A method for selecting and processing targets during vehicle lane change

Abstract

The present invention provides a method for selecting a target object and processing its attributes during a vehicle lane change process. When the driver of the vehicle intends to change lanes, the vehicle's driving trajectory is predicted based on the predicted curvature of the front wheel turning angle and yaw rate of the vehicle, and the nearest vehicle in front of the vehicle that overlaps with the vehicle's driving trajectory is selected as the target object for longitudinal control in combination with the vehicle in front of the vehicle; when the intelligent driving system automatically changes lanes and the target vehicle of longitudinal control cuts in and out, the weight value of the key attribute parameters of the virtual target vehicle is introduced, and the virtual target vehicle of longitudinal control is designed based on the motion state of the front vehicle and the motion state of the vehicle, so that the key attribute parameters such as the speed and distance of the target vehicle when switching are continuous, and the human-like lane change driving function of the intelligent driving system is realized. The above two schemes improve the intelligent ability of the driving system, while taking into account the current target vehicle and the candidate target vehicle, ensuring the safety and comfort of the control end.

Description

Technical Field

[0001] The present invention belongs to the technical field of automatic driving systems, and in particular relates to a method for selecting a target object and processing its attributes during a vehicle lane change process. Background Art

[0002] The lane change assist function can, to some extent, characterize the intelligent capabilities of the intelligent driving system, prompting the intelligent driving system to move from a single lane to multiple lanes. Therefore, the intelligent driving system is no longer just satisfied with the longitudinal (acceleration and deceleration) and lateral (steering) control of the vehicle in the lane, but should also control between multiple lanes, so that the entire intelligent driving process is smooth and uninterrupted.

[0003] In longitudinal control, the selection and timing of the target object that affects the lane change process is very important. In the past lane change process, the selection and timing of the target object is generally to switch the target object at a certain time point / position point during the lane change. That is, if the time point / position point is exceeded during the lane change process, the target object is switched to the vehicle in the target lane to achieve longitudinal control; if the time point / position point is not exceeded, the target vehicle in the lane is still used as the target object for longitudinal control.

[0004] In actual road scenarios, this method has a major flaw - when there is a large difference between the target vehicle attributes of the lane (such as absolute speed, relative distance, etc.) and the vehicle attributes of the target lane, it is very easy to experience uncomfortable longitudinal control during the lane changing process. This is very different from the longitudinal control of human drivers during the lane changing process and is not in line with the lane changing habits of human drivers. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method for selecting a target object and processing its attributes during a vehicle lane change process,

[0006] The technical solution adopted by the present invention to solve the above technical problems is: a method for selecting a target object and processing its attributes during a vehicle lane change, comprising the following steps:

[0007] When the driver of the ego vehicle intends to change lanes, the curvature is predicted based on the front wheel steering angle and yaw rate of the ego vehicle, thereby predicting the ego vehicle's driving trajectory; combined with the vehicle in front of the ego vehicle, the nearest vehicle in front that overlaps with the ego vehicle's driving trajectory is selected as the target vehicle for longitudinal control;

[0008] When the vehicle is in a scenario including automatic lane changing by the self-vehicle intelligent driving system and the entry or exit of a longitudinally controlled target vehicle, the target vehicle is virtualized, and weight values ​​of key attribute parameters of the virtual target vehicle are introduced. The longitudinally controlled virtual target vehicle is designed based on the motion state of the preceding vehicle and the motion state of the self-vehicle, so that the key attribute parameters of the target vehicle are continuous when switching.

[0009] According to the above scheme, in the step of "combining the vehicle in front of the vehicle, selecting the nearest vehicle in front that overlaps with the vehicle's driving trajectory as the target vehicle for longitudinal control", in the longitudinal control of a single lane, the vehicle directly in front of the vehicle is used as the target object.

[0010] According to the above scheme, in the step of "when the driver of the vehicle intends to change lanes, the curvature is predicted based on the front wheel turning angle and yaw rate of the vehicle, thereby predicting the driving trajectory of the vehicle", the specific steps of curvature calculation are:

[0011] When the vehicle speed is less than the first threshold, let k be the ratio of the steering wheel angle to the front wheel angle, θ be the steering wheel angle, then the front wheel angle β is:

[0012] β=k·θ,

[0013] Assuming L is the wheelbase, the curvature ρ is calculated using the front wheel turning angle β and the wheelbase L as:

[0014] ρ = (sinβ) / L;

[0015] When the vehicle speed is greater than the second threshold, the yaw rate ω and the vehicle speed v are used. h The curvature ρ is calculated as:

[0016] ρ = ω / vh;

[0017] When the vehicle speed is between the first threshold and the second threshold, let μ be the weight coefficient of the front wheel steering angle to calculate the curvature ρ, and the calculation result using the front wheel steering angle β and the wheelbase L is compared with the calculation result using the yaw rate ω and the vehicle speed v h The weighted average curvature ρ calculated by the calculation result is:

[0018]

[0019] According to the above scheme, the step of "when the vehicle is in a scenario including automatic lane changing by the vehicle's intelligent driving system and the longitudinally controlled target vehicle cutting in or out" also includes the scenario where the overtaking function of the intelligent driving system is activated.

[0020] According to the above scheme, the step of "when the vehicle is in a scene including the automatic lane change of the self-vehicle intelligent driving system and the longitudinally controlled target vehicle cutting in or out, the target vehicle is virtualized, the weight values ​​of the key attribute parameters of the virtual target vehicle are introduced, and the longitudinally controlled virtual target vehicle is designed based on the motion state of the preceding vehicle and the motion state of the self-vehicle, so that the key attribute parameters of the target vehicle when switching are continuous" has the following specific steps:

[0021] The target vehicles include the current target vehicle and the candidate target vehicles;

[0022] Assume d1 is the distance of the current target vehicle, v1 is the speed of the current target vehicle, d2 is the distance of the candidate target vehicle, and v2 is the speed of the candidate target vehicle; α is the weight value of the key attribute parameter of the virtual target vehicle, which is related to the motion state of the preceding vehicle and the self-vehicle; then the distance d of the virtual target vehicle is CIPV for:

[0023] d CIPV =α·d1+(1-α)·d2;

[0024] The speed v of the virtual target vehicle CIPV for:

[0025] v CIPV =α·v1+(1-α)·v2.

[0026] Furthermore, in the step of "when the vehicle is in a scenario including automatic lane change by the self-vehicle intelligent driving system and the longitudinally controlled target vehicle cutting in or out, the target vehicle is virtualized, the weight values ​​of the key attribute parameters of the virtual target vehicle are introduced, and the longitudinally controlled virtual target vehicle is designed based on the motion state of the preceding vehicle and the motion state of the self-vehicle, so that the key attribute parameters of the target vehicle are continuous when switching", in the conventional driving mode and the scenario without the preceding vehicle cutting in or out, the current target vehicle is used as input for the longitudinal control of the intelligent driving system, then the distance d of the virtual target vehicle CIPv and the speed v of the virtual target vehicle CIPV They are:

[0027] d CIPV =d1,

[0028] v CIPV =v1.

[0029] Furthermore, in the step of "when the vehicle is in a scenario including automatic lane change by the self-vehicle intelligent driving system and the longitudinally controlled target vehicle cutting in or out, the target vehicle is virtualized, the weight values ​​of the key attribute parameters of the virtual target vehicle are introduced, and the longitudinally controlled virtual target vehicle is designed based on the motion state of the front vehicle and the motion state of the self-vehicle, so that the key attribute parameters of the target vehicle are continuous when switching", in the scenario of dangerous driving mode and close cutting in by the front vehicle, the candidate target vehicle is used as input for the longitudinal control of the intelligent driving system; let d 2y is the lateral distance of the candidate target vehicle, v 2y is the lateral speed of the candidate target vehicle, Thred2, Thred 2y , Threv 2yare the distance threshold, lateral distance threshold, and lateral speed threshold for triggering dangerous driving mode for the candidate target vehicle. The above thresholds are strongly related to the vehicle speed and the distance of the candidate target vehicle and are calibrated according to the actual vehicle debugging. 2y 、v 2y The triggering conditions for dangerous driving mode are:

[0030] (d2<Thred2)&&(v2<v h )&&(d 2y <Thred 2y )&&(v 2y <Threv 2y );

[0031] The distance d of the virtual target vehicle CIPV and the speed v of the virtual target vehicle CIPV They are:

[0032] a CIPV =a2,

[0033] v CIPV =v2.

[0034] Furthermore, in the step of "when the vehicle is in a scenario including automatic lane changing by the self-vehicle intelligent driving system and the longitudinally controlled target vehicle cutting in or out, the target vehicle is virtualized, the weight values ​​of the key attribute parameters of the virtual target vehicle are introduced, and the longitudinally controlled virtual target vehicle is designed based on the motion state of the leading vehicle and the motion state of the self-vehicle, so that the key attribute parameters of the target vehicle are continuous when switching", in the scenario of the leading vehicle smoothly cutting in and out, a virtual target vehicle is established according to the motion states of the current target vehicle and the candidate target vehicle; the α value change curve is calibrated through actual vehicle debugging, and t0 is set to represent that the leading vehicle starts to cut in or out or the intelligent driving system starts to change lanes automatically, and t1 represents that the leading vehicle ends cutting in or out or the intelligent driving system ends automatic lane changing.

[0035] Furthermore, in the step of "in the scenario where the front vehicle smoothly cuts in and out of the mode, a virtual target vehicle is established according to the motion states of the current target vehicle and the candidate target vehicle; the α value change curve is calibrated by real vehicle debugging, and t0 indicates that the front vehicle starts to cut in or out or the intelligent driving system starts to change lanes automatically, and t1 indicates that the front vehicle ends to cut in or out or the intelligent driving system ends to change lanes automatically", the specific steps of determining t0 and t1 are:

[0036] A line graph is established with the lateral speed Vy of cutting into or out of the target as the horizontal axis and the lateral distance dy as the vertical axis. Assume that there are two points A and B on the line.

[0037] The lateral distance when the target vehicle just cuts in or out of the lane or the self-vehicle just starts changing lanes is dy0.

[0038] The lateral speed of the target vehicle when it just cuts in or out of the lane or when the vehicle just starts to change lanes is Vy0.

[0039] The lateral distance when the target vehicle completely enters or exits the lane or the self-vehicle completes lane change is dy1.

[0040] The lateral speed of the target vehicle when it completely enters or exits the lane or the ego vehicle completes lane change is Vy1;

[0041] When the point (Vy, dy) representing the lateral distance and lateral speed of the cut-in target just meets the lower left of the broken line dy1→A→B→Vy1, that is, the target vehicle just cuts into the lane, or when the point (Vy, dy) representing the lateral distance and lateral speed of the cut-out target just meets the upper right of the broken line dy1→A→B→Vy1, that is, the target vehicle just cuts out of the lane, it is considered that the horizontal coordinate of the point corresponding to the α value change curve at this time is t0;

[0042] When the lateral distance of the cut-in target is less than the threshold set according to the vehicle speed and the cut-in target distance, it is considered that the cut-in target has completely entered the lane. At this time, the horizontal coordinate of the point corresponding to the α value change curve is t1;

[0043] Or when the lateral distance of the cut-out target is greater than the threshold set according to the vehicle speed and the cut-out target distance, it is considered that the cut-out target has completely driven out of the lane, and the horizontal coordinate of the point in the α value change curve corresponding here is t1;

[0044] When the intelligent driving system automatically changes lanes, the horizontal coordinate of the point in the α value change curve corresponding to when the system starts to automatically change lanes is set to t0; the horizontal coordinate of the point in the α value change curve corresponding to when the center of the rear axle of the vehicle crosses the lane line is set to t1.

[0045] Furthermore, the step "when the center of the rear axle of the vehicle crosses the lane line, the horizontal coordinate of the point in the α value change curve diagram corresponding to the time when the vehicle's left front wheel, right front wheel, left rear wheel or right rear wheel crosses the lane line is replaced by t1 based on the performance requirements of the automatic lane changing of the intelligent driving system, the sensor configuration and performance, and the vehicle configuration.

[0046] The beneficial effects of the present invention are:

[0047] 1. The present invention provides a method for selecting a target object and processing its attributes during a vehicle lane change process. Based on the scenarios of the driver's intention to change lanes, the intelligent driving system automatically changing lanes, and the longitudinally controlled target vehicle cutting in and out, the method proposes a corresponding method for selecting a target object and processing its attributes, thereby realizing the human-like lane change driving function of the intelligent driving system.

[0048] 2. When the driver of the vehicle intends to change lanes, the present invention predicts the curvature based on the front wheel angle and yaw rate of the vehicle, thereby predicting the vehicle's driving trajectory, and combining the vehicle in front of the vehicle to select the nearest vehicle in front that overlaps with the vehicle's driving trajectory as the target of longitudinal control. In the longitudinal control of a single lane, the vehicle directly in front of the vehicle is used as the target, and the driving intention of the vehicle is not taken into account. This method solves the problem of target vehicle selection for longitudinal control considering the driver's intention, and achieves a key technical breakthrough in the conversion of longitudinal control from a single lane to multiple lanes.

[0049] 3. In the scenarios where the intelligent driving system automatically changes lanes or the longitudinally controlled target vehicle cuts in and out, the present invention introduces the weight values ​​of the key attribute parameters of the virtual target vehicle, and designs the longitudinally controlled virtual target vehicle based on the motion state of the preceding vehicle and the motion state of the own vehicle, so that the key attribute parameters such as the speed and distance when the target vehicle switches are continuous, while taking into account the current target vehicle and the candidate target vehicles, to ensure the safety and comfort of the control end.

[0050] 4. The present invention aims at the problem that the target object selection and timing strategy of the intelligent driving system during the lane changing process is either 0 or 1, that is, the target vehicle in the current lane is selected before a certain time point / position point of the lane changing, and the vehicle in the target lane is selected as the target object after the time point / position point. Through the above two schemes, the intelligent capability of the driving system is enhanced, and the longitudinal control safety and comfort of the intelligent driving system during the lane changing process are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a scene diagram of a normal driving mode without a leading vehicle cutting in or out in an embodiment of the present invention.

[0052] Figure 2 This is a scene diagram of a front vehicle cutting in at close range in a dangerous driving mode in an embodiment of the present invention.

[0053] Figure 3 This is a scene diagram of the smooth cut-in mode of the leading vehicle according to an embodiment of the present invention.

[0054] Figure 4 This is a scene diagram of the smooth cut-out mode of the leading vehicle in an embodiment of the present invention.

[0055] Figure 5 1 is a curve diagram of the change of α value according to an embodiment of the present invention.

[0056] Figure 6 4 is a graph showing the lateral distance and lateral speed of a target according to an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0058] Embodiment 1:

[0059] See also Figure 1 , a method for selecting a target object and processing its attributes during a vehicle lane change process according to an embodiment of the present invention comprises the following steps:

[0060] S1: When receiving a lane change signal input by the driver of the vehicle, the curvature is predicted based on the front wheel angle and yaw rate of the vehicle, thereby predicting the vehicle's driving trajectory, and combining it with the vehicle in front of the vehicle, the nearest vehicle in front that overlaps with the vehicle's driving trajectory is selected as the target object for longitudinal control. In the longitudinal control of a single lane, the vehicle directly in front of the vehicle is used as the target object.

[0061] The specific steps of predicting curvature based on the front wheel angle and yaw rate of the vehicle are as follows:

[0062] The first threshold is 1 m / s and the second threshold is 1.5 m / s; they can also be preset according to actual application conditions.

[0063] When the vehicle speed is less than 1m / s, let k be the ratio of the steering wheel angle to the front wheel angle, θ be the steering wheel angle, then the front wheel angle β = k·θ; let L be the wheelbase, then the curvature ρ is calculated using the front wheel angle β and the wheelbase L:

[0064] ρ = (sinnβ) / L;

[0065] When the vehicle speed is greater than 1.5 m / s, the yaw rate ω and the vehicle speed v are used. h The curvature ρ is calculated as:

[0066] ρ=ω / v h ;

[0067] When the vehicle speed is between 1m / s and 1.5m / s, let μ be the weight coefficient of the curvature calculated by the front wheel steering angle, and the calculation result using the front wheel steering angle β and wheelbase L is compared with the calculation result using the yaw rate ω and the vehicle speed v h The weighted average curvature ρ calculated by the calculation result is:

[0068]

[0069] S2: When in a scenario including automatic lane changing by the self-vehicle intelligent driving system, the longitudinally controlled target vehicle cutting in or out, and the activation of the overtaking function of the intelligent driving system, the target vehicle is virtualized and the weight values ​​of the key attribute parameters of the virtual target vehicle are introduced. The longitudinally controlled virtual target vehicle is designed based on the motion state of the leading vehicle and the motion state of the self-vehicle, so that the key attribute parameters such as speed and distance of the target vehicle when switching are continuous.

[0070] The target vehicles include the current target vehicle and the candidate target vehicles; let d1 be the distance of the current target vehicle, v1 be the speed of the current target vehicle, d2 be the distance of the candidate target vehicle, and v2 be the speed of the candidate target vehicle; α is the weight value of the key attribute parameter of the virtual target vehicle, which is related to the motion state of the preceding vehicle and the self-vehicle. Then the distance d of the virtual target vehicle is CIPV for:

[0071] d CIPV =α·d1+(1-α)·d2;

[0072] The speed v of the virtual target vehicle CIPV for:

[0073] v CIPV =α·v1+(1-α)·v2.

[0074] See also Figure 1 In the normal driving mode and when there is no front vehicle cutting in or out, the current target vehicle is used as input for the longitudinal control of the intelligent driving system. The distance d of the virtual target vehicle is CIPV and the speed v of the virtual target vehicle CIPV They are:

[0075] d CIPV =d1,

[0076] v CIPV =v1.

[0077] See also Figure 2 , in the dangerous driving mode and when the front vehicle cuts in at a close distance, in order to ensure the safety of the longitudinal control of the intelligent driving system, the candidate target vehicle is used as input for the longitudinal control of the intelligent driving system. Assume d 2y is the lateral distance of the candidate target vehicle, v 2y is the lateral speed of the candidate target vehicle, Thred2, Thred 2y , Threv 2y are the distance threshold, lateral distance threshold, and lateral speed threshold for triggering dangerous driving mode of the candidate target vehicle. The threshold is strongly related to the vehicle speed and the distance of the candidate target vehicle and is calibrated according to the actual vehicle debugging. 2y 、v 2y The triggering conditions for dangerous driving mode are as follows:

[0078] (d2<Thred2)&&(v2<v h )&&(d 2y <Thred 2y )&&(v 2y <Threv 2y )

[0079] At this time, the distance d of the virtual target vehicle CIPV and the speed v of the virtual target vehicle CIPV They are:

[0080] d CIPV =d2,

[0081] v CIPV =v2.

[0082] See also Figure 3 , Figure 4 , when the preceding vehicle smoothly cuts in or out, a virtual target vehicle is established based on the motion states of the current target vehicle and the candidate target vehicles; see Figure 5 , the α value change curve is calibrated through actual vehicle debugging, and t0 represents the start of the front vehicle cutting in or out or the start of automatic lane change by the intelligent driving system, and t1 represents the end of the front vehicle cutting in or out or the end of automatic lane change by the intelligent driving system.

[0083] like Figure 6 As shown, the specific steps for determining t0 and t1 are: establish a broken line graph with the lateral speed Vy of cutting in or out of the target as the horizontal axis and the lateral distance dy as the vertical axis, assuming that there are two points A and B on the broken line, the lateral distance when the target vehicle just cuts in or out of the lane or the ego vehicle just starts to change lanes is dy0, the lateral speed when the target vehicle just cuts in or out of the lane or the ego vehicle just starts to change lanes is Vy0, the lateral distance when the target vehicle completely enters or exits the lane or the ego vehicle completes the lane change is dy1, and the lateral speed when the target vehicle completely enters or exits the lane or the ego vehicle completes the lane change is Vy1.

[0084] When the point (Vy, dy) representing the lateral distance and lateral speed of the target entering the lane just meets the lower left of the broken line dyl→A→B→Vyl, that is, the target vehicle has just entered the lane, or when the point (Vy, dy) representing the lateral distance and lateral speed of the target cutting out the lane just meets the upper right of the broken line dy1→A→B→Vy1, that is, the target vehicle has just cut out of the lane, it is considered that the horizontal coordinate of the point corresponding to the α value change curve at this time is t0.

[0085] When the lateral distance of the cut-in target is less than the threshold set according to the vehicle speed and the cut-in target distance, the cut-in target is considered to have completely entered the lane, and the horizontal coordinate of the point in the α value change curve diagram here is t1; or when the lateral distance of the cut-out target is greater than the threshold set according to the vehicle speed and the cut-out target distance, the cut-out target is considered to have completely left the lane, and the horizontal coordinate of the point in the α value change curve diagram here is t1. When the intelligent driving system automatically changes lanes, the horizontal coordinate of the point in the α value change curve diagram corresponding to the time when the system starts to change lanes automatically is t0; when the center of the rear axle of the vehicle crosses the lane line, the horizontal coordinate of the point in the α value change curve diagram corresponding to the time when the system starts to change lanes automatically is t1.

[0086] Embodiment 2: The principle of this embodiment is the same as that of Embodiment 1, except that:

[0087] Based on the performance requirements of the automatic lane changing of the intelligent driving system, sensor configuration and performance, vehicle configuration, etc., the horizontal coordinate of the point in the α value change curve corresponding to when the left front wheel, right front wheel, left rear wheel or right rear wheel of the vehicle crosses the lane line is set to t1, in order to seek the best overall performance of the automatic lane changing of the intelligent driving system.

[0088] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0089] The above embodiments are only used to illustrate the design ideas and features of the present invention, and their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. The protection scope of the present invention is not limited to the above embodiments. Therefore, any equivalent changes or modifications made based on the principles and design ideas disclosed by the present invention are within the protection scope of the present invention.

Claims

1. A method for selecting a target object and processing its attributes during a vehicle lane change, characterized in that: The following steps are involved: When receiving a lane change signal input by the driver of the vehicle, the curvature is predicted based on the front wheel steering angle and yaw rate of the vehicle, thereby predicting the vehicle's driving trajectory; combined with the vehicle in front of the vehicle, the nearest vehicle in front that overlaps with the vehicle's driving trajectory is selected as the target vehicle for longitudinal control; When in a scenario including automatic lane changing by the ego vehicle's intelligent driving system and the entry or exit of a longitudinally controlled target vehicle, the target vehicle is virtualized, and weight values ​​of key attribute parameters of the virtual target vehicle are introduced. The longitudinally controlled virtual target vehicle is designed based on the motion state of the leading vehicle and the motion state of the ego vehicle, so that the key attribute parameters of the target vehicle are continuous when switching.

2. The method for selecting a target object and processing its attributes during a vehicle lane change according to claim 1, characterized in that: In the step of "combining the vehicle in front of the vehicle, selecting the nearest vehicle in front that overlaps with the vehicle's driving trajectory as the target vehicle for longitudinal control", in the longitudinal control of a single lane, the vehicle directly in front of the vehicle is used as the target object.

3. The method for selecting a target object and processing its attributes during a vehicle lane change according to claim 1, characterized in that: In the step of "when receiving a lane change signal input by the driver of the vehicle, predicting the curvature based on the front wheel steering angle and yaw rate of the vehicle, thereby predicting the driving trajectory of the vehicle", the specific steps of curvature calculation are: When the vehicle speed is less than the first threshold, let k be the ratio of the steering wheel angle to the front wheel angle, θ be the steering wheel angle, then the front wheel angle β is: β=k·θ, Assuming L is the wheelbase, the curvature ρ is calculated using the front wheel turning angle β and the wheelbase L as: ρ = (sinβ) / L; When the vehicle speed is greater than the second threshold, the yaw rate ω and the vehicle speed v are used. h The curvature ρ is calculated as: p=ω / v h ; When the vehicle speed is between the first threshold and the second threshold, let μ be the weight coefficient of the front wheel steering angle to calculate the curvature ρ, and the calculation result using the front wheel steering angle β and the wheelbase L is compared with the calculation result using the yaw rate ω and the vehicle speed v h The weighted average curvature ρ calculated by the calculation result is:

4. The method for selecting a target object and processing its attributes during a vehicle lane change according to claim 1, characterized in that: The step of "when the vehicle is in a scenario including automatic lane changing by the vehicle's intelligent driving system and the longitudinally controlled target vehicle cutting in or out" also includes a scenario in which the overtaking function of the intelligent driving system is activated.

5. The method for selecting a target object and processing its attributes during a vehicle lane change according to claim 1, characterized in that: The step of "when the vehicle is in a scene including automatic lane change by the self-vehicle intelligent driving system and the longitudinally controlled target vehicle cutting in or out, virtualizing the target vehicle, introducing weight values ​​of key attribute parameters of the virtual target vehicle, designing the longitudinally controlled virtual target vehicle based on the motion state of the preceding vehicle and the motion state of the self-vehicle, and making the key attribute parameters of the target vehicle continuous when switching" specifically includes the following steps: The target vehicles include the current target vehicle and the candidate target vehicles; Assume d1 is the distance of the current target vehicle, v1 is the speed of the current target vehicle, d2 is the distance of the candidate target vehicle, and v2 is the speed of the candidate target vehicle; α is the weight value of the key attribute parameter of the virtual target vehicle, which is related to the motion state of the preceding vehicle and the self-vehicle; then the distance d of the virtual target vehicle is CIPV for: d CIPV =α·d1+(1-α)·d2; The speed v of the virtual target vehicle CIPV for: v CIPV = α·v1+(1-α)·v2.

6. The method for selecting a target object and processing its attributes during a vehicle lane change according to claim 5, characterized in that: In the step of "when the vehicle is in a scene including automatic lane change by the self-vehicle intelligent driving system and the longitudinally controlled target vehicle cutting in or out, the target vehicle is virtualized, the weight values ​​of the key attribute parameters of the virtual target vehicle are introduced, and the longitudinally controlled virtual target vehicle is designed based on the motion state of the preceding vehicle and the motion state of the self-vehicle, so that the key attribute parameters of the target vehicle when switching are continuous", in the normal driving mode and when there is no preceding vehicle cutting in or out, the current target vehicle is used as input for the longitudinal control of the intelligent driving system, then the distance d of the virtual target vehicle CIPV and the speed v of the virtual target vehicle CIPV They are: d CIPV =d1, v CIPV =v1.

7. A method for selecting a target object and processing its attributes during a vehicle lane change according to claim 3 or 5, characterized in that: In the step of "when the vehicle is in a scenario including automatic lane change by the self-vehicle intelligent driving system and the longitudinally controlled target vehicle cutting in or out, the target vehicle is virtualized, the weight values ​​of the key attribute parameters of the virtual target vehicle are introduced, and the longitudinally controlled virtual target vehicle is designed based on the motion state of the front vehicle and the motion state of the self-vehicle, so that the key attribute parameters of the target vehicle are continuous when switching", in the dangerous driving mode and when the front vehicle cuts in at a close distance, the candidate target vehicle is used as input for the longitudinal control of the intelligent driving system; d 2y is the lateral distance of the candidate target vehicle, v 2y is the lateral speed of the candidate target vehicle, Thred2, Thred 2y , Threv 2y are the distance threshold, lateral distance threshold, and lateral speed threshold for triggering dangerous driving mode for the candidate target vehicle. The above thresholds are strongly related to the vehicle speed and the distance of the candidate target vehicle and are calibrated according to the actual vehicle debugging. 2y 、v 2y The triggering conditions for dangerous driving mode are: (d2<Thred2)&&(v2<v h )&&(d 2y <Thred 2y )&&(v 2y <Threv 2y );The distance d of the virtual target vehicle CIPV and the speed v of the virtual target vehicle CIPV They are: d CIPV =d2, <h2 style=";text-align:left;direction:ltr">v<h2 style=";text-align:left;direction:ltr"> CIPV <h2 style=";text-align:left;direction:ltr"> <v2。 8. The method for selecting a target object and processing its attributes during a vehicle lane change according to claim 5, characterized in that: In the step of "when the vehicle is in a scenario including automatic lane changing by the self-vehicle intelligent driving system and the longitudinally controlled target vehicle cutting in or out, the target vehicle is virtualized, the weight values ​​of the key attribute parameters of the virtual target vehicle are introduced, and the longitudinally controlled virtual target vehicle is designed based on the motion state of the front vehicle and the motion state of the self-vehicle, so that the key attribute parameters of the target vehicle when switching are continuous", when the front vehicle smoothly cuts in or out, a virtual target vehicle is established according to the motion states of the current target vehicle and the candidate target vehicle; the α value change curve is calibrated through actual vehicle debugging, and t0 is set to represent that the front vehicle starts to cut in or out or the intelligent driving system starts to automatically change lanes, and t1 represents that the front vehicle ends to cut in or out or the intelligent driving system ends to automatically change lanes.

9. The method for selecting a target object and processing its attributes during a vehicle lane change according to claim 8, characterized in that: The specific steps to determine t0 and t1 are: A line graph is established with the lateral speed Vy of cutting into or out of the target as the horizontal axis and the lateral distance dy as the vertical axis. Assume that there are two points A and B on the line. The lateral distance when the target vehicle just cuts in or out of the lane or the self-vehicle just starts changing lanes is dy0. The lateral speed of the target vehicle when it just cuts in or out of the lane or when the vehicle just starts to change lanes is Vy0. The lateral distance when the target vehicle completely enters or exits the lane or the self-vehicle completes lane change is dy1. The lateral speed of the target vehicle when it completely enters or exits the lane or the ego vehicle completes lane change is Vy1; When the point (Vy, dy) representing the lateral distance and lateral speed of the cut-in target just meets the lower left of the broken line dy1→A→B→Vy1, that is, the target vehicle just cuts into the lane, or when the point (Vy, dy) representing the lateral distance and lateral speed of the cut-out target just meets the upper right of the broken line dy1→A→B→Vy1, that is, the target vehicle just cuts out of the lane, it is considered that the horizontal coordinate of the point corresponding to the α value change curve at this time is t0; When the lateral distance of the cut-in target is less than the threshold set according to the vehicle speed and the cut-in target distance, it is considered that the cut-in target has completely entered the lane. At this time, the horizontal coordinate of the point corresponding to the α value change curve is t1; Or when the lateral distance of the cut-out target is greater than the threshold set according to the vehicle speed and the cut-out target distance, it is considered that the cut-out target has completely left the lane, and the horizontal coordinate of the point in the α value change curve corresponding here is t1; When the intelligent driving system changes lanes automatically, the horizontal coordinate of the point in the α value change curve corresponding to when the system starts to change lanes automatically is set to t0; the horizontal coordinate of the point in the α value change curve corresponding to when the center of the rear axle of the vehicle crosses the lane line is set to t1.

10. The method for selecting a target object and processing its attributes during a vehicle lane change according to claim 9, characterized in that: The step "when the center of the rear axle of the vehicle crosses the lane line, the horizontal coordinate of the point in the α value change curve corresponding to the time when the center of the rear axle of the vehicle crosses the lane line is t1" is replaced by: based on the performance requirements of the automatic lane changing of the intelligent driving system, the sensor configuration and performance, and the vehicle configuration, the horizontal coordinate of the point in the α value change curve corresponding to the time when the left front wheel, right front wheel, left rear wheel or right rear wheel of the vehicle crosses the lane line is set to t1.

Images