阅读说明：本技术 基于时序的车辆变道意图预测方法及系统 (Vehicle lane change intention prediction method and system based on time sequence ) 是由 朱亚坤 赖锋 刘义军 余昊 陈永昌 于 2021-07-07 设计创作，主要内容包括：本发明公开了基于时序的车辆变道意图预测方法及系统,涉及智能驾驶技术领域。本发明提供了的基于时序的车辆变道意图预测方法,包括以下步骤：获取临车道车辆轨迹时序信息,轨迹时序信息包括临车道的每个车辆的实时位置、实时车速、实时加速度、实时车速方差和实时车速协方差；根据获取的车辆轨迹时序信息,获取临车道车辆的预测轨迹信息；根据获取的临车道车辆的预测轨迹信息,获取临车道车辆的变道意图。本发明提供的基于时序的车辆变道意图预测方法,通过获取车辆的轨迹时序信息,获取临车道车辆的预测轨迹信息,从而获取临车道车辆的变道意图,提供了一种准确的临车道变道意图预测方法,便于本车道车辆及时作出变道对策,保障行车安全。(The invention discloses a vehicle lane change intention prediction method and system based on time sequence, and relates to the technical field of intelligent driving. The invention provides a vehicle lane change intention prediction method based on time sequence, which comprises the following steps: acquiring track time sequence information of vehicles on the adjacent lane, wherein the track time sequence information comprises the real-time position, the real-time speed, the real-time acceleration, the real-time speed variance and the real-time speed covariance of each vehicle on the adjacent lane; acquiring predicted track information of the vehicle in the adjacent lane according to the acquired vehicle track time sequence information; and acquiring the lane change intention of the vehicle in the adjacent lane according to the acquired predicted track information of the vehicle in the adjacent lane. The method for predicting the lane change intention of the vehicle based on the time sequence obtains the predicted track information of the vehicle in the adjacent lane by obtaining the track time sequence information of the vehicle so as to obtain the lane change intention of the vehicle in the adjacent lane, and provides an accurate prediction method of the lane change intention of the vehicle in the adjacent lane, so that the vehicle in the adjacent lane can take lane change countermeasures in time, and the driving safety is guaranteed.)

1. A time sequence-based vehicle lane change intention prediction method is characterized by comprising the following steps:

acquiring track time sequence information of vehicles on the adjacent lane, wherein the track time sequence information comprises the real-time position, the real-time vehicle speed, the real-time acceleration, the real-time vehicle speed variance and the real-time vehicle speed covariance of each vehicle on the adjacent lane;

acquiring predicted track information of the vehicle in the adjacent lane according to the acquired vehicle track time sequence information;

and acquiring the lane change intention of the vehicle in the adjacent lane according to the acquired predicted track information of the vehicle in the adjacent lane.

2. The time-series-based prediction method for the lane-changing intention of the vehicle as claimed in claim 1, wherein the step of obtaining the lane-changing intention of the lane-adjacent vehicle according to the obtained predicted track information of the lane-adjacent vehicle comprises the following steps:

acquiring a relation model of preset vehicle track time sequence information and cut-in probability;

acquiring the cut-in probability of the vehicle in the adjacent lane according to the acquired vehicle track time sequence information and a relation model between the vehicle track time sequence information and the cut-in probability;

and acquiring the lane change intention of the vehicle in the adjacent lane according to the cut-in probability of the vehicle in the adjacent lane.

3. The time-series-based prediction method for the lane-changing intention of the vehicle as claimed in claim 2, wherein the step of obtaining the lane-changing intention of the lane-adjacent vehicle according to the cut-in probability of the lane-adjacent vehicle comprises the following steps:

comparing the cut-in probability of the vehicle approaching the lane with the lane change probability threshold value to obtain a comparison result;

and acquiring the lane change intention of the adjacent lane vehicle according to the acquired comparison result.

4. The time-series-based vehicle lane-changing intention prediction method according to claim 3, wherein the step of obtaining the lane-changing intention of the oncoming lane vehicle according to the obtained comparison result specifically comprises the steps of:

when the cut-in probability of the vehicle in the adjacent lane is larger than the lane change probability threshold value, judging that the vehicle in the adjacent lane has a lane change intention;

and when the cut-in probability of the lane-adjacent vehicle is not greater than the lane-changing probability threshold value, determining that the lane-adjacent vehicle has no intention of changing lanes.

5. The time-series-based vehicle lane-changing intention prediction method according to claim 4, wherein after the step of obtaining the lane-changing intention of the oncoming lane vehicle according to the obtained comparison result, the method further comprises the following steps:

acquiring the running condition of a vehicle on a temporary lane and the running condition of the vehicle on the current lane;

and controlling the vehicle of the lane to execute a lane change countermeasure aiming at the vehicle of the lane according to the acquired running condition of the vehicle of the lane and the running condition of the vehicle of the lane.

6. The time-series-based prediction method for the lane change intention of the vehicle according to claim 5, wherein the step of controlling the vehicle in the own lane to perform the lane change countermeasure for the vehicle in the adjacent lane according to the acquired driving condition of the vehicle in the adjacent lane and the acquired driving condition of the vehicle in the own lane comprises the following steps:

and when the cut-in probability of the lane-adjacent vehicle is greater than the lane-changing probability threshold value and the speed of the lane-adjacent vehicle is lower than that of the vehicle, controlling the vehicle of the vehicle to decelerate.

7. The time-series-based prediction method for the lane-changing intention of the vehicle according to claim 6, wherein after the step of controlling the vehicle in the own lane to perform the lane-changing countermeasure for the vehicle in the adjacent lane according to the acquired driving condition of the vehicle in the adjacent lane and the driving condition of the vehicle in the own lane, the method further comprises the following steps:

and when the cut-in probability of the vehicle in the adjacent lane is not greater than the lane change probability threshold and the single-wheel line of the vehicle in the adjacent lane, controlling the normal running speed of the vehicle in the lane to follow the vehicle and executing a whistling warning instruction.

8. The time-series-based prediction method for the lane change intention of the vehicle according to claim 2, wherein the step of obtaining the preset relationship model between the vehicle track time-series information and the cut-in probability comprises the following steps:

acquiring vehicle track time sequence information;

weighting and establishing a relation model of each track time sequence information of the vehicle and theoretical cut-in probability according to the obtained vehicle track time sequence information and the weight value letter code of each track time sequence information of the vehicle;

carrying out real vehicle test on the acquired vehicle track time sequence information to acquire an actual cut-in working condition;

comparing the theoretical cut-in probability with the actual cut-in working condition, and adjusting the weight value of each track time sequence information of the vehicle until the theoretical cut-in probability obtained through weighting calculation accords with the actual cut-in working condition;

and establishing a preset relation model of the vehicle track time sequence information and the cut-in probability according to the final weight value of each track time sequence information of the vehicle and each track time sequence information of the vehicle.

9. A time-series-based vehicle lane-change intention prediction system, comprising:

the system comprises a track time sequence information acquisition module, a lane information acquisition module and a lane information processing module, wherein the track time sequence information acquisition module is used for acquiring the track time sequence information of vehicles in the lane, and the track time sequence information comprises the real-time position, the real-time vehicle speed, the real-time acceleration, the real-time vehicle speed variance and the real-time vehicle speed covariance of each vehicle in the lane;

the predicted track information acquisition module is in communication connection with the track time sequence information acquisition module and is used for acquiring predicted track information of the vehicle in the adjacent lane according to the acquired vehicle track time sequence information;

and the lane change intention acquisition module is in communication connection with the predicted track information acquisition module and is used for acquiring the lane change intention of the vehicle in the adjacent lane according to the acquired predicted track information of the vehicle in the adjacent lane.

10. The time-series based vehicle lane-change intention prediction system of claim 9, wherein the lane-change intention acquisition module further comprises:

the system comprises a relation model obtaining unit, a relation model obtaining unit and a relation model judging unit, wherein the relation model obtaining unit is used for obtaining preset vehicle track time sequence information and a relation model of cut-in probability;

the cut-in probability acquisition unit is in communication connection with the relation model acquisition unit and is used for acquiring the cut-in probability of the vehicle in the adjacent lane according to the acquired vehicle track time sequence information and the relation model of the vehicle track time sequence information and the cut-in probability;

and the lane change intention acquisition unit is in communication connection with the cut-in probability acquisition unit and is used for acquiring the lane change intention of the vehicle in the adjacent lane according to the acquired cut-in probability of the vehicle in the adjacent lane.

Technical Field

The invention relates to the technical field of carrying intelligent auxiliary driving, in particular to a vehicle lane change intention prediction method and system based on time sequence.

Background

With the increasing abundance of types of automobile drivers, the safety and convenience of driving are also valued by the majority of users. The high-level driver assist function has become a mainstream safety configuration in recent years.

At present, vehicles which are mainstream in the market and are provided with intelligent auxiliary driving systems can be decelerated only when the vehicles cut into a temporary lane and meet a certain coincidence rate, and the vehicles can not act when the coincidence rate does not reach a certain threshold value. When the oncoming traffic lane has a cut-in intention, an experienced human driver can accelerate or decelerate to avoid the oncoming traffic lane. When the intelligent driving system is used, the intelligent auxiliary driving automobile can generate larger deceleration when encountering sudden congestion of the vehicle in the adjacent lane, but not generate certain reaction action when the congestion intention is generated before the vehicle in the adjacent lane completes the congestion. In order to make an intelligent driving assistance system more intelligent, a vehicle lane change intention prediction method based on time sequence information is provided. The method can continuously observe the vehicle track, and when the vehicle has the intention of cutting into the lane, the vehicle can respond, such as whistling warning, deceleration and avoidance.

The prior art provides an automatic driving vehicle and a lane change control method and system thereof, and discloses the automatic driving vehicle and the lane change control method and system thereof, which solve the problem that a safe and reliable lane change path cannot be planned for the vehicle in the prior art, so that automatic, safe and reliable lane change control of the vehicle is realized. The method comprises the steps of determining a target point of a main vehicle changing a lane to an adjacent lane as a first target point according to the current vehicle state and the target vehicle state of the main vehicle and at least one adjacent vehicle, determining a point of the main vehicle on the lane where the main vehicle is located for executing lane changing operation as a second target point, determining a track of the main vehicle driving from the current position to the second target point as a first track and a track of the main vehicle driving from the second target point to the first target point as a second track, and controlling the vehicle to track the first track and the second track and drive from the current position to the first target point on the adjacent lane. However, the method does not use a large amount of data to train the track, so that detailed description of the characteristics of the target track is lacked, and the track is predicted by using a rule-based method, so that the method is poor in practical applicability, and therefore the method does not have the generalization capability on large-scale scenes.

The prior art discloses a vehicle driving behavior prediction method based on machine learning, mainly provides a vehicle driving behavior prediction method based on machine learning, relates to the field of vehicle networking, and aims to utilize the machine learning technology to mine the relationship between vehicle attributes, road information and driving environment information and vehicle driving behaviors and improve the accuracy of vehicle driving behavior prediction. The method comprises the following specific steps: step 1, defining a feature set: defining vehicle characteristics, road characteristics and vehicle running environment; step 2, a vehicle displacement prediction model: feature extraction and data preprocessing: extracting distance characteristics of the vehicle and a front intersection, extracting turning action characteristics of the intersection and extracting labels; vehicle displacement prediction model: defining a training sample set, and training a vehicle displacement prediction model; step 3, a vehicle driving behavior prediction model: defining a Gaussian component; and 4, predicting the driving behavior of the vehicle. The method uses a full-connection network to predict displacement, does not use a time sequence neural network, and has insufficient utilization of time sequence information; the method mainly predicts the in-situ motionless information of the vehicle without combining the lane change information of the vehicle, namely the vehicle goes straight, turns left, turns right and turns around, and the pain point in the intelligent driving auxiliary system is: the method is lack of a countermeasure scheme for traffic jam of a traffic lane.

Disclosure of Invention

The invention aims to overcome the defects of the background technology and provides a time sequence-based vehicle lane-changing intention prediction method and a time sequence-based vehicle lane-changing intention prediction system.

In a first aspect, the invention provides a time-series-based vehicle lane-change intention prediction method, which comprises the following steps:

acquiring track time sequence information of vehicles on the adjacent lane, wherein the track time sequence information comprises the real-time position, the real-time vehicle speed, the real-time acceleration, the real-time vehicle speed variance and the real-time vehicle speed covariance of each vehicle on the adjacent lane;

acquiring predicted track information of the vehicle in the adjacent lane according to the acquired vehicle track time sequence information;

and acquiring the lane change intention of the vehicle in the adjacent lane according to the acquired predicted track information of the vehicle in the adjacent lane.

According to the first aspect, in a first possible implementation manner of the first aspect, the step of "obtaining a lane change intention of the lane-adjacent vehicle according to the obtained predicted trajectory information of the lane-adjacent vehicle" specifically includes the following steps:

acquiring a relation model of preset vehicle track time sequence information and cut-in probability;

acquiring the cut-in probability of the vehicle in the adjacent lane according to the acquired vehicle track time sequence information and a relation model between the vehicle track time sequence information and the cut-in probability;

and acquiring the lane change intention of the vehicle in the adjacent lane according to the cut-in probability of the vehicle in the adjacent lane.

According to the first possible implementation manner of the first aspect, in the second possible implementation manner of the first aspect, the step of obtaining the lane change intention of the lane-adjacent vehicle according to the obtained cut-in probability of the lane-adjacent vehicle includes the following steps:

comparing the cut-in probability of the vehicle approaching the lane with the lane change probability threshold value to obtain a comparison result;

and acquiring the lane change intention of the adjacent lane vehicle according to the acquired comparison result.

According to the second possible implementation manner of the first aspect, in a third possible implementation manner of the first aspect, the step of "obtaining the lane change intention of the oncoming lane vehicle according to the obtained comparison result" specifically includes the following steps:

when the cut-in probability of the vehicle in the adjacent lane is larger than the lane change probability threshold value, judging that the vehicle in the adjacent lane has a lane change intention;

and when the cut-in probability of the lane-adjacent vehicle is not greater than the lane-changing probability threshold value, determining that the lane-adjacent vehicle has no intention of changing lanes.

According to a third possible implementation manner of the first aspect, in a fourth possible implementation manner of the first aspect, after the step of "obtaining a lane change intention of a vehicle approaching a lane according to the obtained comparison result", the method further includes the following steps:

acquiring the running condition of a vehicle on a temporary lane and the running condition of the vehicle on the current lane;

and controlling the vehicle of the lane to execute a lane change countermeasure aiming at the vehicle of the lane according to the acquired running condition of the vehicle of the lane and the running condition of the vehicle of the lane.

According to the fourth possible implementation manner of the first aspect, in a fifth possible implementation manner of the first aspect, the step of controlling the vehicle in the local lane to execute a lane change countermeasure for the vehicle in the adjacent lane according to the acquired driving condition of the vehicle in the adjacent lane and the acquired driving condition of the vehicle in the local lane specifically includes the following steps:

and when the cut-in probability of the lane-adjacent vehicle is greater than the lane-changing probability threshold value and the speed of the lane-adjacent vehicle is lower than that of the vehicle, controlling the vehicle of the vehicle to decelerate.

According to a fifth possible implementation manner of the first aspect, in a sixth possible implementation manner of the first aspect, after the step of "controlling the host-lane vehicle to perform a lane change countermeasure for the host-lane vehicle according to the acquired driving condition of the host-lane vehicle and the acquired driving condition of the host-lane vehicle", the method further includes the following steps:

and when the cut-in probability of the vehicle in the adjacent lane is not greater than the lane change probability threshold and the single-wheel line of the vehicle in the adjacent lane, controlling the normal running speed of the vehicle in the lane to follow the vehicle and executing a whistling warning instruction.

According to the first possible implementation manner of the first aspect, in a seventh possible implementation manner of the first aspect, the step of obtaining a relationship model between preset vehicle trajectory timing information and cut-in probability specifically includes the following steps:

acquiring vehicle track time sequence information;

weighting and establishing a relation model of each track time sequence information of the vehicle and theoretical cut-in probability according to the obtained vehicle track time sequence information and the weight value letter code of each track time sequence information of the vehicle;

carrying out real vehicle test on the acquired vehicle track time sequence information to acquire an actual cut-in working condition;

comparing the theoretical cut-in probability with the actual cut-in working condition, and adjusting the weight value of each track time sequence information of the vehicle until the theoretical cut-in probability obtained through weighting calculation accords with the actual cut-in working condition;

and establishing a preset relation model of the vehicle track time sequence information and the cut-in probability according to the final weight value of each track time sequence information of the vehicle and each track time sequence information of the vehicle.

In a second aspect, the present invention provides a time-series-based vehicle lane-change intention prediction system, comprising:

the system comprises a track time sequence information acquisition module, a lane information acquisition module and a lane information processing module, wherein the track time sequence information acquisition module is used for acquiring the track time sequence information of vehicles in the lane, and the track time sequence information comprises the real-time position, the real-time vehicle speed, the real-time acceleration, the real-time vehicle speed variance and the real-time vehicle speed covariance of each vehicle in the lane;

the predicted track information acquisition module is in communication connection with the track time sequence information acquisition module and is used for acquiring predicted track information of the vehicle in the adjacent lane according to the acquired vehicle track time sequence information;

and the lane change intention acquisition module is in communication connection with the predicted track information acquisition module and is used for acquiring the lane change intention of the vehicle in the adjacent lane according to the acquired predicted track information of the vehicle in the adjacent lane.

According to the first possible implementation manner of the second aspect, in the first possible implementation manner of the second aspect, the lane change intention acquisition module further includes:

the system comprises a relation model obtaining unit, a relation model obtaining unit and a relation model judging unit, wherein the relation model obtaining unit is used for obtaining preset vehicle track time sequence information and a relation model of cut-in probability;

the cut-in probability acquisition unit is in communication connection with the relation model acquisition unit and is used for acquiring the cut-in probability of the vehicle in the adjacent lane according to the acquired vehicle track time sequence information and the relation model of the vehicle track time sequence information and the cut-in probability;

and the lane change intention acquisition unit is in communication connection with the cut-in probability acquisition unit and is used for acquiring the lane change intention of the vehicle in the adjacent lane according to the acquired cut-in probability of the vehicle in the adjacent lane.

Compared with the prior art, the invention has the following advantages:

the method for predicting the lane change intention of the vehicle based on the time sequence obtains the predicted track information of the vehicle in the adjacent lane by obtaining the track time sequence information of the vehicle so as to obtain the lane change intention of the vehicle in the adjacent lane, and provides an accurate prediction method of the lane change intention of the vehicle in the adjacent lane, so that the vehicle in the adjacent lane can take lane change countermeasures in time, and the driving safety is guaranteed.

Drawings

FIG. 1 is a flowchart of a method for predicting lane change intention of a vehicle based on time sequence according to an embodiment of the present invention;

FIG. 2 is a flowchart of another method for predicting lane change intention of a vehicle based on time sequence according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a timing neural network training process of a timing-based vehicle lane change intention prediction method according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of a characteristic vector and a classifier training process of a time-series-based vehicle lane-change intention prediction method according to an embodiment of the present invention;

FIG. 5 is a functional block diagram of a time-series based vehicle lane change intent prediction system provided by an embodiment of the present invention;

FIG. 6 is a block diagram of another functional block of a time-based vehicle lane change intention prediction system according to an embodiment of the present invention.

In the figure, 100, a track time sequence information acquisition module; 200. a predicted trajectory information acquisition module; 300. a lane change intention acquisition module; 310. a relational model acquisition unit; 320. a cut-in probability acquisition unit; 330. and a lane change intention acquisition unit.

Detailed Description

Reference will now be made in detail to the present embodiments of the invention, examples of which are illustrated in the accompanying drawings. While the invention will be described in conjunction with the specific embodiments, it will be understood that they are not intended to limit the invention to the embodiments described. On the contrary, it is intended to cover alternatives, modifications, and equivalents as may be included within the spirit and scope of the invention as defined by the appended claims. It should be noted that the method steps described herein may be implemented by any functional block or functional arrangement, and that any functional block or functional arrangement may be implemented as a physical entity or a logical entity, or a combination of both.

In order that those skilled in the art will better understand the present invention, the following detailed description of the invention is provided in conjunction with the accompanying drawings and the detailed description of the invention.

Note that: the example to be described next is only a specific example, and does not limit the embodiments of the present invention necessarily to the following specific steps, values, conditions, data, orders, and the like. Those skilled in the art can, upon reading this specification, utilize the concepts of the present invention to construct more embodiments than those specifically described herein.

Referring to fig. 1, an embodiment of the present invention provides a time-series-based method and a time-series-based system for predicting a lane change intention of a vehicle, including the following steps:

s100, acquiring track time sequence information of vehicles approaching a lane, wherein the track time sequence information comprises the real-time position, the real-time vehicle speed, the real-time acceleration, the real-time vehicle speed variance and the real-time vehicle speed covariance of each vehicle of the lane;

s200, acquiring predicted track information of the vehicle in the adjacent lane according to the acquired vehicle track time sequence information;

and S300, acquiring the lane change intention of the vehicle in the adjacent lane according to the acquired predicted track information of the vehicle in the adjacent lane.

The method for predicting the lane change intention of the vehicle based on the time sequence obtains the predicted track information of the vehicle in the adjacent lane by obtaining the track time sequence information of the vehicle so as to obtain the lane change intention of the vehicle in the adjacent lane, and provides an accurate prediction method of the lane change intention of the vehicle in the adjacent lane, so that the vehicle in the adjacent lane can take lane change countermeasures in time, and the driving safety is guaranteed.

In an embodiment, referring to fig. 2, the step of obtaining the lane change intention of the lane-adjacent vehicle according to the obtained predicted track information of the lane-adjacent vehicle includes the following steps:

s310, obtaining a relation model of preset vehicle track time sequence information and cut-in probability;

s320, acquiring the cut-in probability of the vehicle in the adjacent lane according to the acquired vehicle track time sequence information and the relation model of the vehicle track time sequence information and the cut-in probability;

and S330, acquiring the lane change intention of the adjacent lane vehicle according to the cut-in probability of the adjacent lane vehicle.

In an embodiment, the step of obtaining a relationship model between preset vehicle trajectory timing information and cut-in probability specifically includes the following steps:

acquiring vehicle track time sequence information;

weighting and establishing a relation model of each track time sequence information of the vehicle and theoretical cut-in probability according to the obtained vehicle track time sequence information and the weight value letter code of each track time sequence information of the vehicle;

carrying out real vehicle test on the acquired vehicle track time sequence information to acquire an actual cut-in working condition;

comparing the theoretical cut-in probability with the actual cut-in working condition, and adjusting the weight value of each track time sequence information of the vehicle until the theoretical cut-in probability obtained through weighting calculation accords with the actual cut-in working condition;

and establishing a preset relation model of the vehicle track time sequence information and the cut-in probability according to the final weight value of each track time sequence information of the vehicle and each track time sequence information of the vehicle.

In an embodiment, the weight value of each track timing information of the vehicle is trained by a timing neural network to obtain an adjusted weight value.

In one embodiment, the processing flow of the sequential neural network is as shown in fig. 3:

given the initial state of the vehicle, the time sequence neural network is sequentially propagated from left to right, the state of w1 is only related to the initial state and the input t1 at the moment, the predicted output is a p1 state, and the output p1 is a characteristic vector formed by the real-time position, the real-time speed, the real-time acceleration, the real-time speed variance, the real-time speed covariance and the like of the vehicle. The w2 state is only associated with the state w1 at the previous time and the input t2 at that time, while w1 is again associated with the state at the previous time. The time sequence neural network is formed by the circulation. For the current state, the output feature vector of a future period of time is circularly predicted through the output feature vector of a past period of time (for example, the feature vector (tn-tm) in fig. 4 is composed of a series of output vectors of pm to pn). The training process of the time sequence neural network is to train the output of the past time as the weight value of the current input, the weight value of the current input feature vector and the weight value of the current output feature vector.

In one embodiment, the feature vector and classifier training process is as shown in FIG. 4:

the training process is that for data collected by the multi-source sensor, a characteristic vector is formed by actual output of a period of time, the characteristic vector comprises a vehicle real-time position, a real-time vehicle speed, a real-time acceleration, a real-time vehicle speed variance, a real-time vehicle speed covariance and the like, and the condition that whether the vehicle changes the lane to the front of the current vehicle in the period of time after the output of the period is judged by combining the information of the automobile data recorder, a training set which is formed by the characteristic vectors pm to pn and corresponds to a lane change target or not is constructed, and the characteristic vector classifier is trained. The classifier classifies output feature vectors predicted by the time sequence neural network within a period of time in the future, and judges the probability of whether the vehicle changes lane targets in the future so as to realize the function of prediction in advance.

The method comprises the steps of extracting time sequence information of each track of a vehicle, using a deep learning algorithm, predicting the track by using a time sequence neural network on the basis of accumulating a large amount of driving mileage and driving scene data, forming a characteristic vector by using the predicted track, and training a classifier by combining whether the vehicle has cut-in motion or not; after the training is completed, the classifier is deployed in an intelligent driving controller, historical track information of a target vehicle which is possibly cut into the adjacent lane is accumulated, the historical track information is sent to a time sequence neural network to obtain predicted track information, a feature vector is formed, the feature vector is sent to the trained classifier, and the probability of the cut-in of the vehicle in the adjacent lane is obtained. When the probability reaches a certain threshold value, the target is regarded as a key control target or a dangerous target, and the vehicle planning control module generates requests such as early deceleration avoidance, whistling or light reminding and the like, so that the comfort and the safety of the intelligent driving auxiliary system are enhanced.

In an embodiment, the step of obtaining the lane change intention of the lane-adjacent vehicle according to the cut-in probability of the lane-adjacent vehicle includes the following steps:

comparing the cut-in probability of the vehicle approaching the lane with the lane change probability threshold value to obtain a comparison result;

and acquiring the lane change intention of the adjacent lane vehicle according to the acquired comparison result.

In an embodiment, the step of obtaining the lane change intention of the oncoming lane vehicle according to the obtained comparison result specifically includes the following steps:

when the cut-in probability of the vehicle in the adjacent lane is larger than the lane change probability threshold value, judging that the vehicle in the adjacent lane has a lane change intention;

and when the cut-in probability of the lane-adjacent vehicle is not greater than the lane-changing probability threshold value, determining that the lane-adjacent vehicle has no intention of changing lanes.

In an embodiment, after the step of "obtaining the lane change intention of the oncoming lane vehicle according to the obtained comparison result", the method further includes the following steps:

acquiring the running condition of a vehicle on a temporary lane and the running condition of the vehicle on the current lane;

and controlling the vehicle of the lane to execute a lane change countermeasure aiming at the vehicle of the lane according to the acquired running condition of the vehicle of the lane and the running condition of the vehicle of the lane.

In an embodiment, the step of controlling the vehicle in the lane to execute a lane change countermeasure for the vehicle in the lane according to the acquired driving condition of the vehicle in the lane and the driving condition of the vehicle in the lane specifically includes the following steps:

when the cut-in probability of the vehicle in the adjacent lane is larger than the lane change probability threshold value, the speed of the vehicle in the adjacent lane is lower than that of the vehicle in the main lane, and the collision risk exists, the vehicle decelerates in advance to avoid, the deceleration is gentle, the running pause and frustration feeling of the vehicle is reduced, and the vehicle is more intelligent and comfortable.

In an embodiment, after the step of controlling the host-lane vehicle to perform the lane-change countermeasure for the lane-adjacent vehicle according to the acquired driving condition of the lane-adjacent vehicle and the driving condition of the host-lane vehicle, the method further includes the following steps:

when the cut-in probability of the vehicle in the adjacent lane is not greater than the lane change probability threshold and the single-wheel line of the vehicle in the adjacent lane is pressed, the normal running speed of the vehicle in the lane is controlled to follow the vehicle to run and a whistling warning instruction is executed, so that the mistaken deceleration action of the vehicle caused by the vehicle in the adjacent lane is avoided, and the complaint of drivers is not easy to cause.

Based on the same inventive concept, referring to fig. 5, the present invention provides a time-series-based vehicle lane-change intention prediction system, comprising:

the system comprises a track timing information acquisition module 100, a lane-adjacent vehicle track timing information acquisition module and a lane-adjacent vehicle track timing information processing module, wherein the track timing information comprises the real-time position, the real-time vehicle speed, the real-time acceleration, the real-time vehicle speed variance and the real-time vehicle speed covariance of each vehicle in a lane;

the predicted track information acquisition module 200 is in communication connection with the track timing information acquisition module 100 and is used for acquiring predicted track information of the vehicle approaching the lane according to the acquired vehicle track timing information;

and a lane change intention acquisition module 300, which is in communication connection with the predicted trajectory information acquisition module 200 and is used for acquiring a lane change intention of the vehicle in the adjacent lane according to the acquired predicted trajectory information of the vehicle in the adjacent lane.

As described above, the trajectory timing information acquisition module is implemented as a multi-source sensor including a radar, a camera, inertial navigation, and the like.

In an embodiment, please refer to fig. 6, the lane change intention acquiring module further includes:

a relation model obtaining unit 310, configured to obtain a relation model between preset vehicle trajectory timing information and a cut-in probability;

the cut-in probability acquiring unit 320 is in communication connection with the relation model acquiring unit 310 and is used for acquiring the cut-in probability of the vehicle in the adjacent lane according to the acquired vehicle track time sequence information and the relation model of the vehicle track time sequence information and the cut-in probability;

and a lane-changing intention acquisition unit 330, communicatively connected to the cut-in probability acquisition unit 320, for acquiring a lane-changing intention of the oncoming vehicle according to the acquired cut-in probability of the oncoming vehicle.

Based on the same inventive concept, the embodiments of the present application further provide a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements all or part of the method steps of the above method.

The present invention can implement all or part of the processes in the above methods, and can also be implemented by instructing related hardware through a computer program, which can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above method embodiments can be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying computer program code, recording medium, U-disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution media, and the like. It should be noted that the computer readable medium may contain other components which may be suitably increased or decreased as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, in accordance with legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunications signals.

Based on the same inventive concept, an embodiment of the present application further provides an electronic device, which includes a memory and a processor, where the memory stores a computer program running on the processor, and the processor executes the computer program to implement all or part of the method steps in the method.

The processor may be a Central Processing Unit (CP U), or may be other general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, or the like. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, the processor being the control center of the computer device and the various interfaces and lines connecting the various parts of the overall computer device.

The memory may be used to store computer programs and/or modules, and the processor may implement various functions of the computer device by executing or executing the computer programs and/or modules stored in the memory, as well as by invoking data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (e.g., a sound playing function, an image playing function, etc.); the storage data area may store data (e.g., audio data, video data, etc.) created according to the use of the cellular phone. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a flash memory Card (flash Card), at least one magnetic disk storage device, a flash memory device, or other volatile solid state storage device.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, server, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.

The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), servers and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams 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, embedded processor, 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, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.