CN110789528B - Vehicle driving track prediction method, device, equipment and storage medium - Google Patents

Vehicle driving track prediction method, device, equipment and storage medium Download PDF

Info

Publication number
CN110789528B
CN110789528B CN201910809191.9A CN201910809191A CN110789528B CN 110789528 B CN110789528 B CN 110789528B CN 201910809191 A CN201910809191 A CN 201910809191A CN 110789528 B CN110789528 B CN 110789528B
Authority
CN
China
Prior art keywords
target
vehicle
lane
target vehicle
track
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910809191.9A
Other languages
Chinese (zh)
Other versions
CN110789528A (en
Inventor
钱祥隽
周信宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tencent Technology Shenzhen Co Ltd
Original Assignee
Tencent Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Priority to CN201910809191.9A priority Critical patent/CN110789528B/en
Publication of CN110789528A publication Critical patent/CN110789528A/en
Application granted granted Critical
Publication of CN110789528B publication Critical patent/CN110789528B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/10—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to vehicle motion
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/10—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to vehicle motion
    • B60W40/107—Longitudinal acceleration
    • 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
    • B60W2520/00—Input parameters relating to overall vehicle dynamics
    • B60W2520/06—Direction of travel
    • 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
    • B60W2520/00—Input parameters relating to overall vehicle dynamics
    • B60W2520/10—Longitudinal speed
    • 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
    • B60W2520/00—Input parameters relating to overall vehicle dynamics
    • B60W2520/10—Longitudinal speed
    • B60W2520/105—Longitudinal acceleration

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Mathematical Physics (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
  • Traffic Control Systems (AREA)

Abstract

The application discloses a vehicle driving track prediction method, which comprises the following steps: obtaining current driving data of a target vehicle; predicting a rough track of the target vehicle within a set duration after the current time based on the current driving data; determining a target lane of the target vehicle according to the rough track; and generating a running track of the target vehicle to the target lane. By applying the technical scheme provided by the embodiment of the application, the running track of the surrounding vehicle can be generated more efficiently, the finally generated running track is more accurate, the current vehicle can be controlled to run more accurately according to the generated running track, and the running safety of the vehicle is improved. The application also discloses a vehicle running track prediction device, equipment and a storage medium, and the device and the equipment have corresponding technical effects.

