CN115731708A - Real-time vehicle track lane change point monitoring method based on Bayesian theory - Google Patents
Real-time vehicle track lane change point monitoring method based on Bayesian theory Download PDFInfo
- Publication number
- CN115731708A CN115731708A CN202211428816.5A CN202211428816A CN115731708A CN 115731708 A CN115731708 A CN 115731708A CN 202211428816 A CN202211428816 A CN 202211428816A CN 115731708 A CN115731708 A CN 115731708A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- time
- lane change
- change point
- moment
- 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.)
- Granted
Links
- 230000008859 change Effects 0.000 title claims abstract description 81
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000012544 monitoring process Methods 0.000 title claims abstract description 25
- 238000005070 sampling Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 abstract description 2
- 238000007477 logistic regression Methods 0.000 abstract description 2
- 230000006399 behavior Effects 0.000 description 22
- 230000006870 function Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 5
- 230000001133 acceleration Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000003137 locomotive effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000002301 combined effect Effects 0.000 description 1
- 230000004424 eye movement Effects 0.000 description 1
- 230000004886 head movement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention discloses a real-time vehicle track lane change point monitoring method based on Bayesian theory, which comprises the steps of obtaining a running track data set of a vehicle, taking generated track data and historical vehicle statistical data as prior information, considering the influence of traffic environment indexes on the lane change behavior of a driver, providing a self-adaptive risk function calculation method based on a binomial logistic regression model, and finally obtaining the vehicle at t based on the Bayesian theory n Time of day
To obtain the posterior predicted distribution of the vehicle at t n Time of day
When the vehicle is at t n When the posterior prediction probability at the moment is greater than a preset threshold value, considering that a lane change point is generated at the moment t of the vehicle; the method not only considers the influence of the driving habit, the driving skill and the like of the driver, but also considers the influence of the traffic environment indexes on the lane changing behavior of the driver, and the final result can guide and standardize the driving behavior, thereby reducing the occurrence of accidents and improving the driving safety.
Description
Technical Field
The invention belongs to the technical field of safe driving of vehicles, and particularly relates to a real-time vehicle track changing point monitoring algorithm based on a Bayesian theory.
Background
The recognition and prediction of lane-changing behavior of vehicles is one of the main research contents in the field of automobile development today. The following and lane change are two basic states of vehicle operation. The lane change behavior causes interaction with the vehicle more easily than the following behavior. It has been found that if lane change behavior can be detected before the vehicle crosses the centerline, the accident rate will be significantly reduced. Therefore, the traffic accident can be effectively avoided by timely identifying, understanding and predicting the lane changing behavior of the vehicle.
With the further development of the car networking and vehicle road coordination technology, a traffic system monitor can obtain unprecedented individual, high-precision and high-dimensional vehicle track data. The running locus of the vehicle is formed when the driver performs a series of driving operations, and is a result of a combined effect of physiological factors, environmental factors, and unobservable psychological factors. Generally, lane change points are likely to be generated only when a driver generates a lane change tendency and a target lane meets the lane change condition; therefore, currently, special equipment is commonly used for monitoring the operation behaviors and physiological characteristics (such as eye movement, heartbeat, head movement and hand movement) of a driver so as to understand and predict the driving behaviors of the vehicle, and the method not only suffers from the limitations of low data quality and high cost, but also is easy to actively remove or damage, thereby seriously restricting the development of lane change behavior monitoring technology. And the defects of difficult observation of driving behaviors, high cost of monitoring equipment, low precision and the like exist.
Disclosure of Invention
The invention aims to solve the problems of difficulty in observing driving behaviors, high cost of monitoring equipment and low precision, and provides an algorithm which has high accuracy of a prediction result and can monitor the lane changing behavior of a vehicle in real time.
