CN116923397A - Driving risk prediction system and method for coupling people and vehicles - Google Patents
Driving risk prediction system and method for coupling people and vehicles Download PDFInfo
- Publication number
- CN116923397A CN116923397A CN202310929565.7A CN202310929565A CN116923397A CN 116923397 A CN116923397 A CN 116923397A CN 202310929565 A CN202310929565 A CN 202310929565A CN 116923397 A CN116923397 A CN 116923397A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- risk
- driving
- collision
- early warning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000008878 coupling Effects 0.000 title claims abstract description 27
- 238000010168 coupling process Methods 0.000 title claims abstract description 27
- 238000005859 coupling reaction Methods 0.000 title claims abstract description 27
- 238000012502 risk assessment Methods 0.000 claims abstract description 54
- 230000008859 change Effects 0.000 claims description 15
- 238000005314 correlation function Methods 0.000 claims description 14
- 230000001133 acceleration Effects 0.000 claims description 12
- 238000001514 detection method Methods 0.000 claims description 6
- 238000010223 real-time analysis Methods 0.000 claims description 2
- 206010039203 Road traffic accident Diseases 0.000 abstract description 9
- 230000006870 function Effects 0.000 description 7
- 230000006399 behavior Effects 0.000 description 6
- 230000002787 reinforcement Effects 0.000 description 6
- 230000009471 action Effects 0.000 description 5
- 238000011161 development Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000007423 decrease Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/095—Predicting travel path or likelihood of collision
- B60W30/0956—Predicting travel path or likelihood of collision the prediction being responsive to traffic or environmental parameters
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/095—Predicting travel path or likelihood of collision
- B60W30/0953—Predicting travel path or likelihood of collision the prediction being responsive to vehicle dynamic parameters
-
- 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
-
- 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
- B60W2554/00—Input parameters relating to objects
- B60W2554/40—Dynamic objects, e.g. animals, windblown objects
- B60W2554/402—Type
- B60W2554/4029—Pedestrians
-
- 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
- B60W2554/00—Input parameters relating to objects
- B60W2554/80—Spatial relation or speed relative to objects
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Human Computer Interaction (AREA)
- Traffic Control Systems (AREA)
Abstract
The application provides a method and a system for predicting the driving risk of man-vehicle road coupling, comprising the following steps: step S1: analyzing risk characteristics of a following scene and a lane changing scene; step S2: constructing a driving risk assessment method of the coupling of the person and the vehicle, analyzing driving collision points and predicting an uncertainty risk area; step S3: and constructing a driving safety early warning system based on the uncertainty risk analysis result. The application realizes accurate assessment, prediction and early warning of the traffic risk situation, and lays a foundation for improving traffic safety and reducing traffic accident rate.
Description
Technical Field
The application relates to the technical field of intelligent driving, in particular to a driving risk prediction system and method for coupling people and vehicles.
Background
With the steady and rapid development of the macroscopic economy in China and the remarkable improvement of the income level of residents, automobiles become one of the most commonly used travel tools for people. The rapid increase of the quantity of automobiles improves the efficiency and the comfort of the traveling of people in China, but also causes a large number of traffic accidents, threatens the traveling safety of people and brings great economic loss to the China. The national highway traffic safety administration reports that 42,939 people die from motor vehicle accidents on the road in the united states in 2021, which is 10% more than 39,007 people in 2020. Road traffic accidents have been one of the biggest and most serious public safety problems in the world.
And the development of intelligent driving technology is expected to reduce traffic accident rate and improve driving safety. Unmanned vehicles with high safety performance are receiving increasing attention with their potential advantages of adapting to complex scenes, improving economy and increasing traffic. The driving risk prediction system is one of key technologies for ensuring safe driving of the intelligent automobile. The technology can analyze and evaluate traffic conditions in real time so as to predict potential dangers in advance and enable vehicles to respond correspondingly, thereby ensuring the safety of drivers and other pedestrians on roads. Specifically, the driving risk prediction system can collect data through various sensors, identify and analyze various characteristics and factors on the road, such as the positions, actions, speeds, directions and the like of vehicles and pedestrians, so that the system can monitor the changes of the factors in real time, predict the flow, traffic conditions, road conditions and the like in front in time, and avoid the occurrence of potential dangerous situations.
Patent document CN110390451A (application number: 201810347997.6) discloses a road traffic safety risk prediction and early warning index system for solving the problem of road traffic safety risk prediction and early warning. The key points of the technical proposal are as follows: a road driving safety risk prediction early warning index classification method; a road environment monitoring index classification method and standard; a road condition monitoring index classification method and standard; a traffic condition monitoring index classification method and a standard; a classification method and a standard for vehicle condition monitoring indexes are provided, which are road driving safety risk level evaluation indexes.
In conclusion, the driving risk prediction system has very important significance in intelligent driving, can effectively improve driving safety, reduce accident risk, optimize driving experience and promote intelligent development of traffic.