Description

Vehicle driving track prediction method, device, equipment and storage medium
Technical Field
The present application relates to the field of automatic driving technologies, and in particular, to a method, an apparatus, a device, and a storage medium for predicting a vehicle driving trajectory.
Background
With the rapid development of artificial intelligence in various industries, as an important application of artificial intelligence, an automatic driving technique has gradually emerged. The automatic driving is to guide and decide a vehicle driving task without a driver executing physical driving operation, and replace the driver control behavior to enable the vehicle to realize safe driving, namely, the aim is to realize that the vehicle autonomously travels along a road in an unmanned state, and ensure the safety of the vehicle while reaching a target point as soon as possible, and simultaneously ensure that no direct or indirect threat is caused to the safety of other traffic participants.
Under the automatic driving mode of the vehicle, the prediction system of the vehicle is very important for judging the driving trend of the surrounding vehicle within a plurality of seconds in the future, and the correct judgment of the prediction system can lead the automatic driving system of the vehicle to pre-judge the change of the surrounding environment so as to make the automatic driving system of the vehicle respond more quickly. Prediction of the travel path of the surrounding vehicle is one of the main functions of the prediction system. During cruising, the vehicle in the automatic driving mode can understand whether other vehicles are in a normal lane keeping state or have lane changing attempts through predicting the driving track of the vehicle, so that the vehicle is helped to make decisions such as lane keeping, lane changing or overtaking.
How to accurately predict the running track of the surrounding vehicle is a technical problem which needs to be solved urgently by the technical personnel in the field at present.
Disclosure of Invention
The application aims to provide a vehicle running track prediction method, a vehicle running track prediction device and a storage medium, so that running tracks of surrounding vehicles can be accurately predicted.
In order to solve the technical problem, the application provides the following technical scheme:
a vehicle travel track prediction method, comprising:
obtaining current driving data of a target vehicle;
predicting a rough trajectory of the target vehicle within a set duration after a current time based on the current travel data;
determining a target lane of the target vehicle according to the rough track;
and generating a running track of the target vehicle to the target lane.
In one embodiment of the present application, the predicting a rough trajectory of the target vehicle within a set time period after a current time based on the current travel data includes:
determining the acceleration and the angular speed of the target vehicle according to the current running data;
assuming that the target vehicle moves according to the acceleration and the angular velocity, and obtaining a rough track of the target vehicle in a set time length after the current time based on vehicle kinematics model recursion.
In one embodiment of the present application, the determining a target lane of the target vehicle according to the rough track includes:
determining a candidate lane of the target vehicle associated with the rough track according to the rough track;
selecting a target lane of the target vehicle from the candidate lanes.
In one embodiment of the present application, the determining the candidate lane of the target vehicle according to the rough track includes:
discretizing the coarse trajectory into a plurality of location points;
respectively inquiring the lane where each position point is located;
and determining the inquired lane as a candidate lane of the target vehicle.
In one embodiment of the present application, the selecting a target lane of the target vehicle among the candidate lanes includes:
determining a characteristic value of each characteristic of each candidate lane according to a set characteristic table;
calculating the hit probability of each candidate lane based on the characteristic value of each characteristic of each candidate lane;
and determining a target lane from the candidate lanes according to the selected probability.
In one embodiment of the present application, the generating a driving trajectory of the target vehicle to the target lane includes:
predicting a driving state of the target vehicle when the target vehicle reaches the target lane;
and generating a driving track of the target vehicle to the target lane according to the driving state.
In one embodiment of the present application, the driving state includes a target line and a target speed at which the target vehicle reaches the target lane, and the generating the driving trajectory of the target vehicle to the target lane according to the driving state includes:
and assuming that the target vehicle is to reach the target line, generating a running track of the target lane reaching the target line of the target lane by adopting a pure tracking control algorithm.
A vehicle travel track prediction apparatus comprising:
the driving data obtaining module is used for obtaining the current driving data of the target vehicle;
a rough trajectory prediction module for predicting a rough trajectory of the target vehicle within a set duration after a current time based on the current travel data;
a target lane determining module, configured to determine a target lane of the target vehicle according to the rough track;
and the driving track generating module is used for generating a driving track of the target vehicle reaching the target lane.
A vehicle travel track prediction apparatus comprising:
a memory for storing a computer program;
a processor for implementing the steps of the vehicle travel track prediction method of any one of the above when executing the computer program.
A computer-readable storage medium having a computer program stored thereon, the computer program, when being executed by a processor, implementing the steps of the vehicle travel track prediction method according to any one of the preceding claims.
By applying the technical scheme provided by the embodiment of the application, after the current running data of the target vehicle is obtained, the rough track of the target vehicle within the set time length after the current time is predicted based on the current running data, the target lane which the target vehicle is most likely to reach is determined according to the rough track, and the running track of the target vehicle reaching the target lane is generated, so that the running track of the peripheral vehicle is generated more efficiently, and the finally generated running track is more accurate. And the current vehicle can be more accurately controlled according to the generated running track, so that the running safety of the vehicle is improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic structural diagram of a scheme application system in an embodiment of the present application;
FIG. 2 is a schematic diagram of an interaction process implemented by an embodiment of the present application;
FIG. 3 is a schematic diagram illustrating a configuration of a vehicle travel track prediction apparatus according to an embodiment of the present application;
FIG. 4 is a flowchart illustrating an implementation of a method for predicting a driving trajectory of a vehicle according to an embodiment of the present disclosure;
FIG. 5 is a schematic view of a vehicle kinematics model in an embodiment of the present application;
FIG. 6 is a schematic diagram of a candidate lane determined according to a rough track in an embodiment of the present application;
FIG. 7 is a schematic diagram illustrating a vehicle driving state prediction in an embodiment of the present application;
FIG. 8 is a schematic diagram illustrating determination of waypoints in a target line of a target lane in an embodiment of the application;
fig. 9 is a schematic structural diagram of a vehicle travel track prediction apparatus according to an embodiment of the present application.
Detailed Description