The invention adopts the following technical scheme:
a real-time vehicle track lane change point monitoring method based on a Bayesian theory is characterized in that the following steps are executed for a vehicle on a target road section, and the lane change condition of the vehicle on the target road section at the current moment is monitored:
step A: acquiring t within a historical time period based on a preset sampling time interval for vehicles on a target road section 0 To t n The preset track data of the running of the vehicle i at each sampling moment form a running track data set of the vehicle i
wherein ,tn At the current moment, n is more than or equal to 2;
and B: vehicle i-based trajectory dataset
Obtaining vehicle i at sampling time t θ To t n Time series of vertical position coordinates within a time period of (1)
wherein tθ Is a distance t n The generation time of the nearest lane change point satisfies t being more than or equal to 0 θ ＜t n ；
Step C: for vehicles on the target road segment, steps C1-C3 are performed to construct vehicle i at t n Time vertical position coordinate
A posteriori predicted distribution of
And then judging that the vehicle i is on the target road section t n Whether a lane change point is generated at the moment or not is monitored, and the lane change condition of a vehicle i on the target road section at the current moment is monitored:
step C1: firstly, a run function is constructed through the following formula
Wherein when the vehicle i is at t T When the track is changed from the beginning, then
Is 0, i.e. vehicle i is at t T Generating a lane change point at any moment; when vehicle i is at t T When the track is not changed at any time, then
Is composed of
Prior information of (2);
and step C2: based on vehicle i at t θ To t n-1 Time series of longitudinal position coordinates within a time period
Constructing vehicle i at t by the following equation n Time longitudinal position coordinate
A posteriori predicted distribution of
in the formula ,
to represent
UPM prediction of (2);
and C3: based on vehicle i at t n Time vertical position coordinate
A posteriori predicted distribution of
Obtaining the posterior prediction probability P if the vehicle i is at t n Time of day
If the posterior prediction probability P is greater than the preset threshold value U, the vehicle i is predicted to be at t n Generating a lane change point at any moment; if vehicle i is at t n Time of day
Is less than or equal to a preset threshold value U, the vehicle i is predicted at t n No lane change point is generated at that time.
In a preferred embodiment of the present invention, in the step C2,
UPM prediction of
Namely that
Obtained by the following formula:
in the formula, the time sequence of the vertical position coordinates of the vehicle i
Subject to a normal distribution of alpha, with a hyperparameter of
And
alpha 'represents the conjugate index distribution of alpha, and the hyperparameters of alpha' are respectively
And
and
according to the preset
And
and gradually iteratively updating to obtain the target.
As a preferred technical solution of the present invention, in the step C2, the run length
A posteriori predicted distribution of
Namely that
Obtained by the following steps: time series based on i-longitudinal position coordinates of vehicle
Constructed by the following formula
And
joint probability distribution of
in the formula ,
to represent
About
And
UPM prediction of (2);
step 2: based on
And
are distributed over a joint probability
Constructed by the following formula
A posteriori predicted distribution of
As a preferred technical solution of the present invention, the preset threshold U is 0.001.
As a preferred technical scheme of the invention, the hyper-parameter is
And
is preset to be at
wherein 
Representing a time series of longitudinal position coordinates
A starting value of (a).
As a preferred technical scheme of the invention, the hyper-parameter
And
obtained according to the following formula:
in the formula ：
indicates that the vehicle is at t m The ordinate of the moment.
As a preferred technical solution of the present invention, in step 2, the above-mentioned
The risk function is obtained by the following specific formula:
as a preferred technical solution of the present invention, in step 2, the above-mentioned
The value of (d) is set by the following equation:
in the formula, λ represents an average time interval between adjacent lane change points.
As a preferred solution of the invention, said risk function
Constructed by the following formula:
in the formula ,
indicates that vehicle i is at t n-1 Probability of changing lanes at any moment;
indicates that vehicle i is at t n-1 The vector formed by the traffic environment indexes of the time comprises t n-1 Presetting track data of each vehicle i at any moment, and presetting track data of front and rear vehicles adjacent to the same lane of the vehicle i; beta is a k And expressing a parameter vector, and following normal distribution, wherein Q expresses the total number of the traffic environment indexes.