Disclosure of Invention
Aiming at the defects in the prior art, the application aims to provide a driving risk prediction system and method for man-vehicle road coupling.
The application provides a driving risk prediction method for man-vehicle road coupling, which comprises the following steps:
step S1: analyzing risk characteristics of a following scene and a lane changing scene;
step S2: constructing a driving risk assessment method of the coupling of the person and the vehicle, analyzing driving collision points and predicting an uncertainty risk area;
step S3: and constructing a driving safety early warning system based on the uncertainty risk analysis result.
Preferably, the risk features of the following scene relate to the relative position, speed and acceleration of the preceding vehicle;
the risk characteristics of the lane change scenario are related to the driving status of the preceding vehicle, the other vehicle in the lane.
Preferably, the step S2 employs: and analyzing risk influence factors including the driving states of the unexpected pedestrians and Zhou Chejia and the road environment, analyzing the probability of occurrence of unexpected pedestrian collision accidents, other vehicle collision accidents and skidding and rollover accidents, constructing a driving risk assessment method of coupling the pedestrians and the vehicles, and predicting an uncertain risk area.
Preferably, the occurrence of the unexpected pedestrian collision accident probability includes:
R p =f 1 (L b )
L b =L 1 tanθ-L 2 -L 3
wherein R is p Representing a probability of collision with an unexpected pedestrian; blind area L b F is formed into 1 Positive correlation; θ represents an included angle between a connecting line of a head center point and a left rearview mirror of a front car and a vertical direction; l (L) 1 Representing the vertical distance between the center point of the vehicle and the pedestrian; l (L) 2 Representing the vertical distance between the own vehicle and the front vehicle; l (L) 3 Indicating the length of the front vehicle;
the collision accident probability of the other vehicle comprises:
Δv=v i (t)-v i-1 (t)
D=x i (t)-x i-1 (t)-l i-1
R v =f 3 (TTC)
wherein TTC represents a collision time; d represents twoThe relative distance of the vehicle; deltav represents the speed difference between the rear vehicle and the front vehicle; l (L) i-1 Is the length of the front vehicle; x is x i-1 (t) represents the position of the vehicle before time t; x is x i (t) represents the vehicle position after time t; v i-1 (t) represents the speed of the vehicle before time t; v i (t) represents the vehicle speed after time t; r is R v A collision possibility indicating Zhou Chejia driving status; f (f) 3 Representing a negative correlation function;
the probability of the slip rollover accident comprises:
R r =f 4 (r)
wherein r represents a curve radius; f (f) 4 Representing a negative correlation function.
Preferably, the step S3 employs: based on an uncertainty risk detection result in an actual driving environment, a driving scene that the uncertainty probability meets a preset value is analyzed, a road section and a driving time period where a possible accident occurs are determined, and early warning explanation is made for the possible specific risk accident.
Preferably, the unexpected pedestrian risk assessment result R is analyzed in real time p Zhou Chejia Driving risk assessment result R v Road environment risk assessment result R v And respectively with the set threshold valueComparing, judging that the driving is safe and no early warning exists when the risk assessment result is within the threshold value; if the risk assessment result exceeds the threshold value, sending out an early warning and giving out the type of the early warning, wherein the method comprises the following steps: unexpected pedestrian collision early warning, other vehicle collision accident early warning, skid rollover accident early warning and specific indexes (R) p ,R v ,R r )。
The application provides a driving risk prediction system for man-vehicle road coupling, which comprises the following components:
module M1: analyzing risk characteristics of a following scene and a lane changing scene;
module M2: constructing a driving risk assessment method of the coupling of the person and the vehicle, analyzing driving collision points and predicting an uncertainty risk area;
module M3: and constructing a driving safety early warning system based on the uncertainty risk analysis result.
Preferably, the risk features of the following scene relate to the relative position, speed and acceleration of the preceding vehicle;
the risk characteristics of the lane change scenario are related to the driving status of the preceding vehicle, the other vehicle in the lane.