The core of the embodiment of the application is to provide a vehicle driving track prediction method, which can be applied to automatic driving systems of vehicles at various levels, such as an L2 auxiliary driving system, a high-speed automatic driving system requiring human supervision by an L3, a L4 and L5 urban full-automatic driving system, and the like, so as to realize accurate prediction of the driving track of surrounding vehicles by the automatic driving vehicle under different driving conditions, and contribute to further driving decision of the automatic driving system in the automatic driving vehicle.
In order that those skilled in the art will better understand the disclosure, the following detailed description will be given with reference to the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
For the convenience of understanding, the composition architecture of the application system of the solution of the embodiment of the present application is described first, and is shown in fig. 1, which is a schematic structural diagram of the application system of the embodiment of the present application, and the application system includes a vehicle 10 and vehicles 11 and 12 around the vehicle 10. The vehicle 10 may be an autonomous vehicle, in which an autonomous driving system is provided, and the autonomous driving system needs to predict the driving tracks of other surrounding vehicles in the process of controlling the vehicle 10 to drive, and by predicting the driving tracks of other surrounding vehicles, it can be understood whether the other surrounding vehicles are in a normal lane keeping state or have lane changing attempts, so as to make a decision on lane keeping, lane changing or overtaking for the vehicle 10 based on the predicted driving tracks. In addition, in the vicinity of the intersection, by predicting the travel locus of other vehicles around, the travel direction of the other vehicles can be known, and the right to the vehicle can be determined.
The vehicle 10 can obtain current travel data of the vehicles 11, 12 around the vehicle 10 by a sensor, a laser radar, an ultra-wave radar, or the like, and based on the current travel data, rough trajectories of the vehicles 11, 12 within a set time period after the current time can be predicted. Thus, the target lanes of the vehicles 11 and 12 can be determined according to the rough tracks. Specifically, the candidate lanes of the vehicles 11 and 12, that is, the lanes that are likely to arrive, may be determined first, and then the corresponding target lane, that is, the lane that is most likely to arrive, may be selected from the candidate lanes corresponding to the vehicles 11 and 12. And generating the running tracks of the vehicles 11 and 12 reaching the corresponding target lanes. Accurate prediction of the running tracks of the surrounding vehicles 11 and 12 is accomplished, and further decisions on the further running behaviors of the vehicles are made.
Fig. 1 shows only two vehicles around the vehicle 10, and there may be more vehicles around the vehicle during actual driving, and the driving track of each vehicle can be predicted by obtaining driving data of each vehicle. The surrounding vehicle may be an autonomous vehicle or a general vehicle.
The lane in the embodiment of the present application refers to a lane in which a vehicle can travel. In the road, the track that the vehicle normally travels from the entrance to the exit may be considered as a virtual lane, although this lane is not indicated on the actual road.
Specifically, as shown in fig. 2, the automatic driving system provided in the vehicle 10 may include a vehicle travel track prediction device 101 and a vehicle travel control device 102, during normal running of the vehicle 10, the vehicle running track prediction apparatus 101 can obtain current running data of a target vehicle (here, the target vehicle is exemplified as the vehicle 11) by a sensor, a radar, or the like, predict a rough track of the target vehicle within a set period of time after the current time based on the current running data, determine a lane candidate of the target vehicle based on the rough track, the method includes selecting a target lane, to which the target vehicle is most likely to reach, among the candidate lanes, generating a travel trajectory for the target vehicle to reach the target lane, transmitting the generated travel trajectory to the vehicle travel control device 102, and the vehicle travel control device 102 controlling travel of the vehicle 10 based on the travel trajectory.
Fig. 3 is a schematic diagram illustrating a configuration of the vehicle travel track prediction apparatus 101, and the vehicle travel track prediction apparatus 101 may include: a processor 20, a memory 21, a communication interface 22 and a communication bus 23. The processor 20, the memory 21 and the communication interface 22 all communicate with each other via a communication bus 23.
In the embodiment of the present application, the processor 20 may be a Central Processing Unit (CPU), an application specific integrated circuit, a digital signal processor, a field programmable gate array or other programmable logic device, etc.
The processor 20 may call a program stored in the memory 21, and in particular, the processor 20 may perform the operations in the following embodiments of the vehicle travel track prediction method.
The memory 21 is used for storing one or more programs, the program may include program codes, the program codes include computer operation instructions, in this embodiment, the memory 21 stores at least the program for implementing the following functions:
obtaining current driving data of a target vehicle;
predicting a rough track of the target vehicle within a set duration after the current time based on the current driving data;
determining a target lane of the target vehicle according to the rough track;
and generating a running track of the target vehicle to the target lane.
In one possible implementation, the memory 21 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function (such as a sound playing function and an image playing function), and the like; the storage data area may store data created during use, such as travel data, trajectory data, and the like.
Further, the memory 21 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device or other volatile solid state storage device.
The communication interface 23 may be an interface of a communication module for connecting with other devices or systems.
Of course, it should be noted that the structure shown in fig. 3 does not constitute a limitation of the vehicle travel track prediction apparatus in the embodiment of the present application, and the vehicle travel track prediction apparatus may include more or less components than those shown in fig. 3 in practical use, or some components in combination.
With the above generality, referring to fig. 4, an implementation flowchart of a vehicle driving trajectory prediction method provided in the embodiment of the present application is shown, where the method may include the following steps:
s310: current travel data of the target vehicle is obtained.
In the embodiment of the present application, the target vehicle may be any one of vehicles around the target vehicle at the current time during the driving of the autonomous vehicle, or a vehicle with a lane change intention determined by a pre-determination of surrounding vehicles. If the driving tracks of a plurality of vehicles are predicted, the driving tracks of the plurality of vehicles can be simultaneously predicted through multiple threads, and then the prediction results of the driving tracks of the plurality of vehicles can be comprehensively obtained when the driving decision is determined.
The following description of the embodiments of the present application is made in a scenario where the execution subject is an autonomous vehicle and there is a target vehicle.
The automatic driving vehicle can obtain the current driving data of the surrounding target vehicles in real time or according to a set time interval, and particularly can obtain the current driving data of the target vehicles through a sensor, a laser radar, an ultra-meter wave radar and other devices. The current driving data may include speed, acceleration, position, driving direction, and the like.
S320: based on the current travel data, a rough trajectory of the target vehicle within a set period of time after the current time is predicted.
