CN113077619B - Method, device, equipment and storage medium for vehicle motion prediction - Google Patents

Method, device, equipment and storage medium for vehicle motion prediction Download PDF

Info

Publication number
CN113077619B
CN113077619B CN202010649210.9A CN202010649210A CN113077619B CN 113077619 B CN113077619 B CN 113077619B CN 202010649210 A CN202010649210 A CN 202010649210A CN 113077619 B CN113077619 B CN 113077619B
Authority
CN
China
Prior art keywords
vehicle
behavior
determined
probability
value
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
CN202010649210.9A
Other languages
Chinese (zh)
Other versions
CN113077619A (en
Inventor
晏明扬
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.)
China Mobile Communications Group Co Ltd
China Mobile Shanghai ICT Co Ltd
CM Intelligent Mobility Network Co Ltd
Original Assignee
China Mobile Communications Group Co Ltd
China Mobile Shanghai ICT Co Ltd
CM Intelligent Mobility Network 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 China Mobile Communications Group Co Ltd, China Mobile Shanghai ICT Co Ltd, CM Intelligent Mobility Network Co Ltd filed Critical China Mobile Communications Group Co Ltd
Priority to CN202010649210.9A priority Critical patent/CN113077619B/en
Publication of CN113077619A publication Critical patent/CN113077619A/en
Application granted granted Critical
Publication of CN113077619B publication Critical patent/CN113077619B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Traffic Control Systems (AREA)

Abstract

The embodiment of the invention discloses a method, a device, equipment and a storage medium for predicting vehicle motion. The method comprises the steps of obtaining the motion states of a plurality of vehicles within a preset time period; determining at least one first behavior of the vehicle and a first probability of each of the at least one first behavior according to the motion state of the vehicle; determining a first value of at least one first behavior of the vehicle to be determined based on the first probability of each of the at least one first behavior; and determining a second behavior of the vehicle to be determined according to the first numerical value and the historical motion trail of the vehicle to be determined. The embodiment of the invention solves the problem that the motion condition of the vehicle cannot be accurately predicted, and improves the accuracy of predicting the motion condition of the vehicle.

Description

Method, device, equipment and storage medium for vehicle motion prediction
Technical Field
The present invention relates to the field of automatic driving technologies, and in particular, to a method, an apparatus, a device, and a storage medium for predicting vehicle motion.
Background
When the intelligent vehicle runs in a complex traffic environment, the development situation of the surrounding environment and the possible dangerous conditions need to be concerned all the time, so that the intelligent vehicle needs to reasonably predict the change of the surrounding environment.
Most vehicle motion prediction methods predict based on a motion model of a physical mechanism and a motion model of a behavior, but in the current motion prediction methods, each traffic vehicle is researched as an independent individual, the action of a main vehicle on the traffic vehicle is ignored, and the prediction effect is greatly reduced.
In the actual motion prediction process, the used sensing perception is often inaccurate and incomplete, and the state and the condition of the driver of the traffic vehicle are unknown, so that the actual motion trail of the traffic vehicle is full of uncertainty.
Therefore, the problem that the motion condition of the vehicle cannot be accurately predicted exists in the prior art.
Disclosure of Invention
The embodiment of the invention provides a method, a device, equipment and a storage medium for predicting vehicle motion, solves the problem that the motion condition of a vehicle cannot be accurately predicted, and improves the accuracy of predicting the motion condition of the vehicle.
In order to solve the technical problems, the invention comprises the following steps:
in a first aspect, a method for vehicle motion prediction is provided, the method comprising:
acquiring the motion states of a plurality of vehicles within a preset time period;
determining at least one first behavior of the vehicle and a first probability of each of the at least one first behavior according to the motion state of the vehicle;
determining a first value of at least one first behavior of the vehicle to be determined according to the first probability of each of the at least one first behavior;
and determining a second behavior of the vehicle to be determined according to the first numerical value and the historical motion trail of the vehicle to be determined.
In some implementations of the first aspect, the motion state includes at least one of coordinates, speed, acceleration, and direction information of the vehicle; the first behavior comprises at least one of left turn, right turn and straight line; said determining a first value of at least one first behavior of the vehicle to be determined from the first probability of each of said at least one first behavior comprises:
determining a first value of at least one first behavior of the vehicle to be determined based on the first probability of each of the at least one first behavior and the second value of each of the at least one first behavior for the vehicle to be determined.
In some implementations of the first aspect, the determining a first value of the at least one first behavior of the vehicle to be determined according to the first probability of each of the at least one first behavior and the second value of each of the at least one first behavior for the vehicle to be determined satisfies a formula:
Figure BDA0002574275210000021
wherein i is the first behavior, m0，iFor the first behavior of the vehicle to be determined, u (m)0，i) To determine a first value of a first behaviour of the vehicle to be determined, j being the first behaviour, mt，jA first behavior of any one vehicle except the vehicle to be determined in a set T composed of a plurality of vehicles, wherein T is the set composed of the plurality of vehicles, MtSet of first behaviors u for any one vehicle of the set T except the vehicle to be determined0() For determining a function of a second value of the vehicle to be determined, p (m)t，j) A first probability of a first behavior of any one of the vehicles in the set of vehicles T except the vehicle to be determined.
In some implementations of the first aspect, the determining a second behavior of the vehicle to be determined according to the first numerical value and the historical motion trajectory of the vehicle to be determined includes:
determining a second probability of at least one first behavior of the vehicle to be determined according to the first numerical value and the historical motion trail of the vehicle to be determined;
and determining that the corresponding first behavior is the second behavior of the vehicle to be determined when the second probability is the maximum value.
In a second aspect, there is provided an apparatus for vehicle motion prediction, the apparatus comprising:
the acquisition module is used for acquiring the motion states of a plurality of vehicles within a preset time period;
the processing module is used for determining at least one first behavior of the vehicle and a first probability of each first behavior of the at least one first behavior according to the motion state of the vehicle;
the processing module is further configured to determine a first numerical value of at least one first behavior of the vehicle to be determined according to the first probability of each of the at least one first behavior;
the processing module is further configured to determine a second behavior of the vehicle to be determined according to the first numerical value and the historical motion trajectory of the vehicle to be determined.
In some implementations of the second aspect, the motion state includes at least one of coordinates, speed, acceleration, and direction information of the vehicle; the first behavior comprises at least one of left turn, right turn and straight line;
the processing module is further configured to determine a first value of at least one first behavior of the vehicle to be determined according to the first probability of each of the at least one first behavior and the second value of each of the at least one first behavior for the vehicle to be determined.
In some implementations of the second aspect, the determining a first value of the at least one first behavior of the vehicle to be determined according to the first probability of each of the at least one first behavior and the second value of each of the at least one first behavior for the vehicle to be determined satisfies a formula:
Figure BDA0002574275210000031
wherein i is the first behavior, m0，iFor the first behavior of the vehicle to be determined, u (m)0，i) To determine a first value of a first behaviour of the vehicle to be determined, j being the first behaviour, mt，jA first behavior of any one vehicle except the vehicle to be determined in a set T composed of a plurality of vehicles, wherein T is the set composed of the plurality of vehicles, MtSet of first behaviors u for any one vehicle of the set T except the vehicle to be determined0() For determining a function of a second value of the vehicle to be determined, p (m)t，j) A first probability of a first behavior of any one of the vehicles in the set of vehicles T except the vehicle to be determined.
In some implementations of the second aspect, the processing module is further configured to determine a second probability of at least one first behavior of the vehicle to be determined according to the first numerical value and the historical motion trajectory of the vehicle to be determined;
the processing module is further configured to determine that the corresponding first behavior is the second behavior of the vehicle to be determined when the second probability is the maximum value.
In a third aspect, there is provided a vehicle motion prediction apparatus comprising: a processor and a memory storing computer program instructions;
the processor, when executing the computer program instructions, implements the first aspect, and methods of vehicle motion prediction in some implementations of the first aspect.
In a fourth aspect, a computer storage medium is provided, having computer program instructions stored thereon that, when executed by a processor, implement the first aspect and methods of vehicle motion prediction in some implementations of the first aspect.
The embodiment of the invention provides a vehicle motion prediction method, a vehicle motion prediction device, a vehicle motion prediction equipment and a storage medium, wherein at least one first behavior of a vehicle and a first probability of each first behavior in the at least one first behavior are determined according to motion states of the vehicle by acquiring motion states of a plurality of vehicles in a preset time period, and a first numerical value of the at least one first behavior of the vehicle to be determined is determined according to the first probability of each first behavior in the at least one first behavior; and finally, determining a second behavior of the vehicle to be determined according to the first numerical value and the historical motion track of the vehicle to be determined, wherein the second behavior is the behavior of the vehicle to be determined in the future time. The embodiment of the invention solves the problem that the motion condition of the vehicle cannot be accurately predicted in the prior technical scheme, and improves the accuracy of predicting the motion condition of the vehicle.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described below, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart diagram illustrating a method for vehicle motion prediction according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a method for obtaining a vehicle trajectory feature using a sliding window according to an embodiment of the present invention;
FIG. 3 is a schematic illustration of an in-lane vehicle interaction provided by an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an apparatus for predicting vehicle motion according to an embodiment of the present invention;
fig. 5 is a block diagram of an exemplary hardware architecture of a computing device provided by an embodiment of the invention.
Detailed Description
Features and exemplary embodiments of various aspects of the present invention will be described in detail below, and in order to make objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not to be construed as limiting the invention. It will be apparent to one skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present invention by illustrating examples of the present invention.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
With the increasing level of vehicle intelligence, some advanced driving assistance systems and partial automatic driving functions have appeared in mass production vehicles, and the complete intelligent driving function has entered a large-scale test stage. The continuous maturity of the vehicle intelligent systems has great significance for solving the problems of safety, congestion, environmental protection and the like brought by the traditional automobile industry. Automobile intelligence has also been recognized as one of the inevitable directions in automobile technology development.
When the intelligent vehicle runs in a complex traffic environment, the intelligent vehicle needs to pay attention to the development situation of the surrounding environment and possible dangerous situations. When the intelligent vehicle makes a decision and plans a track which is safe and feasible in the future and meets requirements of comfort, economy, driving efficiency and the like, the change of the surrounding environment needs to be reasonably predicted. Human drivers can learn to learn the ability to infer the intent of travel of surrounding traffic vehicles and to predict their future movement by constantly learning in the traffic environment. However, in the case of an intelligent driving system, it is necessary to perform probabilistic inference through observable variables, which cannot be directly observed from a large amount of information in the surrounding driving environment, for example, the driving intention of the surrounding vehicle. Therefore, it is important for an intelligent driving system to study a motion prediction method reflecting the driving intention of a transportation vehicle.
At present, most methods for predicting the motion of the traffic vehicles are motion models based on physical mechanisms and motion models based on behaviors. The two methods only consider the one-way influence of the traffic vehicle on the movement of the main vehicle when performing movement prediction, namely, the movement prediction result of the traffic vehicle is used as the input of the main vehicle decision and plan. In fact, the movement of the main vehicle can generate a large image of the movement of the traffic vehicle, but the methods study each traffic vehicle as an independent individual during movement prediction, and neglect the effect of the main vehicle on the traffic vehicle. Obviously, under the scene with obvious behavior interactivity, such as an expressway entrance or an urban congested road, the method can cause the prediction effect of the model to be greatly reduced. On the other hand, most of the current motion prediction methods are deterministic trajectory prediction, and the methods are usually realized based on dynamic and kinematic models of vehicles, however, sensing perception in the actual motion prediction process is often inaccurate and incomplete, and the state and the condition of drivers of traffic vehicles are unknown, so that the actual motion trajectory of the traffic vehicles is full of uncertainty. In motion prediction over a longer time, it is obviously difficult for such deterministic motion prediction methods to obtain reliable results. In addition, even in the current method considering uncertainty, the uncertainty is simulated only by adding gaussian noise, and the probability distribution of vehicle motion in a real traffic scene is difficult to accurately express.
Therefore, the problem that the motion condition of the vehicle cannot be accurately predicted exists in the prior art.
In order to solve the problem that the motion condition of a vehicle cannot be accurately predicted in the existing technical scheme, embodiments of the present invention provide a method, an apparatus, a device, and a storage medium for predicting vehicle motion, in which motion states of a plurality of vehicles within a preset time period are obtained, at least one first behavior of the vehicle and a first probability of each first behavior of the at least one first behavior are determined according to the motion states of the vehicles, and a first value of at least one first behavior of the vehicle to be determined is determined according to the first probability of each first behavior of the at least one first behavior; and finally, determining a second behavior of the vehicle to be determined according to the first numerical value and the historical motion trail of the vehicle to be determined. The problem of exist among the current technical scheme and carry out accurate prediction to the motion condition of vehicle is solved, the accuracy of predicting the vehicle motion condition has been improved.
The technical solutions provided by the embodiments of the present invention are described below with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a method for predicting vehicle motion according to an embodiment of the present invention.
As shown in fig. 1, the main body of execution of the method may be an in-vehicle terminal device, and the method of vehicle motion prediction may include:
s101: the motion states of a plurality of vehicles within a preset time period are acquired.
Wherein the motion state may include at least one of coordinates, speed, acceleration, and direction information of the vehicle. The obtaining of the motion states of the plurality of vehicles within the preset time period may be the motion states of a plurality of surrounding vehicles that can be obtained within a sensor coverage range obtained by a sensor of the on-board terminal device within the preset time period, and the plurality of surrounding vehicles may also include a vehicle that performs the obtaining action.
Specifically, the preset time period may include an elapsed h time period, and the plurality of vehicles may be represented using n vehicles.
Therefore, in one embodiment, the motion states of a plurality of vehicles within a preset time period may be obtained by obtaining the motion states of n vehicles observed in the past h time, and the data obtained by the process may be as shown in formula (1).
Figure BDA0002574275210000071
Where δ represents a moving state of the vehicle, δ ═ (x, y, v)x，vy，ax，ayZ), x is the x-axis coordinate of the global coordinates of the vehicle, y is the y-axis coordinate of the global coordinates of the vehicle, vxSpeed of the vehicle in the x-axis, vySpeed of the vehicle in the y-axis, axAcceleration of the vehicle in the x-axis, ayIs the acceleration of the vehicle on the y axis, z is the heading information of the vehicle, h is the time length of the time period, t is the starting time of the time period, deltat-h：tThe motion states of n vehicles in the past h time,
Figure BDA0002574275210000072
for the motion state of the vehicle 1 in the elapsed h time,
Figure BDA0002574275210000073
the motion state of the vehicle i in the past h time,
Figure BDA0002574275210000074
the motion state of the vehicle n in the past h time.
Optionally, in an embodiment, because of factors such as accuracy of a sensor on the vehicle-mounted terminal device, a measurement range, and the like, noise and errors may exist in the data, and therefore, the motion state δ of the vehicle may be obtained by state estimation, and a commonly used method may include kalman filtering.
Optionally, in one embodiment, since the motion state of the vehicle is all for a continuous driving process, the behavior of the vehicle cannot be judged only by observing the data of the variables at a single moment. To solve the problem, a sliding time window method can be used to obtain the track characteristics of the vehicle, and the track within a period of time is simultaneously focused at each time t, so as to estimate the behavior probability distribution of the vehicle at the moment. Therefore, judgment errors caused by only paying attention to a certain moment can be avoided, and the fault tolerance rate and the judgment accuracy of the system are increased.
Further, the process of obtaining the moving state of the vehicle may be as shown in fig. 2.
Fig. 2 is a schematic diagram of a method for acquiring a track characteristic of a vehicle by using a sliding time window according to an embodiment of the present invention.
The process of obtaining the trajectory characteristics of the vehicle using the sliding time window method is shown in fig. 2. When the behavior recognition is performed at the time t, the time is Δ tCFor the time interval, n feature points on the sampling trajectory are used as input to the algorithm, and thus the time width of the time window can be expressed as Δ Tc＝nΔtc。
Obtaining a first behavior O of the vehicle which can be directly observed at the time t according to the n collected feature pointstAs shown in equation (2).
Figure BDA0002574275210000081
Wherein d and
Figure BDA0002574275210000082
may each be a vector of n x 1 dimensions:
dt＝[d(t-(n-1)Δtc)，d(t-(n-2)Δtc)，...，d(t)]T∈Rn
Figure BDA0002574275210000083
obtaining observation data on n feature points by sliding a time window, dtIs the speed of the vehicle and is,
Figure BDA0002574275210000084
r is the acceleration of the vehicle and is the set of trajectories of the vehicle. The first sequence of behaviors of the set of vehicles that can be directly observed can be written as: o ═ O1O2...OTFrom this set of observation sequences, the state of motion of the vehicle can be determined.
After obtaining the motion state of the vehicle, S102 may be entered to determine a first behavior of the vehicle and a first probability of the first behavior.
S102: at least one first behavior of the vehicle and a first probability of each of the at least one first behavior are determined based on the state of motion of the vehicle.
Specifically, after obtaining the motion state of the vehicle, a first behavior of each vehicle that is possible in the future can be determined, and the probability of the first behavior can be predicted. The first behavior may be understood as a behavior that may occur in the vehicle for a future period of time.
FIG. 3 is a schematic diagram of an in-lane vehicle interaction provided by an embodiment of the present invention.
Alternatively, the process of determining the first behavior of each vehicle, and the process of determining the probability of the first behavior may be described by taking three vehicles shown in fig. 3 as an example.
As shown in FIG. 3, there are three ABC vehicles in FIG. 3, because the A vehicle is in the middle of the lane, so the A vehicle can have three first behaviors of turning left, turning right and going straight; the B vehicle is on the left side of the lane, so the A vehicle can have two first behaviors of turning right and going straight; the C vehicle is on the left side of the lane, so the C vehicle can have two first behaviors of turning left and going straight.
Specifically, the different first behaviors of the three ABC vehicles may be shown in equations (3), (4), and (5), respectively.
MA＝{mA，k|k∈{1，2}＝{LCL，LK} (3)
MB＝{mB，i|i∈{1，2}＝{LK，LCR} (4)
MC＝{mC，j|j∈{1，2，3}＝{LCL，LK，LCR} (5)
Wherein M isASet of first behaviors for A cars, mA，kThe behaviors of the A vehicles are shown, and k is the number of the behaviors of the A vehicles; mBSet of first behaviors for B cars, mB，iThe behaviors of the B vehicle are shown, i is the number of the behaviors of the B vehicle; mCSet of first behaviors for car C, mC，jThe behaviors of the C vehicle are shown, and j is the number of the behaviors of the C vehicle; LCL is the behavior of left turn, LK is the behavior of straight going, LCR is the behavior of right turn.
Since there is uncertainty about the driver's future behavior during road driving, there is a probability that each behavior of A, B, C three vehicles will be.
In one embodiment, the probabilistic predictive expression for the first behavior of each vehicle may be as shown in equation 6.
p(mi|δt-h：t)∈[0，1] (6)
Where p represents the probability of the first action, miFor the first behaviour of the vehicle, δt-h：tThe motion state of the vehicle in the past h time.
A. B, C the probability of each behavior of the three vehicles can be p (m) respectivelyA，k)、p(mB，i)、p(mC，j) And (4) showing. And p (m) because the vehicle can not be considered to exit the roadA，LCR)＝p(mB，LCL)＝0。
Because Hidden Markov Models (HMMs) refer to time-sequential probabilistic models in which a single discrete random variable describes a process state, Gaussian Mixture Models (GMMs) can be used to represent any form of continuous probability distribution theoretically, have superior computational advantages, and are often used to represent output observation probabilities. And the GMM is used for expressing the HMM outputting the observation probability, various behavior probabilities at each moment are calculated to be discrete points, and a continuous probability distribution model is obtained through Gaussian model smoothing processing, so that the calculation error of the algorithm is reduced, and the judgment accuracy of the algorithm is improved.
Thus, optionally, in one embodiment, a GMM-HMM behavior recognition model may be used to determine the probability of each first behavior of the vehicle and determine a possible sequence of states. The process of determining a possible sequence of states by calculating the probability that the vehicle is in each behavior is shown in equation (7).
γt(i)＝p(qt＝si|O，λ)，i∈[1，N]，t∈[1，T] (7)
Wherein gamma ist(i) Is a first sequence of behaviors of the vehicle, qtIs a set of first behaviors of the vehicle related to time t, siIs a first behavior implied by the vehicle, O is a first behavior that can be directly observed, λ is a preset model for calculating the probability, p () is the probability of the first behavior of the vehicle, i is the number of the first behavior of the vehicle, t represents the vehicle,n may be 3.
By means of bayesian equations, equation (7) can be transformed into:
Figure BDA0002574275210000101
the probability of each first behavior of the vehicle may then be determined by solving a first underlying problem of the hidden markov model to generate an observation sequence probability.
After determining the probability of the first action, the process may proceed to S103.
S103: a first value of at least one first behavior of the vehicle to be determined is determined as a function of the first probability of each of the at least one first behavior.
In particular, a first value of at least one first behavior of the vehicle to be determined may be determined based on a first probability of each of the at least one first behavior and a second value of each of the at least one first behavior for the vehicle to be determined. The second value may be referred to as a profit value, and the first value may be referred to as an expectation value.
The revenue function for A, B, C three vehicles, according to the vehicle shown in fig. 3, can be as shown in equations (8), (9), (10).
UA(mA，k，mB，i，mC，j) (8)
UB(mA，k，mB，i，mC，j) (9)
UC(mA，k，mB，i，mC，j) (10)
Wherein, UA() As a revenue function of vehicle A, UB() As a revenue function for vehicle B, UC() As a revenue function of vehicle C, mA，k∈MA，mA，kIs the first action of vehicle A, mB，i∈MB，mB，kAs the first action of the B car, mC，j∈MC，mC，uThe first behavior of car C.
Specifically, the expression of equation (8) means the first behavior of the vehicle a, the first behavior of the vehicle B, and the first behavior of the vehicle C for the vehicle a. The meaning of equation (9), equation (10) and so on.
After obtaining the revenue function of the respective vehicle, expected values of the at least one first behavior of the vehicle to be determined may be determined based on the first probability of each of the at least one first behavior and the revenue value of each of the at least one first behavior for the vehicle to be determined. The theory of using the expected value for the judgment may also be referred to as an expected utility theory.
Alternatively, for vehicle a, the expected utility resulting from left turn or straight behavior may be as shown in equation (11) and equation (12), respectively, according to the expected utility theory.
Figure BDA0002574275210000111
Figure BDA0002574275210000112
Wherein m isA，LCLFor the left turn behavior of vehicle A, mB，iFor the first behaviour of vehicle B, MBIs a set of first behaviors of the vehicle B, mC，jIs the first action of the C car, mA，LKFor the straight-ahead behavior of the vehicle A, p () is a function of the calculated probability, uA() As a function of the revenue for car a.
Further, it can be deduced therefrom that the scene within the lane can be as shown in equation (13).
Figure BDA0002574275210000113
H in the formula (13) represents a scene of a plurality of vehicles in a lane, mt，jIs the first behavior of any vehicle in the lane, j is the first behavior, t is any one in the laneA vehicle, T is a set of a plurality of vehicles, MtIs a set of first behaviors of any one vehicle in the set T.
Thus, for a vehicle to be determined of the plurality of vehicles, the expected value produced by the vehicle to be determined may be as shown in equation (14), where the expected value may also be referred to as the expected utility, according to equation (13) for the scene within the determined lane.
Figure BDA0002574275210000121
Wherein i is the first behavior, m0，iFor the first behavior of the vehicle to be determined, u (m)0，i) To determine a function of the expected value of a first behaviour of the vehicle to be determined, j being the first behaviour, mt，jA first behavior of any one vehicle except the vehicle to be determined in a set T composed of a plurality of vehicles, wherein T is the set composed of the plurality of vehicles, MtSet of first behaviors, u, of any one vehicle in the set T0() For determining a function of a second value of the vehicle to be determined, p (m)t，j) A first probability of a first behavior of any one of the vehicles in the set of vehicles T except the vehicle to be determined.
It should be noted that, in the embodiment of the present invention, the vehicles in the same scene do not share their respective revenue matrices
Figure BDA0002574275210000122
Each vehicle sees the game G from the perspective of the vehicle, and the decision and display show that the traffic scene is constant and the vehicles are not uniformly and coordinately controlled, so that each vehicle decides independently.
Optionally, after obtaining the expected value of the vehicle to be determined, the prediction accuracy may be further improved by combining historical past data of the vehicle to be determined, and therefore, S104 may be entered.
S104: and determining a second behavior of the vehicle to be determined according to the first numerical value and the historical motion trail of the vehicle to be determined.
Because of the expected utility u (m) produced by each of its behaviors for the predicted vehicle to be determined0，j) The value of (a) represents in fact that the driver of the vehicle selects this row m at the level of the intention for a future period of time0，iThe size of the probability of (c). Thus, can be according to m0，iDetermines a second behavior of the vehicle to be determined.
In one embodiment, the desired utility u (m) may be0，j) The value of the time difference is normalized to obtain the behavior m of the driver in the future period0，iProbability of intention. The process of normalization can be as shown in equation (15).
Figure BDA0002574275210000123
Wherein p isintend() Function representing probability of intention, m0，iFor the vehicle to be determined, deltat-h：tFor the motion states of other vehicles in the surroundings of the vehicle to be determined, g () is a normalization function, u (m)0，i) For determining a function of the expected value of a first behaviour of the vehicle to be determined, i being the first behaviour, M0Is a set of first behaviors of the vehicle to be determined.
According to the desired effect u (m)0，i) To calculate pintend(m0，i|δt-h：t) The process of (1) is to predict the traffic vehicle v by the surrounding traffic situation by using a game method0The game is viewed from the perspective of each vehicle, the decision and the displayed traffic scene are constant, uniform coordinated control is not performed on all vehicles, and each vehicle makes an independent decision, so that the prediction accuracy of the motion condition of the vehicle is improved.
Optionally, in an embodiment, in order to further improve the prediction accuracy by combining historical past data of the vehicle to be determined, the historical motion trajectory p of the vehicle to be determined may be combinedrecog(m0，i|δt-h：t) Determining a second behavior of the vehicle to be determinedThis second behavior may also be referred to as the final behavior.
Specifically, the process of deriving the final behavior of the vehicle to be determined based on the above historical motion trajectory and the inference of the expected value may be as shown in equation (16).
p(m0，i|δt-h：t)＝τ1precog(m0，i|δt-h：t)+τ2pintend(m0，i|δt-h：t) (16)
Where p () is the second probability of the first behavior of the vehicle to be determined, m0，iFor the first behavior of the vehicle to be determined, δt-h：tFor the state of motion of other vehicles in the surroundings of the vehicle to be determined, precog() As a function of the history of the movement track, pintend() As a function of the probability of intention, τ1The weight coefficient, tau, occupied by the historical motion track in the final behavior prediction probability2For the weight coefficient occupied by the intention probability in the final behavior prediction probability, and furthermore τ1+τ2＝1。
After the second probability is obtained by calculating the first behavior of the vehicle to be determined, the first behavior corresponding to the maximum second probability may be determined as the second behavior of the vehicle to be determined.
Optionally, in the process of according to the first value and the historical motion trajectory of the vehicle to be determined, a forward-backward algorithm of an HMM may be further used to extract a motion expectation probability before a current time state of the vehicle, extract a motion expectation probability after the current time state, and solve the maximum motion expectation through an iterative method.
In order to improve the calculation efficiency and reduce the algorithm complexity, a forward and backward algorithm is generally used, and a forward variable α in the algorithmt(i) And a backward variable betat(i) May be represented by equation (17) and equation (18), respectively.
αt(i)＝p(o1o2...ot|qt＝si，λ) (17)
βt(i)＝p(ot+1ot+2...oT|qt＝si，λ) (18)
αt(i) Representing a first sequence of behaviors o from a past initial moment to a time t that can be observed directly given a preset model λ1o2...otAnd state siProbability of occurrence, betat(i) Representing a first sequence of behaviors o from time T to time T that can be observed directly given a preset model λt+1ot+2...oTAnd state siProbability of occurrence of the risk. Wherein alpha ist(i) Representing the historical movement path, beta, of the vehicle to be determinedt(i) The method has the advantages that the influence of surrounding vehicles on the vehicles to be determined is combined in a future period, the obtained movement probability data of the main vehicle is predicted, the historical data and the future probability of the main vehicle are fully utilized to decide the behavior of the main vehicle, the prediction of the movement of the main vehicle in a real scene is more met, and meanwhile, the accuracy of the prediction of the movement of the main vehicle is greatly improved.
Optionally, in one embodiment, the determination may be made based on the in-vehicle travelable space, the collision risk index, and the comfort index when determining the benefit of the at least one first behavior of the vehicle to be determined.
Specifically, the idle distances in front of the vehicle and behind the vehicle can be obtained according to the front sensors and the rear sensors of the vehicle, the left idle distance and the right idle distance of the vehicle are obtained through the sensors on the two sides, the running space in front of the vehicle can be determined according to the idle distances in front of the vehicle and behind the vehicle and the left idle distance and the right idle distance of the vehicle, and meanwhile, the running space coefficient in front of the vehicle can be specified to be ns.
Since safety is often the most important factor during driving. Therefore, the future form track of the vehicle can be predicted, and then the danger assessment is carried out based on the future form track to determine the collision danger index. Specifically, a collision condition may be defined when the distance between the vehicle to be determined and a surrounding vehicle is less than 5 cm. When the sensor senses a specific vehicle distance, collision danger early warning is carried out, and meanwhile, the dangerous collision coefficient can be designated as roc.
Since the most direct motion state of the vehicle affecting ride comfort is the acceleration of the vehicle, the acceleration of the vehicle can be used to determine the comfort index. Specifically, the acceleration of the observed vehicle may be resolved into longitudinal and lateral accelerations to determine the comfort index. And calculating the integral of longitudinal and lateral acceleration in a certain period T, and when the integral is more than 10, judging that the comfort is low and designating a comfort index c.
The determination of the yield by incorporating the coefficients of the terms in equation (7) is shown in equation (19).
γt(i)＝p(qt＝si|O，λ)，i∈[1，N]，t∈[1，T]*0.3ns*0.5roc*0.2c (19)
As shown in equation (19), the front drivable space may account for 30%, the collision risk index may account for 50%, and the comfort index may account for 20%.
The profit is determined according to the formula (17), and completely different behaviors can be decided under different situations, such as the same influence of surrounding vehicles when the distance between vehicles is in a normal state and in a special situation in the driving process and under the same influence of surrounding vehicles by combining various contextual profit coefficients, so that the uncertainty of a traffic scene in real life is better met, and the applicability of the algorithm is increased.
Further, under different scenes, by introducing scene income coefficients into the probability model, expected values under different scenes can be calculated, so that the motion behaviors of the vehicle to be determined are decided by utilizing a maximum expected strategy, the method is suitable for the complex traffic scene in reality, and the robustness and the practicability of the algorithm are improved, wherein the vehicle to be determined comprises a main vehicle with vehicle-mounted terminal equipment.
According to the vehicle motion prediction method provided by the embodiment of the invention, at least one first behavior of a vehicle and a first probability of each first behavior in the at least one first behavior are determined according to the motion states of the vehicles by acquiring the motion states of a plurality of vehicles in a preset time period, and then a first numerical value of the at least one first behavior of the vehicle to be determined is determined according to the first probability of each first behavior in the at least one first behavior; and finally, determining a second behavior of the vehicle to be determined according to the first numerical value and the historical motion trail of the vehicle to be determined. The problem of exist among the current technical scheme and carry out accurate prediction to the motion condition of vehicle is solved, the accuracy of predicting the vehicle motion condition has been improved.
Corresponding to the embodiment of the method for predicting the vehicle motion, the embodiment of the invention also provides a device for predicting the vehicle motion, as shown in fig. 4. Fig. 4 is a schematic structural diagram of a vehicle motion prediction apparatus according to an embodiment of the present invention.
The vehicle motion prediction apparatus shown in fig. 4 may include an obtaining module 401 and a processing module 402.
The obtaining module 401 may be configured to obtain motion states of multiple vehicles within a preset time period.
The processing module 402 may be configured to determine at least one first behavior of the vehicle and a first probability of each of the at least one first behavior according to the motion state of the vehicle.
The processing module 402 may be further configured to determine a first value of at least one first behavior of the vehicle to be determined according to the first probability of each of the at least one first behavior.
The processing module 402 may further be configured to determine a second behavior of the vehicle to be determined according to the first numerical value and the historical motion trajectory of the vehicle to be determined.
The motion state includes at least one of coordinates, speed, acceleration, and direction information of the vehicle.
The first behavior comprises at least one of left turn, right turn and straight line.
The processing module 402 may be further configured to determine a first value of at least one first behavior of the vehicle to be determined according to the first probability of each of the at least one first behavior and the second value of each of the at least one first behavior for the vehicle to be determined.
Determining a first value of at least one first behavior of the vehicle to be determined, according to a first probability of each of the at least one first behavior and a second value of each of the at least one first behavior for the vehicle to be determined, satisfying the formula:
Figure BDA0002574275210000161
wherein i is the first behavior, m0，iFor the first behavior of the vehicle to be determined, u (m)0，i) To determine a first value of a first behaviour of the vehicle to be determined, j being the first behaviour, mt，jA first behavior of any one vehicle except the vehicle to be determined in a set T composed of a plurality of vehicles, wherein T is the set composed of the plurality of vehicles, MtSet of first behaviors u for any one vehicle of the set T except the vehicle to be determined0() For determining a function of a second value of the vehicle to be determined, p (m)t，j) A first probability of a first behavior of any one of the vehicles in the set of vehicles T except the vehicle to be determined.
The processing module 402 may be further configured to determine a second probability of at least one first behavior of the vehicle to be determined according to the first numerical value and the historical motion trajectory of the vehicle to be determined.
The processing module 402 may be further configured to determine that the corresponding first behavior when the second probability is the maximum value is the second behavior of the vehicle to be determined.
It can be understood that each module in the device for predicting vehicle motion shown in fig. 4 has a function of implementing each step from S101 to S104 in fig. 1, and can achieve the corresponding technical effect, and for brevity, no further description is provided here.
According to the vehicle motion prediction device provided by the embodiment of the invention, at least one first behavior of a vehicle and a first probability of each first behavior in the at least one first behavior are determined according to the motion states of the vehicles in a preset time period, and then a first numerical value of the at least one first behavior of the vehicle to be determined is determined according to the first probability of each first behavior in the at least one first behavior; and finally, determining a second behavior of the vehicle to be determined according to the first numerical value and the historical motion trail of the vehicle to be determined. The problem of exist among the current technical scheme and carry out accurate prediction to the motion condition of vehicle is solved, the accuracy of predicting the vehicle motion condition has been improved.
FIG. 5 illustrates a block diagram of an exemplary hardware architecture of a computing device capable of implementing the method of vehicle motion prediction of an embodiment of the present invention. As shown in fig. 5, computing device 500 includes an input device 501, an input interface 502, a central processor 503, a memory 504, an output interface 505, and an output device 506. The input interface 502, the central processing unit 503, the memory 504, and the output interface 505 are connected to each other through a bus 510, and the input device 501 and the output device 506 are connected to the bus 510 through the input interface 502 and the output interface 505, respectively, and further connected to other components of the computing device 500.
Specifically, the input device 501 receives input information from the outside and transmits the input information to the central processor 503 through the input interface 502; the central processor 503 processes input information based on computer-executable instructions stored in the memory 504 to generate output information, temporarily or permanently stores the output information in the memory 504, and then transmits the output information to the output device 506 through the output interface 505; output device 506 outputs the output information outside of computing device 500 for use by a user.
That is, the computing device shown in fig. 5 may also be implemented as a vehicle motion prediction device that may include: a memory storing computer-executable instructions; and a processor which, when executing computer executable instructions, may implement the method of vehicle motion prediction provided by embodiments of the present invention.
An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium has computer program instructions stored thereon; the computer program instructions, when executed by a processor, implement a method for vehicle motion prediction provided by embodiments of the present invention.
It is to be understood that the invention is not limited to the specific arrangements and instrumentality described above and shown in the drawings. A detailed description of known methods is omitted herein for the sake of brevity. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present invention are not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications and additions or change the order between the steps after comprehending the spirit of the present invention.
The functional blocks shown in the above-described structural block diagrams may be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, plug-in, function card, or the like. When implemented in software, the elements of the invention are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine-readable medium or transmitted by a data signal carried in a carrier wave over a transmission medium or a communication link. A "machine-readable medium" may include any medium that can store or transfer information. Examples of a machine-readable medium include electronic circuits, semiconductor memory devices, ROM, flash memory, Erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via computer networks such as the internet, intranet, etc.
It should also be noted that the exemplary embodiments mentioned in this patent describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.
Aspects of the present disclosure are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions/acts specified in the flowchart and/or block diagram block or blocks. Such a processor may be, but is not limited to, a general purpose processor, a special purpose processor, an application specific processor, or a field programmable logic circuit. It will also be understood that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware for performing the specified functions or acts, or combinations of special purpose hardware and computer instructions.
As described above, only the specific embodiments of the present invention are provided, and it can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system, the module and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. It should be understood that the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the present invention, and these modifications or substitutions should be covered within the scope of the present invention.

Claims (8)

1. A method of vehicle motion prediction, the method comprising:
acquiring the motion states of a plurality of vehicles within a preset time period;
determining at least one first behavior of the vehicle and a first probability of each of the at least one first behavior according to the motion state of the vehicle;
determining a first value of at least one first behavior of the vehicle to be determined according to the first probability of each of the at least one first behavior;
determining a second behavior of the vehicle to be determined according to the first numerical value and the historical motion track of the vehicle to be determined;
said determining a first value of at least one first behavior of the vehicle to be determined from the first probability of each of said at least one first behavior comprises:
determining a first value of at least one first behavior of the vehicle to be determined according to a first probability of each of the at least one first behavior and a second value of each of the at least one first behavior for the vehicle to be determined;
the determining of the first value of the at least one first behavior of the vehicle to be determined according to the first probability of each of the at least one first behavior and the second value of each of the at least one first behavior for the vehicle to be determined satisfies the formula:
Figure FDA0003300140890000011
wherein i is the first behavior, m0，iFor the first behavior of the vehicle to be determined, u (m)0，i) To determine a first value of a first behaviour of the vehicle to be determined, j being the first behaviour, mt，jA first behavior of any one vehicle except the vehicle to be determined in a set T composed of a plurality of vehicles, wherein T is the set composed of the plurality of vehicles, MtSet of first behaviors u for any one vehicle of the set T except the vehicle to be determined0() For determining a function of a second value of the vehicle to be determined, p (m)t，j) A first probability of a first behavior of any one of the vehicles in the set of vehicles T except the vehicle to be determined.
2. The method of claim 1, wherein the motion state comprises at least one of coordinates, speed, acceleration, and direction information of the vehicle; the first behavior comprises at least one of left turn, right turn and straight line.
3. The method according to claim 1 or 2, wherein the determining of the second behavior of the vehicle to be determined from the first numerical value and the historical movement trajectory of the vehicle to be determined comprises:
determining a second probability of at least one first behavior of the vehicle to be determined according to the first numerical value and the historical motion trail of the vehicle to be determined;
and determining that the corresponding first behavior is the second behavior of the vehicle to be determined when the second probability is the maximum value.
4. An apparatus for vehicle motion prediction, the apparatus comprising:
the acquisition module is used for acquiring the motion states of a plurality of vehicles within a preset time period;
the processing module is used for determining at least one first behavior of the vehicle and a first probability of each first behavior of the at least one first behavior according to the motion state of the vehicle;
the processing module is further configured to determine a first numerical value of at least one first behavior of the vehicle to be determined according to the first probability of each of the at least one first behavior;
the processing module is further used for determining a second behavior of the vehicle to be determined according to the first numerical value and the historical motion track of the vehicle to be determined;
the processing module is further configured to determine a first value of at least one first behavior of the vehicle to be determined according to a first probability of each of the at least one first behavior and a second value of each of the at least one first behavior for the vehicle to be determined;
the determining of the first value of the at least one first behavior of the vehicle to be determined according to the first probability of each of the at least one first behavior and the second value of each of the at least one first behavior for the vehicle to be determined satisfies the formula:
Figure FDA0003300140890000021
wherein i is the first behavior, m0，iFor the first behavior of the vehicle to be determined, u (m)0，i) To determine a first value of a first behaviour of the vehicle to be determined, j being the first behaviour, mt，jA first behavior of any one vehicle except the vehicle to be determined in a set T composed of a plurality of vehicles, wherein T is the set composed of the plurality of vehicles, MtSet of first behaviors u for any one vehicle of the set T except the vehicle to be determined0() For determining a function of a second value of the vehicle to be determined, p (m)t，j) A first probability of a first behavior of any one of the vehicles in the set of vehicles T except the vehicle to be determined.
5. The apparatus of claim 4, wherein the motion state comprises at least one of coordinates, speed, acceleration, and direction information of the vehicle; the first behavior comprises at least one of left turn, right turn and straight line.
6. The apparatus according to claim 4 or 5,
the processing module is further used for determining a second probability of at least one first behavior of the vehicle to be determined according to the first numerical value and the historical motion trail of the vehicle to be determined;
the processing module is further configured to determine that the corresponding first behavior is the second behavior of the vehicle to be determined when the second probability is the maximum value.
7. An apparatus for vehicle motion prediction, the apparatus comprising: a processor and a memory storing computer program instructions;
the processor, when executing the computer program instructions, implements a method of vehicle motion prediction according to any of claims 1-3.
8. A computer storage medium having computer program instructions stored thereon which, when executed by a processor, implement a method of vehicle motion prediction as claimed in any one of claims 1 to 3.
CN202010649210.9A 2020-07-08 2020-07-08 Method, device, equipment and storage medium for vehicle motion prediction Active CN113077619B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010649210.9A CN113077619B (en) 2020-07-08 2020-07-08 Method, device, equipment and storage medium for vehicle motion prediction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010649210.9A CN113077619B (en) 2020-07-08 2020-07-08 Method, device, equipment and storage medium for vehicle motion prediction

Publications (2)

Publication Number Publication Date
CN113077619A CN113077619A (en) 2021-07-06
CN113077619B true CN113077619B (en) 2021-12-07

Family

ID=76609004

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010649210.9A Active CN113077619B (en) 2020-07-08 2020-07-08 Method, device, equipment and storage medium for vehicle motion prediction

Country Status (1)

Country Link
CN (1) CN113077619B (en)

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006315514A (en) * 2005-05-12 2006-11-24 Toyota Motor Corp Steering control device for vehicle
CN103069466A (en) * 2010-06-18 2013-04-24 本田技研工业株式会社 System for inferring driver's lane change intention
DE102017223222A1 (en) * 2017-12-19 2019-06-19 Zf Friedrichshafen Ag Method for determining a tripping time of an active personnel protection device and a personal protection device
CN109949611A (en) * 2019-03-28 2019-06-28 百度在线网络技术（北京）有限公司 The lane change method, apparatus and storage medium of unmanned vehicle
CN110146100A (en) * 2018-02-13 2019-08-20 华为技术有限公司 Trajectory predictions method, apparatus and storage medium
CN110304063A (en) * 2019-07-15 2019-10-08 广州小鹏汽车科技有限公司 A kind of automatic lane change method and system of vehicle, vehicle
CN110758382A (en) * 2019-10-21 2020-02-07 南京航空航天大学 Surrounding vehicle motion state prediction system and method based on driving intention
CN110796856A (en) * 2019-10-16 2020-02-14 腾讯科技（深圳）有限公司 Vehicle lane change intention prediction method and training method of lane change intention prediction network
CN110908375A (en) * 2019-11-14 2020-03-24 北京三快在线科技有限公司 Method and device for acquiring lane change decision information, storage medium and vehicle
CN111267846A (en) * 2020-02-11 2020-06-12 南京航空航天大学 Game theory-based peripheral vehicle interaction behavior prediction method
CN111284485A (en) * 2019-10-10 2020-06-16 中国第一汽车股份有限公司 Method and device for predicting driving behavior of obstacle vehicle, vehicle and storage medium

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20140121544A (en) * 2013-04-05 2014-10-16 한국전자통신연구원 Apparatus for providing collision information in intersection and method thereof

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006315514A (en) * 2005-05-12 2006-11-24 Toyota Motor Corp Steering control device for vehicle
CN103069466A (en) * 2010-06-18 2013-04-24 本田技研工业株式会社 System for inferring driver's lane change intention
DE102017223222A1 (en) * 2017-12-19 2019-06-19 Zf Friedrichshafen Ag Method for determining a tripping time of an active personnel protection device and a personal protection device
CN110146100A (en) * 2018-02-13 2019-08-20 华为技术有限公司 Trajectory predictions method, apparatus and storage medium
CN109949611A (en) * 2019-03-28 2019-06-28 百度在线网络技术（北京）有限公司 The lane change method, apparatus and storage medium of unmanned vehicle
CN110304063A (en) * 2019-07-15 2019-10-08 广州小鹏汽车科技有限公司 A kind of automatic lane change method and system of vehicle, vehicle
CN111284485A (en) * 2019-10-10 2020-06-16 中国第一汽车股份有限公司 Method and device for predicting driving behavior of obstacle vehicle, vehicle and storage medium
CN110796856A (en) * 2019-10-16 2020-02-14 腾讯科技（深圳）有限公司 Vehicle lane change intention prediction method and training method of lane change intention prediction network
CN110758382A (en) * 2019-10-21 2020-02-07 南京航空航天大学 Surrounding vehicle motion state prediction system and method based on driving intention
CN110908375A (en) * 2019-11-14 2020-03-24 北京三快在线科技有限公司 Method and device for acquiring lane change decision information, storage medium and vehicle
CN111267846A (en) * 2020-02-11 2020-06-12 南京航空航天大学 Game theory-based peripheral vehicle interaction behavior prediction method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
一种基于高斯混合模型的轨迹预测算法;乔少杰，金琨等;《软件学报》;20150228;全文 *
张颖达;基于NGSIM数据的车辆变道行为博弈模型;《中国优秀硕士学位论文数据库（工程科技Ⅱ辑）》;20170115;全文 *
结构化道路中动态车辆的轨迹预测;谢辉等;《汽车安全与节能学报》;20191215(第04期);全文 *

Also Published As

Publication number Publication date
CN113077619A (en) 2021-07-06

Similar Documents

Publication Publication Date Title
Shin et al. Vehicle speed prediction using a Markov chain with speed constraints
CN110834644B (en) Vehicle control method and device, vehicle to be controlled and storage medium
Althoff et al. Comparison of Markov chain abstraction and Monte Carlo simulation for the safety assessment of autonomous cars
US8948995B2 (en) Preceding vehicle state prediction
EP2610836B1 (en) Device and method for the on-line prediction of the driving cycle in an automotive vehicle
CN111736142A (en) System and method for radar cross traffic tracking and maneuver risk assessment
CN112347567A (en) Vehicle intention and track prediction method
CN111976741A (en) Apparatus and method for controlling autonomous vehicle
Han et al. Driving intention recognition and lane change prediction on the highway
CN114312830B (en) Intelligent vehicle coupling decision model and method considering dangerous driving conditions
CN111145552A (en) Planning method for vehicle dynamic lane changing track based on 5G network
Wirthmüller et al. Predicting the time until a vehicle changes the lane using LSTM-based recurrent neural networks
Shin et al. Comparative study of Markov chain with recurrent neural network for short term velocity prediction implemented on an embedded system
Jin et al. Driver intention recognition based on continuous hidden Markov model
Gadepally et al. Driver/vehicle state estimation and detection
CN113128826B (en) Navigation risk prediction method, navigation risk prediction device and storage medium
Wissing et al. Trajectory prediction for safety critical maneuvers in automated highway driving
CN112249008A (en) Unmanned automobile early warning method aiming at complex dynamic environment
Amsalu et al. Driver intention estimation via discrete hidden Markov model
CN115848398B (en) Lane departure early warning system assessment method based on learning and considering driver behavior characteristics
CN112977412A (en) Vehicle control method, device and equipment and computer storage medium
Jeong Predictive lane change decision making using bidirectional long shot-term memory for autonomous driving on highways
Klingelschmitt et al. Using context information and probabilistic classification for making extended long-term trajectory predictions
CN116758741A (en) Multi-dimensional uncertainty perception intelligent automobile collision probability prediction method
Sharma et al. Highway discretionary lane changing behavior recognition using continuous and discrete hidden markov model

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