Vehicle intelligent driving control method, device and equipment and computer storage medium

Abstract

The invention relates to a vehicle intelligent driving control method, device and equipment and a computer storage medium, and belongs to the technical field of intelligent driving. The method comprises the steps that the longitudinal position, the transverse position, the longitudinal speed and the transverse speed of a current vehicle in a current driving scene are obtained; the longitudinal position, the transverse position, the longitudinal speed, the transverse speed, the vehicle length and the vehicle width of the target vehicle in the current driving scene are obtained; calculating longitudinal risk time when the current vehicle collides with the target vehicle in the longitudinal direction; calculating the transverse lane changing time when the current vehicle and the target vehicle are overlapped in the transverse direction; and performing intelligent driving control on the current vehicle through the longitudinal risk time and the transverse lane changing time. According to the method, the longitudinal risk time and the transverse lane changing time of the vehicle in the driving process are determined, so that vehicle collision is avoided, and the accuracy and the reliability of the intelligent driving technology of the vehicle are ensured.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and in particular to a vehicle intelligent driving control method, device, equipment and computer storage medium. Background Art

[0002] As a product of the integration of the automotive industry and information technology, intelligent driving is gradually becoming an important development direction of the automotive industry. The intelligent driving system mainly includes three key links: network navigation, autonomous driving and manual intervention. Among them, network navigation solves problems such as vehicle positioning and path planning; autonomous driving is to complete driving behaviors such as lane keeping, overtaking and merging, and traffic signal recognition under the control of the intelligent system; and manual intervention means that when necessary, the driver can take over the control of the vehicle and respond to the actual road conditions. However, the current vehicle intelligent driving technology also faces some challenges and limitations. For example, in a complex and changing road environment, how to ensure the accuracy and reliability of vehicle intelligent driving technology is an urgent problem to be solved. Summary of the invention

[0003] In view of this, it is necessary to provide a vehicle intelligent driving control method, device, equipment and computer storage medium to ensure the accuracy and reliability of vehicle intelligent driving technology.

[0004] In order to solve the above problems, in a first aspect, the present invention provides a vehicle intelligent driving control method, comprising: Obtaining the longitudinal position, lateral position, longitudinal speed and lateral speed of the current vehicle in the current driving scene, and obtaining the longitudinal position, lateral position, longitudinal speed, lateral speed, vehicle length and vehicle width of the target vehicle in the current driving scene; Calculate the longitudinal risk time when the current vehicle collides with the target vehicle in the longitudinal direction based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle; Calculate the lateral lane change time when the current vehicle and the target vehicle overlap in the lateral direction based on the lateral position of the current vehicle, the lateral position of the target vehicle, the vehicle width of the target vehicle, the lateral speed of the current vehicle and the lateral speed of the target vehicle; Intelligent driving control is performed on the current vehicle through the longitudinal risk time and the lateral lane change time.

[0005] In a possible implementation, the calculating, based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle, and the longitudinal speed of the target vehicle, the longitudinal risk time when the current vehicle collides with the target vehicle in the longitudinal direction includes: Calculating a collision time when the current vehicle collides with the target vehicle in the longitudinal direction based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle; Based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle and the longitudinal speed of the current vehicle, the inter-vehicle time corresponding to when the current vehicle travels to the position of the target vehicle is calculated; The longitudinal risk time when the current vehicle collides with the target vehicle in the longitudinal direction is determined according to the collision time and the vehicle-to-vehicle time.

[0006] In a possible implementation, based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle, and the longitudinal speed of the target vehicle, the collision time when the current vehicle collides with the target vehicle in the longitudinal direction is calculated:

[0007] In the formula, Indicates the collision time when the current vehicle collides with the target vehicle in the longitudinal direction. Indicates the longitudinal position of the target vehicle; Indicates the current longitudinal position of the vehicle; Indicates the length of the target vehicle; the longitudinal speed of the target vehicle; the longitudinal speed of the current vehicle; Based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle and the longitudinal speed of the current vehicle, the inter-vehicle time corresponding to when the current vehicle travels to the position of the target vehicle is calculated;

[0008] In the formula, Indicates the workshop time corresponding to the time when the current vehicle travels to the target vehicle position.

[0009] In a possible implementation, based on the lateral position of the current vehicle, the lateral position of the target vehicle, the vehicle width of the target vehicle, the lateral speed of the current vehicle, and the lateral speed of the target vehicle, a lateral lane change time when the current vehicle and the target vehicle overlap in the lateral direction is calculated;

[0010] In the formula, Indicates the lateral lane change time when the current vehicle and the target vehicle overlap in the lateral direction. Indicates the current lateral position of the vehicle. represents the lateral position of the target vehicle, represents the width of the target vehicle, represents the lateral velocity of the target vehicle, Indicates the current lateral speed of the vehicle.