Preferably, the module M2 employs: analyzing risk influence factors including the states of unexpected pedestrians and Zhou Chejia and road environments, analyzing the probability of unexpected pedestrian collision accidents, other vehicle collision accidents and skidding and rollover accidents, constructing a vehicle-vehicle coupling driving risk assessment method, and predicting an uncertain risk area;
the occurrence of the unexpected pedestrian collision accident probability comprises:
R p =f 1 (L b )
L b =L 1 tanθ-L 2 -L 3
wherein R is p Representing a probability of collision with an unexpected pedestrian; blind area L b F is formed into 1 Positive correlation; θ represents an included angle between a connecting line of a head center point and a left rearview mirror of a front car and a vertical direction; l (L) 1 Representing the vertical distance between the center point of the vehicle and the pedestrian; l (L) 2 Representing the vertical distance between the own vehicle and the front vehicle; l (L) 3 Indicating the length of the front vehicle;
the collision accident probability of the other vehicle comprises:
Δv=v i (t)-v i-1 (t)
D=x i (t)-x i-1 (t)-l i-1
R v =f 3 (TTC)
wherein TTC represents a collision time; d represents the relative distance between two vehicles; deltav represents the speed difference between the rear vehicle and the front vehicle; l (L) i-1 Is the length of the front vehicle; x is x i-1 (t) represents the position of the vehicle before time t; x is x i (t) represents the parking space after the time tPlacing; v i-1 (t) represents the speed of the vehicle before time t; v i (t) represents the vehicle speed after time t; r is R v A collision possibility indicating Zhou Chejia driving status; f (f) 3 Representing a negative correlation function;
the probability of the slip rollover accident comprises:
R r =f 4 (r)
wherein r represents a curve radius; f (f) 4 Representing a negative correlation function.
Preferably, the module M3 employs: based on an uncertainty risk detection result in an actual driving environment, analyzing a driving scene in which the uncertainty probability meets a preset value, determining a road section and a driving time period in which a possible accident occurs, and making early warning explanation for the possible specific risk accident;
real-time analysis of unexpected pedestrian risk assessment result R p Zhou Chejia Driving risk assessment result R v Road environment risk assessment result R v And respectively with the set threshold valueComparing, judging that the driving is safe and no early warning exists when the risk assessment result is within the threshold value; if the risk assessment result exceeds the threshold value, sending out an early warning and giving out the type of the early warning, wherein the method comprises the following steps: unexpected pedestrian collision early warning, other vehicle collision accident early warning, skid rollover accident early warning and specific indexes (R) p ,R v ,R r )。
Compared with the prior art, the application has the following beneficial effects:
1. the application provides a driving risk prediction method for coupling a person and a vehicle, namely, by analyzing risk characteristics of a following scene and a lane changing scene, researching risk influence factors of unexpected pedestrians, zhou Chejia driving states and road environments, analyzing unexpected pedestrian collision accidents, other vehicle collision accidents, slipping rollover accidents and the like which possibly occur, and making corresponding early warning descriptions. Not only is the real-time performance good, but also the driving safety is improved;
2. the application effectively improves the driving safety, reduces accident risk, optimizes driving experience and promotes intelligent development of traffic.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the detailed description of non-limiting embodiments, given with reference to the accompanying drawings in which:
fig. 1 is a schematic flow chart of a traffic risk prediction system for coupling people and vehicles.
Fig. 2 is a schematic view of long distance following driving.
Fig. 3 is a schematic view of a short distance following drive.
Fig. 4 is a schematic view of a lane change scenario driving.
Fig. 5 is a schematic diagram of an unexpected pedestrian scenario.
Fig. 6 is a schematic diagram of TTC.
Fig. 7 is a schematic diagram of a driving risk prediction system.
Detailed Description
The present application will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the present application, but are not intended to limit the application in any way. It should be noted that variations and modifications could be made by those skilled in the art without departing from the inventive concept. These are all within the scope of the present application.
Example 1
As shown in fig. 1, the application discloses a driving risk prediction method for coupling people and vehicles, comprising the following steps:
step S1: analyzing risk characteristics of a following scene and a lane changing scene;
step S2: constructing a driving risk assessment method of the coupling of the person and the vehicle, analyzing driving collision points and predicting an uncertainty risk area;
step S3: and constructing a driving safety early warning system based on the uncertainty risk analysis result.
Specifically, the step S1 includes:
the following scenes mainly analyze long-distance following scenes and short-distance following scenes, and risk characteristics of the following scenes are mainly related to the relative position, speed and acceleration of a front vehicle; in the lane change scenario, the risk features are related to the driving status of the preceding vehicle, which relates to the other vehicle in the lane.
It should be noted that, as shown in fig. 2, in the long-distance following driving scenario, since the vehicle driving in the rear is far away from the vehicle in front, the driving behavior of the vehicle driving in front does not affect the rear vehicle, and since the vehicle driving in the rear has enough time to make a corresponding decision to avoid collision no matter whether the vehicle driving in front makes any driving behavior such as acceleration, deceleration or lane change.
Short-range, on-board scenes are the focus of research, as the driving scene is the most common and one of the most likely to be involved in collision accidents. As shown in fig. 3, in a short-distance following scene, particularly when the front vehicle speed is lower than the rear vehicle speed, it is necessary for the rear vehicle driver to pay attention to the driving operation of the front vehicle at all times. Irrespective of the acceleration, the distance between the two vehicles gradually decreases, and the probability of collision between the two vehicles increases. Of course, even in the case where the speed of the front vehicle is lower than that of the rear vehicle, attention is paid to the driving operation of the front vehicle in order to prevent the front vehicle from being braked urgently to cause collision of the two vehicles. Therefore, in the short-distance following driving scene, the relative distance and the relative speed of two vehicles and the acceleration of the two vehicles are important to consider.
The lane changing scene is more complex than the following scene because the lane changing vehicle does not just move longitudinally but is coupled longitudinally and longitudinally. As shown in fig. 4, not only attention is paid to not collide with the front vehicle but also the driving state of the rear vehicle is estimated in the course of lane change to avoid collision. Under the channel changing scene, the positions of the basically equal-height risk ellipsoids also change correspondingly due to the change of the vehicle body posture, and the interactive vehicles are relatively more, and the considered influence factors are relatively more.
Specifically, the step S2 includes:
and analyzing risk influence factors of unexpected pedestrian and Zhou Chejia driving states and road environments, analyzing unexpected pedestrian collision accidents, other vehicle collision accidents, slipping rollover accidents and the like which possibly occur, constructing a driving risk assessment method of human-vehicle road coupling, and predicting an uncertain risk area.
It should be noted that an unexpected pedestrian refers to an unexpected pedestrian that appears under the undetectable area. As shown in fig. 5, a truck is parked on the right side of the road in front of the right side of the vehicle, and it is difficult for the vehicle to find the pedestrian in front of the right side of the truck and its behavior to traverse the road. The own vehicle travels over an area where no pedestrian is detected, and there is a possibility of collision with the pedestrian. If the speed of the vehicle is high, the situation is worse. In addition, as the yellow car advances, pedestrians are easier to find, and the judgment on whether pedestrians cross the road is more accurate. The risk probability of an unexpected pedestrian can thus be estimated based on the following formula:
R p =f 1 (L b )
L b =L 1 tanθ-L 2 -L 3
wherein R is p Representing collision probability with unexpected pedestrians and vision blind area L b F is formed into 1 Positive correlation, L 1 ,L 2 ,L 3 As shown in fig. 5, and is available through information interaction during intelligent driving.
The risk of Zhou Chejia driving state can be evaluated in combination with the existing study, such as the time to collision TTC, as shown in fig. 6, and the remaining time for the collision between the front vehicle (i-1 vehicle) and the own vehicle (i vehicle) to occur when the vehicle is driven at the current speed in the following scene, specifically shown in the following formula:
Δv=v i (t)-v i-1 (t)
D=x i (t)-x i-1 (t)-l i-1
wherein l i-1 Is the length of the front vehicle, D is the relative distance between the two vehicles, and Δv is the speed difference between the rear vehicle and the front vehicle. Therefore, risk assessment is safe under long-distance following scenes, corresponding TTC values need to be calculated in real time under short-distance following scenes, a safe TTC threshold value is determined, and collision possibility is calculated. The greater the TTC, the lower the possibility of collision, so Zhou Chejia driving statePossibility of collision R v Can be expressed as follows:
R v =f 3 (TTC)
wherein f 3 Is a negative correlation function.
The risk of road environment is focused on the curve driving environment, because a curve is one of traffic scenes in which traffic accidents are extremely easy to occur. The vehicle running on the curve is subjected to the action of centrifugal force, the magnitude of the centrifugal force is closely related to the radius of the flat curve corresponding to the curve, the larger the radius of the flat curve is, the smaller the centrifugal force is, the slower the running direction changes, and the lower the running risk is; the smaller the radius, the greater the centrifugal force, the faster the direction of travel changes and the higher the risk of the vehicle tipping over. Therefore, the probability R of vehicle rollover risk in road environment r Can be expressed as follows:
R r =f 4 (r)
wherein f 4 R is the curve radius, which is a negative correlation function.
It should be further noted that the related correlation function f 1 ,f 2 ,f 3 ,f 4 Etc. may be obtained by reinforcement learning. First, the environment is an external system (driving environment) in which an agent (traffic participant) is able to perceive the risk profile of the system and to take certain actions based on the perceived risk profile. After initializing the function, calculating the risk distribution in the current driving environment, selecting a low risk position by a traffic participant in the driving environment, adopting corresponding driving operations such as holding, accelerating, decelerating, turning left, turning right and the like, obtaining the next state, judging whether traffic accidents such as collision exist in the state, backtracking the result if the accidents happen, and carrying out weight adjustment, otherwise, continuing simulation. If no accident occurs in the scene, feedback is rewarded, and risk accidents such as collision occur, feedback is punishment, and punishment strength is determined based on sensitivity and severity of the accidents. Through multiple reinforcement learning training, the set termination point is reached, the function is set reasonably, and by adopting the function, the intelligent driving vehicle can obtain accurate and reasonable risk assessment results when various driving environment risk factors are considered.