As a preferred solution of the present invention, the risk function may be set as
The invention has the beneficial effects that: the invention provides a real-time vehicle track lane change point monitoring method based on Bayesian theory, which comprises the steps of acquiring a running track data set of a vehicle, taking generated track data and historical vehicle statistical data as prior information, considering the influence of traffic environment indexes on the lane change behavior of a driver, providing a self-adaptive risk function calculation method based on a binomial logistic regression model, considering the influence of the driving habit, the driving skill and the like of the driver, and finally obtaining the vehicle t on the basis of the Bayesian theory n Time of day
To obtain the posterior predicted distribution of the vehicle at the time t
When the vehicle is at t n When the posterior prediction probability at the moment is greater than a preset threshold value, the vehicle t is considered n Generating a lane change point at any moment; and when the peripheral conditions do not meet the lane change conditions, the risk function is reduced, the variable point identification result is influenced, and the monitoring result is more accurate. The method not only considers the influence of the driving habit, the driving skill and the like of the driver, but also considers the influence of the traffic environment indexes on the lane changing behavior of the driver, and the final result can guide and standardize the driving behavior, thereby reducing the occurrence of accidents and improving the driving safety.
Drawings
FIG. 1 is a schematic diagram of a vehicle trajectory acquisition simulation provided by an embodiment of the present invention;
FIG. 2 is a diagram of the running track of an NGSIM straight-going vehicle applied in the embodiment of the invention;
FIG. 3 is a diagram of the NGSIM lane-changing vehicle operation track applied in the embodiment of the invention;
FIG. 4 is a track diagram of a lane-change vehicle in accordance with an embodiment of the present invention;
fig. 5 is a diagram illustrating a detection result of a lane change point of a vehicle running track according to an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are presented to enable one of ordinary skill in the art to more fully understand the present invention and are not intended to limit the invention in any way.
The vehicle trajectory is dynamically changing due to human, vehicle, road and environmental influences. The average value around which the vehicle trajectory fluctuates within a certain range due to differences of factors such as individual driver risk perception, driving experience, driving skill, lane keeping ability and the like. When the vehicle changes from a straight course to a lane-change course, the vehicle trajectory may deviate from the mean. And defining a starting point of the vehicle for starting the conversion from the following process to the lane changing process as a lane changing point. For the running track of the slave vehicle, the lane changing point and the change point corresponding to the track respectively correspond to two running states of the vehicle on two sides of the change point. Therefore, the invention provides a method for monitoring the lane change behavior of the vehicle in real time based on the Bayesian theory, and the lane change behavior of the vehicle is detected in real time based on the acquired vehicle track data.
A real-time vehicle track lane change point monitoring method based on a Bayesian theory is characterized in that the following steps are executed for a vehicle on a target road section, and the lane change condition of the vehicle on the target road section at the current moment is monitored:
step A: acquiring data, and acquiring t in a historical time period based on a preset sampling time interval for a vehicle on a target road section 0 To t n Presetting track data of the running of the vehicle i at each sampling moment to form a running track data set of the vehicle i
wherein ,tn At the current moment, n is more than or equal to 2; the trajectory data set comprises sampling instants, i.e. t 0 To t n Presetting each track data of the vehicle i corresponding to each moment respectively, wherein the presetting each track data comprises speed, longitudinal position coordinates, transverse position coordinates, acceleration, front vehicle speed on the same lane, longitudinal position coordinates of front vehicles on the same lane, transverse position coordinates of front vehicles on the same lane and acceleration of front vehicles on the same lane; the adopted data acquisition technical method comprises the following steps: beidou positioning technology, vehicle-mounted radar, roadside detection equipment and the like; as shown in fig. 1, the coordinate axes are the original point of the initial position of the vehicle on the target road section, the advancing direction of the vehicle is the positive direction of the Y axis, and the X axis is perpendicular to the Y axis; along the X-axis, i.e. expressed as a longitudinal variation; the horizontal position coordinate refers to the advancing direction of the vehicle, and the vertical position coordinate refers to the distribution direction of the lanes, namely the X-axis change direction; the horizontal coordinate of the vehicle refers to the driving direction of the vehicle, the vertical position coordinate refers to the distribution direction of the lanes, and whether lane change generating points occur or not can be judged according to the change amplitude of the vertical position coordinate of the vehicle. Running track data set of vehicles on target road section can be obtained based on collected data
The vehicle running track data set comprises a time sequence corresponding to each preset track data, wherein the time sequence of the longitudinal position coordinate of the vehicle
Where i denotes a vehicle.