[0011] In a possible implementation, performing intelligent driving control on the current vehicle by using the longitudinal risk time and the lateral lane change time includes: Calculating a risk index of contact between the current vehicle and the target vehicle in a current driving scenario based on the longitudinal risk time and the lateral lane change time; Obtaining a first risk index data set of the current vehicle that changes over time in a preset time period and a second risk index data set established for a successful risk avoidance scenario of the current vehicle at a historical moment, and performing a correlation analysis on the first risk index data set and the second risk index data set to obtain a risk correlation coefficient; When the risk correlation coefficient is greater than a first preset threshold and the risk index is less than a second preset threshold, a risk avoidance planning trajectory of the current vehicle is generated, and intelligent driving control is performed on the current vehicle based on the risk avoidance planning trajectory.

[0012] In a possible implementation, generating a risk avoidance planning trajectory of the current vehicle includes: Obtaining first historical track information of the current vehicle and second historical track information of the target vehicle; Using a coding neural network to encode the first historical trajectory information and the second historical trajectory information respectively, to obtain a current vehicle hidden layer vector and a target vehicle hidden layer vector containing historical trajectory features; The current vehicle hidden layer vector and the target vehicle hidden layer vector are used as vertex feature vectors, and the current vehicle and the target vehicle are used as vertices, and the vertices are connected to obtain an edge feature vector; Constructing a graph structure based on the vertex feature vector and the edge feature vector, and determining the connection relationship of the vertices in the graph structure as an adjacency matrix; Based on the adjacency matrix and the vertex feature vector, generating a trajectory interaction feature vector between the current vehicle and the target vehicle; Calculating the attention scores of the current vehicle and the target vehicle according to the current vehicle hidden layer vector and the target vehicle hidden layer vector, and normalizing the attention scores to obtain the attention vectors of the current vehicle and the target vehicle; Concatenating the trajectory interaction feature vector and the attention vector to generate a driving intention recognition vector for the current vehicle; The trajectory interaction feature vector, the attention vector and the driving intention recognition vector of the current vehicle are input into a decoding neural network for decoding, the future trajectory probability distribution under different driving intentions is predicted, and the risk avoidance planning trajectory of the current vehicle is generated based on the future trajectory probability distribution.

[0013] In a possible implementation, the current driving scene includes a straight-line driving scene and / or a lane-changing driving scene, and the target vehicle includes a vehicle that is within a preset distance range from the current vehicle.

[0014] In a second aspect, the present invention further provides a vehicle intelligent driving control device, comprising: A data acquisition module is used to acquire the longitudinal position, lateral position, longitudinal speed and lateral speed of the current vehicle in the current driving scene, and acquire the longitudinal position, lateral position, longitudinal speed, lateral speed, vehicle length and vehicle width of the target vehicle in the current driving scene; A first time calculation module, for calculating a longitudinal risk time when a collision occurs between the current vehicle and the target vehicle in the longitudinal direction based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle; a second time calculation module, for calculating a lateral lane change time when the current vehicle and the target vehicle overlap in the lateral direction based on the lateral position of the current vehicle, the lateral position of the target vehicle, the vehicle width of the target vehicle, the lateral speed of the current vehicle and the lateral speed of the target vehicle; The intelligent driving control module is used to perform intelligent driving control on the current vehicle through the longitudinal risk time and the lateral lane change time.

[0015] In a third aspect, the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps in the above-mentioned vehicle intelligent driving control method when executing the computer program.

[0016] In a fourth aspect, the present invention further provides a computer storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned vehicle intelligent driving control method are implemented.

[0017] The beneficial effects of the present invention are: The present invention first obtains the longitudinal position, lateral position, longitudinal speed and lateral speed of the current vehicle in the current driving scene, as well as the longitudinal position, lateral position, longitudinal speed, lateral speed, vehicle length and vehicle width of the target vehicle in the current driving scene; then, based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the vehicle length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle, the longitudinal risk time when the current vehicle collides with the target vehicle is calculated; at the same time, based on the lateral position of the current vehicle, the lateral position of the target vehicle, the vehicle width of the target vehicle, the lateral speed of the current vehicle and the lateral speed of the target vehicle, the lateral lane change time when the current vehicle overlaps with the target vehicle in the lateral direction is calculated; finally, through the longitudinal risk time and the lateral lane change time, the current vehicle is intelligently driven and controlled so that the current vehicle does not contact the target vehicle in the current driving scene. The accuracy and reliability of the vehicle intelligent driving technology are ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A method flow chart of an embodiment of a vehicle intelligent driving control method provided by the present invention; Figure 2 A schematic diagram of a current vehicle and a target vehicle in a driving scene provided in an embodiment of the present invention; Figure 3 A flow chart of a method for performing intelligent driving control on a current vehicle provided by an embodiment of the present invention; Figure 4 A flow chart of a method for generating a risk avoidance planning trajectory of a current vehicle provided by an embodiment of the present invention; Figure 5 A schematic diagram of a vehicle intelligent driving control device according to an embodiment of the present invention Figure 6 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0021] In the description of the embodiments of the present invention, unless otherwise specified, “plurality” means two or more than two.

[0022] The descriptions of "first", "second", etc. involved in the embodiments of the present invention are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the technical features defined as "first" and "second" may explicitly or implicitly include at least one of the features.

[0023] Reference to an "embodiment" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0024] A specific embodiment of the present invention, as Figure 1 As shown, Figure 1 A method flow chart of an embodiment of a vehicle intelligent driving control method provided by the present invention includes: S101: Acquire the longitudinal position, lateral position, longitudinal speed and lateral speed of the current vehicle in the current driving scene, and acquire the longitudinal position, lateral position, longitudinal speed, lateral speed, vehicle length and vehicle width of the target vehicle in the current driving scene; S102: Calculating a longitudinal risk time when the current vehicle collides with the target vehicle in the longitudinal direction based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle; S103: Calculating a lateral lane change time when the current vehicle and the target vehicle overlap in the lateral direction based on the lateral position of the current vehicle, the lateral position of the target vehicle, the vehicle width of the target vehicle, the lateral speed of the current vehicle, and the lateral speed of the target vehicle; S104: Perform intelligent driving control on the current vehicle through the longitudinal risk time and the lateral lane change time.

[0025] It can be understood that the current driving scene includes a straight-line driving scene and / or a lane-changing driving scene, and the target vehicle includes surrounding vehicles whose distance from the current vehicle is within a preset distance range. The current vehicle can be a manned vehicle or an unmanned vehicle, and the target vehicle can be a manned vehicle or an unmanned vehicle. As an example, please refer to Figure 2 , Figure 2 A schematic diagram of a current vehicle and a target vehicle in a certain driving scene is provided in one embodiment of the present invention.

[0026] The present invention first obtains the longitudinal position, lateral position, longitudinal speed and lateral speed of the current vehicle in the current driving scene, as well as the longitudinal position, lateral position, longitudinal speed, lateral speed, vehicle length and vehicle width of the target vehicle in the current driving scene; then, based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the vehicle length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle, the longitudinal risk time when the current vehicle collides with the target vehicle is calculated; at the same time, based on the lateral position of the current vehicle, the lateral position of the target vehicle, the vehicle width of the target vehicle, the lateral speed of the current vehicle and the lateral speed of the target vehicle, the lateral lane change time when the current vehicle overlaps with the target vehicle in the lateral direction is calculated; finally, through the longitudinal risk time and the lateral lane change time, the current vehicle is intelligently driven and controlled so that the current vehicle does not contact the target vehicle in the current driving scene. The accuracy and reliability of the vehicle intelligent driving technology are ensured.

[0027] In one embodiment of the present invention, based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle, the longitudinal risk time when the current vehicle collides with the target vehicle in the longitudinal direction is calculated, including: Calculating a collision time when the current vehicle collides with the target vehicle in the longitudinal direction based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle; Based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle and the longitudinal speed of the current vehicle, the inter-vehicle time corresponding to when the current vehicle travels to the position of the target vehicle is calculated; According to the collision time and the vehicle-to-vehicle time, the longitudinal risk time when the current vehicle collides with the target vehicle in the longitudinal direction is determined.

[0028] It can be understood that the collision time and the workshop time are used as longitudinal risk indicators for the current vehicle's risk avoidance decision; among them, the collision time can represent the time required before the two vehicles collide, which is the relative distance between the two vehicles divided by the relative speed. The smaller the value, the greater the collision risk. When the speed of the current vehicle is less than that of the target vehicle, the collision time may not be able to be risk assessed, so the workshop time is introduced as a supplement to the longitudinal risk indicator assessment. The workshop time is the time required for the current vehicle to reach the target vehicle position, and the smaller the value, the greater the collision risk. Specifically, based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle, the collision time when the current vehicle collides with the target vehicle in the longitudinal direction is calculated: ; In the formula, Indicates the collision time when the current vehicle collides with the target vehicle in the longitudinal direction. Indicates the longitudinal position of the target vehicle; Indicates the current longitudinal position of the vehicle; Indicates the length of the target vehicle; The longitudinal velocity of the target vehicle; the longitudinal speed of the current vehicle; in addition, based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle and the longitudinal speed of the current vehicle, the inter-vehicle time corresponding to the time when the current vehicle travels to the position of the target vehicle is calculated; ; In the formula, Indicates the workshop time corresponding to the time when the current vehicle travels to the target vehicle position.

