CN117985004A

CN117985004A - Overtaking strategy determining method, apparatus, computer device, storage medium and product

Info

Publication number: CN117985004A
Application number: CN202211334364.4A
Authority: CN
Inventors: 余开江
Original assignee: Beijing Didi Infinity Technology and Development Co Ltd
Current assignee: Ruilian Xingchen Beijing Technology Co ltd
Priority date: 2022-10-28
Filing date: 2022-10-28
Publication date: 2024-05-07

Abstract

The application relates to a method, a device, computer equipment, a storage medium and a product for determining a passing strategy, wherein the method comprises the following steps: and obtaining driving characteristic information of the target vehicle, inputting the driving characteristic information into the overtaking strategy prediction model, and obtaining a target overtaking strategy corresponding to the target vehicle according to the output of the overtaking strategy prediction model. By adopting the embodiment, the target overtaking strategy can be determined by the overtaking strategy prediction model obtained based on the sample driving characteristic information training of various overtaking styles, so that the obtained target overtaking strategy can be suitable for different overtaking styles, the problem that the overtaking strategy is single is solved, and the wide applicability of the overtaking strategy determination method is improved.

Description

Overtaking strategy determining method, apparatus, computer device, storage medium and product

Technical Field

The present application relates to the field of traffic technologies, and in particular, to a method and apparatus for determining an overtaking policy, a computer device, a storage medium, and a product.

Background

With the development of society, the living standard of people is continuously improved, and the vehicle requirements of people are also increased. Usually, people continuously use the vehicle for too long, the problem of driving fatigue can occur, and fatigue driving is one of the important reasons for causing traffic accidents. In order to reduce the occurrence rate of traffic events, auxiliary driving systems have been developed to replace manual driving to some extent, so as to alleviate driving fatigue of drivers. In the assisted driving system, overtaking is one of the key demands of the driver.

However, the conventional driving assistance system has a single overtaking strategy.

Disclosure of Invention

Based on this, it is necessary to provide a method, a device, a computer device, a storage medium and a product for determining an overtaking strategy according to the actual driving characteristic information of the vehicle, so that the determined overtaking strategy is suitable for a corresponding overtaking scene.

In a first aspect, an embodiment of the present application provides a method for determining an overtaking strategy, where the method includes:

Acquiring driving characteristic information of a target vehicle;

Inputting the driving characteristic information into an overtaking strategy prediction model, and obtaining a target overtaking strategy corresponding to the target vehicle according to the output of the overtaking strategy prediction model;

the overtaking strategy prediction model is obtained by training based on sample driving characteristic information of various overtaking styles.

In one embodiment, obtaining a target overtaking strategy corresponding to the target vehicle according to the output of the overtaking strategy prediction model includes:

Acquiring an initial overtaking path output by an overtaking strategy prediction model;

And acquiring a target overtaking strategy corresponding to the target vehicle based on the initial overtaking path.

In one embodiment, based on the initial cut-in path, obtaining a target cut-in strategy corresponding to the target vehicle includes:

Acquiring an overtaking region of the target vehicle, and acquiring an overtaking path of the target vehicle according to the overtaking region;

and determining a target overtaking strategy corresponding to the target vehicle through the initial overtaking path and the overtaking path.

In one embodiment, acquiring an overtaking region of a target vehicle includes:

acquiring an environment image of a target vehicle, and carrying out semantic prediction on the environment image to obtain a candidate overtaking region in a road where the target vehicle is located;

Detecting traffic marking lines in the environment image to obtain attribute information of a road where a target vehicle is located and attribute information of a lane;

and carrying out semantic processing on the environment image based on the attribute information of the road and the attribute information of the lane, and acquiring the overtaking region of the target vehicle from the candidate overtaking region.

In one embodiment, determining a target overtaking strategy corresponding to the target vehicle through the initial overtaking path and the overtaking path includes:

Acquiring a reference driving path of the target vehicle, and adjusting the reference driving path according to the initial overtaking path and the overtaking path to obtain a candidate overtaking path of the target vehicle;

a target cut-in strategy of the target vehicle is determined based on the candidate cut-in path.

In one embodiment, acquiring a reference travel path of a target vehicle includes:

Performing fitting smoothing processing on boundary points of the lane lines based on the position information of the boundary points of the lane lines of the road where the target vehicle is located, so as to obtain the position information of the fitting smoothing boundary points of the lane lines;

And carrying out average processing on the position information of the fitting smooth boundary points of the lane lines to obtain the reference driving path of the target vehicle.

In one embodiment, determining a target cut-in strategy for a target vehicle based on a candidate cut-in path includes:

Acquiring error information related to a target vehicle, and performing fitting smoothing processing on the error information based on a candidate overtaking path to obtain smoothed error information; the error information comprises lane line error information, positioning error information of a target vehicle and dynamics model error information of the target vehicle;

Determining a target overtaking path of the target vehicle through the smoothed error information and the candidate overtaking path;

And determining a target overtaking strategy of the target vehicle according to the target overtaking path.

In one embodiment, determining a target cut-in strategy for a target vehicle from a target cut-in path includes:

Acquiring the driving time length between adjacent overtaking position points in a target overtaking path;

Determining a target overtaking speed corresponding to the target overtaking path according to the target overtaking path and the running duration between adjacent overtaking position points in the target overtaking path;

And determining a target overtaking strategy of the target vehicle according to the target overtaking path and the target overtaking speed.

In one embodiment, the process of constructing the overtaking strategy prediction model includes:

acquiring a plurality of sample driving characteristic information related to a target vehicle and a sample overtaking strategy corresponding to each sample driving characteristic information;

and training the initial overtaking strategy prediction model according to the plurality of sample driving characteristic information and the sample overtaking strategy corresponding to each sample driving characteristic information to obtain an overtaking strategy prediction model.