In this embodiment, the trajectory data of the american public vehicle trajectory data set NGSIM, including the speed, longitudinal and transverse position coordinates, acceleration, vehicle head interval, and vehicle head time distance of the vehicle, is used in the present invention; the horizontal position coordinate of the vehicle refers to the driving direction Local _ x of the vehicle, the vertical position coordinate refers to the distribution direction Local _ y of the lane, and whether a lane change generating point occurs can be judged through the change amplitude of the vertical position coordinate of the vehicle; the trajectories of 330 straight going vehicles and 300 lane changing vehicles are extracted from the trajectory dataset of the NGSIM, respectively, as shown in fig. 2 and 3.
And B, step B: vehicle i-based trajectory dataset
Obtaining vehicle i at sampling time t θ To t n Time series of vertical position coordinates within a time period of (1)
wherein tθ Is a distance t n The generation time of the nearest lane change point satisfies t being more than or equal to 0 θ ＜t n . And obtaining the vehicle i at the sampling instant t 0 To t n Time series of vertical position coordinates within a time period of (1)
And C: for the vehicle on the target road section, executing the steps C1-C3, and for the vehicle i, the track is shown in FIG. 4; build vehicle i at t n Time longitudinal position coordinate
A posteriori predicted distribution of
And then judging that the vehicle i is on the target road section t n And whether a lane change point is generated at the moment or not is monitored, and the lane change condition of the vehicle on the target road section at the current moment is monitored.
Taking vehicle i as an example, extracting vehicle i at t 0 To t n Time series of longitudinal position coordinates of vehicle in time period, if vehicle i is at t 0 To t n If there are multiple lane change points, then the distance t is n-1 The nearest lane change point t θ As a starting point, the vertical position coordinate of each vehicle i is constructed at t n Time of day
A posteriori predicted distribution of
And further judging that each vehicle i is on the target road section t n Time of day
Whether the point is a lane change point; t is not less than 0 θ ＜t n . Theta is a known value if vehicle i is at t 0 To t n-1 There is no lane change point in between, then t θ Value of t n-1 。
Step C1: firstly, a run function is constructed through the following formula
Wherein when the vehicle i is at t T When the track is changed from the moment, then
Is 0, i.e. vehicle i is at t T Generating a lane change point at any moment; when the vehicle i is at t T When the track is not changed at any time, then
Is composed of
A priori information of. t is t 0 ≤t T ＜t n Said t is T Time of day
Can be expressed as
And step C2: based on vehicle i at t θ To t n-1 Time series of longitudinal position coordinates within a time period
Constructing vehicle i at t by the following equation n Time longitudinal position coordinate
A posteriori predicted distribution of
in the formula ,
represent
UPM (underling licensing model) prediction;
representing runs of a run
A posterior prediction distribution of
A priori information of. To calculate
And (4) performing the steps 1 to 2; to calculate
Step 3 may be performed.
Run length gamma of vehicle i i,t-1 Posterior distribution of
Is constructed by the following steps:
step 1: based on the time series of the i-ordinate of the vehicle according to the probabilistic chain rule
Constructed by the following formula
And
joint probability distribution of
in the formula ,
to represent
About
And
UPM prediction of (2);
The above-mentioned
Can be defined by the risk function H (gamma) i,t-2 + 1), the specific formula is as follows:
the risk function
Constructed by the following formula:
in the formula ,
indicates that vehicle i is at t n-1 Probability of changing lanes at any moment;
indicates that vehicle i is at t n-1 The vector formed by the traffic environment indexes of the time comprises t n-1 Presetting track data of each vehicle i at the moment, and presetting track data of front and rear vehicles adjacent to the same lane of the vehicle i; presetting each track data to comprise speed, longitudinal position coordinates, transverse position coordinates, acceleration, locomotive spacing and locomotive time interval information; beta is a beta k And representing a parameter vector, representing regression coefficients of the variables, and Q representing the total number of the traffic environment indexes.
In addition, the risk function can also be set directly as
λ represents the average time interval between adjacent lane change points, i.e. represents the average time interval between adjacent lane change points based on the historical time period, and λ is 1000s in this embodiment.
