍物同向行驶时，危险驾驶盲区为总盲区减去安全驾驶盲区；
    当车辆和障碍物对向行驶时，危险驾驶盲区为总盲区加上安全驾驶盲区。
  4. 根据权利要求1-3任一所述方法，还包括：
    依据所述车辆行驶场景确定车辆所处场景的风险系数；
    至少依据所述危险驾驶盲区和所述车辆所处场景的风险系数确定车辆盲区风险等级。
  5. 根据权利要求1-4任一所述方法，所述至少依据所述危险驾驶盲区和所述车辆所处场景的风险系数确定车辆盲区风险等级，包括：
    依据危险驾驶盲区、车辆距离盲区横向距离、车辆当前车速、车辆到达冲突区域的时间和车辆所处场景的风险系数确定车辆盲区风险等级。
  6. 根据权利要求5所述方法，还包括：
    当所述车辆盲区风险等级高于设定阈值时，指示车辆优先进行横向避让，所述横向避让指车辆在横向方向上进入远离盲区的车道行驶；
    当所述车辆盲区风险等级低于设定阈值时，指示车辆继续按照驾驶辅助系统的规划行驶。
  7. 根据权利要求6所述方法，还包括：
    当所述车辆盲区风险等级高于设定阈值，且车辆无法进行横向避让时，指示车辆进行道内避让，所述道内避让指车辆在当前车道沿横向方向上远离盲区移动。
  8. 根据权利要求7所述方法，还包括：
    当所述车辆盲区风险等级高于设定阈值，且车辆无法进行横向避让和道内避让时，车辆以自动驾驶辅助系统速度规划和安全行驶速度规划中的速度最小值行驶。
  9. 一种自动驾驶辅助系统，包括：
    传感器部，所述传感器部包括：相机、GPS、雷达，所述传感器部用于获取车辆周围环境信息；
    感知部，所述感知部和所述传感器部通信连接，所述感知部接收传感器部获取的信息并依据所获取的信息确定车辆行驶场景；
    决策部，所述决策部与所述感知部通信连接，所述决策部依据所述车辆行驶场景确定危险驾驶盲区和车辆盲区风险等级；
    车辆控制部，所述车辆控制部与所述决策部通信连接，所述车辆控制部至少依据所述车辆盲区风险等级确定驾驶策略。
  10. 根据权利要求9所述系统，所述在有障碍物场景下，至少基于障碍物和障碍物的速度确定危险驾驶盲区，包括：：
    所述车辆行驶场景包括无障碍物场景和有障碍物场景；在有障碍物场景下，至少基于车辆和障碍物的速度确定危险驾驶盲区。
  11. 根据权利要求10所述系统，其中：
    所述有障碍物场景包括障碍物速度为零和不为零；
    当障碍物速度为零时，确定障碍物产生遮挡的区域为危险驾驶盲区；
    当障碍物速度不为零时，障碍物产生遮挡的区域为总盲区，障碍物以最大减速度制动在总盲区中扫过的长度或面积为安全驾驶盲区，依据总盲区和安全驾驶盲区确定危险驾驶盲区。
  12. 根据权利要求11所述系统，所述依据总盲区和安全驾驶盲区确定所述危险驾驶盲区，包括：
    当车辆和障碍物同向行驶时，危险驾驶盲区为总盲区减去安全驾驶盲区；
    当车辆和障碍物对向行驶时，危险驾驶盲区为总盲区加上安全驾驶盲区。
  13. 根据权利要求9-12任一所述系统，其中：
    依据所述车辆行驶场景确定车辆所处场景的风险系数；
    至少依据所述危险驾驶盲区和所述车辆所处场景的风险系数确定车辆盲区风险等级。
  14. 根据权利要求13所述系统，其中：
    依据危险驾驶盲区、车辆距离盲区横向距离、车辆当前车速、车辆到达冲突区域的时间和车辆所处场景的风险系数确定车辆盲区风险等级。
  15. 根据权利要求14所述系统，其中：
    当所述车辆盲区风险等级高于设定阈值时，车辆优先进行横向避让，所述横向避让指车辆在横向方向上进入远离盲区的车道行驶；
    当所述车辆盲区风险等级低于设定阈值时，车辆继续按照驾驶辅助系统的规划行驶。
  16. 根据权利要求15所述系统，其中：
    当所述车辆盲区风险等级高于设定阈值，且车辆无法进行横向避让时，车辆进行道内避让，所述道内避让指车辆在当前车道沿横向方向上远离盲区移动。
  17. 根据权利要求16所述系统，其中：
    当所述车辆盲区风险等级高于设定阈值，且车辆无法进行横向避让和道内避让时，车辆以自动驾驶辅助系统速度规划和安全行驶速度规划中的速度最小值行驶。
  18. 一种智能驾驶车辆，其包括根据权利要求9-17任一所述的自动驾驶辅助系统。
PCT/CN2020/083075 2020-04-02 2020-04-02 一种车辆盲区识别方法、自动驾驶辅助系统以及包括该系统的智能驾驶车辆 WO2021196145A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2020/083075 WO2021196145A1 (zh) 2020-04-02 2020-04-02 一种车辆盲区识别方法、自动驾驶辅助系统以及包括该系统的智能驾驶车辆
CN202080004420.0A CN113348119A (zh) 2020-04-02 2020-04-02 一种车辆盲区识别方法、自动驾驶辅助系统以及包括该系统的智能驾驶车辆

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/083075 WO2021196145A1 (zh) 2020-04-02 2020-04-02 一种车辆盲区识别方法、自动驾驶辅助系统以及包括该系统的智能驾驶车辆

Publications (1)

Publication Number Publication Date
WO2021196145A1 true WO2021196145A1 (zh) 2021-10-07

Family

ID=77468447

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/083075 WO2021196145A1 (zh) 2020-04-02 2020-04-02 一种车辆盲区识别方法、自动驾驶辅助系统以及包括该系统的智能驾驶车辆

Country Status (2)

Country Link
CN (1) CN113348119A (zh)
WO (1) WO2021196145A1 (zh)

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114030483A (zh) * 2021-12-16 2022-02-11 阿波罗智联(北京)科技有限公司 车辆控制方法、装置、电子设备和介质
CN114049779A (zh) * 2021-11-01 2022-02-15 嘉兴学院 一种交通信号全感应实时控制方法
CN114104000A (zh) * 2021-12-16 2022-03-01 智己汽车科技有限公司 一种危险场景的评估与处理系统、方法及存储介质
CN114274979A (zh) * 2022-01-07 2022-04-05 中国第一汽车股份有限公司 自动驾驶的目标关注度等级判别方法、装置及存储介质
CN114274953A (zh) * 2021-11-23 2022-04-05 深圳市易成自动驾驶技术有限公司 车辆防追尾方法、装置、终端设备以及存储介质
CN114312835A (zh) * 2021-12-24 2022-04-12 阿波罗智能技术(北京)有限公司 车辆控制方法、装置、电子设备、介质及自动驾驶车辆
CN114372375A (zh) * 2022-01-13 2022-04-19 华中科技大学 一种越野车安全行驶区域求解方法及系统
CN114643984A (zh) * 2022-05-20 2022-06-21 阿里巴巴达摩院(杭州)科技有限公司 驾驶避险方法、装置、设备、介质及产品
CN114648887A (zh) * 2022-03-24 2022-06-21 上海舟塔新材料科技有限公司 一种基于发光标线的路侧冲出物体警示方法、系统及装置
CN114719867A (zh) * 2022-05-24 2022-07-08 北京捷升通达信息技术有限公司 一种基于传感器的车辆导航方法及系统
CN114801993A (zh) * 2022-06-28 2022-07-29 鹰驾科技（深圳）有限公司 一种汽车盲区监控系统
CN114973669A (zh) * 2022-05-23 2022-08-30 江苏智行未来汽车研究院有限公司 基于车路协同的公路危险环境预警方法、装置及介质
CN115064005A (zh) * 2022-06-08 2022-09-16 中国第一汽车股份有限公司 一种车辆盲区预警方法、系统和存储介质
CN115468578A (zh) * 2022-11-03 2022-12-13 广汽埃安新能源汽车股份有限公司 路径规划方法、装置、电子设备和计算机可读介质
CN115639562A (zh) * 2022-12-22 2023-01-24 天津所托瑞安汽车科技有限公司 商用车盲点探测系统、制动方法、商用车及存储介质
CN116125897A (zh) * 2023-03-02 2023-05-16 重庆长安汽车股份有限公司 自动驾驶决策方法、装置、电子设备及存储介质
CN116142181A (zh) * 2023-04-24 2023-05-23 深圳市科莱德电子有限公司 车辆盲区障碍物的标识系统、方法及计算机可读存储介质
CN117111574A (zh) * 2023-08-03 2023-11-24 南京航空航天大学 一种智能线控底盘的预期功能安全试验分析方法
CN117734680A (zh) * 2024-01-22 2024-03-22 珠海翔越电子有限公司 大型车辆盲区预警方法、系统及存储介质

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113859251B (zh) * 2021-10-29 2023-04-28 广州文远知行科技有限公司 涉及行驶盲区的车速规划方法、驾驶控制方法及相关设备
CN116184992A (zh) * 2021-11-29 2023-05-30 上海商汤临港智能科技有限公司 车辆控制方法、装置、电子设备及存储介质
CN114475651B (zh) * 2021-12-11 2024-05-14 中国电信股份有限公司 基于车路协同的盲区障碍物紧急避让方法和装置
CN114212105B (zh) * 2021-12-16 2024-03-05 中国人民解放军国防科技大学 一种高泛化能力的交互车辆驾驶意图预测方法和装置
CN114290991B (zh) * 2021-12-28 2023-07-04 联通智网科技股份有限公司 一种盲区监测方法、装置、存储介质和边缘云平台
CN114348023A (zh) * 2022-01-25 2022-04-15 北京三快在线科技有限公司 一种基于盲区的无人驾驶设备控制方法及装置
CN114394051B (zh) * 2022-02-28 2023-11-10 东风商用车有限公司 一种车辆间接视野的提供方法及系统
CN114379417B (zh) * 2022-03-23 2023-02-21 蔚来汽车科技(安徽)有限公司 换电站和引导车辆离开换电站的装置、方法、介质
CN117246335A (zh) * 2022-06-10 2023-12-19 宁波路特斯机器人有限公司 一种车辆自动驾驶控制方法、装置和计算机存储介质
CN115565397A (zh) * 2022-08-19 2023-01-03 清智汽车科技(苏州)有限公司 应用于adas产品的风险容错方法和装置
CN115447606A (zh) * 2022-08-31 2022-12-09 九识(苏州)智能科技有限公司 一种基于盲区识别的自动驾驶车辆控制方法和装置
CN115862383B (zh) * 2023-01-20 2023-05-19 北京航空航天大学 一种基于不确定性的车辆盲区潜在风险量化方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20010089103A (ko) * 2000-03-24 2001-09-29 임종팔 차량의 전방 좌측,(혹은 우측)방향을 용이하게 볼 수 있는차량 부착물
CN106476697A (zh) * 2016-10-27 2017-03-08 深圳市元征科技股份有限公司 一种行车指示方法及装置
CN108528445A (zh) * 2018-03-29 2018-09-14 江苏大学 一种智能汽车传感器盲区主动避撞方法
CN108819941A (zh) * 2018-08-10 2018-11-16 吉利汽车研究院（宁波）有限公司 变道行驶预警方法、装置及设备
CN110456796A (zh) * 2019-08-16 2019-11-15 北京百度网讯科技有限公司 视觉盲区检测方法及装置

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6958630B2 (ja) * 2017-11-17 2021-11-02 株式会社アイシン 車両運転補助システム、車両運転補助方法、及び車両運転補助プログラム
CN109017786B (zh) * 2018-08-09 2020-09-22 北京智行者科技有限公司 车辆避障方法
CN109815832A (zh) * 2018-12-28 2019-05-28 深圳云天励飞技术有限公司 行车预警方法及相关产品
CN110316186A (zh) * 2019-07-01 2019-10-11 百度在线网络技术（北京）有限公司 车辆防碰撞预判方法、装置、设备及可读存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20010089103A (ko) * 2000-03-24 2001-09-29 임종팔 차량의 전방 좌측,(혹은 우측)방향을 용이하게 볼 수 있는차량 부착물
CN106476697A (zh) * 2016-10-27 2017-03-08 深圳市元征科技股份有限公司 一种行车指示方法及装置
CN108528445A (zh) * 2018-03-29 2018-09-14 江苏大学 一种智能汽车传感器盲区主动避撞方法
CN108819941A (zh) * 2018-08-10 2018-11-16 吉利汽车研究院（宁波）有限公司 变道行驶预警方法、装置及设备
CN110456796A (zh) * 2019-08-16 2019-11-15 北京百度网讯科技有限公司 视觉盲区检测方法及装置

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114049779A (zh) * 2021-11-01 2022-02-15 嘉兴学院 一种交通信号全感应实时控制方法
CN114049779B (zh) * 2021-11-01 2022-09-20 嘉兴学院 一种交通信号全感应实时控制方法
CN114274953A (zh) * 2021-11-23 2022-04-05 深圳市易成自动驾驶技术有限公司 车辆防追尾方法、装置、终端设备以及存储介质
CN114274953B (zh) * 2021-11-23 2024-01-02 深圳市易成自动驾驶技术有限公司 车辆防追尾方法、装置、终端设备以及存储介质
CN114030483B (zh) * 2021-12-16 2024-06-07 阿波罗智联(北京)科技有限公司 车辆控制方法、装置、电子设备和介质
CN114104000A (zh) * 2021-12-16 2022-03-01 智己汽车科技有限公司 一种危险场景的评估与处理系统、方法及存储介质
CN114030483A (zh) * 2021-12-16 2022-02-11 阿波罗智联(北京)科技有限公司 车辆控制方法、装置、电子设备和介质
CN114104000B (zh) * 2021-12-16 2024-04-12 智己汽车科技有限公司 一种危险场景的评估与处理系统、方法及存储介质
CN114312835A (zh) * 2021-12-24 2022-04-12 阿波罗智能技术(北京)有限公司 车辆控制方法、装置、电子设备、介质及自动驾驶车辆
CN114274979A (zh) * 2022-01-07 2022-04-05 中国第一汽车股份有限公司 自动驾驶的目标关注度等级判别方法、装置及存储介质
CN114372375A (zh) * 2022-01-13 2022-04-19 华中科技大学 一种越野车安全行驶区域求解方法及系统
CN114648887A (zh) * 2022-03-24 2022-06-21 上海舟塔新材料科技有限公司 一种基于发光标线的路侧冲出物体警示方法、系统及装置
CN114643984A (zh) * 2022-05-20 2022-06-21 阿里巴巴达摩院(杭州)科技有限公司 驾驶避险方法、装置、设备、介质及产品
CN114973669B (zh) * 2022-05-23 2023-09-26 江苏智行未来汽车研究院有限公司 基于车路协同的公路危险环境预警方法、装置及介质
CN114973669A (zh) * 2022-05-23 2022-08-30 江苏智行未来汽车研究院有限公司 基于车路协同的公路危险环境预警方法、装置及介质
CN114719867A (zh) * 2022-05-24 2022-07-08 北京捷升通达信息技术有限公司 一种基于传感器的车辆导航方法及系统
CN115064005A (zh) * 2022-06-08 2022-09-16 中国第一汽车股份有限公司 一种车辆盲区预警方法、系统和存储介质
CN114801993A (zh) * 2022-06-28 2022-07-29 鹰驾科技（深圳）有限公司 一种汽车盲区监控系统
CN115468578A (zh) * 2022-11-03 2022-12-13 广汽埃安新能源汽车股份有限公司 路径规划方法、装置、电子设备和计算机可读介质
CN115639562A (zh) * 2022-12-22 2023-01-24 天津所托瑞安汽车科技有限公司 商用车盲点探测系统、制动方法、商用车及存储介质
CN116125897A (zh) * 2023-03-02 2023-05-16 重庆长安汽车股份有限公司 自动驾驶决策方法、装置、电子设备及存储介质
CN116142181A (zh) * 2023-04-24 2023-05-23 深圳市科莱德电子有限公司 车辆盲区障碍物的标识系统、方法及计算机可读存储介质
CN117111574A (zh) * 2023-08-03 2023-11-24 南京航空航天大学 一种智能线控底盘的预期功能安全试验分析方法
CN117734680A (zh) * 2024-01-22 2024-03-22 珠海翔越电子有限公司 大型车辆盲区预警方法、系统及存储介质
CN117734680B (zh) * 2024-01-22 2024-06-07 珠海翔越电子有限公司 大型车辆盲区预警方法、系统及存储介质

Also Published As

Publication number Publication date
CN113348119A (zh) 2021-09-03

Similar Documents

Publication Publication Date Title
WO2021196145A1 (zh) 一种车辆盲区识别方法、自动驾驶辅助系统以及包括该系统的智能驾驶车辆
WO2022105579A1 (zh) 一种基于自动驾驶的控制方法、装置、车辆以及相关设备
US11685360B2 (en) Planning for unknown objects by an autonomous vehicle
WO2021226776A1 (zh) 一种车辆可行驶区域检测方法、系统以及采用该系统的自动驾驶车辆
US10796574B2 (en) Driving assistance method and device
CN109927719B (zh) 一种基于障碍物轨迹预测的辅助驾驶方法和系统
US11804239B2 (en) Detection and classification of siren signals and localization of siren signal sources
JP6800575B2 (ja) 自己の乗り物のドライバを支援する方法およびシステム
KR102613839B1 (ko) 긴급 차량들의 검출
CN111595357B (zh) 可视化界面的显示方法、装置、电子设备和存储介质
US20190035278A1 (en) Driving Assistance Method and Device
CN110497919B (zh) 自动车辆从自主模式转换到手动模式的物体位置历史回放
TW202246093A (zh) 偵測車輛的駕駛行為
US20240025443A1 (en) Path generation based on predicted actions
US20240025395A1 (en) Path generation based on predicted actions
JP5747593B2 (ja) 規範車速算出装置及びプログラム
US20240025444A1 (en) Path generation based on predicted actions
WO2019127076A1 (en) Automated driving vehicle control by collision risk map
CN112970052B (zh) 车辆控制系统
Beresnev et al. Automated Driving System based on Roadway and Traffic Conditions Monitoring.
JP2019095875A (ja) 車両制御装置、車両制御方法、およびプログラム
JP6614508B2 (ja) 車両制御装置、車両制御方法、およびプログラム
JP6827391B2 (ja) 車両制御装置、車両制御方法、およびプログラム
WO2024067174A1 (zh) 一种预测方法、装置和智能驾驶设备
JP7518893B2 (ja) 緊急車両の検出

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20928433

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20928433

Country of ref document: EP

Kind code of ref document: A1