[0029] If the relative distance between the current vehicle and the target vehicle is short and the relative speed is low, if the target vehicle brakes urgently, the current distance is not enough to brake to avoid danger. However, due to the low relative speed, the collision time cannot represent the risk. At this time, the value of the inter-vehicle time is small and can represent a large risk, thus compensating for the disadvantage of the collision time. Therefore, the longitudinal risk time when the current vehicle collides with the target vehicle can be selected based on the collision time and inter-vehicle time. Specifically, ,in, represents the longitudinal risk time; and is a constant, and It can be set according to actual conditions, and this embodiment does not limit the specific numerical value.

[0030] In one embodiment of the present invention, the lateral collision risk between the current vehicle and the target vehicle mainly comes from the vehicle on the lateral side, so the lateral lane change time is set as the lateral risk indicator of the current vehicle's risk avoidance decision. Specifically, based on the lateral position of the current vehicle, the lateral position of the target vehicle, the vehicle width of the target vehicle, the lateral speed of the current vehicle and the lateral speed of the target vehicle, the lateral lane change time when the current vehicle and the target vehicle overlap in the lateral direction is calculated: ; In the formula, Indicates the lateral lane change time when the current vehicle and the target vehicle overlap in the lateral direction. Indicates the current lateral position of the vehicle. represents the lateral position of the target vehicle, represents the width of the target vehicle, represents the lateral velocity of the target vehicle, Indicates the current lateral speed of the vehicle.

[0031] In one embodiment of the present invention, Figure 3 As shown, Figure 3 A flow chart of a method for performing intelligent driving control on a current vehicle provided in one embodiment of the present invention.

[0032] S301: Calculating a risk index of contact between the current vehicle and the target vehicle in the current driving scenario based on the longitudinal risk time and the lateral lane change time; S302: Obtain a first risk index data set of the current vehicle that changes over time in a preset time period and a second risk index data set established for a successful risk avoidance scenario of the current vehicle at a historical moment, and perform a correlation analysis on the first risk index data set and the second risk index data set to obtain a risk correlation coefficient; S303: When the risk correlation coefficient is greater than a first preset threshold and the risk index is less than a second preset threshold, a risk avoidance planning trajectory of the current vehicle is generated based on the first historical trajectory information of the current vehicle and the second historical trajectory information of the target vehicle, and intelligent driving control is performed on the current vehicle based on the risk avoidance planning trajectory.

[0033] It can be understood that when other vehicles on the side change lanes to the lane where the current vehicle is located, the longitudinal risk time can be obtained and lateral lane change time ,like , it means that before the longitudinal positions of the two vehicles overlap, the other side vehicle has completed the lane change, and only the longitudinal risk needs to be considered at this time; if , it means that when the longitudinal positions of the two vehicles overlap, the other vehicle on the side has not completed the lane change. At this time, the risk of collision is relatively high, and the lateral lane change time must be considered. To increase risk considerations. That is, the risk index of the current vehicle and the target vehicle coming into contact in the current driving scenario is defined by the longitudinal risk time and the lateral lane change time: ; In the formula, Indicates the risk index of contact between the current vehicle and the target vehicle in the current driving scenario. represents the longitudinal risk time; represents the error coefficient set in advance or in real time; in this embodiment or other embodiments, the error coefficient Specific numerical limits can be made according to actual conditions, and their role is to prevent calculation errors. For potential collision risks, a risk indicator database of successful risk avoidance scenarios is established based on the data set. During driving, the correlation between the current risk indicator and the successful risk avoidance scenario database is compared to reflect the possibility of future collision risks, and when the correlation is high, risk avoidance planning is triggered to avoid potential collision risks. Therefore, by introducing the risk correlation coefficient to reflect the closeness of the relationship between the two groups of variables and measure their linear relationship, we have:

[0034] In the formula, Represents the risk correlation coefficient, and its value range is between, The closer the absolute value of is to 0, the weaker the correlation is, and the closer it is to 1, the stronger the correlation is. A data set representing the risk index of the current vehicle changing over time during a preset time period; It represents a data set established based on the risk index corresponding to the vehicle's successful risk avoidance scenarios at historical moments; represents the risk index covariance; express The standard deviation of express The standard deviation of Record the risk index of the current vehicle during driving and conduct correlation analysis with the data set at historical moments; is greater than the first threshold and When the distance between the two vehicles is less than the second threshold, the risk avoidance planning trajectory of the current vehicle is generated based on the first historical trajectory information of the current vehicle and the second historical trajectory information of the target vehicle, and the current vehicle is intelligently driven and controlled according to the risk avoidance planning trajectory. In addition, in order to avoid being too aggressive based on the time domain indicator evaluation alone, the spatial domain safety distance can be used as a prerequisite for risk avoidance decision-making, that is, when the distance between the two vehicles is less than the safety distance at the current vehicle speed, the decision is triggered to determine whether the risk avoidance planning trajectory of the current vehicle needs to be generated.

[0035] In one embodiment of the present invention, Figure 4 As shown, Figure 4 A flow chart of a method for generating a risk avoidance planning trajectory of a current vehicle provided in an embodiment of the present invention includes: S401: Encode the first historical trajectory information and the second historical trajectory information respectively using a coding neural network to obtain a current vehicle hidden layer vector and a target vehicle hidden layer vector containing historical trajectory features; S402: taking the current vehicle hidden layer vector and the target vehicle hidden layer vector as vertex feature vectors, and taking the current vehicle and the target vehicle as vertices, and connecting the vertices to obtain an edge feature vector; S403: constructing a graph structure based on the vertex feature vectors and the edge feature vectors, and determining the connection relationship of the vertices in the graph structure as an adjacency matrix; S404: Generate a trajectory interaction feature vector between the current vehicle and the target vehicle based on the adjacency matrix and the vertex feature vector; S405: Calculate the attention scores of the current vehicle and the target vehicle according to the hidden layer vector of the current vehicle and the hidden layer vector of the target vehicle, and normalize the attention scores to obtain the attention vectors of the current vehicle and the target vehicle; S406: concatenating the trajectory interaction feature vector and the attention vector to generate a driving intention recognition vector of the current vehicle; S407: Input the trajectory interaction feature vector, the attention vector and the driving intention recognition vector of the current vehicle into the decoding neural network for decoding, predict the future trajectory probability distribution under different driving intentions, and generate the risk avoidance planning trajectory of the current vehicle based on the future trajectory probability distribution.

[0036] It is understandable that obtaining the moment The historical track information of the current vehicle and the historical track information of the target vehicle are obtained. Specifically, ; In the formula, Indicates the current vehicle at time The horizontal position coordinates of Indicates the current vehicle at time The vertical position coordinates of Indicates The target vehicle at time The lateral relative distance between the current vehicle and the next vehicle. Indicates The target vehicle at time The longitudinal relative distance between the current vehicle and the corresponding position. If there is no vehicle at the corresponding position, the corresponding lateral relative distance and longitudinal relative distance are set to positive infinity.

[0037] The encoding neural network composed of the long short-term memory neural network LSTM is used to encode the first historical trajectory information of the current vehicle and the second historical trajectory information of the target vehicle, and the current vehicle hidden layer vector and the target vehicle hidden layer vector containing historical trajectory features are obtained respectively. Specifically: , where Indicates time The hidden layer state containing the historical trajectory features is as follows. Correspondingly, the current vehicle hidden layer vector containing the historical trajectory features is , the corresponding hidden layer vector of the target vehicle containing historical trajectory features is , Represents the weights of the encoded neural network.

[0038] The obtained current vehicle hidden layer vector and target vehicle hidden layer vector are used as vertex feature vectors, and the current vehicle and target vehicle are used as vertices, and the edge feature vectors are obtained by connecting the vertices; and a graph structure is constructed based on the vertex feature vectors and the edge feature vectors, and the connection relationship of the vertices in the graph structure is represented as an adjacency matrix. Specifically, at time Under this circumstance, the current vehicle and the target vehicle are regarded as graph structures. In Vertices, the current vehicle hidden vector obtained by encoding the long short-term memory neural network LSTM and the target vehicle hidden vector As the vertex eigenvector, , by connecting vertices to get edge features , and then represent the connection relationship of the vertices as an adjacency matrix .

[0039] The trajectory interaction feature vector of the current vehicle and the target vehicle is generated through the adjacency matrix and vertex feature vector. Specifically, by superimposing two layers of graph convolutional neural network, the output is obtained at the moment The trajectory interaction feature vector containing its own features : , where represents the activation function, is the weight of the first layer of graph convolutional neural network, is the weight of the second layer of graph convolutional neural network.

[0040] Based on the hidden layer vector of the current vehicle and the hidden layer vector of the target vehicle, the attention scores of the current vehicle and the target vehicle are calculated, and the attention scores are normalized to obtain the attention vectors of the current vehicle and the target vehicle. The hidden vector of the target vehicle after being encoded by the long short-term memory neural network LSTM As K and V, and the current vehicle hidden vector Considered as Q, the correlation between K and Q is calculated to explain the degree of interaction between the target vehicle and the current vehicle. and The dimensions are the same, and cosine similarity is selected as the attention score calculation function. The obtained attention score After the softmax function is normalized, the output is the interaction weight. , and weighted sum it with V to get the final attention vector , the calculation expression is as follows: , , .

[0041] The trajectory interaction feature vector and the attention vector are concatenated to generate a driving intention recognition vector; and the trajectory interaction feature vector, the attention vector and the driving intention recognition vector are input into the decoding neural network for decoding, and the future trajectory probability distribution under different driving intentions is predicted, and the risk avoidance planning trajectory of the current vehicle is generated based on the future trajectory probability distribution. Specifically, the trajectory interaction feature vector after graph convolution is With attention vector The concatenation is used as the input vector for driving intention prediction, and the output is normalized by the softmax function to obtain the driving intention recognition vector ,in , and They represent the probability of changing lanes to the left, going straight, and changing lanes to the right, respectively. That is, the output of the intention recognition layer is ; Finally, the trajectory interaction feature vector after graph convolution is , attention vector and the intent category vector After splicing, the data is input into the decoding neural network composed of the long short-term memory neural network LSTM to predict the probability distribution of future trajectories under different intentions: ; In the formula, represents the future trajectory predicted by the sliding window method; Predict the bivariate Gaussian distribution parameters for each frame in the future , and are the means of the horizontal and vertical positions, respectively. and are the variances of the horizontal and vertical positions, respectively. is the correlation coefficient.

[0042] It can be seen that this method determines the risk index of longitudinal collision and lateral collision between the current vehicle and the target vehicle based on multiple vehicle parameters, which can provide a judgment basis for the current vehicle to perform risk avoidance planning. At the same time, by real-time monitoring of the driving risk index and then performing correlation analysis with the risk database of successful risk avoidance scenarios, the potential collision risk between the current vehicle and the target vehicle can be judged, thereby generating the risk avoidance planning trajectory of the current vehicle in advance, and then performing intelligent driving control on the current vehicle according to the generated risk avoidance planning trajectory to help the current vehicle avoid collision risks. And when generating the risk avoidance planning trajectory, the graph convolutional neural network and the long short-term memory network can be combined, and the attention mechanism can be added to establish a trajectory prediction model, so as to predict the probability distribution of future trajectories under different driving intentions and generate the risk avoidance planning trajectory of the current vehicle.

[0043] In order to better implement the vehicle intelligent driving control method in the embodiment of the present invention, based on the vehicle intelligent driving control method, correspondingly, Figure 5 As shown, Figure 5 A schematic diagram of the structure of an embodiment of a vehicle intelligent driving control device provided by the present invention includes: The data acquisition module 501 is used to acquire the longitudinal position, lateral position, longitudinal speed and lateral speed of the current vehicle in the current driving scene, and acquire the longitudinal position, lateral position, longitudinal speed, lateral speed, vehicle length and vehicle width of the target vehicle in the current driving scene; A first time calculation module 502, for calculating a longitudinal risk time when the current vehicle collides with the target vehicle in the longitudinal direction based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle; A second time calculation module 503, for calculating a lateral lane change time when the current vehicle and the target vehicle overlap in the lateral direction based on the lateral position of the current vehicle, the lateral position of the target vehicle, the vehicle width of the target vehicle, the lateral speed of the current vehicle and the lateral speed of the target vehicle; The intelligent driving control module 504 is used to perform intelligent driving control on the current vehicle through longitudinal risk time and lateral lane change time.

[0044] The vehicle intelligent driving control device 500 provided in the above embodiment can implement the technical solution described in the above vehicle intelligent driving control method embodiment. The specific implementation principles of the above modules or units can refer to the corresponding contents in the above vehicle intelligent driving control method embodiment, which will not be repeated here.

[0045] like Figure 6 As shown, the present invention also provides an electronic device 600. The electronic device 600 includes a processor 601, a memory 602 and a display 603. Figure 6 Only some components of the electronic device 600 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0046] In some embodiments, the processor 601 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 602, such as the vehicle intelligent driving control method of the present invention.

[0047] In some embodiments, the processor 601 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processor 601 may be local or remote. In some embodiments, the processor 601 may be implemented in a cloud platform. In one embodiment, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-cloud, etc., or any combination thereof.

[0048] In some embodiments, the memory 602 may be an internal storage unit of the electronic device 600, such as a hard disk or memory of the electronic device 600. In other embodiments, the memory 602 may also be an external storage device of the electronic device 600, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the electronic device 600.

[0049] Furthermore, the memory 602 may include both an internal storage unit of the electronic device 600 and an external storage device. The memory 602 is used to store application software installed in the electronic device 600 and various data.

[0050] In some embodiments, the display 603 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 603 is used to display information of the electronic device 600 and to display a visual user interface. The components 601-603 of the electronic device 600 communicate with each other through a system bus.

[0051] In some embodiments, when the processor 601 executes the vehicle intelligent driving control program in the memory 602, the following steps may be implemented: Obtaining the longitudinal position, lateral position, longitudinal speed and lateral speed of the current vehicle in the current driving scene, and obtaining the longitudinal position, lateral position, longitudinal speed, lateral speed, vehicle length and vehicle width of the target vehicle in the current driving scene; Calculate the longitudinal risk time when the current vehicle collides with the target vehicle in the longitudinal direction based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle; Calculate the lateral lane change time when the current vehicle and the target vehicle overlap in the lateral direction based on the lateral position of the current vehicle, the lateral position of the target vehicle, the vehicle width of the target vehicle, the lateral speed of the current vehicle and the lateral speed of the target vehicle; Intelligent driving control of the current vehicle is carried out through longitudinal risk time and lateral lane change time.

[0052] It should be understood that: when the processor 601 executes the vehicle intelligent driving control program in the memory 602, in addition to the above functions, other functions can also be implemented. For details, please refer to the description of the corresponding method embodiment above.

[0053] Furthermore, the embodiment of the present invention does not specifically limit the type of the electronic device 600 mentioned, and the electronic device 600 may be a portable electronic device such as a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer, etc. Exemplary embodiments of portable electronic devices include but are not limited to portable electronic devices equipped with IOS, Android, Microsoft or other operating systems. The above-mentioned portable electronic devices may also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, the electronic device 600 may not be a portable electronic device, but a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0054] In another exemplary embodiment of the present invention, the embodiment further provides a vehicle, which includes a vehicle intelligent driving control device as described in the above embodiment, or includes an electronic device as described in the above embodiment. It should be noted that since the specific manner in which the vehicle intelligent driving control device and the electronic device perform operations has been described in detail in the embodiments, the technical functions and effects of the vehicle provided in this embodiment can be referred to the above embodiments, and will not be repeated here.

[0055] Accordingly, an embodiment of the present application also provides a computer-readable storage medium, which is used to store computer-readable programs or instructions. When the program or instructions are executed by a processor, the steps or functions of the vehicle intelligent driving control method provided in the above-mentioned method embodiments can be implemented.

[0056] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0057] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A vehicle intelligent driving control method, characterized in that: include: Obtaining the longitudinal position, lateral position, longitudinal speed and lateral speed of the current vehicle in the current driving scene, and obtaining the longitudinal position, lateral position, longitudinal speed, lateral speed, vehicle length and vehicle width of the target vehicle in the current driving scene; Calculate the longitudinal risk time when the current vehicle collides with the target vehicle in the longitudinal direction based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle; Calculate the lateral lane change time when the current vehicle and the target vehicle overlap in the lateral direction based on the lateral position of the current vehicle, the lateral position of the target vehicle, the vehicle width of the target vehicle, the lateral speed of the current vehicle and the lateral speed of the target vehicle; Intelligent driving control is performed on the current vehicle through the longitudinal risk time and the lateral lane change time.

2. The vehicle intelligent driving control method according to claim 1, characterized in that: The method of calculating the longitudinal risk time when the current vehicle collides with the target vehicle in the longitudinal direction based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle comprises: Calculating a collision time when the current vehicle collides with the target vehicle in the longitudinal direction based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle; Based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle and the longitudinal speed of the current vehicle, the inter-vehicle time corresponding to when the current vehicle travels to the position of the target vehicle is calculated; The longitudinal risk time when the current vehicle collides with the target vehicle in the longitudinal direction is determined according to the collision time and the vehicle-to-vehicle time.

3. The vehicle intelligent driving control method according to claim 2, characterized in that: Based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle, the collision time when the current vehicle collides with the target vehicle in the longitudinal direction is calculated: In the formula, Indicates the collision time when the current vehicle collides with the target vehicle in the longitudinal direction. Indicates the longitudinal position of the target vehicle; Indicates the current longitudinal position of the vehicle; Indicates the length of the target vehicle; the longitudinal speed of the target vehicle; the longitudinal speed of the current vehicle; Based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle and the longitudinal speed of the current vehicle, the inter-vehicle time corresponding to when the current vehicle travels to the position of the target vehicle is calculated; In the formula, Indicates the workshop time corresponding to the time when the current vehicle travels to the target vehicle position.

4. The vehicle intelligent driving control method according to claim 3, characterized in that: Calculate the lateral lane change time when the current vehicle and the target vehicle overlap in the lateral direction based on the lateral position of the current vehicle, the lateral position of the target vehicle, the vehicle width of the target vehicle, the lateral speed of the current vehicle and the lateral speed of the target vehicle; In the formula, Indicates the lateral lane change time when the current vehicle overlaps the target vehicle in the lateral direction. Indicates the current lateral position of the vehicle, represents the lateral position of the target vehicle, represents the width of the target vehicle, represents the lateral velocity of the target vehicle, Indicates the current lateral speed of the vehicle.

5. The vehicle intelligent driving control method according to claim 4, characterized in that: The intelligent driving control of the current vehicle according to the longitudinal risk time and the lateral lane change time includes: Calculating a risk index of contact between the current vehicle and the target vehicle in a current driving scenario based on the longitudinal risk time and the lateral lane change time; Obtaining a first risk index data set of the current vehicle that changes over time in a preset time period and a second risk index data set established for a successful risk avoidance scenario of the current vehicle at a historical moment, and performing a correlation analysis on the first risk index data set and the second risk index data set to obtain a risk correlation coefficient; When the risk correlation coefficient is greater than a first preset threshold and the risk index is less than a second preset threshold, a risk avoidance planning trajectory of the current vehicle is generated based on the first historical trajectory information of the current vehicle and the second historical trajectory information of the target vehicle, and intelligent driving control is performed on the current vehicle based on the risk avoidance planning trajectory.

6. The vehicle intelligent driving control method according to claim 5, characterized in that: Generating a risk avoidance planning trajectory of the current vehicle based on the first historical trajectory information of the current vehicle and the second historical trajectory information of the target vehicle, including: Using a coding neural network to encode the first historical trajectory information and the second historical trajectory information respectively, to obtain a current vehicle hidden layer vector and a target vehicle hidden layer vector containing historical trajectory features; The current vehicle hidden layer vector and the target vehicle hidden layer vector are used as vertex feature vectors, and the current vehicle and the target vehicle are used as vertices, and the vertices are connected to obtain an edge feature vector; Constructing a graph structure based on the vertex feature vector and the edge feature vector, and determining the connection relationship of the vertices in the graph structure as an adjacency matrix; Based on the adjacency matrix and the vertex feature vector, generating a trajectory interaction feature vector between the current vehicle and the target vehicle; Calculating the attention scores of the current vehicle and the target vehicle according to the current vehicle hidden layer vector and the target vehicle hidden layer vector, and normalizing the attention scores to obtain the attention vectors of the current vehicle and the target vehicle; Concatenating the trajectory interaction feature vector and the attention vector to generate a driving intention recognition vector for the current vehicle; The trajectory interaction feature vector, the attention vector and the driving intention recognition vector of the current vehicle are input into a decoding neural network for decoding, the future trajectory probability distribution under different driving intentions is predicted, and the risk avoidance planning trajectory of the current vehicle is generated based on the future trajectory probability distribution.

7. The vehicle intelligent driving control method according to claim 1, characterized in that: The current driving scene includes a straight-line driving scene and / or a lane-changing driving scene, and the target vehicle is a vehicle whose distance from the current vehicle is within a preset distance range.

8. A vehicle intelligent driving control device, characterized in that: include: A data acquisition module is used to acquire the longitudinal position, lateral position, longitudinal speed and lateral speed of the current vehicle in the current driving scene, and acquire the longitudinal position, lateral position, longitudinal speed, lateral speed, vehicle length and vehicle width of the target vehicle in the current driving scene; A first time calculation module, for calculating a longitudinal risk time when a collision occurs between the current vehicle and the target vehicle in the longitudinal direction based on the longitudinal position of the current vehicle, the longitudinal position of the target vehicle, the length of the target vehicle, the longitudinal speed of the current vehicle and the longitudinal speed of the target vehicle; a second time calculation module, for calculating a lateral lane change time when the current vehicle and the target vehicle overlap in the lateral direction based on the lateral position of the current vehicle, the lateral position of the target vehicle, the vehicle width of the target vehicle, the lateral speed of the current vehicle and the lateral speed of the target vehicle; The intelligent driving control module is used to perform intelligent driving control on the current vehicle through the longitudinal risk time and the lateral lane change time.

9. An electronic device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the vehicle intelligent driving control method described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the vehicle intelligent driving control method described in any one of claims 1 to 7 above.

Images