Step 2: based on the longitudinal position coordinate time series of the vehicle i according to the Bayes inference theorem
And
and
are distributed over a joint probability
Constructed by the following formula
A posteriori predicted distribution of
In addition, the risk function H (γ) i,t-2 + 1), can also be constructed by the following formula:
in the formula, λ represents the average time interval between adjacent lane change points, and λ is 1000s.
And 3, step 3:
UPM prediction of
Namely, it is
Obtained by the following formula:
in the formula, a time sequence of a vertical position coordinate of a vehicle i is preset
Subject to a normal distribution of alpha, with a hyperparameter of
And
alpha 'represents the conjugate index distribution of alpha, and the hyperparameters of alpha' are respectively
And
and
according to the preset
And
and gradually iteratively updating to obtain the target.
The hyper-parameter is
And
is preset as an initial value of
wherein 
Representing a time series of longitudinal position coordinates
The starting value of (c). The hyper-parameter
And
obtained according to the following formula:
in the formula ：
indicates that the vehicle is at t m The ordinate of the time.
And C3: based on vehicle i at t n Time vertical position coordinate
Posterior predictive distribution of
Obtaining the posterior prediction probability P if the vehicle i is at t n Time of day
If the posterior prediction probability is larger than a preset threshold value U, predicting that the vehicle i generates a lane change point at the moment t; if vehicle i is at t n Time of day
Is less than or equal to a preset threshold value U, the vehicle i is predicted to be at t n No lane change point is generated at that time. In this embodiment, the preset threshold U is 0.001, and the detection result is shown in fig. 5.
Since the trajectory of the vehicle is easier to observe and capture than a complex driving behavior. Therefore, studying hard-to-observe complex driving behaviors using simple, readily available vehicle trajectory data would be a new approach to solving the bottleneck problem. The method provided by the invention can realize real-time and online monitoring of the lane changing behavior of the vehicle, and overcomes the defects of difficulty in observing the driving behavior, high cost of monitoring equipment and low precision; the method is oriented to a future vehicle-road cooperative application scene, generated track data and historical vehicle statistical data are used as prior information, posterior distribution of the current moment is predicted according to Bayesian inference, and a prediction result can be gradually updated along with the change of time; generally, lane change points are likely to be generated only when a driver generates a lane change tendency and a target lane meets the lane change condition; compared with the traditional method, the method considers the influence of the driving habit, the driving skill and the like of the driver on the basis of the Bayesian vehicle lane change point detection algorithm on one hand, and considers the influence of the traffic environment index on the lane change behavior of the driver on the other hand. In addition, the method starts from the perspective of a single vehicle, and the final result can be used for guiding and standardizing driving behaviors, so that the occurrence of accidents is reduced, and the driving safety is improved.
Although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described in the foregoing detailed description, or equivalent changes may be made in some of the features of the embodiments described above. All equivalent structures made by using the contents of the specification and the attached drawings of the invention can be directly or indirectly applied to other related technical fields, and are also within the protection scope of the patent of the invention.
Claims (10)
1. A real-time vehicle track lane change point monitoring method based on Bayesian theory is characterized in that: aiming at the vehicles on the target road section, the following steps are carried out, and the lane changing condition of the vehicles on the target road section at the current moment is monitored:
step A: acquiring t within a historical time period based on a preset sampling time interval for vehicles on a target road section 0 To t n The preset track data of the running of the vehicle i at each sampling moment form a running track data set of the vehicle i
wherein ,tn At the current moment, n is more than or equal to 2;
and B, step B: vehicle i-based travel track dataset
Obtaining vehicle i at sampling time t θ To t n Time series of vertical position coordinates within a time period of (1)
wherein tθ Is a distance t n The generation time of the nearest lane change point meets t 0 ≤t θ ＜t n ；
And C: for vehicles on the target road segment, steps C1-C3 are performed to construct vehicle i at t n Time vertical position coordinate
A posteriori predicted distribution of
And then judging that the vehicle i is on the target road section t n Whether a lane change point is generated at the moment or not is monitored, and the lane change condition of a vehicle i on the target road section at the current moment is monitored:
step C1: the run function is constructed by the following formula
Wherein when the vehicle i is at t T When the track is changed from the moment, then
Is 0, i.e. vehicle i is at t T Generating a lane change point at any moment; when the vehicle i is at t T When the track is not changed at any time, then
Is composed of
Prior information of (2);
and step C2: based on vehicle i at t θ To t n-1 Time series of longitudinal position coordinates within a time period
Constructing vehicle i at t by the following equation n Time vertical position coordinate
A posteriori predicted distribution of
in the formula ,
to represent
UPM prediction of (2);
and C3: based on vehicle i at t n Time of dayLongitudinal position coordinate
Posterior predictive distribution of
Obtaining the posterior prediction probability P if the vehicle i is at t n Time of day
If the posterior prediction probability P is greater than the preset threshold value U, the vehicle i is predicted to be at t n Generating a lane change point at any moment; if vehicle i is at t n Time of day
Is less than or equal to a preset threshold value U, the vehicle i is predicted at t n No lane change point is generated at that time.
2. The real-time vehicle track lane change point monitoring method based on the Bayesian theory as recited in claim 1, wherein: in the step (C2), the step (C) is carried out,
UPM prediction of
Namely, it is
Obtained by the following formula:
in the formula, the time sequence of the vertical position coordinates of the vehicle i
Obeying a normal distribution of alpha, with a hyperparameter of
And
alpha 'represents the conjugate index distribution of alpha, and the hyperparameters of alpha' are respectively
And
and
according to the preset
And
and gradually iteratively updating to obtain the target.
3. The Bayesian theory-based real-time vehicle trajectory lane change point monitoring method according to claim 1, wherein: in said step C2, run
A posteriori predicted distribution of
Namely that
Is obtained by the following steps:
step 1: time series based on i-longitudinal position coordinates of vehicle
Constructed by the following formula
And
joint probability distribution of
in the formula ,
to represent
About
And
UPM prediction of (2);
step 2: based on
And
joint probability distribution of
Constructed by the following formula
A posteriori predicted distribution of
4. The Bayesian theory-based real-time vehicle trajectory lane change point monitoring method according to claim 1, wherein: the preset threshold value U is 0.001.
5. The real-time vehicle track lane change point monitoring method based on the Bayesian theory as recited in claim 2, wherein: the hyper-parameter is
And
is preset to be at
wherein 
Representing a time series of longitudinal position coordinates
A starting value of (a).
6. The Bayesian theory-based real-time vehicle trajectory lane change point monitoring method according to claim 2, wherein: hyper-parameter
And
obtained according to the following formula:
in the formula ：
indicates that the vehicle is at t m The ordinate of the moment.
7. The real-time vehicle track lane change point monitoring method based on the Bayesian theory as recited in claim 3, wherein: for step 2, the
Obtained from the risk function, the concrete formula is as follows:
8. the Bayesian theory-based real-time vehicle trajectory lane change point monitoring method according to claim 3, wherein the method comprises the following steps: for step 2, the
The value of (d) is set by the following equation:
in the formula, λ represents an average time interval between adjacent lane change points.
9. The Bayesian theory-based real-time vehicle trajectory lane change point monitoring method according to claim 7, wherein: the risk function
Constructed by the following formula:
in the formula ,
indicates that vehicle i is at t n-1 Probability of changing lanes at any moment;
indicates that vehicle i is at t n-1 The vector formed by the traffic environment indexes of the time comprises t n-1 Presetting track data of each vehicle i at the moment, and presetting track data of front and rear vehicles adjacent to the same lane of the vehicle i; beta is a k And expressing a parameter vector, and following normal distribution, wherein Q expresses the total number of the traffic environment indexes.
10. The real-time vehicle track lane change point monitoring method based on the Bayesian theory as recited in claim 7, wherein: the risk function may be set to
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211428816.5A CN115731708B (en) | 2022-11-15 | 2022-11-15 | Real-time vehicle track lane change point monitoring method based on Bayesian theory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211428816.5A CN115731708B (en) | 2022-11-15 | 2022-11-15 | Real-time vehicle track lane change point monitoring method based on Bayesian theory |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115731708A true CN115731708A (en) | 2023-03-03 |
| CN115731708B CN115731708B (en) | 2023-10-17 |
Family
ID=85295808
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211428816.5A Active CN115731708B (en) | 2022-11-15 | 2022-11-15 | Real-time vehicle track lane change point monitoring method based on Bayesian theory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115731708B (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108387242A (en) * | 2018-02-07 | 2018-08-10 | 西南交通大学 | Automatic Pilot lane-change prepares and executes integrated method for planning track |
| WO2018220418A1 (en) * | 2017-06-02 | 2018-12-06 | Toyota Motor Europe | Driving assistance method and system |
| CN110705388A (en) * | 2019-09-16 | 2020-01-17 | 清华大学 | Target vehicle lane change identification method for auxiliary driving based on prediction feedback |
| CN111079834A (en) * | 2019-12-16 | 2020-04-28 | 清华大学 | Intelligent vehicle safety situation assessment method considering multi-vehicle interaction |
| CN111086514A (en) * | 2018-10-08 | 2020-05-01 | 现代摩比斯株式会社 | Apparatus for notifying driving lane and control method thereof |
| CN112053589A (en) * | 2020-08-18 | 2020-12-08 | 北京航空航天大学 | Target vehicle lane changing behavior adaptive identification model construction method |
| CN112347567A (en) * | 2020-11-27 | 2021-02-09 | 青岛莱吉传动系统科技有限公司 | Vehicle intention and track prediction method |
| CN112819845A (en) * | 2021-02-26 | 2021-05-18 | 华南理工大学 | Flexible packaging substrate contour, line width and line distance defect detection method, medium and equipment |
| CN113096379A (en) * | 2021-03-03 | 2021-07-09 | 东南大学 | Driving style identification method based on traffic conflict |
| CN113178097A (en) * | 2020-01-09 | 2021-07-27 | 大陆汽车有限责任公司 | Method for establishing probability map of unoccupied space comprising static target and dynamic target |
| US20210272018A1 (en) * | 2020-03-02 | 2021-09-02 | Uatc, Llc | Systems and Methods for Training Probabilistic Object Motion Prediction Models Using Non-Differentiable Prior Knowledge |
| US11151880B1 (en) * | 2020-08-21 | 2021-10-19 | Toyota Motor Engineering & Manufacturing North America, Inc. | Systems and methods for providing guidance to vehicle drivers regarding predicted lane-change behavior of other vehicle drivers |
| DE102020214203A1 (en) * | 2020-11-11 | 2022-05-12 | Volkswagen Aktiengesellschaft | Method for determining a lane change intention of another motor vehicle located in the immediate vicinity of a motor vehicle, and assistance system |
| CN115018016A (en) * | 2022-08-03 | 2022-09-06 | 苏州大学 | Method and system for identifying lane changing intention of manually-driven vehicle |
| CN115056798A (en) * | 2022-05-30 | 2022-09-16 | 天津大学 | Automatic driving vehicle lane change behavior vehicle-road cooperative decision algorithm based on Bayesian game |
-
2022
- 2022-11-15 CN CN202211428816.5A patent/CN115731708B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018220418A1 (en) * | 2017-06-02 | 2018-12-06 | Toyota Motor Europe | Driving assistance method and system |
| CN108387242A (en) * | 2018-02-07 | 2018-08-10 | 西南交通大学 | Automatic Pilot lane-change prepares and executes integrated method for planning track |
| CN111086514A (en) * | 2018-10-08 | 2020-05-01 | 现代摩比斯株式会社 | Apparatus for notifying driving lane and control method thereof |
| CN110705388A (en) * | 2019-09-16 | 2020-01-17 | 清华大学 | Target vehicle lane change identification method for auxiliary driving based on prediction feedback |
| CN111079834A (en) * | 2019-12-16 | 2020-04-28 | 清华大学 | Intelligent vehicle safety situation assessment method considering multi-vehicle interaction |
| CN113178097A (en) * | 2020-01-09 | 2021-07-27 | 大陆汽车有限责任公司 | Method for establishing probability map of unoccupied space comprising static target and dynamic target |
| US20210272018A1 (en) * | 2020-03-02 | 2021-09-02 | Uatc, Llc | Systems and Methods for Training Probabilistic Object Motion Prediction Models Using Non-Differentiable Prior Knowledge |
| CN112053589A (en) * | 2020-08-18 | 2020-12-08 | 北京航空航天大学 | Target vehicle lane changing behavior adaptive identification model construction method |
| US11151880B1 (en) * | 2020-08-21 | 2021-10-19 | Toyota Motor Engineering & Manufacturing North America, Inc. | Systems and methods for providing guidance to vehicle drivers regarding predicted lane-change behavior of other vehicle drivers |
| DE102020214203A1 (en) * | 2020-11-11 | 2022-05-12 | Volkswagen Aktiengesellschaft | Method for determining a lane change intention of another motor vehicle located in the immediate vicinity of a motor vehicle, and assistance system |
| CN112347567A (en) * | 2020-11-27 | 2021-02-09 | 青岛莱吉传动系统科技有限公司 | Vehicle intention and track prediction method |
| CN112819845A (en) * | 2021-02-26 | 2021-05-18 | 华南理工大学 | Flexible packaging substrate contour, line width and line distance defect detection method, medium and equipment |
| CN113096379A (en) * | 2021-03-03 | 2021-07-09 | 东南大学 | Driving style identification method based on traffic conflict |
| CN115056798A (en) * | 2022-05-30 | 2022-09-16 | 天津大学 | Automatic driving vehicle lane change behavior vehicle-road cooperative decision algorithm based on Bayesian game |
| CN115018016A (en) * | 2022-08-03 | 2022-09-06 | 苏州大学 | Method and system for identifying lane changing intention of manually-driven vehicle |
Non-Patent Citations (1)
| Title |
|---|
| 赵树恩;柯涛;柳平;: "基于贝叶斯网络的车辆换道决策模型研究", 重庆交通大学学报(自然科学版), no. 05 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115731708B (en) | 2023-10-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107958269B (en) | Driving risk degree prediction method based on hidden Markov model | |
| Tang et al. | A hierarchical prediction model for lane-changes based on combination of fuzzy C-means and adaptive neural network | |
| CN112249008B (en) | Unmanned automobile early warning method aiming at complex dynamic environment | |
| Li et al. | Lane change maneuver recognition via vehicle state and driver operation signals—Results from naturalistic driving data | |
| CN111079590A (en) | Peripheral vehicle behavior pre-judging method of unmanned vehicle | |
| CN111159642B (en) | Online track prediction method based on particle filtering | |
| CN113720346B (en) | Vehicle path planning method and system based on potential energy field and hidden Markov model | |
| CN115056798B (en) | Automatic driving vehicle lane change behavior vehicle-road collaborative decision algorithm based on Bayesian game | |
| CN112258893B (en) | Intelligent vehicle lane change collision probability assessment method based on track prediction | |
| Chen et al. | Advanced driver assistance strategies for a single-vehicle overtaking a platoon on the two-lane two-way road | |
| Sharma et al. | Highway discretionary lane changing behavior recognition using continuous and discrete hidden markov model | |
| Mullakkal-Babu et al. | Empirics and models of fragmented lane changes | |
| Sun et al. | Machine-learning-based hybrid recognition approach for longitudinal driving behavior in noisy environment | |
| Zhang et al. | A lane-changing prediction method based on temporal convolution network | |
| CN115731708A (en) | Real-time vehicle track lane change point monitoring method based on Bayesian theory | |
| Suh et al. | Stochastic predictive control based motion planning for lane change decision using a vehicle traffic simulator | |
| CN115871671A (en) | Intelligent automobile lane change decision and trajectory planning method and system considering surrounding vehicles | |
| CN113033902B (en) | Automatic driving lane change track planning method based on improved deep learning | |
| CN114872727A (en) | Peripheral vehicle track prediction method applied to automatic driving vehicle | |
| Yu et al. | Drivers' car-following correlative behavior with preceding vehicles in multilane driving | |
| Zeng et al. | Risk-aware deep reinforcement learning for decision-making and planning of autonomous vehicles | |
| CN116749925B (en) | Man-machine hybrid driving-oriented artificial driving evidence dynamic accumulation braking method and system | |
| Liu et al. | Identification of driver distraction based on SHRP2 naturalistic driving study | |
| Mechernene et al. | Lane change decision algorithm based on risk prediction and fuzzy logic method | |
| Wu et al. | A Humanlike Lane Change Decision Strategy for Mixed Traffics with Multi-objects |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |