A method and system for identifying vehicle lateral driving conditions based on coordinate system transformation

Abstract

The present invention discloses a vehicle lateral driving condition identification method and system based on coordinate system conversion, aiming to concisely and accurately identify the vehicle lateral condition. The method comprises: converting the vehicle driving trajectory of the global Cartesian coordinate system to the Frenet coordinate system; calculating the identification index data of the vehicle lateral driving condition, including the absolute value of the vehicle trajectory slope based on the Frenet coordinate system, the absolute value of the vehicle trajectory slope based on the global Cartesian coordinate system and the direction factor; based on the calculated condition identification index data, determining the vehicle lateral driving process to which the vehicle driving condition to be identified belongs, including the lane keeping driving condition and the lane changing driving condition. The present invention also provides a vehicle lateral driving condition identification system based on coordinate system conversion. Since the present invention adopts the vehicle trajectory slope of the global Cartesian coordinate system and the Frenet coordinate system for lateral condition identification, it has strong applicability and is more concise and direct.

Description

Technical Field

[0001] The present invention relates to the field of intelligent automobile perception technology, and in particular to a vehicle lateral driving condition recognition method and system based on coordinate system conversion. Background Art

[0002] Vehicle driving condition recognition is a key basic work for predicting driver's driving intention, identifying driving style and evaluating driving skills. By identifying the vehicle's driving condition, the specific driving condition components of the vehicle driving process can be effectively divided, which is helpful for further analysis and research on driving intention, driving style and driving skill evaluation.

[0003] There are currently two main methods for identifying lateral vehicle driving conditions. Their main principles and steps are as follows:

[0004] 1. Step 1: Select state quantities related to the lateral movement of the vehicle as judgment indicators to identify the working condition, such as yaw angular velocity, lateral acceleration, and lateral displacement; Step 2: Set the corresponding threshold according to the selected identification indicator; Step 3: Establish a streamlined judgment process based on the identification indicator to identify the lateral driving condition.

[0005] The invention patent with the announcement number CN109878530B discloses a method and system for identifying the lateral driving condition of a vehicle, that is, using the above technology. The disadvantages of this method are:

[0006] The selected working condition identification indicators are related to the vehicle's motion state, such as yaw angular velocity, lateral acceleration, etc., but they are closely related to the vehicle's driving speed. Fixed thresholds or simple table lookup relationship threshold settings cannot effectively identify lateral driving conditions.

[0007] The selected working condition identification index based on lateral displacement in the geodetic coordinate system cannot effectively reflect the relationship between the vehicle and the geometric characteristics of the road. This also affects the recognition accuracy of the vehicle's lateral driving condition to a certain extent, and its applicability is limited.

[0008] This complex selection of multiple identification indicators and setting of threshold relationships are not conducive to the simplicity of vehicle driving condition identification.

[0009] 2. Step 1: Select characteristic parameters for vehicle driving condition identification; Step 2: Perform correlation analysis on the preliminarily selected characteristic parameters and simplify the characteristic parameters; Step 3: Use clustering method to classify vehicle driving conditions.

[0010] The invention patent with publication number CN103921743A discloses a vehicle driving condition identification system and a method thereof, that is, using the above technology. The disadvantages of this method are:

[0011] A large amount of vehicle driving data needs to be collected to avoid the influence of accidental points on the results and to ensure the accuracy of vehicle driving condition classification, which increases the workload considerably;

[0012] This method of identifying vehicle driving conditions based on clustering of driving condition segments makes it difficult to determine the start and end times of the identified conditions, and is not easy to further calculate the running time of the identified conditions.

[0013] At the same time, with the rapid development of autonomous driving technology, more and more on-board cameras are installed on vehicles, which can obtain more external environmental information such as lane line information, which provides convenient conditions for identifying vehicle driving conditions. Summary of the invention

[0014] In view of the above situation, the object of the present invention is to provide a vehicle lateral driving condition identification method and system based on coordinate system transformation, which is concise and accurate, has a wide range of applications, and can meet the needs of driving intention prediction, driving style identification and driving skill evaluation.

[0015] To achieve the above object, the technical solution adopted by the present invention is:

[0016] An embodiment of the present invention provides a method for identifying a vehicle lateral driving condition based on coordinate system transformation, comprising:

[0017] Convert the vehicle trajectory in the global Cartesian coordinate system to the Frenet coordinate system;

[0018] Calculating identification index data of the vehicle lateral driving condition, wherein the condition identification index data includes an absolute value of a vehicle trajectory slope based on a Frenet coordinate system and an absolute value of a vehicle trajectory slope based on a global Cartesian coordinate system;

[0019] Identify the driving process of the vehicle to be identified according to the set vehicle lateral driving condition identification method;

[0020] Based on the calculated working condition identification index data, determining the vehicle lateral driving condition to which the vehicle driving condition to be identified belongs, wherein the vehicle lateral driving condition includes a lane keeping driving condition and a lane changing driving condition, wherein the lane keeping driving condition and the lane changing driving condition are identified based on a change in the slope of the vehicle trajectory in a Frenet coordinate system;

[0021] Based on the calculated working condition identification index data, further determining the lane keeping driving condition to which the vehicle lateral driving condition belongs, wherein the lane keeping driving condition includes a straight driving condition and a curved driving condition, wherein the straight driving condition and the curved driving condition are identified based on a change in the vehicle trajectory slope in a global Cartesian coordinate system;

[0022] Calculating identification index data of a lateral driving condition of the vehicle, wherein the identification index data of the condition includes a direction factor;

[0023] Based on the calculated working condition identification index data, further determining the lane-changing driving condition and the curve driving condition to which the vehicle lateral driving condition belongs, the lane-changing driving condition includes a left-hand lane-changing driving condition and a right-hand lane-changing driving condition, and the curve driving condition includes a left-hand curve driving condition and a right-hand curve driving condition, wherein the left-hand lane-changing driving condition and the right-hand lane-changing driving condition are identified based on a direction factor, and the left-hand curve driving condition and the right-hand curve driving condition are identified based on a direction factor;

[0024] Optionally, the identifying the driving process of the vehicle to be identified according to the set vehicle lateral driving condition identification method includes:

[0025] When the absolute value of the slope of the vehicle trajectory in the Frenet coordinate system at the first judgment moment is greater than a preset first threshold, the first judgment moment is recorded as the start moment of the vehicle lane-changing driving condition;

[0026] When the absolute value of the vehicle trajectory slope in the Frenet coordinate system at the first judgment moment is less than a preset first threshold, the first judgment moment is recorded as the start moment of the vehicle lane keeping driving condition;

[0027] When the absolute value of the slope of the vehicle trajectory in the Frenet coordinate system at the second judgment moment is less than the preset first threshold, the second judgment moment is recorded as the end moment of the vehicle lane-changing driving condition;

[0028] When the absolute value of the vehicle trajectory slope in the Frenet coordinate system at the second judgment moment is greater than a preset first threshold, the second judgment moment is recorded as the end moment of the vehicle lane keeping driving condition;

[0029] Optionally, the method for further determining the lane keeping driving condition to which the vehicle lateral driving condition belongs based on the calculated condition identification index data includes:

[0030] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the first judgment moment is greater than a preset second threshold, the first judgment moment is recorded as the start moment of the vehicle curve driving condition;

[0031] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the first judgment moment is less than a preset second threshold, the first judgment moment is recorded as the start moment of the vehicle straight-line driving condition;

[0032] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the second judgment moment is less than a preset second threshold, the second judgment moment is recorded as the end moment of the vehicle curve driving condition;

[0033] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the second judgment moment is greater than a preset second threshold, the second judgment moment is recorded as the end moment of the vehicle straight-line driving condition;

[0034] Optionally, when the direction factor at the average judgment moment is greater than a preset third threshold, the lane-changing driving condition belongs to a left-hand lane-changing driving condition; when the direction factor at the average judgment moment is less than the preset third threshold, the lane-changing driving condition belongs to a right-hand lane-changing driving condition;

[0035] Optionally, when the direction factor at the average judgment moment is greater than a preset third threshold, the curve driving condition belongs to a left-hand curve driving condition; when the direction factor at the average judgment moment is less than the preset third threshold, the curve driving condition belongs to a right-hand curve driving condition;

[0036] The average judgment time is the average of the first judgment time and the second judgment time;

[0037] Optionally, the reference line sampling points of the vehicle driving trajectory and the sampling points of the vehicle driving trajectory are smoothed, and the smoothing method is an anti-pulse interference average value filtering algorithm;

[0038] Optionally, the vehicle driving trajectory in the Cartesian coordinate system is converted to the Frenet coordinate system, including:

[0039] Obtaining the lateral coordinates and longitudinal coordinates of the reference line sampling points of the vehicle driving trajectory in the global Cartesian coordinate system; the reference line of the vehicle driving trajectory is the center line of the lane; the sampling frequency of the reference line sampling points of the vehicle driving trajectory is selected according to hardware conditions;

[0040] Obtaining the horizontal and vertical coordinates of the sampling points of the vehicle driving trajectory in the global Cartesian coordinate system; the vehicle driving trajectory is the driving trajectory of the vehicle's center of mass; the sampling frequency of the reference line sampling points of the vehicle driving trajectory is selected according to hardware conditions;

[0041] Calculate the heading angle of the reference line sampling point of the vehicle's driving trajectory;

[0042] Calculate the distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point set of the adjacent vehicle driving trajectory; the distance is the Euclidean distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point of the adjacent vehicle driving trajectory; the reference line sampling point set of the adjacent vehicle driving trajectory is a set of a set number of sampling points before and after the corresponding sampling moment of the selected vehicle driving trajectory reference line sampling point;

[0043] Find the vehicle driving trajectory reference line sampling point with the shortest distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point set of the adjacent vehicle driving trajectory; the vehicle driving trajectory reference line sampling point with the shortest distance is used as the projection point of the global Cartesian coordinate system converted to the Frenet coordinate system;

[0044] The path value of the reference line sampling point of the vehicle driving trajectory corresponding to the reference line projection point of the vehicle driving trajectory is used as the lateral coordinate of the Frenet coordinate system;

[0045] The distance and direction factor from the reference line sampling point of the vehicle driving trajectory to the projection point of the corresponding vehicle driving trajectory reference line constitute the longitudinal coordinate of the Frenet coordinate system;

[0046] Optionally, the preset first threshold is 0.021; the preset second threshold is 0.01; the preset third threshold is 0;

[0047] Another embodiment of the present invention provides a vehicle lateral driving condition identification system based on coordinate system transformation, comprising:

[0048] A data input module is used to obtain vehicle driving trajectory data in a global Cartesian coordinate system and reference line data of the vehicle driving trajectory in the global Cartesian coordinate system; the vehicle driving trajectory data in the global Cartesian coordinate system is acquired by a high-precision positioning system; the vehicle driving trajectory reference line data in the global Cartesian coordinate system is acquired by a high-precision vehicle-mounted camera;

[0049] A data processing and calculation module is used to filter and smooth the direct data obtained by the data input module and convert the vehicle driving trajectory of the global Cartesian coordinate system into the Frenet coordinate system;

[0050] A working condition identification index calculation module, used to calculate identification index data of the vehicle lateral driving working condition, wherein the working condition identification index data includes the absolute value of the vehicle trajectory slope based on the Frenet coordinate system, the absolute value of the vehicle trajectory slope based on the global Cartesian coordinate system, and a direction factor;

[0051] A lateral driving condition identification module, used to determine the vehicle lateral driving condition to which the vehicle driving condition to be identified belongs based on the calculated condition identification index data, wherein the vehicle lateral driving condition includes a lane keeping driving condition and a lane changing driving condition, wherein the lane keeping driving condition and the lane changing driving condition are identified based on the vehicle trajectory slope change in the Frenet coordinate system;

[0052] A lane keeping driving condition identification module, for further determining the lane keeping driving condition to which the vehicle lateral driving condition belongs based on the calculated condition identification index data, wherein the lane keeping driving condition includes a straight driving condition and a curved driving condition, wherein the straight driving condition and the curved driving condition are identified based on a change in the vehicle trajectory slope in a global Cartesian coordinate system;

[0053] A driving direction identification module, used for further determining the lane-changing driving condition and the curve driving condition to which the vehicle lateral driving condition belongs based on the calculated condition identification index data, wherein the lane-changing driving condition includes a left lane-changing driving condition and a right lane-changing driving condition, and the curve driving condition includes a left curve driving condition and a right curve driving condition, wherein the left lane-changing driving condition and the right lane-changing driving condition are identified based on a direction factor, and the left curve driving condition and the right curve driving condition are identified based on a direction factor;

[0054] Optionally, the lateral driving condition recognition module recognizes the vehicle driving process including:

[0055] When the absolute value of the slope of the vehicle trajectory in the Frenet coordinate system at the first judgment moment is greater than a preset first threshold, the first judgment moment is recorded as the start moment of the vehicle lane-changing driving condition;

[0056] When the absolute value of the vehicle trajectory slope in the Frenet coordinate system at the first judgment moment is less than a preset first threshold, the first judgment moment is recorded as the start moment of the vehicle lane keeping driving condition;

[0057] When the absolute value of the slope of the vehicle trajectory in the Frenet coordinate system at the second judgment moment is less than the preset first threshold, the second judgment moment is recorded as the end moment of the vehicle lane-changing driving condition;

[0058] When the absolute value of the vehicle trajectory slope in the Frenet coordinate system at the second judgment moment is greater than a preset first threshold, the second judgment moment is recorded as the end moment of the vehicle lane keeping driving condition;

[0059] Optionally, the specific steps of the lane keeping driving condition recognition module recognizing the lane keeping driving condition to which the vehicle lateral driving condition belongs include:

[0060] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the first judgment moment is greater than a preset second threshold, the first judgment moment is recorded as the start moment of the vehicle curve driving condition;

[0061] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the first judgment moment is less than a preset second threshold, the first judgment moment is recorded as the start moment of the vehicle straight-line driving condition;

[0062] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the second judgment moment is less than a preset second threshold, the second judgment moment is recorded as the end moment of the vehicle curve driving condition;

[0063] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the second judgment moment is greater than a preset second threshold, the second judgment moment is recorded as the end moment of the vehicle straight-line driving condition;

[0064] Optionally, the specific steps of the driving direction recognition module identifying the curve driving condition and the lane change driving condition to which the vehicle lateral driving condition belongs include:

[0065] When the direction factor at the average judgment time is greater than the preset third threshold, the lane-changing driving condition belongs to the left lane-changing driving condition; when the direction factor at the average judgment time is less than the preset third threshold, the lane-changing driving condition belongs to the right lane-changing driving condition;

[0066] When the direction factor at the average judgment time is greater than the preset third threshold, the curve driving condition belongs to the left curve driving condition; when the direction factor at the average judgment time is less than the preset third threshold, the curve driving condition belongs to the right curve driving condition;

[0067] The average judgment time is the average of the first judgment time and the second judgment time;

[0068] Optionally, the method for filtering and smoothing the direct data in the data processing and calculation module is an anti-pulse interference average value filtering algorithm;

[0069] Optionally, the specific steps of converting the vehicle driving trajectory of the global Cartesian coordinate system to the Frenet coordinate system in the data processing and calculation module include:

[0070] Obtaining the lateral coordinates and longitudinal coordinates of the reference line sampling points of the vehicle driving trajectory in the global Cartesian coordinate system; the reference line of the vehicle driving trajectory is the center line of the lane; the sampling frequency of the reference line sampling points of the vehicle driving trajectory is selected according to hardware conditions;

[0071] Obtaining the horizontal and vertical coordinates of the sampling points of the vehicle driving trajectory in the global Cartesian coordinate system; the vehicle driving trajectory is the driving trajectory of the vehicle's center of mass; the sampling frequency of the reference line sampling points of the vehicle driving trajectory is selected according to hardware conditions;

[0072] Calculate the vehicle heading angle at the reference line sampling point of the vehicle's driving trajectory;

[0073] Calculate the distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point set of the adjacent vehicle driving trajectory; the distance is the Euclidean distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point of the adjacent vehicle driving trajectory; the reference line sampling point set of the adjacent vehicle driving trajectory is a set of a set number of sampling points before and after the corresponding sampling time of the selected vehicle driving trajectory reference line sampling point;

[0074] Find the vehicle driving trajectory reference line sampling point with the shortest distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling points of the adjacent vehicle driving trajectories; the vehicle driving trajectory reference line sampling point with the shortest distance is used as the projection point of the global Cartesian coordinate system converted to the Frenet coordinate system;

[0075] The path value of the reference line sampling point of the vehicle driving trajectory corresponding to the reference line projection point of the vehicle driving trajectory is used as the lateral coordinate of the Frenet coordinate system;

[0076] The distance and direction factor from the reference line sampling point of the vehicle driving trajectory to the projection point of the corresponding vehicle driving trajectory reference line constitute the longitudinal coordinate of the Frenet coordinate system;

[0077] Optionally, the preset first threshold is 0.021; the preset second threshold is 0.01; the preset third threshold is 0;

[0078] The vehicle lateral driving condition identification method and system based on coordinate system conversion provided in an embodiment of the present invention first convert the vehicle driving trajectory of the global Cartesian coordinate system into the Frenet coordinate system; calculate the identification index data of the vehicle lateral driving condition, including the absolute value of the vehicle trajectory slope based on the Frenet coordinate system, the absolute value of the vehicle trajectory slope based on the global Cartesian coordinate system, and the direction factor; based on the calculated condition identification index data, determine the vehicle lateral driving process to which the vehicle driving condition to be identified belongs, including lane keeping driving condition and lane changing driving condition; further, the lane keeping driving condition includes a straight driving condition and a curved driving condition; further, the lane changing driving condition includes a left lane changing driving condition and a right lane changing driving condition, and the curved driving condition includes a left curved driving condition and a right curved driving condition. The present invention also provides a vehicle lateral driving condition identification system based on coordinate system conversion, including: a data input module, a data processing calculation module, a condition identification index calculation module, and a lateral driving condition identification module; the lateral driving condition identification module includes a lane keeping driving condition identification module and a driving direction identification module. Since the present invention uses two vehicle trajectory slopes of the global Cartesian coordinate system and the Frenet coordinate system for lateral condition identification, it effectively avoids the influence of vehicle speed changes on the condition identification index, has stronger applicability, and is more accurate, direct and concise. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0080] Figure 1 A schematic flow chart of a method for identifying a vehicle lateral driving condition based on coordinate system transformation according to an embodiment of the present invention;

[0081] Figure 2 This is a specific flow chart for identifying vehicle lateral driving conditions.

[0082] Figure 3 The figure is a specific flow chart for converting the vehicle trajectory in the global Cartesian coordinate system to the Frenet coordinate system.

[0083] Figure 4 It is a schematic diagram of the vehicle's driving trajectory in the Frenet coordinate system;

[0084] Figure 5 The present invention is a block diagram of a vehicle lateral driving condition recognition system based on coordinate system transformation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0085] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0086] To further illustrate the technical problems, technical solutions and advantages to be solved by the present invention, a detailed description is given below in conjunction with the accompanying drawings and specific embodiments.

[0087] Figure 1 FIG. 1 is a flow chart of a method for identifying a vehicle lateral driving condition based on coordinate system transformation according to an embodiment of the present invention. Figure 1 As shown, the method for identifying the lateral driving condition of a vehicle based on coordinate system transformation provided by an embodiment of the present invention may include the following steps:

[0088] Convert the vehicle trajectory in the global Cartesian coordinate system to the Frenet coordinate system;

[0089] Calculating identification index data of the vehicle lateral driving condition, wherein the condition identification index data includes an absolute value of a vehicle trajectory slope based on a Frenet coordinate system and an absolute value of a vehicle trajectory slope based on a global Cartesian coordinate system;

[0090] Identify the driving process of the vehicle to be identified according to the set vehicle lateral driving condition identification method;

[0091] Based on the calculated working condition identification index data, determining the vehicle lateral driving condition to which the vehicle driving condition to be identified belongs, wherein the vehicle lateral driving condition includes a lane keeping driving condition and a lane changing driving condition, wherein the lane keeping driving condition and the lane changing driving condition are identified based on a change in the slope of the vehicle trajectory in a Frenet coordinate system;

[0092] Based on the calculated working condition identification index data, further determining the lane keeping driving condition to which the vehicle lateral driving condition belongs, wherein the lane keeping driving condition includes a straight driving condition and a curved driving condition, wherein the straight driving condition and the curved driving condition are identified based on a change in the vehicle trajectory slope in a global Cartesian coordinate system;

[0093] Calculating identification index data of a lateral driving condition of the vehicle, wherein the identification index data of the condition includes a direction factor;

[0094] Based on the calculated working condition identification index data, further determining the lane-changing driving condition and the curve driving condition to which the vehicle lateral driving condition belongs, the lane-changing driving condition includes a left-hand lane-changing driving condition and a right-hand lane-changing driving condition, and the curve driving condition includes a left-hand curve driving condition and a right-hand curve driving condition, wherein the left-hand lane-changing driving condition and the right-hand lane-changing driving condition are identified based on a direction factor, and the left-hand curve driving condition and the right-hand curve driving condition are identified based on a direction factor;

[0095] Figure 2 The following is a specific flow chart for identifying the vehicle's lateral driving condition. Figure 2 The specific process of identifying the vehicle's lateral driving condition is explained.

[0096] In an embodiment of the present invention, identifying the driving process of the vehicle to be identified according to the set vehicle lateral driving condition identification method includes:

[0097] When the absolute value of the slope of the vehicle trajectory in the Frenet coordinate system at the first judgment moment is greater than a preset first threshold, the first judgment moment is recorded as the start moment of the vehicle lane-changing driving condition;

[0098] When the absolute value of the vehicle trajectory slope in the Frenet coordinate system at the first judgment moment is less than a preset first threshold, the first judgment moment is recorded as the start moment of the vehicle lane keeping driving condition;

[0099] When the absolute value of the slope of the vehicle trajectory in the Frenet coordinate system at the second judgment moment is less than the preset first threshold, the second judgment moment is recorded as the end moment of the vehicle lane-changing driving condition;

[0100] When the absolute value of the vehicle trajectory slope in the Frenet coordinate system at the second judgment moment is greater than a preset first threshold, the second judgment moment is recorded as the end moment of the vehicle lane keeping driving condition;

[0101] The above-mentioned identification method can be completed by a controller with calculation and control functions. The controller performs corresponding calculations and control execution operations at each judgment moment, wherein the first judgment moment is the calculation moment before the start moment of the operating condition identification, and the second judgment moment is the calculation moment after the start moment of the lateral driving condition to which the driving condition to be identified belongs.

[0102] In the actual driving process of a vehicle on a road, the vehicle constantly changes between driving along the lane line and driving across the lane line. It is very necessary to establish a coordinate system between the vehicle's driving trajectory and the lane line, namely the Frenet coordinate system, to identify the vehicle's driving condition.

[0103] The vehicle's driving trajectory is the trajectory of the vehicle's center of mass.

[0104] The trajectory slope can effectively describe the driving direction of the vehicle and help identify the lateral vehicle driving conditions. The absolute value of the vehicle trajectory slope based on the Frenet coordinate system can be determined by the following formula (1):

[0105]

[0106] Among them, α f(k) is the absolute value of the trajectory slope of the kth vehicle trajectory sampling point in the Frenet coordinate system, in units of (1), s (k) is the lateral coordinate of the kth vehicle trajectory sampling point in the Frenet coordinate system, in meters, l (k) is the longitudinal coordinate of the vehicle trajectory sampling point in the kth Frenet coordinate system, in (m);

[0107] In addition, the preset first threshold for identifying the vehicle's lateral driving condition can be determined based on the specific signal quality conditions during the actual vehicle driving process. In the example, the preset first threshold can be set to 0.021.

[0108] In a specific embodiment, the identification process of the lane changing driving condition and the lane keeping driving condition in the embodiment of the present invention can be as follows: Figure 2 As shown. Figure 2 As shown, at the beginning of each vehicle driving condition identification, the identification can be carried out through the following steps:

[0109] Determine whether the absolute value of the trajectory slope of the vehicle driving trajectory sampling point in the Frenet coordinate system at the first current moment is greater than a preset first threshold; if greater, identify it as a lane changing driving condition, and record the first current moment as the start time of the lane changing condition; if not greater, identify it as a lane keeping driving condition, and prepare for further specific condition judgment.

[0110] Continue to judge the absolute value of the trajectory slope of the vehicle driving trajectory sampling point in the Frenet coordinate system identified as the lane changing driving condition; judge whether the absolute value of the trajectory slope of the vehicle driving trajectory sampling point in the Frenet coordinate system at the second current moment is greater than the preset first threshold; if greater, identify it as still in the lane changing driving condition, and continue to judge; if not greater, identify it as the lane keeping driving condition, and record the second current moment as the end moment of the lane changing driving condition.

[0111] Before the recognition starts, the vehicle driving condition at the current moment is assumed to be a lane keeping driving condition.

[0112] When a vehicle is driving on the road, the geometric position relationship between the vehicle's driving trajectory and the lane line determines the vehicle's lateral driving condition. The Frenet coordinate system can intuitively reflect the relationship between the vehicle's driving trajectory and the lane line, and is widely used in the field of autonomous driving vehicle path planning. Therefore, starting from the essential characteristics of the vehicle's lateral driving condition, the vehicle trajectory slope based on the Frenet coordinate system is more concise and effective for identifying lane change driving conditions and lane keeping driving conditions.

[0113] The vehicle trajectory in the lane keeping condition is almost parallel to the surrounding lane lines, and the corresponding absolute value of the slope of the vehicle trajectory in the Frenet coordinate system is within a small range. The vehicle trajectory in the lane changing condition is intersecting with the surrounding lane lines, and the corresponding absolute value of the slope of the vehicle trajectory in the Frenet coordinate system is within a large range.

[0114] Next, a further identification method for the lane keeping driving condition is introduced, combined with Figure 2 The specific process of identification is described.

[0115] In the example of the present invention, the identification of the lane keeping driving condition mainly includes the identification of the straight-line driving condition and the identification of the curved driving condition.

[0116] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the first judgment moment is greater than a preset second threshold, the first judgment moment is recorded as the start moment of the vehicle curve driving condition;

[0117] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the first judgment moment is less than a preset second threshold, the first judgment moment is recorded as the start moment of the vehicle straight-line driving condition;

[0118] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the second judgment moment is less than a preset second threshold, the second judgment moment is recorded as the end moment of the vehicle curve driving condition;

[0119] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the second judgment moment is greater than a preset second threshold, the second judgment moment is recorded as the end moment of the vehicle straight-line driving condition;

[0120] The absolute value of the vehicle trajectory slope based on the global Cartesian coordinate system can be calculated using the following formula (2):

[0121]

[0122] Among them, α c(k)is the absolute value of the trajectory slope of the vehicle trajectory sampling point in the kth global Cartesian coordinate system, in units of (1), x t(k) is the lateral coordinate of the vehicle trajectory sampling point in the kth global Cartesian coordinate system, in meters, y t(k) is the longitudinal coordinate of the vehicle trajectory sampling point in the kth global Cartesian coordinate system, in (m);

[0123] In addition, the preset second threshold for identifying the vehicle's lateral driving condition can be determined based on the specific signal quality conditions during the actual vehicle driving process. In the example, the preset second threshold can be set to 0.01.

[0124] In a specific embodiment, the identification process of the straight-line driving condition and the curved driving condition of the embodiment of the present invention can be as follows: Figure 2 As shown. Identification is done by following the steps below:

[0125] Determine whether the absolute value of the trajectory slope of the vehicle driving trajectory sampling point in the global Cartesian coordinate system at the first current moment is greater than a preset second threshold; if greater, identify it as a curved driving condition, and record the first current moment as the starting moment of the curved driving condition; if not greater, identify it as a straight driving condition, and record the first current moment as the starting moment of the straight driving condition.

[0126] Continue to judge the absolute value of the trajectory slope of the vehicle driving trajectory sampling point in the global Cartesian coordinate system identified as the curved driving condition; judge whether the absolute value of the trajectory slope of the vehicle driving trajectory sampling point in the global Cartesian coordinate system at the second current moment is greater than the preset first threshold; if greater, identify it as still in the curved driving condition, and continue to judge; if not greater, identify it as a straight driving condition, and record the second current moment as the end moment of the curved driving condition.

[0127] Continue to judge the absolute value of the trajectory slope of the vehicle driving trajectory sampling point in the global Cartesian coordinate system identified as a straight-line driving condition; judge whether the absolute value of the trajectory slope of the vehicle driving trajectory sampling point in the global Cartesian coordinate system at the second current moment is greater than the preset first threshold; if greater, identify it as a curve driving condition, and record the second current moment as the end time of the straight-line driving condition; if not greater, identify it as still in the straight-line driving condition, and continue to judge.

[0128] The slope of the vehicle trajectory in the global Cartesian coordinate system reflects the driving direction of the vehicle in the geodetic coordinate system. The vehicle trajectory in a straight-line driving condition is almost parallel to the transverse coordinate axis of the geodetic coordinate system, and the corresponding absolute value of the slope of the vehicle trajectory in the global Cartesian coordinate system is within a small range. The vehicle trajectory in a curved driving condition intersects with the transverse coordinate axis of the geodetic coordinate system, and the corresponding absolute value of the slope of the vehicle trajectory in the global Cartesian coordinate system is within a large range.

[0129] Next, the identification method of the lane change driving condition and the curve driving condition is introduced. Figure 2 The specific process of identification is described.

[0130] In the example of the present invention, the identification of the lane-changing driving condition mainly includes the identification of the left lane-changing driving condition and the identification of the right lane-changing driving condition.

[0131] When the direction factor at the average judgment time is greater than the preset third threshold, the lane-changing driving condition belongs to the left lane-changing driving condition; when the direction factor at the average judgment time is less than the preset third threshold, the lane-changing driving condition belongs to the right lane-changing driving condition;

[0132] When the direction factor at the average judgment time is greater than the preset third threshold, the curve driving condition belongs to the left curve driving condition; when the direction factor at the average judgment time is less than the preset third threshold, the curve driving condition belongs to the right curve driving condition;

[0133] The direction of the lateral vehicle driving condition is identified by the direction factor, and the direction factor can be calculated by the following formula (3):

[0134] d (k) =sign((y t(k) -y r(h) )·cos(θ r(k) )-(x t(k) -x r(h) )·sin(θ r(k) )) (3)

[0135] Among them, d (k) is the direction factor of the kth vehicle trajectory sampling point in the Frenet coordinate system, in units of (1), x t(k) is the lateral coordinate of the kth Frenet coordinate system vehicle trajectory sampling point under the global Cartesian, in (m), y t(k) x is the longitudinal coordinate of the kth vehicle trajectory sampling point in the Frenet coordinate system under the global Cartesian, in meters; r(h)is the lateral coordinate of the corresponding reference line projection point of the vehicle driving trajectory sampling point in the kth Frenet coordinate system under the global Cartesian, in (m), y r(h) is the longitudinal coordinate of the corresponding reference projection point of the vehicle driving trajectory sampling point in the kth Frenet coordinate system under the global Cartesian, in meters; θ r(h) is the heading angle of the corresponding reference projection point of the vehicle trajectory sampling point in the kth Frenet coordinate system, in degrees;

[0136] The calculation of the heading angle of the corresponding reference projection point of the vehicle trajectory sampling point in the Frenet coordinate system can be determined by the following formula (4):

[0137]

[0138] Among them, θ r(h) is the heading angle of the vehicle driving trajectory sampling point of the kth Frenet coordinate system, that is, the vehicle driving trajectory reference line sampling point of the hth global Cartesian coordinate system, in degrees, x r(h) is the lateral coordinate of the sampling point of the vehicle trajectory reference line in the hth global Cartesian coordinate system, in meters, y r(h) is the longitudinal coordinate of the sampling point of the vehicle trajectory reference line in the hth global Cartesian coordinate system, in (m);

[0139] The average judgment time is the average of the first judgment time and the second judgment time;

[0140] In addition, the preset second threshold for identifying the vehicle's lateral driving condition can be determined based on the specific signal quality conditions during the actual vehicle driving process. In the example, the preset third threshold can be set to 0.

[0141] In a specific embodiment, the identification process of the lane change driving condition and the curve driving condition in the embodiment of the present invention can be as follows: Figure 2 As shown. It can be identified by the following steps:

[0142] For a lane-changing driving condition whose start and end times have been determined, an average judgment time is calculated based on the start and end times of the lane-changing driving condition, that is, the average time of the start and end times; it is determined whether the direction factor of the average judgment time is greater than a preset third threshold; if so, it is identified as a left-hand lane-changing driving condition; if not, it is identified as a right-hand lane-changing driving condition.

[0143] For a curved driving condition whose start and end times have been determined, calculate the average judgment time according to the start and end times of the curved driving condition, that is, the average time of the start and end times; judge whether the direction factor of the average judgment time is greater than a preset third threshold; if greater, identify it as a left-hand curved driving condition; if not greater, identify it as a right-hand curved driving condition.

[0144] Next, a method for smoothing the reference line sampling points of the vehicle driving trajectory and the sampling points of the vehicle driving trajectory is introduced.

[0145] The smoothing method used is the pulse interference-proof average filtering algorithm. The specific calculation steps are as follows:

[0146] Select a certain number of sampling points to be processed;

[0147] For the selected sampling point set, sort them according to the value size;

[0148] Remove the largest sampling point and the smallest sampling point;

[0149] Calculate the arithmetic mean of the remaining sampling point sets;

[0150] The arithmetic mean is used as the output value after filtering.

[0151] Next, the specific process of converting the vehicle trajectory in the Cartesian coordinate system to the Frenet coordinate system is introduced, such as Figure 3 shown. Figure 3 The following is a specific flow chart for converting the vehicle trajectory in the global Cartesian coordinate system to the Frenet coordinate system. Taking the conversion of the sampling point of the k-th vehicle trajectory from the global Cartesian coordinate system to the Frenet coordinate system as an example, the calculation process is specifically introduced. The main steps include:

[0152] Input the lateral coordinate x of the sampling point set of the vehicle trajectory reference line in the global Cartesian coordinate system r(k) and the vertical coordinate y r(k) ; Input the lateral coordinate x of the vehicle trajectory sampling point set in the global Cartesian coordinate system t(k) and the vertical coordinate y t(k) ; Calculate the heading angle θ of the data point set of the vehicle's trajectory reference line in the global Cartesian coordinate system r(k) ; Calculate the vehicle trajectory data points (x r(k) ,y r(k) ) to the data point set (x) adjacent to the reference line r(k-m) ,y r(k-m) ),...,(x r(k-1) ,y r(k-1) ),(xr(k) ,y r(k) ),(x r(k+1) ,y r(k+1) ),...(x r(k+m) ,y r(k+m) ) of the Euclidean distance; find the data point of the vehicle's driving trajectory reference line with the closest Euclidean distance as the projection point (x r(h) ,y r(h) ); Set the path value ss of the projection point (h) As the horizontal coordinate s of the vehicle's driving trajectory point in the Frenet coordinate system (k) ; The Euclidean distance between the projection point and the vehicle's driving trajectory point is taken as the absolute value of the ordinate of the vehicle's driving trajectory point in the Frenet coordinate system|l (k) |; According to the horizontal and vertical coordinates (x t(k) ,y t(k) ), the horizontal and vertical coordinates of the corresponding projection point (x r(h) ,y r(h) ) and the heading angle θ of the projection point r(h) Determine the direction factor and finally obtain the longitudinal coordinates of the vehicle's driving trajectory points in the Frenet coordinate system.

[0153] The reference line of the vehicle's driving trajectory is the center line of the lane, and the sampling frequency of the reference line sampling points of the vehicle's driving trajectory is selected based on hardware conditions.

[0154] The number of data sets for the vehicle driving trajectory reference lines is selected based on the hardware conditions of the controller. The better the controller performance, the more reference line data sets corresponding to the same vehicle driving trajectory sampling point. Among them, 2m is the number of reference line data sets in this example.

[0155] Among them, h is the sampling sequence of the reference line projection points corresponding to the sampling points of the k-th vehicle driving trajectory, that is, the h-th sampling point in the selected reference line data set is closest to the sampling point of the k-th vehicle driving trajectory.

[0156] Figure 4 Schematic diagram of vehicle driving trajectory in Frenet coordinate system.

[0157] The path value of the projection point is used as the horizontal coordinate of the vehicle's driving trajectory point in the Frenet coordinate system. The calculation of the path value of the projection point can be determined by the following formula (5):

[0158]

[0159] Among them, ss (h) is the path value of the vehicle trajectory reference line sampling point in the hth global Cartesian coordinate system, in meters, x r(h)is the lateral coordinate of the sampling point of the vehicle trajectory reference line in the hth global Cartesian coordinate system, in meters, y r(h) is the longitudinal coordinate of the sampling point of the vehicle trajectory reference line in the hth global Cartesian coordinate system, in meters.

[0160] The longitudinal coordinate in the Frenet coordinate system includes the Euclidean distance between the projection point and the vehicle's driving trajectory point and the direction factor. The calculation of the Euclidean distance can be determined by the following formula (6):

[0161]

[0162] Among them, |l (k) | is the Euclidean distance between the projection point and the kth vehicle trajectory point, in meters, x t(k) is the lateral coordinate of the kth vehicle trajectory sampling point in the global Cartesian coordinate system, in meters, y t(k) is the lateral coordinate of the kth vehicle trajectory sampling point in the global Cartesian coordinate system, in meters, x r(h) is the lateral coordinate of the hth vehicle trajectory reference line sampling point in the global Cartesian coordinate system, in meters, y r(h) is the longitudinal coordinate of the hth vehicle trajectory reference line sampling point in the global Cartesian coordinate system, in meters.

[0163] The calculation of the direction factor of the vehicle driving trajectory sampling point can be determined by the following formula (7):

[0164] d (k) =sign((y t(k) -y r(h) )·cos(θ r(k) )-(x t(k) -x r(h) )·sin(θ r(k) )) (7)

[0165] Among them, d (k) is the direction factor of the kth vehicle trajectory sampling point in the Frenet coordinate system, in units of (1), x t(k) is the lateral coordinate of the kth Frenet coordinate system vehicle trajectory sampling point under the global Cartesian, in (m), y t(k) x is the longitudinal coordinate of the kth vehicle trajectory sampling point in the Frenet coordinate system under the global Cartesian, in meters; r(h)is the lateral coordinate of the corresponding reference line projection point of the vehicle driving trajectory sampling point in the kth Frenet coordinate system under the global Cartesian, in (m), y r(h) is the longitudinal coordinate of the corresponding reference projection point of the vehicle trajectory sampling point in the kth Frenet coordinate system under the global Cartesian, in meters; θ r(h) is the heading angle of the corresponding reference projection point of the vehicle trajectory sampling point in the kth Frenet coordinate system, in degrees.

[0166] Therefore, the calculation of the longitudinal coordinate in the Frenet coordinate system can be determined by the following formula (8):

[0167] l (k) =d (k) ·|l (k) | (8)

[0168] In summary, the vehicle lateral driving condition identification method based on coordinate system conversion provided by the embodiment of the present invention first converts the vehicle driving trajectory of the global Cartesian coordinate system to the Frenet coordinate system; calculates the identification index data of the vehicle lateral driving condition, including the absolute value of the vehicle trajectory slope based on the Frenet coordinate system, the absolute value of the vehicle trajectory slope based on the global Cartesian coordinate system and the direction factor; based on the calculated condition identification index data, determines the vehicle lateral driving process to which the vehicle driving condition to be identified belongs, including lane keeping driving condition and lane changing driving condition; further, the lane keeping driving condition includes straight driving condition and curved driving condition; further, the lane changing driving condition includes left lane changing driving condition and right lane changing driving condition, and the curved driving condition includes left curved driving condition and right curved driving condition. Since the two vehicle trajectory slopes of the global Cartesian coordinate system and the Frenet coordinate system are used for lateral condition identification, the influence of vehicle speed change on the condition identification index is effectively avoided, and it has stronger applicability, and is more accurate, direct and concise.

[0169] Figure 5 is a structural block diagram of a vehicle lateral driving condition recognition system based on coordinate system transformation according to an embodiment of the present invention. Based on the same inventive concept, an embodiment of the present invention further provides a vehicle lateral driving condition recognition system based on coordinate system transformation, such as Figure 5 As shown:

[0170] A data input module is used to obtain vehicle driving trajectory data in a global Cartesian coordinate system and reference line data of the vehicle driving trajectory in the global Cartesian coordinate system; the vehicle driving trajectory data in the global Cartesian coordinate system is acquired by a high-precision positioning system; the reference line information of the vehicle driving trajectory in the global Cartesian coordinate system is acquired by a high-precision vehicle-mounted camera;

[0171] A data processing and calculation module is used to filter and smooth the direct data obtained by the data input module and convert the vehicle driving trajectory of the global Cartesian coordinate system into the Frenet coordinate system;

[0172] A working condition identification index calculation module, used to calculate identification index data of the vehicle lateral driving working condition, wherein the working condition identification index data includes the absolute value of the vehicle trajectory slope based on the Frenet coordinate system, the absolute value of the vehicle trajectory slope based on the global Cartesian coordinate system, and a direction factor;

[0173] A lateral driving condition identification module, used to determine the vehicle lateral driving condition to which the vehicle driving condition to be identified belongs based on the calculated condition identification index data, wherein the vehicle lateral driving condition includes a lane keeping driving condition and a lane changing driving condition, wherein the lane keeping driving condition and the lane changing driving condition are identified based on the vehicle trajectory slope change in the Frenet coordinate system;

[0174] A lane keeping driving condition identification module, for further determining the lane keeping driving condition to which the vehicle lateral driving condition belongs based on the calculated condition identification index data, wherein the lane keeping driving condition includes a straight driving condition and a curved driving condition, wherein the straight driving condition and the curved driving condition are identified based on a change in the vehicle trajectory slope in a global Cartesian coordinate system;

[0175] A driving direction identification module, used for further determining the lane-changing driving condition and the curve driving condition to which the vehicle lateral driving condition belongs based on the calculated condition identification index data, wherein the lane-changing driving condition includes a left lane-changing driving condition and a right lane-changing driving condition, and the curve driving condition includes a left curve driving condition and a right curve driving condition, wherein the left lane-changing driving condition and the right lane-changing driving condition are identified based on a direction factor, and the left curve driving condition and the right curve driving condition are identified based on a direction factor;

[0176] Furthermore, the lateral driving condition recognition module recognizes the vehicle driving process specifically through the following steps:

[0177] When the absolute value of the slope of the vehicle trajectory in the Frenet coordinate system at the first judgment moment is greater than a preset first threshold, the first judgment moment is recorded as the start moment of the vehicle lane-changing driving condition;

[0178] When the absolute value of the vehicle trajectory slope in the Frenet coordinate system at the first judgment moment is less than a preset first threshold, the first judgment moment is recorded as the start moment of the vehicle lane keeping driving condition;

[0179] When the absolute value of the slope of the vehicle trajectory in the Frenet coordinate system at the second judgment moment is less than the preset first threshold, the second judgment moment is recorded as the end moment of the vehicle lane-changing driving condition;

[0180] When the absolute value of the vehicle trajectory slope in the Frenet coordinate system at the second judgment moment is greater than a preset first threshold, the second judgment moment is recorded as the end moment of the vehicle lane keeping driving condition;

[0181] Furthermore, the specific steps of the lane keeping driving condition recognition module recognizing the lane keeping driving condition to which the vehicle lateral driving condition belongs include:

[0182] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the first judgment moment is greater than a preset second threshold, the first judgment moment is recorded as the start moment of the vehicle curve driving condition;

[0183] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the first judgment moment is less than a preset second threshold, the first judgment moment is recorded as the start moment of the vehicle straight-line driving condition;

[0184] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the second judgment moment is less than a preset second threshold, the second judgment moment is recorded as the end moment of the vehicle curve driving condition;

[0185] When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the second judgment moment is greater than a preset second threshold, the second judgment moment is recorded as the end moment of the vehicle straight-line driving condition;

[0186] Furthermore, the specific steps of the driving direction recognition module for recognizing the curve driving condition and the lane change driving condition to which the vehicle lateral driving condition belongs include:

[0187] When the direction factor at the average judgment time is greater than the preset third threshold, the lane-changing driving condition belongs to the left lane-changing driving condition; when the direction factor at the average judgment time is less than the preset third threshold, the lane-changing driving condition belongs to the right lane-changing driving condition;

[0188] When the direction factor at the average judgment time is greater than the preset third threshold, the curve driving condition belongs to the left curve driving condition; when the direction factor at the average judgment time is less than the preset third threshold, the curve driving condition belongs to the right curve driving condition;

[0189] The average judgment time is the average of the first judgment time and the second judgment time.

[0190] Furthermore, the method for filtering and smoothing the direct data in the data processing and calculation module is an anti-pulse interference average value filtering algorithm.

[0191] Furthermore, in the data processing and calculation module, the specific steps of converting the vehicle driving trajectory of the global Cartesian coordinate system to the Frenet coordinate system include:

[0192] Obtaining the lateral coordinates and longitudinal coordinates of the reference line sampling points of the vehicle driving trajectory in the global Cartesian coordinate system; the reference line of the vehicle driving trajectory is the center line of the lane; the sampling frequency of the reference line sampling points of the vehicle driving trajectory is selected according to hardware conditions;

[0193] Obtaining the horizontal and vertical coordinates of the sampling points of the vehicle driving trajectory in the global Cartesian coordinate system; the vehicle driving trajectory is the driving trajectory of the vehicle's center of mass; the sampling frequency of the reference line sampling points of the vehicle driving trajectory is selected according to hardware conditions;

[0194] Calculate the distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point set of the adjacent vehicle driving trajectory; the distance is the absolute value of the Euclidean distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point of the adjacent vehicle driving trajectory; the reference line sampling point set of the adjacent vehicle driving trajectory is a set of a set number of sampling points before and after the corresponding sampling time of the selected vehicle driving trajectory reference line sampling point;

[0195] Find the vehicle driving trajectory reference line sampling point with the shortest distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point set of the adjacent vehicle driving trajectory; the vehicle driving trajectory reference line sampling point with the shortest distance is used as the projection point of the global Cartesian coordinate system converted to the Frenet coordinate system;

[0196] The path value of the reference line sampling point of the vehicle driving trajectory corresponding to the projection point of the reference line of the vehicle driving trajectory is used as the horizontal coordinate value of the Frenet coordinate system. The distance from the reference line sampling point of the vehicle driving trajectory to the projection point of the corresponding vehicle driving trajectory reference line is used as the vertical coordinate value of the Frenet coordinate system.

[0197] Further, the preset first threshold is 0.021; the preset second threshold is 0.01; the preset third threshold is 0;

[0198] The functions of the above modules can correspond to Figures 1 to 5 The corresponding processing steps in the shown process will not be repeated here.

[0199] The above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, it should be understood by those skilled in the art that any person familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or perform equivalent replacements on some of the technical features thereof; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

[0200] The embodiments described above are only descriptions of the preferred modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for identifying vehicle lateral driving conditions based on coordinate system transformation. It is characterized in that include: Convert the vehicle trajectory in the global Cartesian coordinate system to the Frenet coordinate system; Calculating identification index data of the vehicle lateral driving condition, wherein the condition identification index data includes an absolute value of a vehicle trajectory slope based on a Frenet coordinate system and an absolute value of a vehicle trajectory slope based on a global Cartesian coordinate system; Identify the driving process of the vehicle to be identified according to the set vehicle lateral driving condition identification method; Based on the calculated working condition identification index data, determining the vehicle lateral driving condition to which the vehicle driving condition to be identified belongs, wherein the vehicle lateral driving condition includes a lane keeping driving condition and a lane changing driving condition, wherein the lane keeping driving condition and the lane changing driving condition are identified based on a change in the slope of the vehicle trajectory in a Frenet coordinate system; Based on the calculated working condition identification index data, further determining the lane keeping driving condition to which the vehicle lateral driving condition belongs, wherein the lane keeping driving condition includes a straight driving condition and a curved driving condition, wherein the straight driving condition and the curved driving condition are identified based on a change in the vehicle trajectory slope in a global Cartesian coordinate system; Calculating identification index data of a lateral driving condition of the vehicle, wherein the identification index data of the condition includes a direction factor; Based on the calculated working condition identification index data, further determining the lane-changing driving condition and the curve driving condition to which the vehicle lateral driving condition belongs, the lane-changing driving condition includes a left-hand lane-changing driving condition and a right-hand lane-changing driving condition, and the curve driving condition includes a left-hand curve driving condition and a right-hand curve driving condition, wherein the left-hand lane-changing driving condition and the right-hand lane-changing driving condition are identified based on a direction factor, and the left-hand curve driving condition and the right-hand curve driving condition are identified based on a direction factor; Wherein, the identifying of the driving process of the vehicle to be identified according to the set vehicle lateral driving condition identification method includes: When the absolute value of the slope of the vehicle trajectory in the Frenet coordinate system at the first judgment moment is greater than a preset first threshold, the first judgment moment is recorded as the start moment of the vehicle lane-changing driving condition; When the absolute value of the vehicle trajectory slope in the Frenet coordinate system at the first judgment moment is less than a preset first threshold, the first judgment moment is recorded as the start moment of the vehicle lane keeping driving condition; When the absolute value of the slope of the vehicle trajectory in the Frenet coordinate system at the second judgment moment is less than the preset first threshold, the second judgment moment is recorded as the end moment of the vehicle lane-changing driving condition; When the absolute value of the vehicle trajectory slope in the Frenet coordinate system at the second judgment moment is greater than a preset first threshold, the second judgment moment is recorded as the end moment of the vehicle lane keeping driving condition; The calculation formula of the absolute value of the vehicle trajectory slope based on the Frenet coordinate system is: Among them, α f(k) is the absolute value of the trajectory slope of the kth vehicle trajectory sampling point in the Frenet coordinate system, in units of (1), s (k) is the lateral coordinate of the kth vehicle trajectory sampling point in the Frenet coordinate system, in meters, l (k) is the longitudinal coordinate of the vehicle trajectory sampling point in the kth Frenet coordinate system, in (m); Wherein, the method for further determining the lane keeping driving condition to which the vehicle lateral driving condition belongs based on the calculated working condition identification index data includes: When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the first judgment moment is greater than a preset second threshold, the first judgment moment is recorded as the start moment of the vehicle curve driving condition; When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the first judgment moment is less than a preset second threshold, the first judgment moment is recorded as the start moment of the vehicle straight-line driving condition; When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the second judgment moment is less than a preset second threshold, the second judgment moment is recorded as the end moment of the vehicle curve driving condition; When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the second judgment moment is greater than a preset second threshold, the second judgment moment is recorded as the end moment of the vehicle straight-line driving condition; The calculation formula of the absolute value of the vehicle trajectory slope based on the global Cartesian coordinate system is: Among them, α c(k) is the absolute value of the trajectory slope of the vehicle trajectory sampling point in the kth global Cartesian coordinate system, in units of (1), x t(k) is the lateral coordinate of the vehicle trajectory sampling point in the kth global Cartesian coordinate system, in meters, y t(k) is the longitudinal coordinate of the vehicle trajectory sampling point in the kth global Cartesian coordinate system, in (m); When the direction factor at the average judgment time is greater than the preset third threshold, the lane-changing driving condition belongs to the left lane-changing driving condition; when the direction factor at the average judgment time is less than the preset third threshold, the lane-changing driving condition belongs to the right lane-changing driving condition; When the direction factor at the average judgment time is greater than the preset third threshold, the curve driving condition belongs to the left curve driving condition; when the direction factor at the average judgment time is less than the preset third threshold, the curve driving condition belongs to the right curve driving condition; The calculation formula of the direction factor is: d (k) =sign((and t(k) -and r(h) )·cos(θ r(k) )-(x t(k) -x r(h) )·sin(θ r(k) )) Among them, d (k) is the direction factor of the kth vehicle trajectory sampling point in the Frenet coordinate system, in units of (1), x t(k) is the lateral coordinate of the kth Frenet coordinate system vehicle trajectory sampling point under the global Cartesian, in (m), y t(k) x is the longitudinal coordinate of the kth vehicle trajectory sampling point in the Frenet coordinate system under the global Cartesian, in meters; r(h) is the lateral coordinate of the corresponding reference line projection point of the vehicle driving trajectory sampling point in the kth Frenet coordinate system under the global Cartesian, in (m), y r(h) is the longitudinal coordinate of the corresponding reference projection point of the vehicle trajectory sampling point in the kth Frenet coordinate system under the global Cartesian, in meters; θ r(k) is the vehicle heading angle of the corresponding reference projection point of the vehicle trajectory sampling point in the kth Frenet coordinate system, in degrees; The average judgment time is the average value of the first judgment time and the second judgment time.

2. The vehicle lateral driving condition identification method based on coordinate system transformation according to claim 1, It is characterized in that The vehicle trajectory in the Cartesian coordinate system is converted to the Frenet coordinate system, including: Obtaining the lateral coordinates and the longitudinal coordinates of the reference line sampling points of the vehicle driving trajectory in the global Cartesian coordinate system; the reference line of the vehicle driving trajectory is the center line of the lane; the sampling frequency of the reference line sampling points of the vehicle driving trajectory is selected according to the hardware conditions; Obtaining the horizontal and vertical coordinates of the sampling points of the vehicle driving trajectory in the global Cartesian coordinate system; the vehicle driving trajectory is the driving trajectory of the vehicle's center of mass; the sampling frequency of the reference line sampling points of the vehicle driving trajectory is selected according to hardware conditions; Calculate the heading angle of the reference line sampling point of the vehicle's driving trajectory; the calculation formula for the heading angle of the reference line sampling point of the vehicle's driving trajectory is: Among them, θ r(k) is the heading angle of the reference line sampling point of the kth vehicle trajectory, in degrees, x r(k) is the lateral coordinate of the reference line sampling point of the kth vehicle trajectory in the global Cartesian coordinate system, in meters, y r(k) is the longitudinal coordinate of the reference line sampling point of the kth vehicle trajectory in the global Cartesian coordinate system, in (m); Calculate the distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point set of the adjacent vehicle driving trajectory; the distance is the Euclidean distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point of the adjacent vehicle driving trajectory; the reference line sampling point set of the adjacent vehicle driving trajectory is a set of a set number of sampling points before and after the corresponding sampling moment of the selected vehicle driving trajectory reference line sampling point; Find the vehicle driving trajectory reference line sampling point with the shortest distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling points of the adjacent vehicle driving trajectories; the vehicle driving trajectory reference line sampling point with the shortest distance is used as the projection point of the global Cartesian coordinate system converted to the Frenet coordinate system; The path value of the reference line projection point of the vehicle driving trajectory corresponding to the reference line sampling point of the vehicle driving trajectory is used as the transverse coordinate of the Frenet coordinate system. The transverse coordinate of the Frenet coordinate system is: s (k) =ss (k) Among them, s (k) is the lateral coordinate of the kth vehicle trajectory sampling point in the Frenet coordinate system, in meters, ss (k) is the path value of the projection point of the vehicle driving trajectory reference line corresponding to the kth vehicle driving trajectory sampling point, in (m); The Euclidean distance from the reference line sampling point of the vehicle driving trajectory to the projection point of the corresponding vehicle driving trajectory reference line is used as the longitudinal coordinate value of the Frenet coordinate system. The longitudinal coordinate value and the direction factor constitute the longitudinal coordinate of the Frenet coordinate system. The longitudinal coordinate of the Frenet coordinate system is: Among them, l (k) is the longitudinal coordinate of the kth vehicle trajectory sampling point in the Frenet coordinate system, in meters, d (k) is the direction factor of the kth vehicle trajectory sampling point in the Frenet coordinate system, in units of (1), x t(k) is the lateral coordinate of the kth vehicle trajectory sampling point in the global Cartesian coordinate system, in meters, y t(k) is the lateral coordinate of the kth vehicle trajectory sampling point in the global Cartesian coordinate system, in meters, x r(h) is the lateral coordinate of the hth vehicle trajectory reference line sampling point in the global Cartesian coordinate system, in meters, y r(h) is the longitudinal coordinate of the hth vehicle trajectory reference line sampling point in the global Cartesian coordinate system, in (m); Wherein, h is the sampling sequence number of the projection point of the vehicle driving trajectory reference line corresponding to the kth vehicle driving trajectory sampling point.

3. The vehicle lateral driving condition identification method based on coordinate system transformation according to claim 1, It is characterized in that The preset first threshold is 0.021; the preset second threshold is 0.01; and the preset third threshold is 0.

4. A vehicle lateral driving condition recognition system based on coordinate system transformation, It is characterized in that include: A data input module is used to obtain vehicle driving trajectory data in a global Cartesian coordinate system and reference line data of the vehicle driving trajectory in the global Cartesian coordinate system; the vehicle driving trajectory data in the global Cartesian coordinate system is collected and obtained by a high-precision positioning system; The data of the vehicle driving trajectory reference line in the global Cartesian coordinate system is acquired by collecting the high-precision vehicle-mounted camera; A data processing and calculation module is used to filter and smooth the direct data obtained by the data input module and convert the vehicle driving trajectory of the global Cartesian coordinate system into the Frenet coordinate system; A working condition identification index calculation module, used to calculate identification index data of the vehicle lateral driving working condition, wherein the working condition identification index data includes the absolute value of the vehicle trajectory slope based on the Frenet coordinate system, the absolute value of the vehicle trajectory slope based on the global Cartesian coordinate system, and a direction factor; A lateral driving condition identification module, used to determine the vehicle lateral driving condition to which the vehicle driving condition to be identified belongs based on the calculated condition identification index data, wherein the vehicle lateral driving condition includes a lane keeping driving condition and a lane changing driving condition, wherein the lane keeping driving condition and the lane changing driving condition are identified based on the vehicle trajectory slope change in the Frenet coordinate system; A lane keeping driving condition identification module, for further determining the lane keeping driving condition to which the vehicle lateral driving condition belongs based on the calculated condition identification index data, wherein the lane keeping driving condition includes a straight driving condition and a curved driving condition, wherein the straight driving condition and the curved driving condition are identified based on a change in the vehicle trajectory slope in a global Cartesian coordinate system; A driving direction identification module, used for further determining the lane-changing driving condition and the curve driving condition to which the vehicle lateral driving condition belongs based on the calculated condition identification index data, wherein the lane-changing driving condition includes a left lane-changing driving condition and a right lane-changing driving condition, and the curve driving condition includes a left curve driving condition and a right curve driving condition, wherein the left lane-changing driving condition and the right lane-changing driving condition are identified based on a direction factor, and the left curve driving condition and the right curve driving condition are identified based on a direction factor; The lateral driving condition recognition module recognizes the vehicle driving process including: When the absolute value of the slope of the vehicle trajectory in the Frenet coordinate system at the first judgment moment is greater than a preset first threshold, the first judgment moment is recorded as the start moment of the vehicle lane-changing driving condition; When the absolute value of the vehicle trajectory slope in the Frenet coordinate system at the first judgment moment is less than a preset first threshold, the first judgment moment is recorded as the start moment of the vehicle lane keeping driving condition; When the absolute value of the slope of the vehicle trajectory in the Frenet coordinate system at the second judgment moment is less than the preset first threshold, the second judgment moment is recorded as the end moment of the vehicle lane-changing driving condition; When the absolute value of the vehicle trajectory slope in the Frenet coordinate system at the second judgment moment is greater than a preset first threshold, the second judgment moment is recorded as the end moment of the vehicle lane keeping driving condition; The specific steps of the lane keeping driving condition recognition module recognizing the lane keeping driving condition to which the vehicle lateral driving condition belongs include: When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the first judgment moment is greater than a preset second threshold, the first judgment moment is recorded as the start moment of the vehicle curve driving condition; When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the first judgment moment is less than a preset second threshold, the first judgment moment is recorded as the start moment of the vehicle straight-line driving condition; When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the second judgment moment is less than a preset second threshold, the second judgment moment is recorded as the end moment of the vehicle curve driving condition; When the absolute value of the vehicle trajectory slope in the global Cartesian coordinate system at the second judgment moment is greater than a preset second threshold, the second judgment moment is recorded as the end moment of the vehicle straight-line driving condition; The specific steps of the driving direction recognition module for recognizing the curve driving condition and the lane change driving condition to which the vehicle lateral driving condition belongs include: When the direction factor at the average judgment time is greater than the preset third threshold, the lane-changing driving condition belongs to the left lane-changing driving condition; when the direction factor at the average judgment time is less than the preset third threshold, the lane-changing driving condition belongs to the right lane-changing driving condition; When the direction factor at the average judgment time is greater than the preset third threshold, the curve driving condition belongs to the left curve driving condition; when the direction factor at the average judgment time is less than the preset third threshold, the curve driving condition belongs to the right curve driving condition; The calculation process of the working condition identification index calculation module includes the calculation of the absolute value of the vehicle trajectory slope based on the Frenet coordinate system, the absolute value of the vehicle trajectory slope based on the global Cartesian coordinate system, and the direction factor, specifically including: The calculation formula of the absolute value of the vehicle trajectory slope based on the Frenet coordinate system is: Among them, α f(k) is the absolute value of the trajectory slope of the kth vehicle trajectory sampling point in the Frenet coordinate system, in units of (1), s (k) is the lateral coordinate of the kth vehicle trajectory sampling point in the Frenet coordinate system, in meters, l (k) is the longitudinal coordinate of the vehicle trajectory sampling point in the kth Frenet coordinate system, in (m); The calculation formula of the absolute value of the vehicle trajectory slope based on the global Cartesian coordinate system is: Among them, α c(k) is the absolute value of the trajectory slope of the vehicle trajectory sampling point in the kth global Cartesian coordinate system, in units of (1), x t(k) is the lateral coordinate of the vehicle trajectory sampling point in the kth global Cartesian coordinate system, in meters, y t(k) is the longitudinal coordinate of the vehicle trajectory sampling point in the kth global Cartesian coordinate system, in (m); The calculation formula of the direction factor is: d (k) =sign((and t(k) -and r(h) )·cos(θ r(k) )-(x t(k) -x r(h) )·sin(θ r(k) )) Among them, d (k) is the direction factor of the kth vehicle trajectory sampling point in the Frenet coordinate system, in units of (1), x t(k) is the lateral coordinate of the kth Frenet coordinate system vehicle trajectory sampling point under the global Cartesian, in (m), y t(k) x is the longitudinal coordinate of the kth Frenet coordinate system vehicle trajectory sampling point under the global Cartesian, in meters; r(h) is the lateral coordinate of the corresponding reference line projection point of the vehicle driving trajectory sampling point in the kth Frenet coordinate system under the global Cartesian, in (m), y r(h) is the longitudinal coordinate of the corresponding reference projection point of the vehicle driving trajectory sampling point in the kth Frenet coordinate system under the global Cartesian, in meters; θ r(k) is the heading angle of the corresponding reference projection point of the vehicle trajectory sampling point in the kth Frenet coordinate system, in degrees; The average judgment time is the average value of the first judgment time and the second judgment time.

5. According to the vehicle lateral driving condition identification system based on coordinate system transformation according to claim 4, It is characterized in that The specific steps of converting the vehicle trajectory in the global Cartesian coordinate system to the Frenet coordinate system in the data processing and calculation module include: Obtaining the lateral coordinates and the longitudinal coordinates of the reference line sampling points of the vehicle driving trajectory in the global Cartesian coordinate system; the reference line of the vehicle driving trajectory is the center line of the lane; the sampling frequency of the reference line sampling points of the vehicle driving trajectory is selected according to the hardware conditions; Obtaining the horizontal and vertical coordinates of the sampling points of the vehicle driving trajectory in the global Cartesian coordinate system; the vehicle driving trajectory is the driving trajectory of the vehicle's center of mass; the sampling frequency of the reference line sampling points of the vehicle driving trajectory is selected according to hardware conditions; Calculate the heading angle of the reference line sampling point of the vehicle's driving trajectory; the calculation formula for the heading angle of the reference line sampling point of the vehicle's driving trajectory is: Among them, θ r(k) is the heading angle of the reference line sampling point of the kth vehicle trajectory, in degrees, x r(k) is the lateral coordinate of the reference line sampling point of the kth vehicle trajectory in the global Cartesian coordinate system, in meters, y r(k) is the longitudinal coordinate of the reference line sampling point of the kth vehicle trajectory in the global Cartesian coordinate system, in (m); Calculate the distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point set of the adjacent vehicle driving trajectory; the distance is the Euclidean distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point of the adjacent vehicle driving trajectory; the reference line sampling point set of the adjacent vehicle driving trajectory is a set of a set number of sampling points before and after the corresponding sampling moment of the selected vehicle driving trajectory reference line sampling point; Find the vehicle driving trajectory reference line sampling point with the shortest distance from the reference line sampling point of the vehicle driving trajectory to the reference line sampling point set of the adjacent vehicle driving trajectory; the vehicle driving trajectory reference line sampling point with the shortest distance is used as the projection point of the global Cartesian coordinate system converted to the Frenet coordinate system; The path value of the reference line projection point of the vehicle driving trajectory corresponding to the reference line sampling point of the vehicle driving trajectory is used as the transverse coordinate of the Frenet coordinate system. The transverse coordinate of the Frenet coordinate system is: s (k) =ss (k) Among them, s (k) is the lateral coordinate of the kth vehicle trajectory sampling point in the Frenet coordinate system, in meters, ss (k) is the path value of the projection point of the vehicle driving trajectory reference line corresponding to the kth vehicle driving trajectory sampling point, in (m); The distance from the reference line sampling point of the vehicle driving trajectory to the projection point of the corresponding vehicle driving trajectory reference line is taken as the longitudinal coordinate value of the Frenet coordinate system. The longitudinal coordinate of the Frenet coordinate system is: Among them, l (k) is the longitudinal coordinate of the kth vehicle trajectory sampling point in the Frenet coordinate system, in meters, d (k) is the direction factor of the kth vehicle trajectory sampling point in the Frenet coordinate system, in units of (1), x t(k) is the lateral coordinate of the kth vehicle trajectory sampling point in the global Cartesian coordinate system, in meters, y t(k) is the lateral coordinate of the kth vehicle trajectory sampling point in the global Cartesian coordinate system, in meters, x r(h) is the lateral coordinate of the hth vehicle trajectory reference line sampling point in the global Cartesian coordinate system, in meters, y r(h) is the longitudinal coordinate of the hth vehicle trajectory reference line sampling point in the global Cartesian coordinate system, in (m); Wherein, h is the sampling sequence number of the projection point of the vehicle driving trajectory reference line corresponding to the kth vehicle driving trajectory sampling point.

6. According to the vehicle lateral driving condition identification system based on coordinate system transformation according to claim 4, It is characterized in that The preset first threshold is 0.021; the preset second threshold is 0.01; and the preset third threshold is 0.

Images