In one embodiment, before obtaining the plurality of sample driving feature information related to the target vehicle and the sample overtaking strategy corresponding to each sample driving feature information, the method further includes:

acquiring candidate driving characteristic information of a plurality of sample vehicles and reference driving characteristic information of a target vehicle;

Candidate driving characteristic information matched with the reference driving characteristic information is screened out from the candidate driving characteristic information, and the candidate driving characteristic information is determined to be sample driving characteristic information.

In a second aspect, an embodiment of the present application provides an overtaking strategy determination apparatus, including:

the characteristic information acquisition module is used for acquiring driving characteristic information of the target vehicle;

the prediction module is used for inputting the driving characteristic information into the overtaking strategy prediction model, and obtaining a target overtaking strategy corresponding to the target vehicle according to the output of the overtaking strategy prediction model;

In a third aspect, the application also provides a computer device comprising a memory storing a computer program and a processor implementing the steps of the method of any of the first aspects described above when the computer program is executed by the processor.

In a fifth aspect, the present application also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method of any of the first aspects described above.

In a sixth aspect, the application also provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method of any of the first aspects described above.

According to the overtaking strategy determination method, the overtaking strategy determination device, the computer equipment, the storage medium and the product, driving characteristic information of the target vehicle can be obtained, the driving characteristic information is input into the overtaking strategy prediction model, and the target overtaking strategy corresponding to the target vehicle is obtained according to the output of the overtaking strategy prediction model. According to the method, the target overtaking strategy can be determined through the overtaking strategy prediction model obtained through training based on the sample driving characteristic information of various overtaking styles, so that the obtained target overtaking strategy can be suitable for different overtaking styles, the problem that the overtaking strategy is single is solved, the wide applicability of the overtaking strategy determination method is improved, and the use scenes of different overtaking styles are increased; meanwhile, the method can determine the corresponding overtaking strategies according to different overtaking styles, so that the traffic accident rate of overtaking operation according to the overtaking strategies of a single style can be reduced, and the safety in the overtaking process of the vehicle is improved.

Drawings

FIG. 1 is an application environment diagram of a method of determining an overtaking strategy in one embodiment;

FIG. 2 is a flow diagram of a method of determining an override strategy in one embodiment;

FIG. 3 is a flow chart of a method of determining an override strategy in another embodiment;

FIG. 4 is a flow chart of a method of determining an override strategy in another embodiment;

FIG. 5 is a flow chart of a method of determining an override strategy in another embodiment;

FIG. 6 is a flow chart of a method of determining an override strategy in another embodiment;

FIG. 7 is a flow chart of a method of determining an override strategy in another embodiment;

FIG. 8 is a flow chart of a method of determining an override strategy in another embodiment;

FIG. 9 is a flow chart of a method of determining an override strategy in another embodiment;

FIG. 10 is a block diagram of an apparatus for determining an override strategy in one embodiment;

FIG. 11 is an internal block diagram of a computer device in one embodiment.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

In the assisted driving system, overtaking is one of the key demands of the driver. The auxiliary driving system is mainly used for completing overtaking operation based on a planned overtaking strategy. In the conventional technology, an auxiliary driving system completes the overtaking operation according to a fixed overtaking strategy. However, the conventional driving assistance system has a problem that the overtaking strategy is relatively single, so that the overtaking strategy obtained in the conventional manner is not suitable for various overtaking styles.

Based on the above, the embodiment of the application provides a method for determining the overtaking strategy, which can determine the corresponding overtaking strategy according to the actual driving characteristic information of the vehicle, so that the determined overtaking strategy is applicable to the corresponding overtaking style. The overtaking strategy determination method can be applied to the application scene shown in fig. 1, wherein the application scene comprises: a target vehicle and a computer device having a driving assistance function. The target vehicle and the computer equipment are in communication connection, and the connection mode can be Bluetooth, a mobile network, wifi and the like. Alternatively, the target vehicle may be a car, truck, bus, etc., and the specific form of the target vehicle is not limited in this embodiment; the above-mentioned computer device may be implemented by, but not limited to, an in-vehicle central control, a notebook computer, a smart phone, a tablet computer, and a portable wearable device, and the specific form of the computer device is not limited in this embodiment.

Fig. 2 is a schematic flow chart of an overtaking strategy determination method according to an embodiment of the present application, and in the following embodiment, a computer device is used as an execution body to describe the overtaking strategy determination method. The overtaking strategy determination method may include the steps of:

s100, driving characteristic information of the target vehicle is acquired.

The computer device can process driving style information of the target vehicle to obtain driving characteristic information, and can also directly receive driving characteristic information of the target vehicle sent by other electronic devices with information processing functions.

In practical application, a sensor is installed on a target vehicle and is used for monitoring driving style information of the target vehicle, the driving style information is sent to a computer device, and then the computer device performs parameter optimization processing on the driving style information of the target vehicle to obtain driving characteristic information. Or the sensor arranged on the target vehicle sends the monitored driving style information to other electronic equipment with an information processing function, then the electronic equipment performs parameter optimization processing on the driving style information of the target vehicle to obtain driving characteristic information, and then the electronic equipment sends the driving characteristic information to the computer equipment.

In the embodiment of the present application, the driving characteristic information may include a historical driving path of the target vehicle, a collision time of the target vehicle with other environmental vehicles, a collision time of the target vehicle with a lane line, a time period required for the vehicle head to travel to the tail of the environmental vehicle, an overlapping rate of the target vehicle with other environmental vehicles in a unified lane, a lateral driving speed of the target vehicle, a lateral acceleration difference of the target vehicle, a heading angle difference of the target vehicle, a lane association difference of the target vehicle, a lane line curvature difference of the target vehicle, and the like.

Wherein the lateral acceleration difference characterizes a difference between a lateral acceleration of the target vehicle and a lateral acceleration of the ambient vehicle; the course angle difference represents the angle difference between the curve of the lane where the target vehicle is located and the curve of the running direction of the target vehicle; the lane association difference represents the difference between the number of the lane where the target vehicle is located and the number of the lane where the environmental vehicle is located; the lane line curvature difference characterizes a difference between a boundary line curvature of a lane where the target vehicle is located on the road map and a boundary line curvature of a lane where the environmental vehicle is located on the road map.

S200, inputting the driving characteristic information into an overtaking strategy prediction model, and obtaining a target overtaking strategy corresponding to the target vehicle according to the output of the overtaking strategy prediction model. The overtaking strategy prediction model is obtained by training based on sample driving characteristic information of various overtaking styles.

In the embodiment of the application, the overtaking strategy prediction model is a pre-trained neural network model which is obtained based on the sample driving characteristic information training of various overtaking styles, namely, the overtaking strategy prediction model can carry out overtaking strategy prediction on the driving characteristic information of different overtaking styles so as to obtain corresponding target overtaking strategies of different overtaking styles.

Specifically, the cut-in strategy prediction model may be composed of at least one of a convolutional neural network model, a cyclic neural network model, and an antagonistic neural network model. The driving characteristic information can be input into the overtaking strategy prediction model, and then the output of the overtaking strategy prediction model is processed to obtain the target overtaking strategy corresponding to the target vehicle. Optionally, the target cut-in strategy may include a target cut-in speed, a target cut-in path, a target cut-in acceleration, and/or a target cut-in heading angle.

Alternatively, the cut-in style may be understood as the driving requirement for completing cut-in operations according to different drivers. If the driving speed, the driving course angle, the lateral offset distance and other information requirements of different drivers when the overtaking operation is completed are different, namely the overtaking styles of the drivers are different.

In the embodiment of the application, the overtaking strategy prediction model consists of an input layer, a deep convolution pooling layer, a multi-layer perceptron layer, a multi-head attention mechanism layer, two full-connection layers, a judging activation layer and an output layer. The input layer is used for carrying out feature extraction on the driving feature information to obtain extracted feature information; the deep convolution pooling layer is used for carrying out blocking processing on the extracted characteristic information to obtain blocked characteristic information so as to reduce the information operand; the multi-layer perceptron layer is used for extracting depth characteristic information in the segmented characteristic information; the multi-head attention mechanism layer is used for weighting the depth characteristic information so as to improve the calculation speed and the accuracy of the calculation result; the two full-connection layers are used for classifying the weighted depth characteristic information to obtain a classification result; the judging and activating layer is used for extracting effective characteristic information from the classification result; the output layer is used for outputting effective characteristic information, namely outputting the overtaking strategy prediction model.

The following describes the construction process of the overtaking strategy prediction model in S200. As shown in fig. 3, the process of constructing the overtaking strategy prediction model includes:

s10, acquiring a plurality of sample driving characteristic information related to the target vehicle and a sample overtaking strategy corresponding to each sample driving characteristic information.

It should be noted that the plurality of sample driving characteristic information related to the target vehicle may be understood as a plurality of sample driving characteristic information related to different driving styles in the target vehicle history period. Alternatively, the plurality of sample driving characteristic information may be a plurality of sample driving characteristic information of a plurality of sample vehicles; the sample vehicle is a vehicle different from the target vehicle, and may or may not be the same type as the target vehicle. Alternatively, the sample driving characteristic information may include a historical driving path of the sample vehicle, a collision time with other environmental vehicles, a collision time of the sample vehicle with a lane line, a length of time required for the head of the sample vehicle to travel to the tail of the other environmental vehicles, a rate of overlap of the sample vehicle with the other environmental vehicles in a unified lane, a lateral driving speed of the sample vehicle, a lateral acceleration difference of the sample vehicle, a heading angle difference of the sample vehicle, a lane association difference of the sample vehicle, a lane line curvature difference of the sample vehicle, and the like.

In practical applications, sensors are also mounted on each sample vehicle for monitoring sample driving style information of the sample vehicle. Alternatively, the sample driving style information may include a lane change time of the sample vehicle, a collision time and a headway of the sample vehicle with a front vehicle of the sample vehicle at a start of the lane change of the sample vehicle, a collision time and a headway of the sample vehicle with a rear vehicle of the sample vehicle at a start of the lane change of the sample vehicle, a collision time and a headway of the sample vehicle with a front vehicle of the sample vehicle at a start of a cut-in return, a collision time and a headway of the sample vehicle with a rear vehicle of the sample vehicle at a start of the cut-in return, a collision time and a headway of the sample vehicle at an end of the cut-in, a lateral speed of the sample vehicle, a lateral acceleration of the sample vehicle, and a lateral jerk of the sample vehicle.

In one embodiment, the computer device may perform parameter optimization processing on the sample driving style information of each sample vehicle in the historical time period, and determine a plurality of sample driving feature information. Alternatively, the method of parameter optimization processing may be polynomial fitting, curve fitting, kalman filtering, particle filtering, and/or least squares, among others.

In still another embodiment, the other electronic device may perform parameter optimization processing on the sample driving style information of each sample vehicle, determine a plurality of sample driving feature information, and then send the acquired plurality of sample driving feature information to the computer device.

Meanwhile, the computer equipment can also acquire a sample overtaking strategy corresponding to each piece of sample driving characteristic information, and the sample overtaking strategy can be understood as an overtaking strategy gold standard corresponding to each piece of sample driving characteristic information.

Wherein, before executing the step in S10, the method further includes: acquiring candidate driving characteristic information of a plurality of sample vehicles and reference driving characteristic information of a target vehicle; candidate driving characteristic information matched with the reference driving characteristic information is screened out from the candidate driving characteristic information, and the candidate driving characteristic information is determined to be sample driving characteristic information.

In the embodiment of the application, the computer equipment can perform parameter optimization processing on the driving style information of the target vehicle in the historical time period to obtain the reference driving characteristic information of the target vehicle while acquiring the candidate driving characteristic information of a plurality of sample vehicles. Based on the obtained candidate driving feature information of the plurality of sample vehicles and the reference driving feature information of the target vehicle, the reference driving feature information and the candidate driving feature information can be subjected to matching processing, candidate driving feature information matched with the reference driving feature information is screened out from the candidate driving feature information, and the matched candidate driving feature information is determined to be the sample driving feature information.

And S20, training the initial overtaking strategy prediction model according to the plurality of sample driving characteristic information and the sample overtaking strategy corresponding to each sample driving characteristic information to obtain an overtaking strategy prediction model.

Based on the plurality of sample driving characteristic information and the sample overtaking strategy corresponding to each sample driving characteristic information obtained in the steps, the sample overtaking strategy corresponding to the plurality of sample driving characteristic information and each sample driving characteristic information is input into an initial overtaking strategy prediction model to obtain a predicted overtaking strategy, a loss value between the predicted overtaking strategy and the sample overtaking strategy is calculated through a loss function, and network parameters of the initial overtaking strategy prediction model are adjusted according to the loss value until the initial overtaking strategy prediction model converges, so that the overtaking strategy prediction model is obtained. Alternatively, the above-described loss function may be a logarithmic loss function, an exponential loss function, a perceptual loss function, a cross entropy loss function, or the like.

The overtaking strategy determination method provided by the embodiment of the application can acquire the driving characteristic information of the target vehicle, input the driving characteristic information into the overtaking strategy prediction model, and acquire the target overtaking strategy corresponding to the target vehicle according to the output of the overtaking strategy prediction model; according to the method, the target overtaking strategy can be determined through the overtaking strategy prediction model obtained based on the sample driving characteristic information training of various overtaking styles, so that the obtained target overtaking strategy can be suitable for different overtaking styles, the problem that the overtaking strategy is single is solved, the wide applicability of the overtaking strategy determination method is improved, and the use scenes of different overtaking styles are increased; meanwhile, the method can determine the corresponding overtaking strategies according to different overtaking styles, so that the traffic accident rate of overtaking operation according to the overtaking strategies of a single style can be reduced, and the safety in the overtaking process of the vehicle is improved.

In some scenarios, in order to improve the accuracy of the obtained target overtaking strategy, an initial overtaking path of the target vehicle may be obtained through an overtaking strategy prediction model, and then the initial overtaking path is processed to obtain a more accurate target overtaking strategy. Based on this, in an embodiment, the step of obtaining the target overtaking strategy corresponding to the target vehicle according to the output of the overtaking strategy prediction model in S200 may include the following steps: acquiring an initial overtaking path output by an overtaking strategy prediction model; and acquiring a target overtaking strategy corresponding to the target vehicle based on the initial overtaking path.

Specifically, the computer device may obtain an initial cut-in path output by the cut-in strategy prediction model. Further, a target overtaking strategy corresponding to the target vehicle is obtained based on the initial overtaking path. In the embodiment of the application, the target overtaking strategy comprises a target overtaking speed and a target overtaking path.

In an embodiment, the method for obtaining the target overtaking strategy corresponding to the target vehicle based on the initial overtaking path may be that an algorithm model is trained in advance, the initial overtaking path is input into the algorithm model, the target overtaking path is output after the initial overtaking path is adjusted by the algorithm model, and then the target overtaking speed is determined based on the target overtaking path, so as to obtain the target overtaking strategy.

In still another embodiment, the method of obtaining the target overtaking strategy corresponding to the target vehicle based on the initial overtaking path may be further that the target overtaking strategy is obtained by adjusting the initial overtaking path based on the historical overtaking path corresponding to the current overtaking style of the target vehicle, and then determining the target overtaking speed based on the target overtaking path.

According to the technical scheme, the initial overtaking path output by the overtaking strategy prediction model can be obtained, and the target overtaking strategy corresponding to the target vehicle is obtained based on the initial overtaking path; according to the method, the obtained initial overtaking path of the target vehicle can be processed to further obtain a target overtaking strategy, so that the accuracy of the obtained target overtaking strategy is higher, and when overtaking operation is executed based on the obtained target overtaking strategy, the traffic accident rate of overtaking operation executed according to a single style overtaking strategy can be reduced, and the safety in the overtaking process of the vehicle is improved; in addition, the method executes the overtaking operation based on the obtained target overtaking strategy, and can be suitable for overtaking styles of different drivers, so that the driving experience of the drivers on the auxiliary driving system can be improved.

In some scenarios, other environmental vehicles running around the target vehicle may exist, and the individual area around the target vehicle is a prohibited overtaking area, so that in order to improve the accuracy of the obtained target overtaking strategy, these problems need to be taken into consideration, the overtaking path corresponding to the overtaking area of the target vehicle may be obtained first, and then the target overtaking strategy is determined based on the overtaking path and the initial overtaking path. In the following, a description will be given of the above process of acquiring the target overtaking strategy corresponding to the target vehicle based on the initial overtaking path, and in an embodiment, as shown in fig. 4, the step of acquiring the target overtaking strategy corresponding to the target vehicle based on the initial overtaking path may include the following steps:

S210, acquiring an overtaking region of the target vehicle, and acquiring an overtaking path of the target vehicle according to the overtaking region.

It will be appreciated that the computer device may obtain an overtaking region around the target vehicle and then obtain an overtaking path for the target vehicle based on the overtaking region. The manner of acquiring the overtaking region around the target vehicle may be determined by the driver according to the actual driving environment and driving experience of the target vehicle. Optionally, the overtaking area may be understood as an area where no vehicle is driving in at the present moment; the overtaking region may include a partial region of a lane in which the target vehicle is currently located and a partial region of an adjacent lane of the lane in which the target vehicle is located, and the partial region of the lane in which the target vehicle is currently located is communicated with the partial region of the adjacent lane of the lane in which the target vehicle is located. The shape of the overtaking area may be a curve area, and the center line of the curve area may be referred to as an overtaking path.

Naturally, the manner in which the overtaking path of the target vehicle is acquired from the overtaking region may be determined from the coordinates of the boundary line lattice of the overtaking region; the boundary line lattice of the overtaking area can be a set formed by a plurality of discrete points on boundary lines on two sides of the overtaking area corresponding to the vehicle advancing direction. Alternatively, the way to obtain the overtaking path of the target vehicle according to the overtaking area may also be obtained according to the driving requirement of the driver on the target vehicle, where the obtained overtaking path may be any overtaking path in the overtaking area.

S220, determining a target overtaking strategy corresponding to the target vehicle through the initial overtaking path and the overtaking path.

The method for determining the target overtaking strategy corresponding to the target vehicle through the initial overtaking path and the overtaking path may be to adjust the overtaking path through the initial overtaking path to obtain the target overtaking path of the target vehicle, and then determine the target overtaking speed based on the target overtaking path to obtain the target overtaking strategy.

In addition, the method for determining the target overtaking strategy corresponding to the target vehicle through the initial overtaking path and the overtaking path can also be that the initial overtaking path is adjusted through the overtaking path to obtain the target overtaking path of the target vehicle, and then the target overtaking speed is determined based on the target overtaking path to obtain the target overtaking strategy.

Of course, the method of determining the target overtaking strategy corresponding to the target vehicle through the initial overtaking path and the overtaking path may also be to average the initial overtaking path and the overtaking path to obtain the target overtaking path of the target vehicle, and then determine the target overtaking speed based on the target overtaking path to obtain the target overtaking strategy.

The process of acquiring the overtaking region of the target vehicle will be described below. In an embodiment, as shown in fig. 5, the step of obtaining the overtaking region of the target vehicle in S210 may include:

s211, acquiring an environment image of the target vehicle, and carrying out semantic prediction on the environment image to obtain a candidate overtaking region in the road where the target vehicle is located.

In the actual driving process, if the target vehicle is going to overtake to the front vehicle, the overtaking rule is usually that the target vehicle is going to travel from the current lane to the adjacent lane, the passing operation is completed after the lane change from the adjacent lane to the current lane to the corresponding front vehicle, and the front vehicle may be referred to as a rear vehicle of the target vehicle.

Wherein, a plurality of image pickup apparatuses are also installed on the target vehicle, and the coverage view angle of the plurality of image pickup apparatuses may be 360 degrees. The computer equipment can acquire the environment image acquired by the camera equipment, and performs semantic prediction on the environment image to obtain a candidate overtaking region in the road where the target vehicle is located. Alternatively, the above-described environmental image may be an image of an environmental vehicle around the target vehicle and an ambient road. Meanwhile, the mode of carrying out semantic prediction on the environment image can be to carry out semantic prediction on the environment image based on the overtaking rule by adopting a semantic prediction algorithm, or can be to carry out semantic prediction on the environment image by adopting a pre-trained semantic prediction network.

In an actual overtaking environment of the target vehicle, if other lanes exist on both sides of the lane where the target vehicle is located, the obtained candidate overtaking region can comprise a left lane region of the lane where the target vehicle is located and a right lane region of the lane where the target vehicle is located; if only other lanes exist on the left side of the lane where the target vehicle is located, the acquired candidate overtaking region can comprise a lane region on the left side of the lane where the target vehicle is located; if there are only other lanes to the right of the lane in which the target vehicle is located, the acquired candidate overtaking region may include a lane region to the right of the lane in which the target vehicle is located. It should be noted that, when there are other lanes on the side of the lane where the target vehicle is located, the candidate overtaking regions obtained correspondingly on the side may be one or more.

S212, detecting traffic marking lines in the environment image to obtain attribute information of a road where the target vehicle is located and attribute information of a lane.

Meanwhile, the computer equipment can adopt a target detection algorithm to detect traffic marking in the environment image sent by the image pickup equipment, so as to obtain the attribute information of the road where the target vehicle is located and the attribute information of the lane. Alternatively, the target detection algorithm may be a single-stage detection algorithm or a dual-stage detection algorithm. The traffic markings may include indicator markings, forbidden markings, and warning markings.

After the detection processing is performed, the obtained attribute information of the road where the target vehicle is located may include a boundary line lattice of the road, a size of the road, and a shape of the road, and correspondingly, the obtained attribute information of the lane may include a boundary line lattice of the lane, a size of the lane, and a shape of the lane.

S213, carrying out semantic processing on the environment image based on the attribute information of the road and the attribute information of the lane, and acquiring the overtaking region of the target vehicle from the candidate overtaking region.

In an embodiment, based on the attribute information of the road and the attribute information of the lane acquired in the previous step, a semantic processing algorithm is adopted to perform semantic processing on each candidate overtaking region in the environment image sent by the image capturing device, so as to obtain an overtaking region of the target vehicle acquired from the candidate overtaking region.

In still another embodiment, the attribute information of the road, the attribute information of the lane, and the environmental image sent by the image capturing device obtained in the previous step may be input into a pre-trained semantic processing model, semantic processing is performed on each candidate overtaking region in the environmental image, and the overtaking region of the target vehicle is obtained from the candidate overtaking region.

Alternatively, the above-mentioned obtaining the overtaking regions of the target vehicle from the candidate overtaking regions may be understood as obtaining the deletion processed regions from deleting the overtaking prohibition regions from among the candidate overtaking regions, then screening out the deletion processed regions which meet the passing width of the target vehicle and correspond to the overtaking path being shortest from among the deletion processed regions, and determining the screened deletion processed regions as the overtaking regions of the target vehicle.

According to the method and the device for obtaining the overtaking regions of the target vehicles, the candidate overtaking regions of the target vehicles can be initially obtained, and then the overtaking regions of the target vehicles are obtained from the candidate overtaking regions, so that the overtaking paths corresponding to the overtaking regions are finally obtained to be shortest, and overtaking operation is guaranteed not to violate traffic rules.

According to the technical scheme, the corresponding overtaking path can be obtained according to the running environment of the target vehicle, then the target overtaking strategy corresponding to the target vehicle is determined through the initial overtaking path and the overtaking path, so that the accuracy of the obtained target overtaking strategy is higher, and when overtaking operation is executed based on the obtained target overtaking strategy, the traffic accident rate of overtaking operation executed according to the single-style overtaking strategy can be reduced, and the safety in the overtaking process of the vehicle is improved; in addition, the method executes the overtaking operation based on the obtained target overtaking strategy, and can be suitable for overtaking styles of different drivers, so that the driving experience of the drivers on the auxiliary driving system can be improved.

In some scenarios, in order to improve the accuracy of the obtained target overtaking strategy, the current running path of the target vehicle may be considered, and the target overtaking strategy may be comprehensively determined through the current running path, the initial overtaking path and the overtaking path of the target vehicle. In an embodiment, as shown in fig. 6, the step of determining the target overtaking policy corresponding to the target vehicle through the initial overtaking path and the overtaking path in S220 may include:

S221, acquiring a reference driving path of the target vehicle, and adjusting the reference driving path according to the initial overtaking path and the overtaking path to obtain a candidate overtaking path of the target vehicle.

Specifically, the computer device may perform lane detection processing on the environmental image acquired by the image capturing device, to obtain a lane boundary lattice where the target vehicle is located, and determine a reference travel path of the target vehicle according to the lane boundary lattice where the target vehicle is located, that is, a current travel path of the target vehicle, that is, a travel path of the target vehicle before the overtaking operation is performed.

Further, the computer device may adjust the reference travel path according to the initial overtaking path and the overtaking path of the target vehicle obtained in the above steps, so as to obtain a candidate overtaking path of the target vehicle. Optionally, the adjustment processing of the reference travel path according to the initial overtaking path and the overtaking path may be performed by calculating an average value of the initial overtaking path, the overtaking path and the reference travel path, or may be performed by adjusting a position of a part of travel points in the reference travel path based on the initial overtaking path and the overtaking path.

In one scenario, the current travel path of the target vehicle may be determined by lane lines around the target vehicle. As shown in fig. 7, the step of acquiring the reference travel path of the target vehicle in S221 described above may be implemented by:

S2211, fitting smoothing is carried out on boundary points of the lane lines based on the position information of the boundary points of the lane lines of the road where the target vehicle is located, and the position information of the fitting smoothing boundary points of the lane lines is obtained.

Optionally, based on the position information of the boundary point of the lane line of the road where the target vehicle is located, performing fitting smoothing processing on the boundary point of the lane line by adopting a curve fitting method to obtain the position information of the fitting smoothing boundary point of the lane line. Alternatively, the curve fitting method may be a least square method, or a method of approximating discrete data by an analytical expression. The above-mentioned position information of the fitting smooth boundary point may be coordinates of the fitting smooth boundary point.

S2212, carrying out average processing on the position information of the fitting smooth boundary points of the lane lines to obtain a reference driving path of the target vehicle.

Further, the position information of the fitted smooth boundary point of the lane line obtained in the above step may be averaged to obtain a center point of the fitted smooth boundary point of the lane line, and the center points of the fitted smooth boundary points may be sequentially combined according to the position order to obtain the reference driving path of the target vehicle.

According to the embodiment of the application, fitting smoothing and averaging can be carried out according to the obtained position information of the lane line boundary point of the road where the target vehicle is located, so that the current running path of the target vehicle with higher precision can be obtained.

S222, determining a target overtaking strategy of the target vehicle based on the candidate overtaking path.

The target overtaking path of the target vehicle can be obtained by adjusting the candidate overtaking path again according to the requirement of the driver on the overtaking path, and then the target overtaking speed is determined based on the target overtaking path, so that the target overtaking strategy is obtained.

In one scenario, to further improve the accuracy of the determined target cut-in strategy, error information of the target vehicle may also be taken into account to determine the target cut-in strategy. In an embodiment, as shown in fig. 8, the step of determining the target overtaking strategy of the target vehicle based on the candidate overtaking path in S222 may be implemented by:

S2221, acquiring error information related to a target vehicle, and carrying out fitting smoothing processing on the error information based on a candidate overtaking path to obtain smoothed error information; the error information includes lane line error information, positioning error information of the target vehicle, and dynamics model error information of the target vehicle.

Specifically, the computer device may acquire lane line error information, positioning error information of the target vehicle, kinetic model error information of the target vehicle. The lane line error information is obtained by calculating the difference between the position information of the boundary point of the actual lane line where the target vehicle is located and the position information of the boundary point of the lane line obtained in the previous step; the positioning error information of the target vehicle is obtained by calculating the difference between the actual position information of the target vehicle and the position information detected by the positioning system mounted on the target vehicle; the dynamics model error information of the target vehicle is obtained by a difference between the dynamics model of the target vehicle simulated in the actual process and the actual model of the target vehicle.

Further, based on the obtained candidate overtaking path, fitting smoothing algorithm is adopted to carry out fitting smoothing processing on the error information, and smoothed error information is obtained. Meanwhile, the candidate overtaking path and the error information of the target vehicle can be input into a pre-trained algorithm model, and the algorithm model carries out fitting smoothing processing on the error information and then outputs the smoothed error information.

S2222, determining a target overtaking path of the target vehicle through the smoothed error information and the candidate overtaking path.

In an embodiment, the candidate overtaking path may be adjusted based on the smoothed error information, and the target overtaking path with the shortest overtaking path may be selected from the adjusted overtaking paths.

In yet another embodiment, the overtaking path matched with the candidate overtaking path may be searched from the mapping relation according to the smoothed error information, and the overtaking path matched with the candidate overtaking path is determined as the target overtaking path. Alternatively, the mapping relationship may include the smoothed error information and the overtaking path and a one-to-one correspondence therebetween.

S2223, determining a target overtaking strategy of the target vehicle according to the target overtaking path.

The target overtaking speed can be determined according to the target overtaking path, and the target overtaking path and the target overtaking speed are determined to be a target overtaking strategy of the target vehicle. In the embodiment of the application, the target overtaking path can be understood as a target overtaking running lattice, and correspondingly, the target overtaking speed can be understood as the running speed corresponding to each target overtaking running point.

As shown in fig. 9, the following step of determining the target cut-in strategy of the target vehicle according to the target cut-in path in S2223 may be implemented by the following steps:

S2223a, obtaining the driving duration between each two adjacent overtaking position points in the target overtaking path.

Specifically, the computer device may predict a travel duration between each adjacent cut-in location point in the target cut-in path by the each adjacent cut-in location point in the target cut-in path and the current travel speed of the target vehicle. The driving time length between the adjacent overtaking position points can be equal or unequal. The adjacent overtaking position point can be understood as an adjacent target overtaking driving point in the target overtaking driving lattice (target overtaking path).

S2223b, determining the target overtaking speed corresponding to the target overtaking path according to the target overtaking path and the running duration between each two adjacent overtaking position points in the target overtaking path.

According to the position information of each adjacent target overtaking traveling point in the target overtaking traveling point array (target overtaking path), calculating the overtaking distance between each adjacent target overtaking traveling point by adopting a Euclidean distance method, obtaining the traveling speed corresponding to each target overtaking traveling point by using the overtaking distance between each adjacent target overtaking traveling point and the corresponding traveling time length as a quotient, and combining the traveling speeds corresponding to each target overtaking traveling point to obtain the target overtaking speed corresponding to the target overtaking path. The target cut-in speed is understood here as a lattice of the travel speeds of the target vehicle when cut-in.

S2223c, determining a target overtaking strategy of the target vehicle according to the target overtaking path and the target overtaking speed.

According to the technical scheme, the reference driving path of the target vehicle can be obtained, the reference driving path is adjusted according to the initial overtaking path and the overtaking path, the candidate overtaking path of the target vehicle is obtained, and then the target overtaking strategy of the target vehicle is determined based on the candidate overtaking path; according to the method, the current running path, the preliminary overtaking path and the overtaking path of the target vehicle are comprehensively considered, the target overtaking path which is shortest in overtaking path and highest in overtaking success rate is determined, and the target overtaking strategy is determined based on the target overtaking path, so that the overtaking operation speed can be improved, the overtaking operation success rate and the overtaking operation passing efficiency can be improved.

In one embodiment, the present application further provides a method for determining an overtaking strategy, the method comprising the following steps:

(1) And constructing a overtaking strategy prediction model.

The process for constructing the overtaking strategy prediction model comprises the following steps:

Screening candidate driving characteristic information matched with the reference driving characteristic information from the candidate driving characteristic information, and determining the candidate driving characteristic information as sample driving characteristic information;

(2) Inputting driving characteristic information into an overtaking strategy prediction model, and acquiring an initial overtaking path output by the overtaking strategy prediction model; the overtaking strategy prediction model is obtained by training based on sample driving characteristic information of various overtaking styles.

(3) And acquiring an environment image of the target vehicle, and carrying out semantic prediction on the environment image to obtain a candidate overtaking region in the road where the target vehicle is located.

(4) And detecting traffic marking lines in the environment image to obtain attribute information of a road where the target vehicle is located and attribute information of a lane.

(5) And carrying out semantic processing on the environment image based on the attribute information of the road and the attribute information of the lane, acquiring an overtaking region of the target vehicle from the candidate overtaking region, and acquiring an overtaking path of the target vehicle according to the overtaking region.

(6) And carrying out fitting smoothing treatment on the boundary points of the lane lines based on the position information of the boundary points of the lane lines of the road where the target vehicle is located, so as to obtain the position information of the fitting smoothing boundary points of the lane lines.

(7) And carrying out average processing on the position information of the fitted smooth boundary points of the lane lines to obtain a reference driving path of the target vehicle, and carrying out adjustment processing on the reference driving path according to the initial overtaking path and the overtaking path to obtain a candidate overtaking path of the target vehicle.

(8) Acquiring error information related to a target vehicle, and performing fitting smoothing processing on the error information based on a candidate overtaking path to obtain smoothed error information; the error information includes lane line error information, positioning error information of the target vehicle, and dynamics model error information of the target vehicle.

(9) And determining the target overtaking path of the target vehicle through the smoothed error information and the candidate overtaking path.

(10) And acquiring the driving time length between each two adjacent overtaking position points in the target overtaking path.

(11) And determining the target overtaking speed corresponding to the target overtaking path according to the target overtaking path and the running duration between each two adjacent overtaking position points in the target overtaking path.

(12) And determining a target overtaking strategy of the target vehicle according to the target overtaking path and the target overtaking speed.

The implementation process of the above (1) to (12) may be specifically referred to the description of the above embodiment, and its implementation principle and technical effects are similar, and will not be described herein again.

It should be understood that, although the steps in the flowcharts related to the above embodiments are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides an overtaking strategy determination device for realizing the overtaking strategy determination method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in the embodiments of the one or more overtaking policy determining devices provided below may be referred to the limitation of the overtaking policy determining method hereinabove, and will not be described herein.

In one embodiment, fig. 10 is a schematic structural diagram of an overtaking policy determining device according to one embodiment of the present application, where the overtaking policy determining device provided by the embodiment of the present application may be applied to a computer device. As shown in fig. 10, the overtaking strategy determination device according to the embodiment of the present application may include: a feature information acquisition module 11 and a prediction module 12; wherein:

a feature information acquisition module 11 for acquiring driving feature information of the target vehicle;

the prediction module 12 is configured to input driving feature information into an overtaking strategy prediction model, and obtain a target overtaking strategy corresponding to the target vehicle according to output of the overtaking strategy prediction model;

The overtaking strategy determination device provided by the embodiment of the application can be used for executing the technical scheme in the overtaking strategy determination method embodiment of the application, and the implementation principle and the technical effect are similar, and are not repeated here.

In one embodiment, the prediction module 12 includes: an initial path acquisition unit and a target policy acquisition unit, wherein:

The initial path acquisition unit is used for acquiring an initial overtaking path output by the overtaking strategy prediction model;

The target strategy acquisition unit is used for acquiring a target overtaking strategy corresponding to the target vehicle based on the initial overtaking path.

In one embodiment, the target policy obtaining unit is specifically configured to:

In one embodiment, the target strategy acquisition unit comprises an overtaking region acquisition subunit; the overtaking region acquisition subunit is configured to:

In one embodiment, the target policy obtaining unit further comprises a target policy obtaining subunit; the target policy acquisition subunit is configured to:

In one embodiment, the target policy acquisition subunit is specifically configured to:

In one embodiment, the overtaking strategy determination device further comprises: a prediction model construction module; the prediction model construction module is specifically used for:

In one embodiment, the overtaking strategy determination device further comprises: a sample information acquisition module; the sample information acquisition module is specifically configured to:

For specific limitations of the overtaking strategy determination means, reference may be made to the above limitation of the overtaking strategy determination method, and no further description is given here. The respective modules in the above-described passing policy determination device may be implemented in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 11. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and an information base. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The information base of the computer device is used for storing driving characteristic information. The network interface of the computer device is for communicating with an external endpoint via a network connection. The computer program is executed by a processor to implement a method of determining a cut-in strategy.

It will be appreciated by those skilled in the art that the structure shown in FIG. 11 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.

In an embodiment, a computer device is further provided, including a memory and a processor, where the memory stores a computer program, and the processor implements the technical solution in the above embodiment of the method for determining an overtaking policy according to the present application when executing the computer program, and the implementation principle and technical effects are similar, and are not repeated herein.

In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, where the computer program when executed by a processor implements the technical scheme of the above-mentioned overtaking strategy determination method of the present application, and the implementation principle and technical effect are similar, and are not repeated herein.

In one embodiment, a computer program product is provided, which includes a computer program, where the computer program when executed by a processor implements the technical solution of the above-mentioned overtaking strategy determination method of the present application, and the implementation principle and technical effects are similar, and are not repeated herein.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, information storage, or other medium used in embodiments provided herein can include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, or the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory. By way of illustration, and not limitation, RAM can be in various forms such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.

Claims

1. A method of determining a cut-in strategy, the method comprising:

Acquiring driving characteristic information of a target vehicle;

the overtaking strategy prediction model is trained based on sample driving characteristic information of various overtaking styles.

2. The method of claim 1, wherein the obtaining the target cut-in strategy corresponding to the target vehicle according to the output of the cut-in strategy prediction model comprises:

Acquiring an initial overtaking path output by the overtaking strategy prediction model;

3. The method of claim 2, wherein the obtaining, based on the initial cut-in path, a target cut-in strategy corresponding to the target vehicle comprises:

4. The method of claim 3, wherein the acquiring the overtaking region of the target vehicle comprises:

acquiring an environment image of the target vehicle, and carrying out semantic prediction on the environment image to obtain a candidate overtaking region in a road where the target vehicle is located;

detecting traffic marking lines in the environment image to obtain attribute information of a road where the target vehicle is located and attribute information of a lane;

5. The method of claim 4, wherein the determining a target cut-in strategy corresponding to the target vehicle via the initial cut-in path and the cut-in path comprises:

6. The method of claim 5, wherein the obtaining the reference travel path of the target vehicle comprises:

And carrying out average processing on the position information of the fitting smooth boundary point of the lane line to obtain the reference driving path of the target vehicle.

7. The method of claim 5, wherein the determining a target cut-in strategy for the target vehicle based on the candidate cut-in path comprises:

acquiring error information related to the target vehicle, and performing fitting smoothing processing on the error information based on the candidate overtaking path to obtain smoothed error information; the error information comprises lane line error information, positioning error information of the target vehicle and dynamics model error information of the target vehicle;

8. The method of claim 7, wherein the determining a target cut-in strategy for the target vehicle based on the target cut-in path comprises:

Acquiring the driving time length between adjacent overtaking position points in the target overtaking path;

determining a target overtaking speed corresponding to the target overtaking path according to the target overtaking path and the running duration between each adjacent overtaking position point in the target overtaking path;

9. The method according to any one of claims 1 to 8, wherein the process of constructing the cut-in strategy prediction model comprises:

Acquiring a plurality of sample driving characteristic information related to the target vehicle and a sample overtaking strategy corresponding to each sample driving characteristic information;

And training an initial overtaking strategy prediction model according to the plurality of sample driving characteristic information and the sample overtaking strategy corresponding to each sample driving characteristic information to obtain the overtaking strategy prediction model.

10. The method of claim 9, wherein prior to the obtaining a plurality of sample driving characteristic information associated with the target vehicle and the sample cut-in strategy corresponding to each of the sample driving characteristic information, the method further comprises:

Acquiring candidate driving characteristic information of a plurality of sample vehicles and reference driving characteristic information of the target vehicle;

And screening candidate driving characteristic information matched with the reference driving characteristic information from the candidate driving characteristic information, and determining the candidate driving characteristic information as the sample driving characteristic information.

11. An overtaking strategy determination apparatus, the apparatus comprising:

The prediction module is used for inputting the driving characteristic information into an overtaking strategy prediction model, and obtaining a target overtaking strategy corresponding to the target vehicle according to the output of the overtaking strategy prediction model;

12. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1-10 when the computer program is executed.

13. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1-10.

14. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any one of claims 1-10.