After obtaining the current travel data of the target vehicle, the rough trajectory of the target vehicle within a set time period after the current time may be predicted based on the current travel data. The set time period can be set and adjusted according to actual conditions, such as 3 seconds.
In a specific embodiment of the present application, step S320 may include the following steps:
the method comprises the following steps: determining the acceleration and the angular velocity of the target vehicle according to the current driving data;
step two: assuming that the target vehicle moves according to the acceleration and the angular velocity, a rough track of the target vehicle within a set time length after the current time is obtained by recursion based on the vehicle kinematics model.
For convenience of description, the above two steps are combined for illustration.
The obtained current traveling data of the target vehicle includes data such as speed, acceleration, position, traveling direction, and the like of the target vehicle, and the acceleration and angular velocity of the target vehicle can be determined from the current traveling data.
When the rough track is predicted, assuming that the target vehicle moves according to the current acceleration and angular velocity, the rough track of the target vehicle within a set time length after the current time can be obtained by recursion based on the vehicle kinematics model.
The vehicle kinematic model on which the embodiments of the present application are based will be described first.
Fig. 5 is a schematic diagram of a vehicle kinematic model. Generally speaking, the kinematic characteristics of a four-wheeled vehicle can be simplified to some extent, and a pair of front wheels and a pair of rear wheels are considered in combination respectively, so as to form a "bicycle model" shown in fig. 5, and the kinematic equations of the vehicle are shown in equations (1) - (5):
Figure BDA0002184552510000071
Figure BDA0002184552510000072
Figure BDA0002184552510000081
Figure BDA0002184552510000082
Figure BDA0002184552510000083
wherein the content of the first and second substances,
Figure BDA0002184552510000084
representing the x-position, y-position, speed and direction of travel of the vehicle, respectively, a and delta representing the acceleration and front wheel steering angle of the vehicle, respectively, which is directly related to the steering wheel steering angle, lfIndicating the distance between the center of gravity of the vehicle and the front wheel,/rIndicating the vehicle center of gravity and rear wheel distance, and CoM is the vehicle center of gravity.
In brief, the kinematic equation described above means that by controlling the vehicle acceleration a and the front wheel steering angle δ, the states of the position, speed, orientation, and the like of the vehicle can be controlled. The kinematic equation can well describe the real motion process of the vehicle.
In the embodiment of the present application, the above kinematic equation is discretized, and is described by formula (6):
x (n +1) ═ f (X (n), u (n))) formula (6)
Wherein the content of the first and second substances,
Figure BDA0002184552510000085
for the vehicle state quantity, u ═ a, δ is the vehicle control quantity, and f is a discretized version of the above differential equation. Equation (6) expresses that the state of the vehicle at time n +1 is equal to the state of the vehicle at time n and control u is used. By this discrete equation, given the vehicle initial state quantity and the controlled quantity, the state of the vehicle at the next time can be obtained.
After obtaining the current travel data of the target vehicle and determining the Acceleration and angular velocity of the target vehicle, it may be assumed that the target vehicle moves at the current Acceleration and angular velocity, that is, a future set time period, such as a rough trajectory of the target vehicle within 3 seconds, is approximately predicted from the current state quantity and the current control quantity of the target vehicle based on a vehicle kinematics model using a CACR (Constant Acceleration Constant velocity) concept. This rough trajectory is not the final predicted trajectory of the target vehicle, but is used to roughly relate the target vehicle to the lanes it may reach.
Specifically, it is assumed that the target vehicle will move according to the current acceleration and the current front wheel steering angle, i.e., keeping u ═ a0,δ0) And (4) carrying out recursion according to the formula (6) without changing, and iteratively calculating a rough track. a is0Indicating the current acceleration, δ0Indicating the current front wheel steering angle.
S330: and determining a target lane of the target vehicle according to the rough track.
As described previously, the purpose of predicting the rough trajectory of the target vehicle within a set length of time after the current time is to associate the target vehicle with the lanes that it may reach. After the rough track of the target vehicle is predicted, a way may be to determine a lane to which the rough track is finally reached as a target lane of the target vehicle.
In one embodiment of the present application, according to the rough trajectory, candidate lanes of the target vehicle associated with the rough trajectory may be determined, and then the target lane of the target vehicle may be selected from the candidate lanes.
All lanes traversed by the rough trajectory may be determined as candidate lanes for the target vehicle, i.e., the target vehicle is likely to later arrive in one of the candidate lanes. There may be one or more candidate lanes.
In a specific embodiment of the present application, step S330 may include the following steps:
the first step is as follows: discretizing the coarse trajectory into a plurality of location points;
the second step is that: respectively inquiring the lane where each position point is located;
the third step: and determining the inquired lane as a candidate lane of the target vehicle.
For convenience of description, the above three steps are combined for illustration.
After determining the coarse trajectory of the target vehicle, the coarse trajectory may be discretized into a plurality of location points. Specifically, the dispersion may be performed in such a manner that the time intervals are equal or the distance intervals are equal.
Taking the example of performing the discretization in a manner of equal time intervals, assuming that the rough trajectory is a trajectory of 3 seconds, the rough trajectory may be discretized at equal time intervals of 30 points, and the time interval between two adjacent points is 0.1 seconds. Each point represents the state of the target vehicle at a corresponding point in time, including the target vehicle's position, direction of travel, speed, etc. Therefore, the corresponding position of each point can be obtained, and the corresponding position point can be obtained.
After the rough track is dispersed into a plurality of position points, the lane where each position point is located can be inquired and obtained through a high-precision map. That is, a query is made for 30 points in the above example, one or more candidate lanes can be obtained, and the candidate lanes are all lanes that the target vehicle is likely to reach in the future.
That is, when the vehicle travels on a regular road, it is possible to keep traveling in the current lane, and to change lanes to the left or to the right. At the crossroad, the vehicle can turn left, go straight and turn right. The determined candidate lanes of the target vehicle are the lanes that the target vehicle is likely to travel in the future.
As shown in fig. 6, the schematic diagram of candidate lanes determined according to the rough track is that the vehicle position is roughly predicted after 3 seconds from the current vehicle position is determined by the rough track, the rough track passes through three lanes, the three lanes can be determined as candidate lanes of the target vehicle, and the waiting lane is a lane which the target vehicle is likely to reach.
After determining the candidate lanes of the target vehicle associated with the coarse trajectory, the target lane of the target vehicle may be selected among the candidate lanes.