Specifically, the step S3 includes:
as shown in fig. 7, based on the uncertainty risk detection result in the actual driving environment, a driving scene with large uncertainty is analyzed, a road section and a driving time period where a possible accident occurs are determined, and early warning explanation is specifically made for specific risk accidents which may occur, such as unexpected pedestrian collision accidents, other vehicle collision accidents, skid rollover accidents, and the like.
It should be noted that, first, based on the description of step S2, the unexpected pedestrian risk assessment result R is analyzed in real time p Zhou Chejia Driving risk assessment result R v Road environment risk assessment result R v And respectively with the set threshold valueComparing, the threshold value is obtained through reinforcement learning training in the step S2, and aiming at the risk assessment result in the threshold value, the safety driving is judged without any early warning, if the risk assessment result exceeds the threshold value, the early warning is sent out, and the type of early warning (unexpected pedestrian collision early warning, other vehicle collision accident early warning, skid rollover accident early warning) and the specific index (R) are given p ,R v ,R r )。
Example 2
Example 2 is a preferred example of example 1
As shown in fig. 1, the application discloses a driving risk prediction method for coupling people and vehicles, comprising the following steps:
module M1: analyzing risk characteristics of a following scene and a lane changing scene;
module M2: constructing a driving risk assessment method of the coupling of the person and the vehicle, analyzing driving collision points and predicting an uncertainty risk area;
module M3: and constructing a driving safety early warning system based on the uncertainty risk analysis result.
Specifically, the module M1 includes:
the following scenes mainly analyze long-distance following scenes and short-distance following scenes, and risk characteristics of the following scenes are mainly related to the relative position, speed and acceleration of a front vehicle; in the lane change scenario, the risk features are related to the driving status of the preceding vehicle, which relates to the other vehicle in the lane.
It should be noted that, as shown in fig. 2, in the long-distance following driving scenario, since the vehicle driving in the rear is far away from the vehicle in front, the driving behavior of the vehicle driving in front does not affect the rear vehicle, and since the vehicle driving in the rear has enough time to make a corresponding decision to avoid collision no matter whether the vehicle driving in front makes any driving behavior such as acceleration, deceleration or lane change.
Short-range, on-board scenes are the focus of research, as the driving scene is the most common and one of the most likely to be involved in collision accidents. As shown in fig. 3, in a short-distance following scene, particularly when the front vehicle speed is lower than the rear vehicle speed, it is necessary for the rear vehicle driver to pay attention to the driving operation of the front vehicle at all times. Irrespective of the acceleration, the distance between the two vehicles gradually decreases, and the probability of collision between the two vehicles increases. Of course, even in the case where the speed of the front vehicle is lower than that of the rear vehicle, attention is paid to the driving operation of the front vehicle in order to prevent the front vehicle from being braked urgently to cause collision of the two vehicles. Therefore, in the short-distance following driving scene, the relative distance and the relative speed of two vehicles and the acceleration of the two vehicles are important to consider.
The lane changing scene is more complex than the following scene because the lane changing vehicle does not just move longitudinally but is coupled longitudinally and longitudinally. As shown in fig. 4, not only attention is paid to not collide with the front vehicle but also the driving state of the rear vehicle is estimated in the course of lane change to avoid collision. Under the channel changing scene, the positions of the basically equal-height risk ellipsoids also change correspondingly due to the change of the vehicle body posture, and the interactive vehicles are relatively more, and the considered influence factors are relatively more.
Specifically, the module M2 includes:
and analyzing risk influence factors of unexpected pedestrian and Zhou Chejia driving states and road environments, analyzing unexpected pedestrian collision accidents, other vehicle collision accidents, slipping rollover accidents and the like which possibly occur, constructing a driving risk assessment method of human-vehicle road coupling, and predicting an uncertain risk area.
It should be noted that an unexpected pedestrian refers to an unexpected pedestrian that appears under the undetectable area. As shown in fig. 5, a truck is parked on the right side of the road in front of the right side of the vehicle, and it is difficult for the vehicle to find the pedestrian in front of the right side of the truck and its behavior to traverse the road. The own vehicle travels over an area where no pedestrian is detected, and there is a possibility of collision with the pedestrian. If the speed of the vehicle is high, the situation is worse. In addition, as the yellow car advances, pedestrians are easier to find, and the judgment on whether pedestrians cross the road is more accurate. The risk probability of an unexpected pedestrian can thus be estimated based on the following formula:
R p =f 1 (L b )
L b =L 1 tanθ-L 2 -L 3
wherein R is p Representing collision probability with unexpected pedestrians and vision blind area L b F is formed into 1 Positive correlation, L 1 ,L 2 ,L 3 As shown in fig. 5, and is available through information interaction during intelligent driving.
The risk of Zhou Chejia driving state can be evaluated in combination with the existing study, such as the time to collision TTC, as shown in fig. 6, and the remaining time for the collision between the front vehicle (i-1 vehicle) and the own vehicle (i vehicle) to occur when the vehicle is driven at the current speed in the following scene, specifically shown in the following formula:
Δv=v i (t)-v i-1 (t)
D=x i (t)-x i-1 (t)-l i-1
wherein l i-1 Is the length of the front vehicle, D is the relative distance between the two vehicles, and Δv is the speed difference between the rear vehicle and the front vehicle. Therefore, risk assessment is safe under long-distance following scenes, corresponding TTC values need to be calculated in real time under short-distance following scenes, a safe TTC threshold value is determined, and collision possibility is calculated. The greater the TTC, the lower the possibility of collision, so the possibility of collision R in the Zhou Chejia driving state v Can be expressed as follows:
R v =f 3 (TTC)
wherein f 3 Is a negative correlation function.
The risk of road environment is focused on the curve driving environment, because a curve is one of traffic scenes in which traffic accidents are extremely easy to occur. The vehicle running on the curve is subjected to the action of centrifugal force, the magnitude of the centrifugal force is closely related to the radius of the flat curve corresponding to the curve, the larger the radius of the flat curve is, the smaller the centrifugal force is, the slower the running direction changes, and the lower the running risk is; the smaller the radius, the greater the centrifugal force, the faster the direction of travel changes and the higher the risk of the vehicle tipping over. Therefore, the probability R of vehicle rollover risk in road environment r Can be expressed as follows:
R r =f 4 (r)
wherein f 4 R is the curve radius, which is a negative correlation function.
It should be further noted that the related correlation function f 1 ,f 2 ,f 3 ,f 4 Etc. may be obtained by reinforcement learning. First, the environment is an external system (driving environment) in which an agent (traffic participant) is able to perceive the risk profile of the system and to take certain actions based on the perceived risk profile. After initializing the function, calculating the risk distribution in the current driving environment, selecting a low risk position by a traffic participant in the driving environment, adopting corresponding driving operations such as holding, accelerating, decelerating, turning left, turning right and the like, obtaining the next state, judging whether traffic accidents such as collision exist in the state, backtracking the result if the accidents happen, and carrying out weight adjustment, otherwise, continuing simulation. If no accident occurs in the scene, feedback is rewarded, and risk accidents such as collision occur, feedback is punishment, and punishment strength is determined based on sensitivity and severity of the accidents. Through multiple reinforcement learning training, the set termination point is reached, the function is set reasonably, and by adopting the function, the intelligent driving vehicle can obtain accurate and reasonable risk assessment results when various driving environment risk factors are considered.
Specifically, the module M3 includes:
based on an uncertainty risk detection result in an actual driving environment, a driving scene with large uncertainty is analyzed, road sections and driving time periods where possible accidents occur are determined, and early warning explanation is specifically made for possible specific risk accidents, such as unexpected pedestrian collision accidents, other vehicle collision accidents, skidding rollover accidents and the like.
It should be noted that, first, based on the description of the module M2, the unexpected pedestrian risk assessment result R is analyzed in real time p Zhou Chejia Driving risk assessment result R v Road environment risk assessment result R v And respectively with the set threshold valueComparing, the threshold value is obtained through reinforcement learning training described in the module M2, and aiming at the risk assessment result in the threshold value, the safety driving is judged without any early warning, if the risk assessment result exceeds the threshold value, the early warning is sent out, and the type of early warning (unexpected pedestrian collision early warning, other vehicle collision accident early warning, skid rollover accident early warning) and the specific index (R) are given p ,R v ,R r )。
Those skilled in the art will appreciate that the systems, apparatus, and their respective modules provided herein may be implemented entirely by logic programming of method steps such that the systems, apparatus, and their respective modules are implemented as logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc., in addition to the systems, apparatus, and their respective modules being implemented as pure computer readable program code. Therefore, the system, the apparatus, and the respective modules thereof provided by the present application may be regarded as one hardware component, and the modules included therein for implementing various programs may also be regarded as structures within the hardware component; modules for implementing various functions may also be regarded as being either software programs for implementing the methods or structures within hardware components.
The foregoing describes specific embodiments of the present application. It is to be understood that the application is not limited to the particular embodiments described above, and that various changes or modifications may be made by those skilled in the art within the scope of the appended claims without affecting the spirit of the application. The embodiments of the application and the features of the embodiments may be combined with each other arbitrarily without conflict.
Claims (10)
1. A driving risk prediction method for coupling people and vehicles is characterized by comprising the following steps:
step S1: analyzing risk characteristics of a following scene and a lane changing scene;
step S2: constructing a driving risk assessment method of the coupling of the person and the vehicle, analyzing driving collision points and predicting an uncertainty risk area;
step S3: and constructing a driving safety early warning system based on the uncertainty risk analysis result.
2. The human-vehicle coupled driving risk prediction method according to claim 1, wherein the risk characteristics of the following scene are related to the relative position, speed and acceleration of the preceding vehicle;
the risk characteristics of the lane change scenario are related to the driving status of the preceding vehicle, the other vehicle in the lane.
3. The method for predicting the risk of driving coupled to a vehicle according to claim 1, wherein the step S2 uses: and analyzing risk influence factors including the driving states of the unexpected pedestrians and Zhou Chejia and the road environment, analyzing the probability of occurrence of unexpected pedestrian collision accidents, other vehicle collision accidents and skidding and rollover accidents, constructing a driving risk assessment method of coupling the pedestrians and the vehicles, and predicting an uncertain risk area.
4. A method of predicting risk of a pedestrian to vehicle coupling as set forth in claim 3, wherein said occurrence of an unexpected pedestrian collision event probability includes:
R p =f 1 (L b )
L b =L 1 tanθ-L 2 -L 3
wherein R is p Representing a probability of collision with an unexpected pedestrian; blind area L b F is formed into 1 Positive correlation; θ represents an included angle between a connecting line of a head center point and a left rearview mirror of a front car and a vertical direction; l (L) 1 Representing the vertical distance between the center point of the vehicle and the pedestrian; l (L) 2 Representing the vertical distance between the own vehicle and the front vehicle; l (L) 3 Indicating the length of the front vehicle;
the collision accident probability of the other vehicle comprises:
Δv=v i (t)-v i-1 (t)
D=x i (t)-x i-1 (t)-l i-1
R v =f 3 (TTC)
wherein TTC represents a collision time; d represents the relative distance between two vehicles; deltav represents the speed difference between the rear vehicle and the front vehicle; l (L) i-1 Is the length of the front vehicle; x is x i-1 (t) represents the position of the vehicle before time t; x is x i (t) represents the vehicle position after time t; v i-1 (t) represents the speed of the vehicle before time t; v i (t) represents the vehicle speed after time t; r is R v A collision possibility indicating Zhou Chejia driving status; f (f) 3 Representing a negative correlation function;
the probability of the slip rollover accident comprises:
R r =f 4 (r)
wherein r represents a curve radius; f (f) 4 Representing a negative correlation function.
5. The method for predicting the risk of driving coupled to a vehicle according to claim 1, wherein the step S3 uses: based on an uncertainty risk detection result in an actual driving environment, a driving scene that the uncertainty probability meets a preset value is analyzed, a road section and a driving time period where a possible accident occurs are determined, and early warning explanation is made for the possible specific risk accident.
6. The method for predicting the risk of driving coupled to a pedestrian and a vehicle according to claim 5, wherein the result R of the risk assessment of the unexpected pedestrian is analyzed in real time p Zhou Chejia Driving risk assessment result R v Road environment risk assessment result R v And respectively with the set threshold valueComparing, judging that the driving is safe and no early warning exists when the risk assessment result is within the threshold value; if the risk assessment result exceeds the threshold value, sending out an early warning and giving out the type of the early warning, wherein the method comprises the following steps: unexpected pedestrian collision early warning, other vehicle collision accident early warning, skid rollover accident early warning and specific indexes (R) p ,R v ,R r )。
7. A traffic risk prediction system for man-vehicle road coupling, comprising:
module M1: analyzing risk characteristics of a following scene and a lane changing scene;
module M2: constructing a driving risk assessment method of the coupling of the person and the vehicle, analyzing driving collision points and predicting an uncertainty risk area;
module M3: and constructing a driving safety early warning system based on the uncertainty risk analysis result.
8. The human-vehicle coupled driving risk prediction system according to claim 7, wherein the risk characteristics of the following scene are related to the relative position, speed, and acceleration of the preceding vehicle;
the risk characteristics of the lane change scenario are related to the driving status of the preceding vehicle, the other vehicle in the lane.
9. The human-vehicle coupled driving risk prediction system according to claim 7, wherein the module M2 employs: analyzing risk influence factors including the states of unexpected pedestrians and Zhou Chejia and road environments, analyzing the probability of unexpected pedestrian collision accidents, other vehicle collision accidents and skidding and rollover accidents, constructing a vehicle-vehicle coupling driving risk assessment method, and predicting an uncertain risk area;
the occurrence of the unexpected pedestrian collision accident probability comprises:
R p =f 1 (L b )
L b =L 1 tanθ-L 2 -L 3
wherein R is p Representing a probability of collision with an unexpected pedestrian; blind area L b F is formed into 1 Positive correlation; θ represents an included angle between a connecting line of a head center point and a left rearview mirror of a front car and a vertical direction; l (L) 1 Representing the vertical distance between the center point of the vehicle and the pedestrian; l (L) 2 Representing the vertical distance between the own vehicle and the front vehicle; l (L) 3 Indicating the length of the front vehicle;
the collision accident probability of the other vehicle comprises:
Δv=v i (t)-v i-1 (t)
D=x i (t)-x i-1 (t)-l i-1
R v =f 3 (TTC)
wherein TTC represents a collision time; d represents the relative distance between two vehicles; deltav represents the speed difference between the rear vehicle and the front vehicle; l (L) i-1 Is the length of the front vehicle; x is x i-1 (t) represents the position of the vehicle before time t; x is x i (t) represents the vehicle position after time t; v i-1 (t) represents the speed of the vehicle before time t; v i (t) represents the vehicle speed after time t; r is R v A collision possibility indicating Zhou Chejia driving status; f (f) 3 Representing a negative correlation function;
the probability of the slip rollover accident comprises:
R r =f 4 (r)
wherein r represents a curve radius; f (f) 4 Representing a negative correlation function.
10. The human-vehicle coupled driving risk prediction system according to claim 7, wherein the module M3 employs: based on an uncertainty risk detection result in an actual driving environment, analyzing a driving scene in which the uncertainty probability meets a preset value, determining a road section and a driving time period in which a possible accident occurs, and making early warning explanation for the possible specific risk accident;
real-time analysis of unexpected pedestrian risk assessment result R p Zhou Chejia Driving risk assessment result R v Road environment risk assessment result R v And respectively with the set threshold valueComparing, judging that the driving is safe and no early warning exists when the risk assessment result is within the threshold value; if the risk assessment result exceeds the threshold value, sending out an early warning and giving out the type of the early warning, wherein the method comprises the following steps: unexpected pedestrian collision early warning, other vehicle collision accident early warning, skid rollover accident early warning and specific indexes (R) p ,R v ,R r )。/>
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310929565.7A CN116923397A (en) | 2023-07-25 | 2023-07-25 | Driving risk prediction system and method for coupling people and vehicles |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310929565.7A CN116923397A (en) | 2023-07-25 | 2023-07-25 | Driving risk prediction system and method for coupling people and vehicles |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116923397A true CN116923397A (en) | 2023-10-24 |
Family
ID=88390611
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310929565.7A Pending CN116923397A (en) | 2023-07-25 | 2023-07-25 | Driving risk prediction system and method for coupling people and vehicles |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116923397A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117873118A (en) * | 2024-03-11 | 2024-04-12 | 中国科学技术大学 | Storage logistics robot navigation method based on SAC algorithm and controller |
-
2023
- 2023-07-25 CN CN202310929565.7A patent/CN116923397A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117873118A (en) * | 2024-03-11 | 2024-04-12 | 中国科学技术大学 | Storage logistics robot navigation method based on SAC algorithm and controller |
CN117873118B (en) * | 2024-03-11 | 2024-05-28 | 中国科学技术大学 | Storage logistics robot navigation method based on SAC algorithm and controller |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112435474B (en) | Expressway variable speed limit control system and control method for rear-end collision risk prevention and control | |
WO2021217752A1 (en) | Vehicle-pedestrian collision risk region calculation method and safety evaluation system | |
CN110001647B (en) | Vehicle triggering lane change method, system and computer readable storage medium | |
CN110085042B (en) | Vehicle driving early warning system and method based on information fusion | |
CN108657066B (en) | Plug early warning system and plug early warning method | |
CN103093636B (en) | Expressway turning segment variable velocity-limiting control method | |
KR20140057583A (en) | Safety device for motor vehicles | |
CN109455183B (en) | Vehicle collision pre-judging method and system | |
CN116923397A (en) | Driving risk prediction system and method for coupling people and vehicles | |
CN112026761A (en) | Automobile auxiliary driving method based on data sharing | |
CN112622937A (en) | Pass right decision method for automatically driving automobile in face of pedestrian | |
CN103093635A (en) | Expressway crest vertical curve road segment variable velocity-limiting control method | |
CN113879211B (en) | Reminding method and system for preventing collision of motor vehicle with Cheng Yufei turning right of dregs car | |
WO2017220176A1 (en) | Method for determining an emergency braking situation of a vehicle and emergency brake system | |
CN106530827A (en) | Corner driving early warning device, system and method | |
CN114506317B (en) | Automatic lane changing safety processing method, device, equipment and readable storage medium | |
CN104842714A (en) | Automotive short-distance auxiliary emergency braking system and braking method thereof | |
CN201882025U (en) | Anti-collision device for automobile | |
CN204605408U (en) | A kind of automobile short range auxiliary emergent brake system | |
CN114312787A (en) | Intelligent vehicle control method for mixed traffic flow congestion working condition | |
CN113119945A (en) | Automobile advanced auxiliary driving system based on environment model | |
CN114038238B (en) | Collision early warning system and method based on running state of second vehicle in front | |
CN106515577B (en) | Device and method for reducing false alarm rate of channel change early warning | |
CN111275986B (en) | Risk decision device and method for vehicle to autonomously merge into main road in acceleration lane | |
Rongqiang et al. | Based on binocular identification technology of automobile active braking safety distance model research |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |