WO2023092490A1 - Procédé de détermination d'urgence de prise en charge et procédé d'avertissement de prise en charge pour véhicule autonome, et système - Google Patents

Procédé de détermination d'urgence de prise en charge et procédé d'avertissement de prise en charge pour véhicule autonome, et système Download PDF

Info

Publication number
WO2023092490A1
WO2023092490A1 PCT/CN2021/133666 CN2021133666W WO2023092490A1 WO 2023092490 A1 WO2023092490 A1 WO 2023092490A1 CN 2021133666 W CN2021133666 W CN 2021133666W WO 2023092490 A1 WO2023092490 A1 WO 2023092490A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
driver
level
vehicle
car
Prior art date
Application number
PCT/CN2021/133666
Other languages
English (en)
Chinese (zh)
Inventor
张玉新
马海涛
缪宝杰
祝彬
Original Assignee
吉林大学
深圳市大疆创新科技有限公司
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 吉林大学, 深圳市大疆创新科技有限公司 filed Critical 吉林大学
Priority to PCT/CN2021/133666 priority Critical patent/WO2023092490A1/fr
Priority to CN202180100555.1A priority patent/CN117642321A/zh
Publication of WO2023092490A1 publication Critical patent/WO2023092490A1/fr

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
    • 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
    • B60W50/08—Interaction between the driver and the control system
    • B60W50/14—Means for informing the driver, warning the driver or prompting a driver intervention

Definitions

  • the present application relates to the field of vehicle technology, and in particular to a method for judging the urgency of taking over of an automatic driving vehicle, a warning method, a system, a vehicle and a storage medium.
  • the present application provides a method for judging the urgency of takeover of an autonomous vehicle, a warning method, a method for determining the risk of the vehicle, a system, the vehicle, and a storage medium.
  • the present application provides a method for alerting a self-driving car to take over, the method comprising:
  • the state information of the driver is obtained based on the first sensor in the car;
  • the environmental information includes: road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information and communication information;
  • the present application provides a method for judging the urgency of a takeover of an autonomous vehicle, the method comprising:
  • the state information of the driver is obtained based on the first sensor in the car;
  • the target take-over urgency is determined, and the target take-over urgency is used for generating takeover prompt information instructing the driver to take over driving control of the vehicle;
  • the take-over urgency changes as the environmental information changes;
  • the takeover urgency changes as the state information changes.
  • the present application provides a method for determining the risk of an automobile, the method comprising:
  • the state information of the driver is obtained based on the first sensor in the car;
  • the environmental information includes: road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information and communication information;
  • the risk level information of the car is generated.
  • the present application provides an automatic driving vehicle takeover warning system, the system comprising: a memory and a processor;
  • the memory is used to store computer programs
  • the processor is configured to execute the computer program and when executing the computer program, implement the following steps:
  • the state information of the driver is obtained based on the first sensor in the car;
  • the environmental information includes: road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information and communication information;
  • the present application provides a system for judging the urgency of taking over of an autonomous vehicle, the system including: a memory and a processor;
  • the memory is used to store computer programs
  • the processor is configured to execute the computer program and when executing the computer program, implement the following steps:
  • the state information of the driver is obtained based on the first sensor in the car;
  • the target take-over urgency is determined, and the target take-over urgency is used for generating takeover prompt information instructing the driver to take over driving control of the vehicle;
  • the take-over urgency changes as the environmental information changes;
  • the takeover urgency changes as the state information changes.
  • the present application provides an automobile risk determination system, the system comprising: a memory and a processor;
  • the memory is used to store computer programs
  • the processor is configured to execute the computer program and when executing the computer program, implement the following steps:
  • the state information of the driver is obtained based on the first sensor in the car;
  • the environmental information includes: road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information and communication information;
  • the risk level information of the car is generated.
  • the present application provides an automatic driving vehicle, the vehicle includes a first sensor, a second sensor, and the automatic driving vehicle takeover warning system as described in the fourth aspect above.
  • the present application provides an automatic driving vehicle, which includes a first sensor, a second sensor, and the automatic driving vehicle takeover urgency judging system as described in the fifth aspect above.
  • the present application provides an automobile, which includes a first sensor, a second sensor, and the automobile risk determination system described in the sixth aspect above.
  • the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor realizes the automatic Driving a car takes over the warning method.
  • the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the above-mentioned second aspect.
  • the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the above-mentioned third aspect Automobile Risk Determination Methods.
  • the environmental information includes: road information, traffic equipment information, road and traffic equipment temporary change information, traffic participation
  • traffic participation basically include environment-related multi-layer risk sources that affect the safety of driving, so in this way, more accurate and realistic risk level information of vehicles can be obtained , to provide technical support for safe driving.
  • the present application after obtaining more accurate and realistic risk level information of the car, it can also generate warning information for instructing the driver to take over the car according to the risk level information, so that the safety of driving can be guaranteed. Due to different risk levels The warning information corresponding to the information is different, so that the warning information with a reasonable warning degree matching the risk level can be generated, which can prevent the driver from not responding in time or the warning effect from excessively affecting the driving experience, and can improve the reliability and safety of car driving.
  • trigger Takeover threshold target takeover urgency is combined with the external environment and the driver’s state, so that a more accurate and realistic target takeover urgency can be obtained. Due to any value of the corresponding driver’s state information, the takeover urgency varies with the environment Information changes, corresponding to any value of environmental information, the urgency of takeover changes with the change of state information, the external environment and driver status trigger takeover is quantitatively evaluated, so that the target takeover urgency can be made more objective, through the above It can provide technical support for the safety of autonomous driving.
  • Fig. 1 is a schematic flow chart of an embodiment of a risk determination method for an automobile of the present application
  • Fig. 2 is a schematic diagram of an embodiment of the driver's concentration-scene suitability two-dimensional model in the risk determination method of the automobile of the present application;
  • Fig. 3 is a schematic structural diagram of an embodiment of a system applying the risk determination method of an automobile of the present application
  • Fig. 4 is a schematic structural diagram of another embodiment of the system applying the risk determination method of the automobile of the present application.
  • Fig. 5 is a schematic flow chart of an embodiment of the automatic driving vehicle takeover warning method of the present application.
  • FIG. 6 is a schematic flow chart of an embodiment of a method for judging the urgency of taking over of an autonomous vehicle according to the present application
  • Fig. 7 is a schematic structural diagram of an embodiment of the automatic driving vehicle takeover warning system of the present application.
  • FIG. 8 is a schematic structural diagram of an embodiment of a system for judging the urgency of taking over of an autonomous vehicle according to the present application
  • Fig. 9 is a schematic structural diagram of an embodiment of an automobile risk determination system of the present application.
  • the environmental information includes: road information, traffic equipment information, road and traffic equipment temporary change information, traffic participation
  • traffic participation basically include environment-related multi-layer risk sources that affect the safety of driving, so in this way, more accurate and realistic risk level information of vehicles can be obtained , to provide technical support for safe driving.
  • FIG. 1 is a schematic flow chart of an embodiment of the method for determining the risk of a car in the present application. It should be noted that the method in the embodiment of the present application can be applied to ordinary car driving scenarios, and can also be applied to automatic driving scenarios. , the scene with more applications is the automatic driving scene.
  • the method includes: step S101, step S102 and step S103.
  • Step S101 When the car is in a driving state, obtain the driver's state information based on a first sensor in the car.
  • the driver's state information may refer to information related to the driver's state.
  • the driver's state information includes, but is not limited to, the driver's information perception state information, the driver's information processing state information, and the driver's action execution state information.
  • the driver's information perception state information includes but is not limited to: eyes open, eyes closed, eyes off (the sight of the eyes deviates from the front of the driver), eye closing time, eye off time, etc., which can reflect the quality of the driver's information perception state .
  • the information processing state information of the driver includes, but is not limited to: the driver's participation in tasks other than driving tasks (for example: operating a mobile phone, listening to music, etc.), the degree of interaction between the driver and other passengers (for example: the driver chat with other crew members), etc. They can reflect the quality of the driver's information processing status.
  • the driver's action execution state information includes, but is not limited to: the driver's hand-off time, foot-off time, and so on. It can reflect the performance status of the driver's actions.
  • the driver’s information perception state, driver’s information processing state, and driver’s action execution state are detected by the first sensor, and data is generated to obtain the driver’s information perception state information, driver’s information processing state information, and driver’s information processing state information. Action execution status information.
  • the first sensor may be a plurality of independent sensors, or may be a shared sensor to detect various signals.
  • the camera can simultaneously detect the driver's eyes, mouth, and hands.
  • Step S102 Collect environmental information based on the second sensor of the car, the environmental information includes: road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information and communication information.
  • the road information includes but not limited to road geometry, road grade, lane line clarity, and road surface adhesion; the road geometry includes but not limited to curve curvature and slope gradient.
  • the traffic facility information includes, but is not limited to, the status of traffic lights and signs; the temporary change information of roads and traffic facilities includes, but is not limited to, road changes caused by road maintenance and construction, or road surface changes caused by water and snow on the road;
  • the traffic participant information includes, but is not limited to, surrounding vehicle types, pedestrians, non-motorized vehicles, traffic volume, distance to traffic participants, and pedestrian density;
  • the climate information includes, but is not limited to, rainfall, smog, Visibility, humidity;
  • the communication information includes but not limited to positioning signal strength, signal connection smoothness.
  • the above environmental information basically includes all factors related to the environment, which is relatively comprehensive.
  • the second sensor collects and generates data to obtain road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information and communication information.
  • the second sensor can be an independent sensor, or a shared sensor to collect multiple signals.
  • cameras can simultaneously capture roads, traffic facilities and traffic participants.
  • Step S103 Generate risk level information of the car according to the driver's state information and the environmental information.
  • the risk level information of the car is generated. Different driver status information and different environmental information, the same driver status information and different environmental information, different driver status information and the same environmental information can generate different risks grade. Wherein, corresponding to any value of the state information of the driver, the risk level of the car changes with the change of the environment information; corresponding to any value of the environment information, the risk level of the car changes with the change of the state information.
  • the driver in the car can be prompted accordingly. If it is an automatic driving system, it can automatically control the car, and it can also remind traffic participants outside the car, etc. Therefore, the risk level information can be used to instruct the car to remind the driver according to the in-vehicle prompt mode corresponding to the current risk level, and/or can be used to instruct the car to execute a motion mode corresponding to the current risk level to reduce traffic risks, And/or, it can be used to instruct the car to remind other traffic participants except the car according to the outside prompt mode corresponding to the current risk level.
  • the environmental information includes: road information, traffic equipment information, road and traffic equipment temporary change information, traffic participation
  • traffic participation basically include environment-related multi-layer risk sources that affect the safety of driving, so in this way, more accurate and realistic risk level information of vehicles can be obtained , to provide technical support for safe driving.
  • step S103, generating the risk level information of the car according to the driver's state information and the environmental information may include: according to the driver's state information and the environmental information, and The corresponding relationship between the driver's state information and environmental information and various preset risk levels generates the risk level information of the car.
  • This embodiment comprehensively considers multiple risk sources related to the driver that have an impact on the safety of car driving, so that more accurate and realistic risk level information of the car can be obtained to provide technical support for safe driving.
  • step S103 generating the risk level information of the car according to the driver's state information and the environment information, may include: sub-step S1031, sub-step S1032 and sub-step S1033.
  • Sub-step S1031 Obtain the concentration of the driver according to the state information of the driver.
  • Sub-step S1032 According to the environmental information, the scene suitability is obtained.
  • Sub-step S1033 Generate risk level information of the car according to the driver's concentration and the scene suitability.
  • the driver's concentration can be obtained according to the driver's state information, and the scene suitability can be obtained according to the environment information.
  • the driver's concentration and scene suitability are quantified numerical forms, which can make the car's risk level information more objective and accurate; the driver's concentration and scene suitability are two key factors of the car's risk level, which can be further increased The objectivity and accuracy of the risk level information of the car, thereby reducing the error of the risk level information.
  • the higher the driver's driving proficiency the higher the driver's concentration
  • the automatic driving level The higher the value, the higher the concentration of the driver.
  • parameters such as the maximum allowable eye-off time and the maximum hand-off time will change with different needs.
  • the maximum eye-off time will change with the proficiency of the driver.
  • the maximum eye-off time allowed by a driver with high proficiency in the system will be longer than that of a driver with a low level of proficiency in the system. Under the same eye-off time , drivers with high system proficiency will be more attentive than drivers with low system proficiency.
  • L2 autonomous driving According to regulations, the driver is not allowed to let go of his hands, eyes are not allowed to take off, and his eyes are focused on the road.
  • L3 autonomous driving after the automatic driving function is turned on, the car is responsible for the driving process, and the driver can let go , Take off your feet. Therefore, under the same driver's status information (hands off, feet off), the driver's concentration under L3 automatic driving is higher than that under L2 automatic driving.
  • the higher the scene suitability the lower the risk level of the car
  • the higher the driver's concentration the lower the risk level of the car
  • the quantification method adopts a normalized calculation method, that is, sub-step S1031, the obtaining the driver's concentration according to the driver's state information may include: The status information is normalized, and the driver's concentration expressed by the normalized value is obtained.
  • Normalization is a dimensionless processing method, which makes the absolute value of the physical system numerical value into a certain relative value relationship, which can simplify calculation and facilitate data processing, and is an effective way to reduce the value.
  • the driver's status information includes the driver's information perception status information, the driver's information processing status information, and the driver's action execution status information.
  • substep S1031 said normalizing the state information of the driver to obtain the concentration of the driver expressed by the normalized value may also include: sub-step S10311 and sub-step S10312.
  • Sub-step S10311 Normalize the driver's information perception state information, the driver's information processing state information, and the driver's action execution state information to obtain the corresponding driver's information perception state information.
  • the state normalized value, the driver information processing state normalized value and the driver action execution state normalized value are normalized values.
  • Sub-step S10312 According to the normalized value of the driver's information perception state, the normalized value of the driver's information processing state, the normalized value of the driver's action execution state and their respective weights, obtain the normalized The driver's concentration indicated by the value.
  • the data can be changed into decimals between (0, 1), and the dimensioned expression can be changed into a dimensionless expression, so that indicators of different units or magnitudes can be compared and weighted.
  • the driver's information perception state information, the driver's information processing state information and the driver's action execution state information can be made into dimensionless decimals with data between (0, 1) , which is convenient for comparison and weighting processing.
  • the driver's information perception state information includes the driver's eye-off time, the longer the eye-off time, the smaller the normalized value of the driver's information perception state;
  • the driver's information processing state information includes the driver's participation in other tasks except driving tasks and the driver's interaction with other occupants. The more the participation, the normalized driver information processing state The smaller the value, the more the degree of interaction, and the smaller the normalized value of the driver's information processing state;
  • the driver's action execution state information includes the driver's hand-off time and the driver's foot-off time Time; the longer the hand-off time, the smaller the normalized value of the driver's action execution state, and the longer the foot-off time, the smaller the normalized value of the driver's action execution state.
  • the specific method for the normalization calculation of the driver's state information can be:
  • X is the concentration of the driver represented by the normalized value
  • a 1 is the normalized value of driver information perception state
  • ⁇ 1 is the weight that the state of information perception affects the concentration of the driver
  • a 2 is the normalized value of the driver information processing state
  • ⁇ 2 is the weight that the information processing state affects the concentration of the driver
  • a 3 is the normalized value of the driver's action execution state
  • ⁇ 3 is the weight of the driver's action execution state affecting the driver's concentration.
  • the value ranges of X, a 1 , a 2 , and a 3 are all [0,1].
  • the judgment method of a i is: taking the normalized value of driver information perception state as an example, taking the driver’s eyes-off time to represent the quality of the driver’s perception state, and t is the driver’s eyes-off time obtained from the actual test during the driving process Time, t ⁇ [0,t max ], where t max is the set maximum value forbidding drivers to take their eyes off, and t max is obtained through actual tests.
  • t max is not a fixed value, it may vary with the age of the driver, the driver's driving proficiency, and the level of automatic driving of the system. For example, the value of t max for a driver with high driving proficiency will be larger than that for a driver with low driving proficiency.
  • driver states there are many combinations of driver states. In addition to obtaining reasonable combinations of driver states, those unpredictable combinations of driver states should also be considered. For example, for L2 autonomous driving, according to regulations, the driver is not allowed to let go of his hands, eyes are not allowed to take off, and his eyes are focused on the road. Scenes.
  • the quantification method adopts a normalized calculation method, that is, sub-step S1032, and obtaining the scene suitability according to the environmental information may include: performing normalization processing on the environmental information to obtain Scene suitability expressed as a normalized value.
  • Normalization is a dimensionless processing method, which makes the absolute value of the physical system numerical value into a certain relative value relationship, which can simplify calculation and facilitate data processing, and is an effective way to reduce the value.
  • the sub-step S1032, performing normalization processing on the environmental information to obtain the scene suitability represented by a normalized value may further include: sub-steps S10321 and S10322.
  • Sub-step S10321 Normalize the road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information, and communication information to obtain the corresponding normalized value of road information, traffic Normalized value of equipment information, normalized value of road and traffic equipment temporary change information, normalized value of traffic participant information, normalized value of climate information, normalized value of communication information.
  • Sub-step S10322 According to the normalized value of road information, the normalized value of traffic equipment information, the normalized value of road and traffic equipment temporary change information, the normalized value of traffic participant information, the normalized value of climate information, The normalized value of the communication information and their respective weights are used to obtain the scene suitability represented by the normalized value.
  • road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information, and communication information can be turned into dimensionless decimals with data between (0, 1), which is convenient Comparison and weighting processing.
  • the road information includes road geometry, road grade, lane line clarity, road surface adhesion; the road geometry includes curve curvature and slope gradient; the traffic facility information includes traffic light status, indication
  • the temporary change information of roads and traffic facilities includes road changes caused by road maintenance and construction, or road surface changes caused by road surface water and snow accumulation;
  • the traffic participant information includes surrounding vehicle types, pedestrians, non-motor vehicles, etc. , traffic volume, distance to traffic participants, and pedestrian density;
  • the climate information includes rainfall, smog, visibility, and humidity;
  • the communication information includes: positioning signal strength, and smoothness of signal connection.
  • Y is the scene suitability represented by the normalized value
  • b 1 is the normalized value of road information
  • ⁇ 1 is the weight of road information affecting scene suitability
  • b2 is the normalized value of traffic facility information
  • ⁇ 2 is the weight of traffic facility information affecting the suitability of the scene
  • b 3 is the normalized value of temporary change information of roads and traffic facilities
  • ⁇ 3 is the weight that changes the temporary information of roads and traffic facilities to affect the suitability of the scene
  • b4 is the normalized value of traffic participant information
  • ⁇ 4 is the weight of traffic participant information affecting scene suitability
  • b 5 is the normalized value of climate information
  • ⁇ 5 is the weight of climate information affecting the suitability of the scenario
  • b 6 is the normalized value of communication information
  • ⁇ 6 is the weight of communication information affecting the suitability of the scene.
  • the normalization method of b i is: taking the climate information of b 5 as an example, taking the influence factors of b 5 as visibility level V, rainfall M, and wind speed W as an example:
  • V is the normalized value of visibility level of climate information
  • ⁇ 1 is the weight of visibility affecting the normalized value of climate information
  • M is the normalized value of rainfall of climate information
  • ⁇ 2 is the weight of rainfall affecting the normalized value of climate information
  • W is the wind speed normalized value of climate information
  • ⁇ 3 is the weight of wind speed affecting the normalized value of climate information.
  • V, M, and W are all [0,1], and the weight is a variable that must be determined through each experiment, but the sum of the weights is 1.
  • the sub-step S1033, generating the risk level information of the car according to the driver's concentration and the scene suitability may include: according to the driver's concentration represented by a normalized value and the scene suitability represented by the normalized value to generate the risk level information of the car. In this way, the risk level information of the automobile can be made more objective through this normalized quantification method.
  • the sub-step S1033 is to generate the risk level information of the car according to the concentration of the driver represented by the normalized value and the suitability of the scene represented by the normalized value, which may also include: sub-steps S10331, Sub-step S10332 and sub-step S10333.
  • Sub-step S10331 Determine the position of the driver's concentration expressed by the normalized value and the scene suitability expressed by the normalized value in the driver's concentration-scene suitability two-dimensional model, the two-dimensional model includes A plurality of regions, the plurality of regions correspond to a plurality of different risk levels.
  • Sub-step S10332 Determine the area of the location in the two-dimensional model.
  • Sub-step S10333 Generate risk level information of the car according to the risk level corresponding to the area.
  • a two-dimensional model of driver concentration-scene suitability is obtained through experiments in advance.
  • the abscissa of the two-dimensional model is the driver's concentration or scene suitability
  • the ordinate is the scene suitability or driver's concentration.
  • the area formed by the abscissa and the ordinate is divided into multiples according to the risk level of the car to represent different risks. level area.
  • the points formed by different driver concentration and scene suitability are located in different areas, and based on this, the risk level information of the car can be generated according to the specific driver concentration and scene suitability, as well as the two-dimensional model.
  • the warning information is used to perform one or more of the following operations:
  • generating warning information for instructing the driver to take over the car according to the risk level information includes:
  • An early warning signal is generated based on the target early warning information level.
  • the plurality of early warning signal levels include one or more of the following early warning signals:
  • the first-level early warning signal uses a combination of high-urgency auditory signal, visual signal and tactile signal to remind the driver to take over the driving right of the vehicle urgently;
  • the second-level early warning signal uses a combination of visual signals and auditory signals with moderate urgency to remind the driver to take over the driving right of the vehicle urgently;
  • the third-level early warning signal uses a combination of visual signal and auditory signal with higher comfort level to remind the driver to take over the driving right of the vehicle or improve driving concentration.
  • controlling the vehicle to execute a motion mode corresponding to the current risk level to reduce traffic risks includes performing one or more of the following operations:
  • the vehicle is controlled to decelerate quickly and stop in the lane urgently.
  • instructing the vehicle to prompt other traffic participants except the car according to the outside prompt mode corresponding to the current risk level includes performing one or more of the following operations:
  • the third level warning signal outside the vehicle is used to remind other traffic participants;
  • the first level warning signal outside the vehicle is used to remind other traffic participants.
  • the two-dimensional model is divided into multiple regions by arc lines. As shown in Figure 2, the two-dimensional model is divided into three areas by circular arcs with different radii. Level 2, Level 3), the greater the driver's concentration, the greater the scene suitability, the greater the risk level, and the lower the risk of the car.
  • Level 2 Level 3
  • the risk level can be replaced by the driver's takeover urgency to take over the car.
  • the risk level is from the third level to the first level, and the driver's takeover urgency to take over the car is also from the third level to the first level. , when the takeover urgency is the first level, the takeover urgency is the largest, and when the takeover urgency is the third level, the takeover urgency is the least.
  • the method further includes: Step S104.
  • Step S104 Send out corresponding warning information according to the risk level information, wherein the warning information corresponding to different risk level information is different.
  • the warning information for instructing the driver to take over the car is generated according to the risk level information, which can ensure the safety of driving the car, because the warning information corresponding to different risk level information is different.
  • warning information with a reasonable warning level matching the risk level can be generated, which can prevent the driver from responding too late for insufficient warning effects or excessively affect the driving experience, and can improve the reliability and safety of car driving.
  • step S104 sending corresponding warning information according to the risk level information of the car, may include: sub-step S1041 and/or sub-step S1042 and/or sub-step S1043.
  • Sub-step S1041 According to the risk level information of the car, remind the driver according to the in-vehicle prompt mode corresponding to the current risk level.
  • sub-step S1042 controlling the car to execute a motion mode corresponding to the current risk level to reduce traffic risk.
  • sub-step S1042 may be included.
  • sub-step S1043 Prompt other traffic participants except the car according to the outside prompt mode corresponding to the current risk level.
  • Reminding the driver through the in-vehicle prompting mode, the car executing the sports mode, and the external prompting mode prompting other traffic participants can fully ensure the safety of the driver, vehicle, and other traffic participants.
  • the risk level of the car includes three levels, namely the first level, the second level and the third level.
  • the prompting the driver according to the in-vehicle prompt mode corresponding to the current risk level may include: if the current risk level is the third level, adopting the third level in-vehicle early warning signal to remind the driver; if the current risk level is the second level, then take the second level early warning signal in the car to remind the driver; if the current risk level is the first level, then take the The first-level early warning signal in the car reminds the driver.
  • the risk level is divided into three levels, and the corresponding in-vehicle prompt modes are also divided into three levels, which can easily and conveniently distinguish the in-vehicle prompt modes, avoiding too complicated distinction and reducing the reminding effect on the driver.
  • the first-level early warning signal in the vehicle uses a combination of high-urgency auditory signal, visual signal and tactile signal to remind the driver to take over the driving right of the vehicle;
  • the early warning signal uses a combination of visual signals and auditory signals with moderate urgency to remind the driver to take over the driving right of the vehicle in an emergency; Ways to remind the driver to take over the driving right of the vehicle or improve driving concentration.
  • this embodiment also considers the comfort of the driver, and establishes a man-machine friendly early warning method, which can effectively remind the driver and improve user experience .
  • the tactile signal in the first-level early warning signal in the vehicle is a vibration signal of the driver's seat
  • the auditory signal in the first-level early warning signal in the vehicle is an audio signal with a high frequency and a high peak value ( For example: "current danger, please pay attention to driving", "the surrounding environment is complicated, please increase your driving concentration")
  • the visual signal in the first-level early warning signal in the car is a highlighted emergency signal displayed on the center console screen.
  • Information or symbols; the auditory signal in the second-level early warning signal in the car is an audio signal with moderate comfort, and the visual signal in the second-level early warning signal in the car is a light with moderate brightness displayed on the screen of the center console.
  • Information or symbols; the auditory signal in the third-level early warning signal in the car is a more comfortable audio signal, and the visual signal in the third-level early warning signal in the car is a comfortable brightness displayed on the center console screen. information or symbols.
  • the controlling the car to execute a motion mode corresponding to the current risk level to reduce traffic risks may include: if the current risk level is the third level, then control the car to slow down; if the current risk level is the second level, then control the car to slow down, change lanes slowly and pull over; if the current risk level is the first level, then control the car to quickly decelerate and stop in this lane urgently.
  • the risk level is divided into three levels, and the corresponding movement modes for controlling the execution of the car are also divided into three levels, so that the movement modes can be easily and conveniently distinguished, and avoid too complicated distinction and execution of the car.
  • the vehicle is controlled to execute corresponding differentiated motion patterns, thereby reducing traffic risks.
  • the prompting other traffic participants except the car according to the prompt mode outside the vehicle corresponding to the current risk level may include: if the current risk level is the third level , then use the third-level early warning signal outside the vehicle to remind other traffic participants; if the current risk level is the second level, use the second-level early warning signal outside the vehicle to remind other traffic participants; if the current If the risk level is the first level, the first level early warning signal outside the vehicle will be used to remind other traffic participants.
  • the risk level is divided into three levels, and the corresponding warning modes outside the vehicle are also divided into three levels, which can easily and conveniently distinguish the warning modes outside the vehicle, avoiding too complicated distinction and reducing the reminder effect on other traffic participants.
  • the first-level early warning signal outside the vehicle uses a combination of high-urgency auditory signal and visual signal to remind other traffic participants; the second-level early warning signal outside the vehicle uses a visual signal with moderate urgency.
  • the combined early warning mode of signal and auditory signal reminds other traffic participants; the first-level early warning signal outside the vehicle adopts the combined early warning mode of visual signal and auditory signal with higher comfort to remind other traffic participants.
  • this embodiment also considers the driver's comfort and establishes a man-machine friendly early warning method, which can effectively remind other traffic participants while also improving user experience.
  • the auditory signal in the first-level early warning signal outside the vehicle is an audio signal with high frequency and high peak value (for example: "the vehicle is malfunctioning, please pay attention"), and the first-level early warning signal outside the vehicle
  • the visual signal in the vehicle is a highlighted and urgent information or symbol displayed on the display screen outside the vehicle;
  • the auditory signal in the second-level warning signal outside the vehicle is an audio signal with moderate comfort, and the second-level warning signal outside the vehicle is an audio signal with moderate comfort.
  • the visual signal in the signal is information or symbols with moderate brightness displayed on the display screen outside the vehicle;
  • the auditory signal in the third-level warning signal outside the vehicle is an audio signal with better comfort (for example: "the surrounding environment is complex, Please be careful")
  • the visual signal in the third-level warning signal outside the vehicle is information or symbols with comfortable brightness displayed on the center console screen.
  • the thresholds of specific warning signals such as the frequency and peak value of the auditory signal, the color and size of the font displayed on the screen to remind traffic participants outside the car, the speed that needs to be reduced for automatic deceleration, etc., are related to different needs and need to be based on special needs. Subsequent design and repeated experimental demonstration.
  • the warning information can vary according to different needs.
  • the warning information can be self-learned based on the characteristics of the driver (such as driving aggressiveness, takeover ability, preference for perceived information, etc.), and then adaptively set; and/or, the warning The information can be set based on driver customization; and/or, the warning information can vary with the level of automatic driving.
  • a set of warning information corresponds to a driver with high system proficiency
  • a set of warning information corresponds to a driver with low system proficiency.
  • the two sets of warning information may have the same method but different parameter thresholds, or two sets of warning information.
  • corresponding warning information can also be configured for L2 and L3 automatic driving systems; when the system triggers warning information for L2 automatic driving, it will trigger the warning information corresponding to L2 and not trigger the warning information corresponding to L3.
  • the system includes: a driver state detection module 1, an external environment detection module 2, a driver concentration calculation module 3, a scene suitability calculation module 4, a comprehensive judgment module 5, and a prompt control module 6 And warning module 7.
  • the driver state detection module 1 may include a driver information perception state detection unit 11 , a driver information processing state detection unit 12 , and a driver action execution state detection unit 13 .
  • the driver information perception state detection unit 11 detects information such as the driver's eyes-off time, generates data and transmits it to the driver concentration calculation module 3 .
  • the driver information processing state detection unit 12 detects information such as the degree of driver's participation in secondary tasks, the degree of interaction between the driver and other passengers, generates data and transmits it to the driver's concentration calculation module 3 .
  • the driver's action execution state detection unit 13 detects information such as the driver's hands-off time and foot-off time, generates data and transmits it to the driver's concentration calculation module 3 .
  • the driver's concentration calculation module 3 can include a driver's information perception state normalization calculation unit 31, a driver's information processing state normalization calculation unit 32, a driver's action execution state normalization calculation unit 33 and a driver's concentration normalization calculation unit.
  • a calculation unit 34 .
  • the driver information perception state normalization calculation unit 31 receives the data sent from the driver information perception state unit 11 and performs normalization processing, and finally outputs a value in the range of 0 to 1, and 1 represents the driver information perception state.
  • the state is good, and 0 represents that the driver's information perception state is poor, and this value is transmitted to the normalized calculation unit 34 of the driver's concentration.
  • the driver information processing state normalization calculation unit 32 receives the data sent from the driver information processing state detection unit 12 and performs normalization processing, and finally outputs a value in the range of 0 to 1, and 1 represents the driver information processing state Good, 0 means that the driver information processing state is poor, and this value is transmitted to the normalized calculation unit 34 of the driver's concentration.
  • the driver action execution state normalization calculation unit 33 receives the data sent from the driver action execution state detection unit 13 and performs normalization processing, and finally outputs a value in the range of 0 to 1, and 1 represents the driver action execution state.
  • the state is good, and 0 represents that the driver's action execution state is poor, and this value is transmitted to the normalized calculation unit 34 of the driver's concentration.
  • the driver concentration calculation unit 34 receives the above data and performs normalized calculation, and finally outputs a value in the range of 0 to 1, where 1 represents a high concentration of the driver, and 0 represents a low concentration of the driver.
  • the external environment detection module 2 may include a road detection unit 21 , a traffic facility detection unit 22 , a road and traffic facility temporary change detection unit 23 , a traffic participant detection unit 24 , a climate information detection unit 25 and a communication information detection unit 26 .
  • the external environment detection module 2 is mainly used to detect external environmental information, including road information, traffic facility information, road and traffic temporary change information, traffic participant information, climate information, communication information, generate data and transmit it to the scene suitability calculation Module 4.
  • the road detection unit 21 detects information such as road geometry (such as curve curvature, slope gradient), road grade, lane line clarity, road surface adhesion, generates road data and transmits it to the scene suitability calculation module 4;
  • the detection unit 22 detects information such as the state of traffic lights, signs, etc., generates data and transmits it to the scene suitability calculation module 4;
  • the temporary change detection unit 23 of roads and traffic facilities detects road changes caused by road maintenance and construction or due to road surface water, accumulated water, etc.
  • Information such as road surface changes caused by snow generates data and transmits it to the scene suitability calculation module 4;
  • traffic participant detection unit 24 detects information such as traffic flow, distance from traffic participants, pedestrian density, etc., generates data and transmits it to the scene The suitability calculation module 4;
  • the climate information detection unit 25 detects the amount of rainfall, haze, visibility, humidity and other information, generates data and transmits it to the scene suitability calculation module 4;
  • the communication information detection unit 26 detects the positioning signal strength, signal Connect information such as smoothness, generate data and transmit it to the scene suitability calculation module 4.
  • the scene suitability calculation module 4 includes a road information normalization calculation unit 41, a traffic facility information normalization calculation unit 42, a road and traffic facility temporary change information normalization calculation unit 43, and a traffic participant information normalization calculation unit 44 , a climate information normalization calculation unit 45 , a communication information normalization calculation unit 46 and a scene suitability normalization calculation unit 47 .
  • the road information normalization calculation unit 41 receives the data sent from the road detection unit 21 and performs normalization processing, and finally outputs a value in the range of 0 to 1, 1 represents ideal road conditions, 0 represents bad road conditions, And transmit this value to the scene suitability normalization calculation unit 47.
  • the traffic facility information normalization calculation unit 42 receives the data sent from the traffic facility detection unit 22 and performs normalization processing, and finally outputs a value in the range of 0 to 1, 1 represents the condition of the traffic facility is good, and 0 represents the condition of the traffic facility bad, and transmit this value to the scene suitability normalization calculation unit 47.
  • the road and traffic facility temporary change information normalization calculation unit 43 receives the data sent from the road and traffic facility temporary change detection unit 23 and performs normalization processing, and finally outputs a value in the range of 0 to 1, and 1 represents road and traffic facility.
  • the traffic facilities have not been temporarily changed, and 0 means that the road and traffic facilities have been temporarily changed, and this value is transmitted to the scene suitability normalization calculation unit 47 .
  • the traffic participant information normalization calculation unit 44 receives the data sent from the traffic participant detection unit 24 and performs normalization processing, and finally outputs a value in the range of 0 to 1, 1 means that the traffic participant situation is simple, and 0 means The situation of the traffic participant is complicated, and this value is transmitted to the scene suitability normalization calculation unit 47 .
  • the climate information normalization calculation unit 45 receives the data sent from the climate information detection unit 25 and performs normalization processing, and finally outputs a value in the range of 0 to 1, where 1 represents good weather conditions, 0 represents bad weather conditions, and This numerical value is transmitted to the scene suitability normalization calculation unit 47 .
  • the communication information normalization calculation unit 46 receives the data sent from the communication information detection unit 26 and performs normalization processing, and finally outputs a value in the range of 0 to 1, where 1 represents good communication conditions, 0 represents bad communication conditions, and This numerical value is transmitted to the scene suitability normalization calculation unit 47 .
  • the scene suitability normalization calculation unit 47 receives the above data and performs normalization calculation, and finally outputs a value in the range of 0 to 1, 1 means the scene suitability is good, and 0 means the scene suitability is bad.
  • the comprehensive judgment module 5 contains a preset two-dimensional model of driver concentration-scene suitability, and the comprehensive judgment module 5 judges this point data according to the point data combined by the driver concentration calculation module 3 and the scene suitability calculation module 4 position in the 2D model.
  • Prompt control module 6 judges and outputs the signal of first-level, second-level or third-level takeover urgency according to the position of point data in two-dimensional model.
  • Warning module 7 is made up of visual warning unit 71, auditory warning unit 72, tactile warning unit 73, and warning module 7 receives and prompts the multi-level (such as first level, second level, third level) that control module 6 sends to take over emergency. According to different levels of signals, different takeover reminders are sent.
  • FIG. 5 is a schematic flow chart of an embodiment of an automatic driving vehicle takeover warning method of the present application.
  • the method of this embodiment is the application of the above-mentioned automobile risk determination method to the automatic driving system.
  • the above-mentioned automobile risk determination method please refer to the above-mentioned automobile risk determination method, which will not be repeated here.
  • the method includes: step S201, step S202, step S203 and step S204.
  • Step S201 when the car is in an automatic driving state, obtain the driver's state information based on the first sensor in the car.
  • Step S202 Collect environmental information based on the second sensor of the car, the environmental information includes: road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information and communication information.
  • Step S203 Generate risk level information of the car according to the driver's state information and the environment information.
  • Step S204 Generate warning information for instructing the driver to take over the car according to the risk level information, wherein the warning information corresponding to different risk level information is different.
  • the embodiment of the present application obtains more accurate and realistic risk level information of the car, when it is applied to the automatic driving system, it can also generate warning information for instructing the driver to take over the car according to the risk level information, so that the driving safety of the car can be guaranteed.
  • Safety because the warning information corresponding to different risk level information is different, so it can generate warning information with a reasonable warning level matching the risk level, which can avoid insufficient warning effect for the driver to react in time or excessive warning effect that affects the driving experience, and can improve car driving. reliability and safety.
  • the warning information is used to perform one or more of the following operations:
  • generating warning information for instructing the driver to take over the car according to the risk level information includes:
  • An early warning signal is generated based on the target early warning information level.
  • the plurality of early warning signal levels include one or more of the following early warning signals:
  • the first-level early warning signal uses a combination of high-urgency auditory signal, visual signal and tactile signal to remind the driver to take over the driving right of the vehicle urgently;
  • the second-level early warning signal uses a combination of visual signals and auditory signals with moderate urgency to remind the driver to take over the driving right of the vehicle urgently;
  • the third-level early warning signal uses a combination of visual signal and auditory signal with higher comfort level to remind the driver to take over the driving right of the vehicle or improve driving concentration.
  • controlling the vehicle to execute a motion mode corresponding to the current risk level to reduce traffic risks includes performing one or more of the following operations:
  • the vehicle is controlled to decelerate quickly and stop in the lane urgently.
  • prompting other traffic participants other than the car according to the prompt mode outside the vehicle corresponding to the current risk level includes performing one or more of the following operations:
  • the third level warning signal outside the vehicle is used to remind other traffic participants;
  • the first level warning signal outside the vehicle is used to remind other traffic participants.
  • the warning information can be self-learned based on the characteristics of the driver and then adaptively set; and/or, the warning information can be set based on the driver's customization; and/or, the The above warning information can vary with the level of automatic driving.
  • Fig. 6 is a schematic flowchart of an embodiment of a method for judging the urgency of takeover of an automatic driving vehicle according to the present application.
  • the method of this embodiment is a method applied to an automatic driving system, and is basically the same as the content in the above-mentioned method for determining the risk of a vehicle. The same, but the difference is: in the automatic driving system, the risk level of the car is not used, but the takeover urgency of the car is used. value, the takeover urgency changes as the other changes.
  • the method includes: step S301, step S302 and step S303.
  • Step S301 when the car is in an automatic driving state, obtain the driver's state information based on the first sensor in the car.
  • Step S302 collecting environmental information based on the second sensor of the car.
  • Step S303 Determine the target takeover urgency according to the driver's state information and the environment information, and the corresponding relationship between the driver's state information and environment information and various preset takeover urgencies, and the target takeover urgency The degree is used to generate takeover prompt information indicating that the driver takes over the driving control of the car; wherein, in the corresponding relationship between the driver's state information and environmental information and various preset takeover urgencies, the corresponding driver's state information For any value of , the takeover urgency changes with the change of the environment information; corresponding to any value of the environment information, the takeover urgency changes with the change of the state information.
  • trigger Takeover threshold target takeover urgency is combined with the external environment and the driver’s state, so that a more accurate and realistic target takeover urgency can be obtained. Due to any value of the corresponding driver’s state information, the takeover urgency varies with the environment Information changes, corresponding to any value of environmental information, the urgency of takeover changes with the change of state information, the external environment and driver status trigger takeover is quantitatively evaluated, so that the target takeover urgency can be made more objective, through the above It can provide technical support for the safety of autonomous driving.
  • the environment information includes: road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information, and communication information.
  • the target takeover urgency is determined according to the driver's state information and the environment information, and the corresponding relationship between the driver's state information and environment information and various preset takeover urgencies, Including: obtaining the driver's concentration according to the driver's state information; obtaining the scene suitability according to the environment information; according to the driver's concentration and the scene suitability, and the driver's concentration
  • the target takeover urgency is determined based on the corresponding relationship between the scene suitability and the preset multiple takeover urgency.
  • the higher the driver's driving proficiency the higher the driver's concentration
  • the automatic driving level The higher the value, the higher the concentration of the driver.
  • FIG. 7 is a schematic structural diagram of an embodiment of an automatic driving vehicle takeover warning system of the present application.
  • the system 700 includes: a memory 701 and a processor 702; the memory 701 and the processor 702 are connected through a bus.
  • the processor 702 may be a microcontroller unit, a central processing unit, or a digital signal processor, among others.
  • the memory 701 may be a Flash chip, a read-only memory, a magnetic disk, an optical disk, a U disk or a mobile hard disk, and the like.
  • the self-driving car takeover warning system of this embodiment can execute the steps in the above-mentioned self-driving car takeover warning method.
  • the relevant content of the above-mentioned self-driving car takeover warning method please refer to the relevant content of the above-mentioned self-driving car takeover warning method, which will not be repeated here.
  • the memory is used to store a computer program; the processor is used to execute the computer program and when executing the computer program, implement the following steps:
  • the driver's state information is obtained based on the first sensor in the car; the environmental information is collected based on the second sensor of the car, and the environmental information includes: road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information and communication information; according to the driver’s status information and the environmental information, generate the risk level information of the car; The driver takes over the warning information of the car, wherein the warning information corresponding to different risk level information is different.
  • warning information is used to perform one or more of the following operations:
  • An early warning signal is generated based on the target early warning information level.
  • the plurality of early warning signal levels include one or more of the following early warning signals: the first level early warning signal uses a combination of high urgency auditory signal, visual signal and tactile signal to remind the driver to take over the driving right of the vehicle urgently;
  • the second-level early warning signal uses a combination of visual signals and auditory signals with moderate urgency to remind the driver to take over the driving right of the vehicle urgently;
  • the third-level early warning signal uses a combination of visual signals and auditory signals with higher comfort. Remind the driver to take over the driving right of the vehicle or improve driving concentration.
  • the processor executes the computer program, the following steps are implemented: if the current risk level is the third level, control the vehicle to slow down; if the current risk level is the second level, then Control the vehicle to slow down, slowly change lanes and pull over to stop; if the current risk level is the first level, control the vehicle to slow down quickly and park in the lane urgently.
  • the processor executes the computer program, the following steps are implemented: if the current risk level is the third level, use the third level warning signal outside the vehicle to remind other traffic participants; If the risk level is the second level, use the second level early warning signal outside the vehicle to remind other traffic participants; if the current risk level is the first level, use the first level early warning signal outside the vehicle to remind other traffic participants remind.
  • the warning information can be self-learned based on the characteristics of the driver and then adaptively set; and/or, the warning information can be set based on the driver's customization; and/or, the warning information can be Varies with autopilot level.
  • FIG. 8 is a schematic structural diagram of an embodiment of a system for judging the urgency of taking over of an autonomous vehicle according to the present application.
  • the system 800 includes: a memory 801 and a processor 802 ; the memory 801 and the processor 802 are connected through a bus.
  • the processor 802 may be a microcontroller unit, a central processing unit, or a digital signal processor, among others.
  • the memory 801 may be a Flash chip, a read-only memory, a magnetic disk, an optical disk, a U disk or a mobile hard disk, and the like.
  • the system for judging the urgency of takeover of an autonomous vehicle in this embodiment can execute the steps in the method for judging the urgency of takeover of an autonomous vehicle. No more details.
  • the memory is used to store a computer program; the processor is used to execute the computer program and when executing the computer program, implement the following steps:
  • the state information of the driver is acquired based on the first sensor in the car; the environmental information is collected based on the second sensor of the car; according to the state information of the driver and the environmental information, and the corresponding relationship between the driver's status information and environment information and various preset takeover urgency levels, determine the target takeover urgency level, and the target takeover urgency level is used to generate takeover prompt information instructing the driver to take over the driving control of the car ;
  • the take-over urgency changes as the environmental information changes;
  • the takeover urgency changes as the state information changes.
  • the environmental information includes: road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information and communication information.
  • the processor executes the computer program, the following steps are implemented: obtaining the concentration degree of the driver according to the state information of the driver; obtaining the suitability of the scene according to the environmental information; The degree of concentration of the driver and the appropriateness of the scene, as well as the corresponding relationship between the degree of concentration of the driver, the appropriateness of the scene and the various preset takeover urgencies, determine the target takeover urgency.
  • the higher the driver's driving proficiency the higher the driver's concentration
  • the higher the driver's concentration under the same driver's state information, the higher the automatic driving level, the The driver's concentration is higher.
  • FIG. 9 is a schematic structural diagram of an embodiment of an automobile risk determination system of the present application.
  • the system 900 includes: a memory 901 and a processor 902; the memory 901 and the processor 902 are connected through a bus.
  • the processor 902 may be a microcontroller unit, a central processing unit, or a digital signal processor, among others.
  • the memory 901 may be a Flash chip, a read-only memory, a magnetic disk, an optical disk, a U disk or a mobile hard disk, and the like.
  • the automobile risk determination system of this embodiment can execute the steps in the above automobile risk determination method.
  • relevant content please refer to the relevant content of the above automobile risk determination method, which will not be repeated here.
  • the memory is used to store a computer program; the processor is used to execute the computer program and when executing the computer program, implement the following steps:
  • the state information of the driver is obtained based on the first sensor in the car; the environmental information is collected based on the second sensor of the car, and the environmental information includes: road information, traffic equipment information, road and Temporary change information of traffic equipment, traffic participant information, climate information and communication information; according to the driver's status information and the environmental information, the risk level information of the car is generated.
  • the driver's status information includes the driver's information perception status information, the driver's information processing status information, and the driver's action execution status information.
  • the risk level information is used to instruct the car to remind the driver according to the in-vehicle prompt mode corresponding to the current risk level, and/or to instruct the car to execute a motion mode corresponding to the current risk level to reduce traffic risks, and/or Or, it is used to instruct the car to remind other traffic participants except the car according to the outside prompt mode corresponding to the current risk level.
  • the risk level of the car changes with the change of the environment information; corresponding to any value of the environment information, the risk level of the car changes with the change of the state information.
  • the processor executes the computer program, the following steps are implemented: obtaining the concentration degree of the driver according to the state information of the driver; obtaining the suitability of the scene according to the environmental information; The driver's concentration and the suitability of the scene generate the risk level information of the car.
  • the driver's status information includes the driver's information perception status information, the driver's information processing status information, and the driver's action execution status information
  • the processor executes the computer program , the following steps are implemented: the driver’s information perception state information, the driver’s information processing state information and the driver’s action execution state information are respectively normalized to obtain the corresponding driver’s
  • the normalized value of the information perception state, the normalized value of the driver information processing state, and the normalized value of the driver's action execution state according to the normalized value of the driver information perception state, the driver information Process the normalized value of the state, the normalized value of the driver's action execution state and their respective weights to obtain the driver's concentration expressed by the normalized value.
  • the driver's information perception state information includes the driver's eye-off time, the longer the eye-off time, the smaller the normalized value of the driver's information perception state;
  • the state information includes the degree of participation of the driver in tasks other than the driving task and the degree of interaction between the driver and other passengers, the more the degree of participation, the smaller the normalized value of the driver information processing state, The more the degree of interaction, the smaller the normalized value of the driver information processing state;
  • the driver's action execution state information includes the driver's hand-off time and the driver's foot-off time; The longer the hands-off time, the smaller the normalized value of the driver's action execution state, and the longer the foot-off time, the smaller the normalized value of the driver's action execution state.
  • the driver with higher driving proficiency has higher degree of concentration; under the same driver's status information, the higher the automatic driving level, the higher the driver's concentration. high.
  • the processor executes the computer program, the following steps are implemented: respectively normalizing the road information, traffic equipment information, road and traffic equipment temporary change information, traffic participant information, climate information and communication information normalized value of road information, normalized value of traffic equipment information, normalized value of road and traffic equipment temporary change information, normalized value of traffic participant information, normalized value of climate information, communication Information normalized value; according to the road information normalized value, traffic equipment information normalized value, road and traffic equipment temporary change information normalized value, traffic participant information normalized value, climate information normalized value value, the normalized value of communication information and their respective weights, and the scene suitability expressed by the normalized value is obtained.
  • the road information includes road geometry, road grade, lane line clarity, and road surface adhesion;
  • the road geometry includes curve curvature and slope gradient;
  • the traffic facility information includes traffic light status, signs;
  • Temporary change information of roads and traffic facilities includes road changes caused by road maintenance and construction, or road surface changes caused by road surface water and snow accumulation;
  • the traffic participant information includes surrounding vehicle types, pedestrians, non-motor vehicles, traffic volume, The distance between traffic participants and pedestrian density;
  • the climate information includes rainfall, smog, visibility, and humidity;
  • the communication information includes: positioning signal strength, signal connection smoothness.
  • the processor executes the computer program, the following steps are implemented: generating the risk level information of the car according to the concentration of the driver represented by the normalized value and the suitability of the scene represented by the normalized value.
  • the processor executes the computer program, the following steps are implemented: determining the concentration of the driver represented by the normalized value and the scene suitability represented by the normalized value in the ratio of driver concentration-scene suitability
  • the position in the degree two-dimensional model, the two-dimensional model includes a plurality of regions, and the plurality of regions correspond to a plurality of different risk levels; determine the region of the position in the two-dimensional model; according to the corresponding The risk level of the car is generated to generate the risk level information of the car.
  • the two-dimensional model is divided into multiple regions by arc lines.
  • the processor executes the computer program, the following step is implemented: issuing corresponding warning information according to the risk level information, wherein the warning information corresponding to different risk level information is different.
  • the processor executes the computer program, the following steps are implemented: according to the risk level information of the car, remind the driver according to the in-vehicle prompt mode corresponding to the current risk level, and/or, control the car Executing the exercise mode corresponding to the current risk level to reduce the traffic risk, and/or prompting other traffic participants except the car according to the outside prompt mode corresponding to the current risk level.
  • the warning information is used to perform one or more of the following operations:
  • generating warning information for instructing the driver to take over the car according to the risk level information includes:
  • An early warning signal is generated based on the target early warning information level.
  • the plurality of early warning signal levels include one or more of the following early warning signals:
  • the first-level early warning signal uses a combination of high-urgency auditory signal, visual signal and tactile signal to remind the driver to take over the driving right of the vehicle urgently;
  • the second-level early warning signal uses a combination of visual signals and auditory signals with moderate urgency to remind the driver to take over the driving right of the vehicle urgently;
  • the third-level early warning signal uses a combination of visual signal and auditory signal with higher comfort level to remind the driver to take over the driving right of the vehicle or improve driving concentration.
  • controlling the vehicle to execute a motion mode corresponding to the current risk level to reduce traffic risks includes performing one or more of the following operations:
  • the vehicle is controlled to decelerate quickly and stop in the lane urgently.
  • prompting other traffic participants other than the car according to the prompt mode outside the vehicle corresponding to the current risk level includes performing one or more of the following operations:
  • the third level warning signal outside the vehicle is used to remind other traffic participants;
  • the first level warning signal outside the vehicle is used to remind other traffic participants.
  • the risk level of the automobile includes three levels, and when the processor executes the computer program, the following steps are implemented: if the current risk level is the third level, then adopt the third level early warning signal in the vehicle to remind the driver; if the current risk level is the second level, then take the second level early warning signal in the car to remind the driver; if the current risk level is the first level, then take the first level warning signal in the car The primary warning signal reminds the driver.
  • the first-level early warning signal in the car uses a combination of high-urgency auditory signal, visual signal and tactile signal to remind the driver to take over the driving right of the vehicle;
  • the second-level early warning signal in the car uses an urgent A combination of visual signals and auditory signals with a moderate degree of early warning reminds the driver to take over the driving right of the vehicle in an emergency; The driver takes over the control of the vehicle or improves driving concentration.
  • the tactile signal in the first-level early warning signal in the vehicle is the vibration signal of the driver's seat
  • the auditory signal in the first-level early warning signal in the vehicle is an audio signal with high frequency and high peak value.
  • the visual signal in the first-level early warning signal is a highlighted and urgent information or symbol displayed on the center console screen
  • the auditory signal in the second-level early warning signal in the car is an audio signal with moderate comfort
  • the vehicle The visual signal in the second-level early warning signal in the car is information or symbols with moderate brightness displayed on the screen of the center console
  • the auditory signal in the third-level early warning signal in the car is an audio signal with better comfort
  • the The visual signal in the third-level early warning signal in the car is information or symbols with comfortable brightness displayed on the screen of the center console.
  • the risk level of the car includes three levels, and the processor implements the following steps when executing the computer program: if the current risk level is the third level, then control the car to slow down; If the current risk level is the second level, control the car to slow down, change lanes slowly and pull over; parking.
  • the risk level of the automobile includes three levels, and when the processor executes the computer program, the following steps are implemented: if the current risk level is the third level, the third level of early warning signal outside the vehicle is used to warn other Traffic participants are reminded; if the current risk level is the second level, the second level warning signal outside the vehicle is used to remind other traffic participants; if the current risk level is the first level, the first level warning signal outside the vehicle is used The first-level early warning signal reminds other traffic participants.
  • the first-level early warning signal outside the vehicle uses a combination of high-urgency auditory signal and visual signal to remind other traffic participants;
  • the second-level early warning signal outside the vehicle uses a moderate urgency visual signal and auditory signal
  • the combined early warning mode reminds other traffic participants;
  • the first-level early warning signal outside the vehicle uses a combination of visual signals and auditory signals with a higher degree of comfort to remind other traffic participants.
  • the auditory signal in the first-level early warning signal outside the vehicle is a high-frequency and high-peak audio signal
  • the visual signal in the first-level early warning signal outside the vehicle is a highlighted emergency signal displayed on a display screen outside the vehicle.
  • the auditory signal in the second-level warning signal outside the vehicle is an audio signal with moderate comfort
  • the visual signal in the second-level warning signal outside the vehicle is a moderate brightness displayed on the display screen outside the vehicle information or symbols
  • the auditory signal in the third-level warning signal outside the vehicle is an audio signal with better comfort
  • the visual signal in the third-level warning signal outside the vehicle is the brightness displayed on the screen of the center console Cozy message or symbol.
  • the warning information may vary according to different demands.
  • the warning information can be self-learned based on the characteristics of the driver and then adaptively set; and/or, the warning information can be set based on the driver's customization; and/or, the warning information can be Varies with autopilot level.
  • the present application also provides an automatic driving vehicle, the vehicle includes a first sensor, a second sensor, and any one of the automatic driving vehicle takeover warning systems described above.
  • the present application also provides another automatic driving vehicle, the vehicle includes a first sensor, a second sensor, and any of the automatic driving vehicle takeover urgency judging systems described above.
  • the present application also provides an automobile, which includes a first sensor, a second sensor, and any risk determination system for the automobile described above.
  • the present application also provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor realizes the above-mentioned self-driving car takeover warning method.
  • the relevant content please refer to the relevant content above, and will not repeat it here.
  • the computer-readable storage medium may be an internal storage unit of the above-mentioned self-driving car takeover warning system, such as a hard disk or a memory.
  • the computer-readable storage medium can also be an external storage device, such as a plug-in hard disk provided, a smart memory card, a secure digital card, a flash memory card, and the like.
  • the present application also provides another computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor realizes the self-driving car takeover as described above.
  • Urgency Judgment Method For a detailed description of the relevant content, please refer to the relevant content above, and will not repeat it here.
  • the computer-readable storage medium may be an internal storage unit of the above-mentioned self-driving car takeover urgency determination system, such as a hard disk or a memory.
  • the computer-readable storage medium can also be an external storage device, such as a plug-in hard disk provided, a smart memory card, a secure digital card, a flash memory card, and the like.
  • the present application also provides yet another computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor realizes the risk determination of an automobile as described above. method.
  • the relevant content please refer to the relevant content above, and will not repeat it here.
  • the computer-readable storage medium may be an internal storage unit of the above-mentioned automobile risk determination system, such as a hard disk or a memory.
  • the computer-readable storage medium can also be an external storage device, such as a plug-in hard disk provided, a smart memory card, a secure digital card, a flash memory card, and the like.

Abstract

Procédé de détermination d'urgence de prise en charge et procédé d'avertissement de prise en charge pour un véhicule autonome, procédé et système de détermination de risque pour le véhicule, et véhicule et support d'enregistrement. Le procédé d'avertissement de prise en charge consiste : lorsqu'un véhicule est dans un état autonome, à acquérir des informations d'état d'un conducteur sur la base d'un premier capteur dans le véhicule (S101, S201) ; à collecter des informations d'environnement sur la base d'un second capteur du véhicule, les informations d'environnement comprenant des informations routières, des informations de dispositif de trafic, des informations de changement temporaire de route et de dispositif de trafic, des informations de participants au trafic, des informations météorologiques et des informations de communication (S102, S202) ; à générer des informations de niveau de risque du véhicule en fonction des informations d'état du conducteur, et des informations d'environnement (S103, S203) ; et en fonction des informations de niveau de risque, à générer des informations d'avertissement utilisées pour ordonner au conducteur de prendre en charge le véhicule, des informations d'avertissement correspondant à différents éléments d'informations de niveau de risque étant différentes (S204).
PCT/CN2021/133666 2021-11-26 2021-11-26 Procédé de détermination d'urgence de prise en charge et procédé d'avertissement de prise en charge pour véhicule autonome, et système WO2023092490A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2021/133666 WO2023092490A1 (fr) 2021-11-26 2021-11-26 Procédé de détermination d'urgence de prise en charge et procédé d'avertissement de prise en charge pour véhicule autonome, et système
CN202180100555.1A CN117642321A (zh) 2021-11-26 2021-11-26 自动驾驶汽车接管紧迫度判别方法和警示方法及系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/133666 WO2023092490A1 (fr) 2021-11-26 2021-11-26 Procédé de détermination d'urgence de prise en charge et procédé d'avertissement de prise en charge pour véhicule autonome, et système

Publications (1)

Publication Number Publication Date
WO2023092490A1 true WO2023092490A1 (fr) 2023-06-01

Family

ID=86538623

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/133666 WO2023092490A1 (fr) 2021-11-26 2021-11-26 Procédé de détermination d'urgence de prise en charge et procédé d'avertissement de prise en charge pour véhicule autonome, et système

Country Status (2)

Country Link
CN (1) CN117642321A (fr)
WO (1) WO2023092490A1 (fr)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109933062A (zh) * 2017-12-15 2019-06-25 百度（美国）有限责任公司 自动驾驶车辆的报警系统
US20200216080A1 (en) * 2019-01-04 2020-07-09 Byton North America Corporation Detecting and diagnosing anomalous driving behavior using driving behavior models
CN112141119A (zh) * 2020-09-23 2020-12-29 上海商汤临港智能科技有限公司 智能驾驶控制方法及装置、车辆、电子设备和存储介质
CN112158204A (zh) * 2020-09-30 2021-01-01 重庆长安汽车股份有限公司 一种l2级自动驾驶车辆接管报警系统及方法
CN112561170A (zh) * 2020-12-18 2021-03-26 青岛未来网络创新技术有限公司 一种智能驾驶汽车用交通安全风险动态预测方法及其预测系统
CN113386786A (zh) * 2021-07-29 2021-09-14 阿波罗智联(北京)科技有限公司 信息提示方法、装置、设备、介质、云控平台和车辆

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109933062A (zh) * 2017-12-15 2019-06-25 百度（美国）有限责任公司 自动驾驶车辆的报警系统
US20200216080A1 (en) * 2019-01-04 2020-07-09 Byton North America Corporation Detecting and diagnosing anomalous driving behavior using driving behavior models
CN112141119A (zh) * 2020-09-23 2020-12-29 上海商汤临港智能科技有限公司 智能驾驶控制方法及装置、车辆、电子设备和存储介质
CN112158204A (zh) * 2020-09-30 2021-01-01 重庆长安汽车股份有限公司 一种l2级自动驾驶车辆接管报警系统及方法
CN112561170A (zh) * 2020-12-18 2021-03-26 青岛未来网络创新技术有限公司 一种智能驾驶汽车用交通安全风险动态预测方法及其预测系统
CN113386786A (zh) * 2021-07-29 2021-09-14 阿波罗智联(北京)科技有限公司 信息提示方法、装置、设备、介质、云控平台和车辆

Also Published As

Publication number Publication date
CN117642321A (zh) 2024-03-01

Similar Documents

Publication Publication Date Title
JP6690649B2 (ja) 情報処理装置、情報処理方法、およびプログラム
US9852625B2 (en) Method and system for providing a tutorial message to a driver of a vehicle
CN104730949B (zh) 自主车辆中的情感型用户界面
US11491994B2 (en) Systems and methods for detecting and dynamically mitigating driver fatigue
US10636301B2 (en) Method for assisting operation of an ego-vehicle, method for assisting other traffic participants and corresponding assistance systems and vehicles
CN111353471A (zh) 安全驾驶监测方法、装置、设备和可读存储介质
JP2019510677A (ja) 運転者のルールベース型支援のための制御データ作成方法
US20140125474A1 (en) Adaptive actuator interface for active driver warning
EP2138987A1 (fr) Procédé pour déterminer une propriété d'un état de conducteur/véhicule/environnement
JP2016145016A (ja) 車間マネジメント装置及び車間マネジメント方法
CN109859509A (zh) 车道通行状态提醒方法及设备
JP5712712B2 (ja) 車載機能制御装置
EP2163450B1 (fr) Procédé pour autoriser la suppression d'une requête pour la présentation d'informations à un utilisateur
EP2138988A1 (fr) Procédé pour déterminer une valeur de demande de conduite
EP4140795A1 (fr) Assistant de transfert pour les transitions entre machine et conducteur
JP2021163345A (ja) 運転支援装置及びデータ収集システム
JP2010182085A (ja) 運転行動誘導システム
CN112330972A (zh) 车联关键信息分类预警方法及车载装置
CN111951548A (zh) 一种车辆驾驶风险确定方法、装置、系统及介质
US20180022357A1 (en) Driving recorder system
WO2023092490A1 (fr) Procédé de détermination d'urgence de prise en charge et procédé d'avertissement de prise en charge pour véhicule autonome, et système
JP2018097479A (ja) 運転支援装置、運転支援方法、運転支援プログラム、及び運転支援システム
CN111717221B (zh) 自动驾驶接管风险评估和人机友好预警方法及预警系统
CN113428176B (zh) 无人车驾驶策略的调整方法、装置、设备和存储介质
WO2023092611A1 (fr) Procédé et appareil de diffusion d'informations, procédé et appareil d'indication d'état de circulation, et véhicule

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: 21965224

Country of ref document: EP

Kind code of ref document: A1