There may be one or more determined candidate lanes of the target vehicle. If there is only one candidate lane, the candidate lane may be directly taken as the target lane of the target vehicle. If there are multiple candidate lanes, at least one lane may be selected among the candidate lanes as a target lane for the target vehicle. The target lane is the lane that the target vehicle is most likely to reach at last, i.e., the target vehicle is more likely to reach the target lane.
In one embodiment of the present application, there are a plurality of candidate lanes, and the target lane of the target vehicle may be selected from the candidate lanes by:
the method comprises the following steps: determining a characteristic value of each characteristic of each candidate lane according to a set characteristic table;
step two: calculating the selection probability of each candidate lane based on the characteristic value of each characteristic of each candidate lane;
step three: and determining a target lane from the candidate lanes according to the selected probability.
For convenience of description, the above three steps are combined for illustration.
In the embodiment of the present application, a feature table may be preset, where the feature table includes a plurality of features related to lanes and vehicles, as shown in table 1:
Figure BDA0002184552510000101
Figure BDA0002184552510000111
TABLE 1
Table 1 includes characteristics such as an angle between the vehicle and the candidate lane, a vehicle speed, a vehicle length, a vehicle width, a vehicle-to-front distance, and a vehicle-to-rear distance. The magnitude of the eigenvalues of these features may be indicative of the likelihood of the vehicle entering the candidate lane. In practice, more features may be added or portions may be selected.
And (3) extracting the characteristics of each candidate lane according to the characteristics in the table 1, and determining the characteristic value of each characteristic of each candidate lane. Based on the feature value of each feature of each candidate lane, the hit probability of each candidate lane may be calculated.
Specifically, a certain weight may be given to each feature in advance, after the feature value of each feature of each candidate lane is determined, the weighted sum of the feature values corresponding to the candidate lane may be calculated for each candidate lane, the calculated weighted sum is determined as the candidate score of the candidate lane, and the candidate score is normalized to obtain the selection probability of each candidate lane. Of course, after the feature value of each feature of each candidate lane is determined, the features may be quantized to [0,1], which facilitates comparison and calculation.
And obtaining the selection probability of each candidate lane by adopting a machine learning algorithm. For example, a large amount of training data may be acquired in advance, each piece of training data includes a feature value of each feature of a candidate lane corresponding to the vehicle and a lane result to which the vehicle finally arrives, and different pieces of training data may correspond to the same or different vehicles, different candidate lanes of the same vehicle, and the like. Machine learning is carried out through training data to obtain a selection model, and the feature value of each feature of each candidate lane is input into the selection model, so that the probability that each candidate lane is selected as a target lane, namely the selection probability, can be obtained.
The machine learning algorithm can be replaced by other algorithms such as neural network and logistic regression. The embodiment of the present application is not described in detail.
According to the selected probability, a target lane can be determined from the candidate lanes. If the candidate lane with the highest hit probability is determined as the target lane, or the candidate lane with the hit probability greater than a preset probability threshold is determined as the target lane, in this case, one or more determined target lanes may be provided.
S340: and generating a running track of the target vehicle to the target lane.
After the target lane of the target vehicle is selected from the candidate lanes of the target vehicle, a driving track of the target vehicle reaching the target lane may be generated.
In a specific embodiment of the present application, step S340 may include the following steps:
the first step is as follows: predicting a driving state of the target vehicle when the target vehicle reaches the target lane;
the second step is that: and generating a running track of the target vehicle reaching the target lane according to the running state.
For convenience of description, the above two steps are combined for illustration.
After determining the target lane of the target vehicle, a future driving state of the target vehicle may be further predicted, and the driving state may include a target line and a target speed at which the target vehicle reaches the target lane. If a target line is predicted to be reached by the target vehicle in the future, this can be characterized by a deviation Δ from the center line of the target lane, and a target speed v of the vehicle at a predicted end time, e.g., 5 seconds1. Both the lane and the center line of the lane can be obtained by high-precision map query.
One of the simpler prediction methods is to assume that the target vehicle finally reaches the center line of the target lane and keep the vehicle speed constant, i.e., Δ ═ 0, v1＝v0，v0The speed of the target vehicle at the present moment.
Alternatively, a machine learning algorithm may be employed to predict Δ and v1. For example, a large amount of training data may be acquired in advance, each piece of training data includes a current driving state of the vehicle and a driving state corresponding to the current driving state after a period of time, and different pieces of training data may correspond to the same or different vehicles, or different lanes of the same vehicle. And performing machine learning on the training data to obtain a state prediction model. The current running state information of the target vehicle is input into the state prediction model, and the running state after a period of time can be output.
The machine learning algorithm can be replaced by other algorithms such as neural network and logistic regression. The embodiment of the present application is not described in detail.
After the driving state of the target vehicle when reaching the target lane is predicted, the position of the target vehicle reaching the target lane can be obtained according to the driving state, and the driving track of the target vehicle reaching the target lane can be generated according to the current position of the target vehicle and the predicted position of the target lane.
Specifically, after the center line of the target lane is determined and the offset and the target speed in the target lane are obtained, the three quantities can be converted into a track which is possible to be driven by the target vehicle in the future. In the embodiment of the application, it may be assumed that the target vehicle is to reach a target line of the target lane, where the target line of the target lane is the center line of the target lane + the offset in the target lane, and the target vehicle is simulated to run, and a running track of the target lane to reach the target line of the target lane may be generated by using a pure tracking control algorithm, as shown in fig. 7. It should be noted that the pure tracking control algorithm can be replaced by other control algorithms such as Stanley (nonlinear feedback function based on lateral tracking error), PID (proportional control, integral control, derivative control), etc.
Taking the pure tracking control algorithm to generate the track of the target line of the target vehicle reaching the target lane as an example, the acceleration of the target vehicle may be calculated first, and the speed of the target vehicle at the current moment is assumedIs v is0Target velocity v1The target vehicle is kept accelerating uniformly for a period of 0 to 5 seconds, and the acceleration a ═ v can be obtained1-v0)/5.0。
As shown in fig. 8, the target vehicle is represented by a bicycle model, the target line of the target lane is path, and the track generation using the pure tracking control algorithm aims to enable the target vehicle to reach a certain waypoint (g) on the target line path of the target lane according to a circular arc trackx,gy)。
The waypoint (g) may be acquired by the following stepsx,gy)：
Assuming that the target vehicle needs to reach the target line within t seconds, t can be a value within a range of 2s to 4s according to the vehicle model, and the current speed of the target vehicle is v0Calculating to obtain the pre-aiming distance ld＝v0T, in ldThe radius is taken as the center of a circle drawn by the center of the rear wheel axle of the target vehicle, and the point intersecting the target line path is the waypoint (g)x,gy)。
After the road position is obtained, the curvature k of the current position of the target vehicle can be obtained according to the geometric relation and a formula (7):
Figure BDA0002184552510000131
wherein alpha is the included angle between the current direction of the target vehicle and the direction from the target vehicle to the waypoint, and ldIs the pre-aiming distance.
The target vehicle front wheel slip angle δ can be obtained by equation (8):
δ＝tan-1(kL) formula (8)
Wherein L is the target vehicle length.
The target vehicle front wheel slip angle δ represents how much the front wheel of the target vehicle needs to be turned if it is desired to reach the target lane, given the current vehicle state.
If the target vehicle is assumed to be controlled according to the acceleration a and the front wheel slip angle δ obtained above, a new state of the target vehicle at the next moment, including a position, an orientation and the like, and a position of the target vehicle at the next moment can be calculated through the previous discrete state equation X (n +1) ═ f (X (n), u (n)), and the new front wheel slip angle can be continuously obtained by using the pure tracking control algorithm, and the iteration is carried out, so that the possible running track of the target vehicle can be obtained.
In practical application, it can be assumed that each control step is 0.1 second, and the total track time length is 5 seconds, so that 50 iterations can obtain a target vehicle future path containing 50 points, and a continuous driving track can be obtained through interpolation.
If there are a plurality of target lanes of the target vehicle, a plurality of traveling tracks of the target vehicle can be obtained by the above method.
According to the embodiment of the application, a rough track of a target vehicle is obtained through a CACR (Constant Acceleration Constant angular velocity) mode and a vehicle kinematic model in a recursion mode, the rough track is used for associating the target vehicle to some candidate lanes, the candidate lanes are selected, the best one or two of the rough tracks are selected as the target lanes, and then a vehicle track generating method of the vehicle kinematic model is used for generating a final driving track by predicting the future state of the target vehicle.
By applying the method provided by the embodiment of the application, after the current running data of the target vehicle is obtained, the rough track of the target vehicle within the set time length after the current time is predicted based on the current running data, the target lane which the target vehicle is most likely to reach is determined according to the rough track, and the running track of the target vehicle reaching the target lane is generated, so that the running track of the peripheral vehicle is generated more efficiently, and the finally generated running track is more accurate. And the current vehicle can be more accurately controlled according to the generated running track, so that the running safety of the vehicle is improved.
Corresponding to the above method embodiments, the present application further provides a vehicle driving track prediction device, and a vehicle driving track prediction device described below and a vehicle driving track prediction method described above may be referred to in correspondence with each other.
Referring to fig. 9, the apparatus may include the following modules:
a driving data obtaining module 410 for obtaining current driving data of the target vehicle;
a rough trajectory prediction module 420 for predicting a rough trajectory of the target vehicle within a set time period after the current time based on the current travel data;
a target lane determining module 430, configured to determine a target lane of the target vehicle according to the rough trajectory;
and a driving track generating module 440, configured to generate a driving track of the target vehicle reaching the target lane.
By applying the device provided by the embodiment of the application, after the current running data of the target vehicle is obtained, the rough track of the target vehicle within the set time length after the current time is predicted based on the current running data, the target lane which the target vehicle is most likely to reach is determined according to the rough track, and the running track of the target vehicle reaching the target lane is generated, so that the running track of the peripheral vehicle is generated more efficiently, and the finally generated running track is more accurate. And the current vehicle can be more accurately controlled according to the generated running track, so that the running safety of the vehicle is improved.
In an embodiment of the present application, the rough trajectory prediction module 420 is specifically configured to:
determining the acceleration and the angular velocity of the target vehicle according to the current driving data;
assuming that the target vehicle moves according to the acceleration and the angular velocity, a rough track of the target vehicle within a set time length after the current time is obtained by recursion based on the vehicle kinematics model.
In an embodiment of the present application, the target lane determining module 430 is specifically configured to:
determining a candidate lane of the target vehicle associated with the rough track according to the rough track;
a target lane of the target vehicle is selected from the candidate lanes.
In an embodiment of the present application, the target lane determining module 430 is specifically configured to:
discretizing the coarse trajectory into a plurality of location points;
respectively inquiring the lane where each position point is located;
and determining the inquired lane as a candidate lane of the target vehicle.
In an embodiment of the present application, there are multiple candidate lanes, and the target lane determining module 430 is specifically configured to:
determining a characteristic value of each characteristic of each candidate lane according to a set characteristic table;
calculating the selection probability of each candidate lane based on the characteristic value of each characteristic of each candidate lane;
and determining a target lane from the candidate lanes according to the selected probability.
In an embodiment of the present application, the driving trajectory generating module 440 is specifically configured to:
predicting a driving state of the target vehicle when the target vehicle reaches the target lane;
and generating a running track of the target vehicle reaching the target lane according to the running state.
In an embodiment of the application, the driving state includes a target line and a target speed of the target vehicle reaching the target lane, and the driving trace generating module 440 is specifically configured to:
and assuming that the target vehicle is to reach the target line, generating a running track of the target lane to the target line of the target lane by adopting a pure tracking control algorithm.
Corresponding to the above method embodiments, the present application further provides a computer readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the vehicle travel track prediction method described above.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The principle and the implementation of the present application are explained in the present application by using specific examples, and the above description of the embodiments is only used to help understanding the technical solution and the core idea of the present application. It should be noted that, for those skilled in the art, it is possible to make several improvements and modifications to the present application without departing from the principle of the present application, and such improvements and modifications also fall within the scope of the claims of the present application.

Claims (8)

1. A method for predicting a travel locus of a vehicle, comprising:
obtaining current driving data of a target vehicle;
predicting a rough trajectory of the target vehicle within a set duration after a current time based on the current travel data;
determining a target lane of the target vehicle according to the rough track;
predicting a driving state of the target vehicle when the target vehicle reaches the target lane; the driving state includes a target line and a target speed at which the target vehicle reaches the target lane;
and assuming that the target vehicle is to reach the target line, generating a running track of the target lane reaching the target line of the target lane by adopting a pure tracking control algorithm.
2. The method according to claim 1, wherein the predicting a rough trajectory of the target vehicle for a set period of time after a current time based on the current travel data comprises:
determining the acceleration and the angular speed of the target vehicle according to the current running data;
assuming that the target vehicle moves according to the acceleration and the angular velocity, and obtaining a rough track of the target vehicle in a set time length after the current time based on vehicle kinematics model recursion.
3. The method of claim 1, wherein said determining a target lane of said target vehicle from said coarse trajectory comprises:
determining a candidate lane of the target vehicle associated with the rough track according to the rough track;
selecting a target lane of the target vehicle from the candidate lanes.
4. The method of claim 3, wherein said determining a candidate lane for the target vehicle from the coarse trajectory comprises:
discretizing the coarse trajectory into a plurality of location points;
respectively inquiring the lane where each position point is located;
and determining the inquired lane as a candidate lane of the target vehicle.
5. The method of claim 3, wherein there are a plurality of the candidate lanes, and wherein selecting the target lane of the target vehicle among the candidate lanes comprises:
determining a characteristic value of each characteristic of each candidate lane according to a set characteristic table;
calculating the hit probability of each candidate lane based on the characteristic value of each characteristic of each candidate lane;
and determining a target lane from the candidate lanes according to the selected probability.
6. A vehicle travel track prediction apparatus characterized by comprising:
the driving data obtaining module is used for obtaining the current driving data of the target vehicle;
a rough trajectory prediction module for predicting a rough trajectory of the target vehicle within a set duration after a current time based on the current travel data;
a target lane determining module, configured to determine a target lane of the target vehicle according to the rough track;
a driving track generation module for predicting a driving state of the target vehicle when the target vehicle reaches the target lane; the driving state includes a target line and a target speed at which the target vehicle reaches the target lane; and assuming that the target vehicle is to reach the target line, generating a running track of the target lane reaching the target line of the target lane by adopting a pure tracking control algorithm.
7. A vehicle travel track prediction apparatus characterized by comprising:
a memory for storing a computer program;
a processor for implementing the steps of the vehicle travel track prediction method according to any one of claims 1 to 5 when executing the computer program.
8. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the vehicle travel track prediction method according to any one of claims 1 to 5.
CN201910809191.9A 2019-08-29 2019-08-29 Vehicle driving track prediction method, device, equipment and storage medium Active CN110789528B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910809191.9A CN110789528B (en) 2019-08-29 2019-08-29 Vehicle driving track prediction method, device, equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910809191.9A CN110789528B (en) 2019-08-29 2019-08-29 Vehicle driving track prediction method, device, equipment and storage medium

Publications (2)

Publication Number Publication Date
CN110789528A CN110789528A (en) 2020-02-14
CN110789528B true CN110789528B (en) 2022-03-25

Family

ID=69427076

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910809191.9A Active CN110789528B (en) 2019-08-29 2019-08-29 Vehicle driving track prediction method, device, equipment and storage medium

Country Status (1)

Country Link
CN (1) CN110789528B (en)

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111341150B (en) * 2020-02-28 2021-01-26 长安大学 Reminding method and device for preventing ultrahigh vehicle from entering limited-height road section
EP4030403A4 (en) * 2020-03-04 2022-10-19 Huawei Technologies Co., Ltd. Method and device for predicting exit for vehicle
CN111488674B (en) * 2020-03-12 2024-01-16 上海理工大学 Plane intersection vehicle running track simulation method
CN111609846A (en) * 2020-04-08 2020-09-01 延锋伟世通电子科技（上海）有限公司 Vehicle travel track prediction method, system, medium, and vehicle-mounted terminal
CN114056347A (en) * 2020-07-31 2022-02-18 华为技术有限公司 Vehicle motion state identification method and device
US11814075B2 (en) * 2020-08-26 2023-11-14 Motional Ad Llc Conditional motion predictions
CN114312770B (en) * 2020-10-09 2023-07-07 宇通客车股份有限公司 Vehicle, vehicle running track prediction method and device
CN112308171A (en) * 2020-11-23 2021-02-02 浙江天行健智能科技有限公司 Vehicle position prediction modeling method based on simulated driver
CN112833903B (en) * 2020-12-31 2024-04-23 广州文远知行科技有限公司 Track prediction method, device, equipment and computer readable storage medium
CN112644518B (en) * 2020-12-31 2021-11-05 广州文远知行科技有限公司 Vehicle track prediction method, device, equipment and storage medium
CN113156963B (en) * 2021-04-29 2022-08-12 重庆大学 Deep reinforcement learning automatic driving automobile control method based on supervision signal guidance
CN113963535B (en) * 2021-09-30 2023-01-13 华为技术有限公司 Driving decision determination method and device and electronic equipment storage medium
FR3128177B1 (en) * 2021-10-18 2024-01-12 Psa Automobiles Sa Method and device for predicting a change of lane for a vehicle
CN113968243B (en) * 2021-11-11 2024-02-23 北京三快在线科技有限公司 Obstacle track prediction method, device, equipment and storage medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040143386A1 (en) * 2002-11-06 2004-07-22 Kiyohito Yoshihara Path predicting method for contents delivery apparatus
EP1835473A2 (en) * 2001-02-09 2007-09-19 MINTZ, Yosef Improved method and system for mapping traffic predictions with respect to telematics and route guidance applications
CN109166140A (en) * 2018-07-27 2019-01-08 长安大学 A kind of vehicle movement track estimation method and system based on multi-line laser radar
CN109583151A (en) * 2019-02-20 2019-04-05 百度在线网络技术（北京）有限公司 The driving trace prediction technique and device of vehicle
CN109789875A (en) * 2016-09-26 2019-05-21 日产自动车株式会社 Driving path setting method and driving path setting device
CN109887334A (en) * 2017-12-01 2019-06-14 奥迪股份公司 Vehicle drive assist system and method

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1835473A2 (en) * 2001-02-09 2007-09-19 MINTZ, Yosef Improved method and system for mapping traffic predictions with respect to telematics and route guidance applications
US20040143386A1 (en) * 2002-11-06 2004-07-22 Kiyohito Yoshihara Path predicting method for contents delivery apparatus
CN109789875A (en) * 2016-09-26 2019-05-21 日产自动车株式会社 Driving path setting method and driving path setting device
CN109887334A (en) * 2017-12-01 2019-06-14 奥迪股份公司 Vehicle drive assist system and method
CN109166140A (en) * 2018-07-27 2019-01-08 长安大学 A kind of vehicle movement track estimation method and system based on multi-line laser radar
CN109583151A (en) * 2019-02-20 2019-04-05 百度在线网络技术（北京）有限公司 The driving trace prediction technique and device of vehicle

Also Published As

Publication number Publication date
CN110789528A (en) 2020-02-14

Similar Documents

Publication Publication Date Title
CN110789528B (en) Vehicle driving track prediction method, device, equipment and storage medium
JP7143250B2 (en) A method and system for assisting an operator of an ego-vehicle in controlling the ego-vehicle by determining the future behavior and associated trajectory of the ego-vehicle
JP7361775B2 (en) Personal driving style learning for autonomous driving
CN111775949B (en) Personalized driver steering behavior auxiliary method of man-machine co-driving control system
CN112099496B (en) Automatic driving training method, device, equipment and medium
CN111284485B (en) Method and device for predicting driving behavior of obstacle vehicle, vehicle and storage medium
US10831208B2 (en) Vehicle neural network processing
CN112034834A (en) Offline agent for accelerating trajectory planning for autonomous vehicles using reinforcement learning
Chae et al. Virtual target-based overtaking decision, motion planning, and control of autonomous vehicles
US11465650B2 (en) Model-free reinforcement learning
CN113267199B (en) Method and device for planning driving track
CN113805572A (en) Method and device for planning movement
CN111868641A (en) Method for generating a training data set for training an artificial intelligence module of a vehicle control unit
CN113515038B (en) Vehicle lane changing method, device, equipment and readable storage medium
CN114667545A (en) Method for training at least one algorithm for a control unit of a motor vehicle, computer program product and motor vehicle
CN112859838B (en) Automatic driving control method, device, equipment and medium
CN113655789A (en) Path tracking method, device, vehicle and storage medium
CN113682312A (en) Autonomous lane changing method and system integrating deep reinforcement learning
Wei et al. Game theoretic merging behavior control for autonomous vehicle at highway on-ramp
CN112124310B (en) Vehicle path transformation method and device
CN114519931B (en) Method and device for predicting behavior of target vehicle in intersection environment
CN111491854A (en) Method, device, computer program and computer program product for driving maneuver assistance for a vehicle
CN115123217B (en) Mine obstacle vehicle driving track generation method and device and computer equipment
CN116653957A (en) Speed changing and lane changing method, device, equipment and storage medium
Arbabi et al. Lane-change initiation and planning approach for highly automated driving on freeways

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
REG Reference to a national code

Ref country code: HK

Ref legal event code: DE

Ref document number: 40021080

Country of ref document: HK

SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant