Method and system for selecting car following target during lane changing in self-adaptive cruise process

Abstract

The invention discloses a method for selecting a vehicle following target during lane changing in an adaptive cruise process. The method comprises the following steps: receiving detection signals of front objects of a vehicle in a lane and an adjacent lane, and judging whether the front objects are vehicles or not; detecting the transverse deviation degree of the own vehicle relative to the own lane to detect whether the own vehicle is changing the lane or not; if the own vehicle changes the lane, detecting the distance between the own lane and a front object in an adjacent lane to which the own vehicle needs to change and the own vehicle and the coincidence rate between the front object in the own lane and the own track of the own vehicle, and selecting the front object in the own lane or the adjacent lane as a preliminary vehicle following target; when it is judged that the preliminary car-following target is the car, determining the preliminary car-following target as a final car-following target; and repeatedly executing the steps at a fixed time interval until the own vehicle is completely driven to the adjacent lane. The system for selecting the car following target during lane changing in the self-adaptive cruise process comprises a sensor, a processor and a self-adaptive cruise control module, and the processor is configured to execute the method.

Description

Technical Field

[0001] The present invention relates to the adaptive cruise control of vehicles, and more particularly to a method and system for selecting a following vehicle target during lane change in the adaptive cruise process. Background Art

[0002] In recent years, with the improvement of vehicle intelligence, the application of Advanced Driving Assistance System (ADAS) has become more and more common. Among them, the Adaptive Cruise Control (ACC) module can make the vehicle automatically drive at a set cruise speed or stably follow the vehicle ahead. Such a vehicle is typically equipped with 1V1R (one camera plus one radar, mainly based on the camera) to detect the relative longitudinal distance and relative speed between the host vehicle and the target vehicle ahead. Based on this, the longitudinal speed of the host vehicle can be automatically adjusted to maintain a certain distance following the target vehicle.

[0003] In some cases, the host vehicle needs to change lanes (by the driver manually turning the steering wheel), and the target vehicle in the original lane may not be suitable to continue as the following vehicle target. Therefore, it is necessary to quickly select a new following vehicle target. In the prior art, when selecting a following vehicle target during the lane change of the host vehicle, the longitudinal distance between the front objects in the host lane and the adjacent lanes and the host vehicle is mainly considered, and the one with the shorter longitudinal distance is selected as the possible following vehicle target. In addition, multiple cycles (for example, at least 4 cycles) of judgment are required to confirm whether the possible following vehicle target can be used as the following vehicle target. This judgment process needs to comprehensively consider the existence probability of the target (the probability of judging that it is a vehicle rather than other interference objects), the coincidence rate between the predicted driving trajectories of the target and the host vehicle (or the host lane), the cut-in or cut-out state of the target relative to the predicted driving (or the host lane) trajectory of the host vehicle, etc. Therefore, it is relatively time-consuming to select a following vehicle target during the lane change of the host vehicle in the prior art, and the lane change of the vehicle usually needs to be completed in a short time, thus there is a risk of collision between the host vehicle and the front object.

[0004] Therefore, there is a need for an improved method and system for selecting a following vehicle target during lane change in the adaptive cruise process, which should quickly select the following vehicle target to minimize the collision risk as much as possible. Summary of the Invention

[0005] The present invention provides a method for selecting a following vehicle target during lane change in the adaptive cruise process, the method comprising:

[0006] Step S1: Receive the detection signals of the front objects existing in the host lane and the adjacent lanes, and judge whether each front object is a vehicle;

[0007] Step S2: Detect whether the host vehicle is performing a lane change behavior by detecting the lateral deviation degree of the host vehicle relative to its own lane;

[0008] Step S3: When the host vehicle is performing a lane change behavior, detect the distance between the host vehicle and the front object in its own lane and the adjacent lane to which the host vehicle is about to change lanes, and the coincidence rate between the front object in its own lane and the self-trajectory of the host vehicle. Select the front object in its own lane or the adjacent lane as the preliminary following target, where the self-trajectory is the driving trajectory in the future period predicted based on the current driving action of the host vehicle;

[0009] Step S4: Determine whether the preliminary following target is a vehicle based on the judgment result of Step S1. If it is determined that the preliminary following target is a vehicle, determine the preliminary following target as the final following target; and

[0010] Repeat the above steps at fixed time intervals until the host vehicle completely drives into the adjacent lane.

[0011] The present invention also provides a processor module for a vehicle with an adaptive cruise function, which is configured to execute the method according to the present invention as described above.

[0012] The present invention also provides a system for selecting a following target during a lane change in an adaptive cruise process, including:

[0013] The above-mentioned processor module;

[0014] A sensor module, which is configured to obtain information including the position of the host vehicle in its own lane, the distance and relative position between the front object of the host vehicle existing in its own lane and the adjacent lane and the host vehicle, and the image of the front object, and send the obtained information to the processor module; and

[0015] An adaptive cruise control module, which is configured to receive the distance and relative speed between the final following target determined by the processor module and the host vehicle from the sensor module, and automatically adjust the speed of the host vehicle according to the distance and relative speed.

[0016] By using the method and system of the present invention, it is realized to quickly select a following target during a lane change in the vehicle adaptive cruise process, and the collision risk during the lane change process is reduced. Description of the Drawings

[0017] Figure 1A A plan view schematically showing the host vehicle driving without changing lanes;

[0018] Figure 1B A plan view schematically showing the host vehicle driving while changing lanes;

[0019] Figures 2A to 2E Schematically shows the situations of the ego-vehicle relative to the front objects in the case of lane change; and

[0020] Figure 3 Shows a flowchart of a method for selecting a following target during lane change in an adaptive cruise according to the present invention. Detailed implementation manners

[0021] The following describes in detail a method for selecting a following target during lane change in an adaptive cruise according to the present invention with reference to the accompanying drawings.

[0022] The method according to the present invention first includes step S1: receiving detection signals of front objects existing in the ego-lane and the adjacent lane, and determining whether each front object is a vehicle. The front objects of the ego-vehicle include the lane where the ego-vehicle is located (i.e., the ego-lane) and the objects in front of the ego-vehicle in the adjacent lane. The front objects of the ego-vehicle may be vehicles, pedestrians, or various objects existing in the lane, and determining whether it is a vehicle can be, for example, identified by sensors equipped on the ego-vehicle itself, which is known in the art and will not be elaborated here.

[0023] Then, the method according to the present invention includes step S2: detecting whether the ego-vehicle is performing a lane change behavior by detecting the lateral deviation degree of the ego-vehicle relative to the ego-lane. This detection is described in combination with Figure 1A and Figure 1B where Figure 1A Schematically shows a plan view of the ego-vehicle traveling without lane change, while Figure 1B Schematically shows a plan view of the ego-vehicle traveling in the case of lane change.

[0024] In the figure, the box 100 represents the ego-vehicle, the solid lines 111 and 112 respectively represent the left and right physical lane lines of the lane where the ego-vehicle 100 is currently located, and the space therebetween is the ego-lane 110, that is, the lane where the ego-vehicle 100 is currently located. The dashed lines 121 and 122 respectively represent the predicted travel trajectories of the ego-vehicle 100, that is, the travel trajectories in a future period predicted based on the current driving actions of the ego-vehicle 100, which can be referred to as the ego-trajectory 120. The method for generating the ego-trajectory 120 is known in the art and will not be elaborated here. In each figure, it is schematically shown that the ego-trajectory 120 is composed of a straight line located in front of the ego-vehicle 100 and slightly larger than the lateral extension range of the ego-vehicle 100 in the lateral direction, but the present invention is not limited thereto, and the ego-trajectory 120 may also be generated as a curve as long as it can accurately predict the trajectory that the ego-vehicle may follow in a future period when the ego-vehicle travels according to the current state.

[0025] As Figure 1A shown, when the ego-vehicle 100 travels without lane change, the ego-trajectory 120 will be completely accommodated in the ego-lane 110, while asFigure 1B As shown, when the self-vehicle 100 is traveling in a lane change situation, the self-trajectory 120 will deviate from the self-lane 110. Therefore, it is possible to detect whether the self-vehicle 100 is changing lanes based on the degree to which the self-vehicle 100 or the self-trajectory 120 deviates from the self-lane 110.

[0026] In order to simplify the detection process, the present invention proposes to detect whether the self-vehicle 100 is changing lanes based on the lateral deviation degree of the self-vehicle 100 relative to the self-lane 110. If the self-vehicle 100 reaches a predetermined lateral deviation degree relative to the self-lane 110, it can be determined that the self-vehicle 100 is changing lanes.

[0027] Specifically, it is possible to determine whether the vehicle 100 is changing lanes by detecting the lateral distance of the center 101 of the front bumper assembly of the vehicle 100 from the left and right physical lane lines 111 and 112 of the lane 110. More specifically, the sensor module of the vehicle itself can detect the lateral distance between the center 101 of the front bumper assembly and the left and right physical lane lines 111 and 112 of the lane 110 in real time. If it is detected that the lateral distance is greater than a certain set distance threshold, it can be determined that the vehicle 100 is changing lanes. For example, in the case where the vehicle 100 may change to the adjacent right lane, the lateral distance between the center 101 of the front bumper assembly and the left lane line 111 of the lane 110 (corresponding to Figure 1B In the case where the ego vehicle 100 may change to the adjacent left lane, the lateral distance between the center 101 of the front bumper assembly and the right lane line 112 of the ego lane 110 is detected.

[0028] It should be noted that the distance threshold set for the lateral distance is adjustable. For example, when the ego lane 110 is wider, that is, the lateral distance between the left and right physical lane lines 111 and 112 is larger, the distance threshold can also be set larger, so as to avoid as much as possible the misjudgment of whether the ego vehicle 100 is changing lanes.

[0029] It should also be noted that in addition to detecting the lateral distance between the center of the front bumper assembly 101 and the left and right physical lane lines 111 and 112 of the own lane 110, other parts of the own vehicle 100, such as the relative position between one or more wheels and the left and right physical lane lines 111 and 112 of the own lane 110, can also be detected to detect whether the own vehicle 100 is changing lanes, but the present invention is not limited to this.

[0030] After it has been detected in the previous step S2 that the host vehicle is performing a lane change maneuver, the method of the present invention further includes step S3: detecting the distance between the host vehicle and the front object in the host lane and the adjacent lane to which the host vehicle is about to change lanes, and the overlap rate between the front object in the host lane and the host vehicle's self-trajectory. The front object in either the host lane or the adjacent lane is selected as the preliminary following target, where the self-trajectory is the driving trajectory of the host vehicle predicted for a period of time in the future based on the host vehicle's current driving actions. The premise for selecting the preliminary following target is that the host vehicle will not collide with the preliminary following target during the lane change process, and to achieve this premise, the distance between the front object of the host vehicle in the host lane and the adjacent lane to which the host vehicle is about to change lanes and the host vehicle, as well as the overlap rate between the front object in the host lane and the host vehicle's self-trajectory, are comprehensively considered.

[0031] Since there may be front objects in both the host lane 110 where the host vehicle 100 is currently located and the adjacent lane to which it is about to change lanes, the host vehicle 100 needs to quickly select a following target during the lane change process to quickly change lanes to the adjacent lane and cruise following the new target in the adjacent lane, and the entire process needs to avoid possible collisions between the host vehicle 100 and the front objects existing in the host lane or the adjacent lane.

[0032] In the prior art, conventionally, the longitudinal distance between the front object of the host vehicle existing in the host lane and the adjacent lane and the host vehicle is sensed by sensors equipped on the vehicle, and the one with the shorter longitudinal distance is selected as the preliminary following target. That is to say, when the host vehicle determines to change lanes, in the case where there are front objects in both the host lane and the adjacent lane, the vehicle that is longitudinally closer to the host vehicle is always used as the current preliminary following target. Thus, the adaptive cruise control module ACC can adjust the longitudinal speed of the host vehicle according to the relative longitudinal distance and relative speed between the host vehicle and the preliminary following target, so as to ensure that the host vehicle will not collide with the preliminary following target during the lane change process. However, the above strategy has the problem of low lane change efficiency of the host vehicle.

[0033] To solve this problem, when the method of the present invention selects the preliminary following target, in addition to considering the longitudinal distance between the front object of the host vehicle in the host lane and the adjacent lane and the host vehicle, the overlap rate between the front object in the host lane and the host vehicle's self-trajectory is also considered. Next, in combination with Figures 2A to 2E The detailed description of how to select the preliminary following target according to the method of the present invention after it has been detected that the host vehicle has performed a lane change maneuver.

[0034] As Figure 2AAs shown, it shows that it is detected that the host vehicle 100 is about to change lanes from the host lane 110 to the adjacent lane 130, and at this time, there is no front object of the host vehicle 100 in the host lane 110, while there is a front object 2 of the host vehicle 100 in the adjacent lane 130. At this time, the object 2 can be directly selected as the preliminary following target of the host vehicle 100 during the lane change process. The adaptive cruise control module ACC can adjust the longitudinal speed of the host vehicle 100 according to the relative distance and relative speed between the host vehicle 100 and the object 2, and the driver manually turns the steering wheel to make the host vehicle 100 drive into the adjacent lane 130 where the object 2 is located.

[0035] As Figure 2B shown, it shows that it is detected that the host vehicle 100 is about to change lanes from the host lane 110 to the adjacent lane 130, and at this time, there is a front object 1 of the host vehicle 100 in the host lane 110, while there is no front object of the host vehicle 100 in the adjacent lane 130. At this time, the object 1 can be directly selected as the preliminary following target of the host vehicle 100 during the lane change process. The adaptive cruise control module ACC can adjust the longitudinal speed of the host vehicle 100 according to the relative distance and relative speed between the host vehicle 100 and the object 1, and the driver manually turns the steering wheel to make the host vehicle 100 drive into the adjacent lane 130. Since there is temporarily no front object of the host vehicle 100 in the adjacent lane 130, the adaptive cruise control module ACC can control the host vehicle 100 to drive at a default speed after completely driving into the adjacent lane 130 until a front object that can be selected as a following target appears in the adjacent lane 130.

[0036] As Figure 2C shown, it shows that it is detected that the host vehicle 100 is about to change lanes from the host lane 110 to the adjacent lane 130, and at this time, there is a front object 1 of the host vehicle 100 in the host lane 110, and at the same time, there is a front object 2 of the host vehicle 100 in the adjacent lane 130, and the distance between the object 1 and the host vehicle 100 is greater than or equal to the distance between the object 2 and the host vehicle 100. At this time, since the host vehicle 100 is far from the object 1 in the host lane 110, there is no risk of collision with the object 1 during the entire process of changing lanes to the adjacent lane 130 and following the object 2. Therefore, the object 2 can be selected as the preliminary following target of the host vehicle 100 during the lane change process. The adaptive cruise control module ACC can adjust the longitudinal speed of the host vehicle 100 according to the relative distance and relative speed between the host vehicle 100 and the object 2, and the driver manually turns the steering wheel to make the host vehicle 100 drive into the adjacent lane 130 where the object 2 is located.

[0037] Figure 2D and Figure 2EIt is shown that it is detected that the host vehicle 100 is about to change lanes from the host lane 110 to the adjacent lane 130. At this time, there is a front object 1 of the host vehicle 100 in the host lane 110, and there is a front object 2 of the host vehicle 100 in the adjacent lane 130, and the distance between the object 1 and the host vehicle 100 is less than the distance between the object 2 and the host vehicle 100. At this time, since the object 1 in the host lane 110 is relatively close to the host vehicle 100, it is necessary to consider whether there is a risk of collision with the object 1 in the host lane 110 during the process of the host vehicle 100 changing lanes from the host lane 110 to the adjacent lane 130.

[0038] Specifically, the present invention proposes to detect whether the ratio of the width of the intersection part of the front object 1 in the host lane 110 and the host trajectory 120 of the host vehicle 100 (the width in the lateral direction of the host lane 110) to the total width of the object 1 reaches a set coincidence rate threshold in this case. If it is detected that the ratio of the width of the intersection part of the object 1 in the host lane 110 and the host trajectory 120 of the host vehicle 100 to the total width of the object 1 reaches this coincidence rate threshold (corresponding to Figure 2D ), it indicates that there may be a collision between the host vehicle 100 and the object 1 in the host lane 110 during the process of the host vehicle 100 changing lanes from the host lane 110 to the adjacent lane 130. Therefore, the object 1 is selected as the preliminary following target of the host vehicle 100 during the lane change process, and the adaptive cruise control module ACC can adjust the longitudinal speed of the host vehicle 100 according to the relative distance and relative speed between the host vehicle 100 and the object 1, while the driver manually turns the steering wheel to make the host vehicle 100 travel to the adjacent lane 130. Since the adjustment of the longitudinal speed of the host vehicle 100 during the lane change process takes into account the object 1 closest to the host vehicle, it can be ensured that the host vehicle will not collide with the object 1 during the lane change process.

[0039] If it is detected that the ratio of the width of the intersection part of the front object 1 in the host lane 110 and the host trajectory 120 of the host vehicle 100 to the total width of the object 1 is less than the set threshold (corresponding to Figure 2E ), it indicates that there is almost no collision between the host vehicle 100 and the object 1 in the host lane 110 during the process of the host vehicle 100 changing lanes from the host lane 110 to the adjacent lane 130. Therefore, the object 2 in the adjacent lane 130 is selected as the preliminary following target of the host vehicle 100 during the lane change process, and the adaptive cruise control module ACC can adjust the longitudinal speed of the host vehicle 100 according to the relative distance and relative speed between the host vehicle 100 and the object 2, while the driver manually turns the steering wheel to make the host vehicle 100 travel to the adjacent lane 130 where the object 2 is located.

[0040] In the prior art, in Figure 2EIn this case, object 1 will be selected as the initial following target because the distance between object 1 and the host vehicle 100 is less than the distance between object 2 and the host vehicle 100, which can ensure that the host vehicle 100 will not collide with object 1, the object closest to it, during the lane change process. However, in fact, by using the method in the present invention, it has been calculated through the comparison of the ratio of the width of the intersection part of the self-trajectory 120 of object 1 and the host vehicle 100 to the total width of object 1 with a threshold that even if the host vehicle 100 does not consider object 1 during the lane change, it will hardly collide with object 1. Although it is predicted at the current moment that the trajectory of the host vehicle 100 slightly overlaps with object 1, since the position and direction of the host vehicle 100 are changing in real time during the lane change, and thus the self-trajectory 120 is also changing in real time, theoretically, the host vehicle 100 will move further and further away from object 1 in the self-lane 110 during the lane change. Therefore, if it has been calculated at the current moment that the ratio of the width of the intersection part of object 1 in the self-lane 110 and the self-trajectory 120 of the host vehicle 100 to the total width of object 1 is less than the set threshold, then as the lane change occurs and the host vehicle 100 moves away from the self-lane 110, the probability of the host vehicle 100 colliding with object 1 will increasingly approach 0. Therefore, different from the prior art, the method of the present invention, when facing Figure 2E the situation shown, will directly select the front object 2 in the adjacent lane as the initial following target, so that the host vehicle 100 can more quickly follow vehicle 2 in the adjacent lane it is about to change into.

[0041] Preferably, the threshold for comparison with the ratio of the width of the intersection part of the front object 1 in the self-lane 110 and the self-trajectory 120 of the host vehicle 100 to the total width of the vehicle 1 can be set to 0.1. The applicant has found that if the distance between the front object 1 in the self-lane 110 and the host vehicle 100 is less than the distance between object 2 in the adjacent lane 130 and the host vehicle 100 and the ratio of the width of the intersection part of object 1 in the self-lane 110 and the self-trajectory 120 of the host vehicle 100 to the total width of object 1 is less than 0.1, then selecting object 2 as the initial following target for lane change will hardly cause the host vehicle 100 to collide with object 1 in the self-lane. However, this application is not limited thereto, and the aforementioned threshold can also be adaptively adjusted to other values.

[0042] It should be noted that the selection of the preliminary target changes over time. The front objects of the host vehicle 100 are re-detected at fixed time intervals (usually 66 ms), including the front objects in the host lane 110 and the adjacent lanes, and the distances between the front objects in the host lane 110 and the host vehicle 100 and between the front objects in the adjacent lane 130 and the host vehicle 100 are re-compared. In addition, the relative magnitude of the ratio of the width of the intersection between the front object in the host lane 110 and the host vehicle's own trajectory 120 to the total width of the object and the set threshold is re-detected. Therefore, with the method of the present invention, the preliminary following target is re-selected at fixed time intervals to ensure that there is no risk of collision between the host vehicle and the front object at each moment during the lane change process.

[0043] Figures 2A to 2E Taking the case where the host vehicle intends to change lanes to the adjacent right lane as an example, the method of the present invention is equally applicable to the case where the host vehicle changes lanes to the adjacent left lane. In this case, the front objects to be considered include, in addition to the front objects in the host lane, the front objects in the adjacent left lane to which the vehicle intends to change lanes. The method of selecting the preliminary following target is the same as described above and will not be elaborated here.

[0044] In addition, Figures 2A to 2E It is schematically shown that there is at most one front object in each of the host lane and the adjacent lane, but the method of the present invention is equally applicable to the case where there are more than one front objects in each of the host lane and the adjacent lane. Specifically, if there are more than one front objects in the host lane and / or the adjacent lane, during the lane change process of the host vehicle, there may be a risk of collision only with the object having the shortest distance from the host vehicle in each lane. Therefore, only one object having the shortest distance from the host vehicle in each lane can be selected as the front object in that lane and the above method can be executed to determine the preliminary following target.

[0045] During the vehicle driving process, the front objects of the host vehicle existing in the host lane and the adjacent lanes may be other interferences in addition to vehicles, such as moving people or objects, billboards, etc. Therefore, it is necessary to determine whether the preliminary following target is a vehicle, and only after determining that it is a vehicle can it be determined as the final following target.

[0046] In the prior art, after determining a preliminary following target, it is then determined whether the preliminary following target is a vehicle, and only when it is determined that it is a vehicle can it be determined as the final following target. Generally, determining whether a certain target is a vehicle requires at least 4 cycles (judgment periods) to be executed. The time interval between two cycles is the time interval between two consecutive selections of the preliminary following target, which is usually 66 ms. Moreover, this judgment process needs to comprehensively consider the existence probability of the target (the probability of judging that it is a vehicle rather than other interference objects), the coincidence rate of the target with the self-trajectory (predicted driving trajectory) of the self-vehicle, the cut-in or cut-out state of the target relative to the self-trajectory of the self-vehicle, etc. Therefore, in the prior art, it usually takes at least 0.264 s to determine whether the target can be selected as the final following target after determining the preliminary following target. Thus, the above judgment process is time-consuming. And, in the prior art, the number of cycles required to determine whether a certain target is a vehicle is mainly affected by the existence probability of the target. The lower the existence probability, the more cycles are required for confirmation. Therefore, in some cases in the prior art, after determining the preliminary following target, it may be necessary to execute more than 4 cycles to determine whether it is a vehicle, thereby further reducing the efficiency of selecting the final following target.

[0047] To solve the above problems, the method of the present invention includes step S4: determining whether the preliminary following target is a vehicle. In the case where it is determined that the preliminary following target is a vehicle, the preliminary following target is determined as the final following target. Among them, determining whether the preliminary following target is a vehicle includes retrieving the judgment result of step S1. Since the preliminary following target is one of the objects in front of the self-vehicle, and in step S1, it has been determined whether each object in front is a vehicle. Therefore, once the preliminary following target is determined in step S3, it can be determined whether the preliminary following target is a vehicle in step S4 by retrieving the judgment result of step S1.

[0048] The method of the present invention further includes: repeatedly executing the above steps S1 - S4 at a fixed time interval until the self-vehicle 100 completely drives into the adjacent lane 130 to which it wants to change lanes.

[0049] Before the ego-vehicle 100 changes lanes, there are already front objects of the ego-vehicle 100 in the ego-lane 110 and the adjacent lane 130, and it is already possible to determine whether these front objects are vehicles (for example, through the sensor module of the ego-vehicle 100, such as a camera, etc.). Therefore, the method of the present invention actually determines whether each object in front of the ego-vehicle is a vehicle before the step S3 of determining the preliminary following target and even before the step S2 of detecting whether the ego-vehicle is performing a lane-changing behavior compared with the prior art. Therefore, once the preliminary following target is determined, it can be immediately known whether the preliminary following target is a vehicle, and as long as it is a vehicle, it can be directly determined as the final following target. Therefore, compared with the prior art, the steps of performing multiple loop judgments on whether it is a vehicle after determining the preliminary following target are omitted, and the selection of the final following target can be significantly accelerated.

[0050] In the case where the front objects existing in the ego-lane and the adjacent lane have been detected and it has been determined whether these front objects are vehicles before the ego-vehicle changes lanes, according to the method of the present invention, it is actually possible to select a following target for each lane before the ego-vehicle changes lanes. Specifically, in the case where the ego-vehicle has two adjacent lanes on the left and right, following targets can be selected for the left lane, the ego-lane, and the right lane. For example, the following target can be selected as the front object that is the closest to the ego-vehicle and has been confirmed as a vehicle in each lane. In this way, when it is detected that the ego-vehicle is performing a lane-changing behavior, if there is a following target that has been selected for the lane to which it is changing lanes, then this following target can be directly used as the final following target of the ego-vehicle, without having to perform at least 4 loop judgments on whether it is a vehicle as in the prior art.

[0051] Here, the concept of a detection value is introduced, which represents the number of times an object is continuously detected as a vehicle after several fixed time intervals (i.e., the time interval between continuously executing each of steps S1 - S4). Each object has a corresponding detection value for the lane (one or more of the ego-lane 110, the ego-trajectory 120, and the adjacent lane 130) in which it is located.

[0052] Specifically, when the host vehicle maintains driving in its own lane before changing lanes, the sensor module of the host vehicle detects whether each front object of the host vehicle in its own lane, its own trajectory, and the adjacent lane is a vehicle at fixed time intervals (usually 66 ms, which is the same as the time interval for executing steps S1 - S4 of the method of the present invention). This detection mainly determines whether the image content of the object obtained by the sensor module is a vehicle, and can comprehensively consider the coincidence rate of the object with the own trajectory of the host vehicle, the cut-in or cut-out state of the object relative to the own trajectory of the host vehicle, etc. The method of determining whether an object is a vehicle is known in the art and will not be elaborated here. If it is determined that the object is a vehicle in one detection, the detection value of the object in the lane where it is located is recorded as 1. If the object is continuously detected as a vehicle n times at fixed time intervals, the detection value of the object in the lane where it is located is recorded as n (n is an integer greater than or equal to 1). If the object is continuously detected as a vehicle n times at fixed time intervals and is detected as not a vehicle in the next detection, the detection value of the object in the lane where it is located is refreshed to 0.

[0053] Take Figure 2E as an example. Whether the host vehicle 100 has changed lanes as shown in the figure or not, the sensor module of the host vehicle 100 detects whether the front objects 1 and 2 of the host vehicle 100 in its own lane 110, its own trajectory 120, and the adjacent lane 130 are vehicles at fixed time intervals. Assume that the host vehicle 100 has been driving in its own lane 110 for 10 fixed time intervals before changing lanes, and the objects 1 and 2 are continuously detected as vehicles 10 times. Then, before changing lanes, the detection value of object 1 in its own lane 110 is assigned 10, and the detection value of object 2 in the adjacent lane 130 where it is located is assigned 10. When a lane change occurs, as Figure 2E shown, at this time, object 2 may enter the range of the own trajectory 120. Since object 2 was not in the range of the own trajectory 120 before, it is only assigned the detection value 10 in the adjacent lane 130 and not the detection value in its own trajectory 120. If object 2 is selected as the preliminary following target at this time, since it has been continuously determined that object 2 is a vehicle 10 times, even if it has just entered the own trajectory 120, it can be directly selected as the final following target without performing additional steps to determine whether the preliminary following target is a vehicle. In other words, when the host vehicle 100 starts to change lanes and object 2 enters the range of the own trajectory 120, the detection value of the preliminary following target originally in the adjacent lane 130 can be directly assigned to its detection value in the own trajectory 120. As long as this detection value is greater than or equal to a certain set detection value threshold, it indicates that object 2 has been continuously determined as a vehicle multiple times, so it can be determined as the final following target.

[0054] There is also another situation. When the host vehicle 100 is traveling in the host lane 110 before changing lanes, the sensor module of the host vehicle 100 may not continuously capture the object 2 in the adjacent lane 130, resulting in a relatively small detection value of the object 2 in the adjacent lane 130 when the host vehicle 100 changes lanes, and it may even be 0. Moreover, the detection value of the object 2 in the adjacent lane 130 may be smaller than its detection value in the host trajectory. At this time, the detection value of the object 2 in the host trajectory is compared with the set detection value threshold. If it is greater than or equal to the set detection value threshold, the object 2 can be determined as the final following target.

[0055] In short, when determining whether the preliminary following target is a vehicle in step S4 of the method of the present invention, it is necessary to compare the detection value of the preliminary following target in the host trajectory with its detection value in the adjacent lane. If the larger detection value of the two is greater than or equal to the set detection value threshold, the preliminary following target can be determined as the final following target, where the detection value is determined as the number of times the preliminary following target is continuously determined to be a vehicle at fixed time intervals.

[0056] Preferably, the set detection value threshold can be 4. If the larger value of the detection value of the preliminary following target in the host trajectory and its detection value in the adjacent lane is greater than or equal to 4, it indicates that the preliminary following target has been determined to be a vehicle in the last 4 fixed time intervals. Therefore, it can be determined as the final following target. Thus, after selecting the preliminary following target, only the size of its detection value needs to be quickly judged to determine the final following target, and the additional step of repeatedly judging whether it is a vehicle after selecting the preliminary following target is omitted. Thereby, the efficiency of selecting the following target during lane change in the adaptive cruise process can be greatly improved. However, the set detection value threshold can also be other values, and the present invention is not limited thereto.

[0057] Figure 3 The flowchart showing the method for selecting a following target during lane change in the adaptive cruise according to the present invention is as follows. The steps included in the method of the present invention are summarized as follows:

[0058] Step S1: Receive the detection signals of the front objects existing in the host lane and the adjacent lane, and judge whether each front object is a vehicle;

[0059] Step S2: Detect whether the host vehicle is performing a lane change behavior by detecting the lateral deviation degree of the host vehicle relative to the host lane;

[0060] Step S3: When the host vehicle is performing a lane change maneuver, detect the distances between the host vehicle and the front objects in the host lane and the adjacent lane to which the host vehicle is about to change lanes, and the overlap rate between the front object in the host lane and the host vehicle's self-trajectory, where the self-trajectory is the predicted driving trajectory of the host vehicle over a period of time based on its current driving actions; select the front object in the host lane or the adjacent lane as the preliminary following target.

[0061] Step S4: Based on the judgment result in Step S1, determine whether the preliminary following target is a vehicle. If it is determined that the preliminary following target is a vehicle, then determine the preliminary following target as the final following target; and

[0062] Repeat the above steps at fixed time intervals until the host vehicle has completely driven into the adjacent lane.

[0063] According to the method including the foregoing steps, a following target can be quickly selected when the vehicle changes lanes during adaptive cruise. After determining the final following target, the adaptive cruise control module ACC of the vehicle can automatically adjust the speed of the host vehicle according to the distance and relative speed between the final following target and the host vehicle, thereby keeping the distance between the host vehicle and the final following target within a suitable range to achieve stable following driving.

[0064] The present invention also provides a processor module for a vehicle with an adaptive cruise function, which is configured to execute the method according to the present invention described above.

[0065] The present invention also provides a system for selecting a following target when a vehicle changes lanes during adaptive cruise, including:

[0066] The above-mentioned processor module;

[0067] A sensor module, which is configured to obtain information including the position of the host vehicle in the host lane, the distances and relative positions between the front objects of the host vehicle existing in the host lane and the adjacent lane and the host vehicle, and the images of the front objects, and send the obtained information to the processor module; and

[0068] An adaptive cruise control module ACC, which is configured to receive the distance and relative speed between the final following target determined by the processor module and the host vehicle from the sensor module, and automatically adjust the speed of the host vehicle according to the distance and relative speed.

[0069] Since both the sensor module and the adaptive cruise control module ACC are hardware systems equipped in existing vehicles, therefore, the system according to the present invention can quickly select a following target when the vehicle changes lanes during adaptive cruise without making major modifications to the hardware system of the existing vehicle.

[0070] The present invention also provides a machine-readable storage medium storing executable instructions which, when executed by a processor module, implement the method for selecting a following vehicle target during lane change in the adaptive cruise according to the present invention.

[0071] The feasible but non-limiting embodiments of the method and system for selecting a following vehicle target during lane change in the adaptive cruise according to the present invention have been described in detail above with reference to the accompanying drawings. For those of ordinary skill in the art, modifications and supplements to the technology and structure, as well as recombination of the features in each embodiment, should be considered to be included within the scope of the present invention without departing from the scope and essence of the present disclosure set forth in the following claims. Therefore, these modifications and supplements that can be conceived under the teachings of the present invention should be regarded as a part of the present disclosure. The scope of the present disclosure is defined by the following appended claims and includes equivalent technologies known at the filing date of the present disclosure and equivalent technologies not yet foreseen.

Claims

1. Method for selecting a following vehicle target during a lane change in an adaptive cruise process, comprising: Step (S1): Receive detection signals of front objects existing in the own lane (110) and the adjacent lane (130) of the own vehicle (100), and determine whether each front object is a vehicle; Step (S2): Detect whether the own vehicle (100) is performing a lane change behavior by detecting the lateral deviation degree of the own vehicle (100) relative to the own lane (110); Step (S3): When the own vehicle (100) is performing a lane change behavior, detect the distance between the front object of the own vehicle (100) in the own lane (110) and the adjacent lane (130) to which the own vehicle (100) is about to change lanes and the own vehicle (100), and the coincidence rate between the front object in the own lane (110) and the own trajectory of the own vehicle (100), and select the front object in the own lane (110) or the adjacent lane (130) as the preliminary following vehicle target, where the own trajectory is the driving trajectory of the own vehicle (100) in a future period predicted based on the current driving action of the own vehicle (100); Step (S4): Determine whether the preliminary following vehicle target is a vehicle based on the judgment result of step (S1), and if it is determined that the preliminary following vehicle target is a vehicle, determine the preliminary following vehicle target as the final following vehicle target; and Repeat the above steps at a fixed time interval until the own vehicle (100) completely drives into the adjacent lane (130).

2. The method according to claim 1, wherein step (S2) includes detecting the lateral distance between the center (101) of the front bumper assembly of the own vehicle (100) and the left and right solid lane lines (111, 112) of the own lane (110), and if the lateral distance is greater than a set distance threshold, it is determined that the own vehicle (100) is performing a lane change behavior.

3. The method according to claim 1, wherein in step (S3), if there is only one front object of the own vehicle (100) in the own lane (110) and the adjacent lane (130), this front object is selected as the preliminary following vehicle target.

4. The method according to claim 1, wherein in step (S3), if there are front objects of the own vehicle (100) in both the own lane (110) and the adjacent lane (130) and the distance between the front object in the own lane (110) and the own vehicle (100) is greater than or equal to the distance between the front object in the adjacent lane (130) and the own vehicle (100), the front object in the adjacent lane (130) is selected as the preliminary following vehicle target.

5. The method according to claim 1, wherein in the step (S3), if there are front objects of the host vehicle (100) in both the host lane (110) and the adjacent lane (130), and the distance between the front object in the host lane (110) and the host vehicle (100) is less than the distance between the front object in the adjacent lane (130) and the host vehicle (100), and the ratio of the width of the intersection part between the front object in the host lane (110) and the host trajectory of the host vehicle (100) to the total width of the front object in the host lane (110) reaches a set coincidence rate threshold, then the front object in the host lane (110) is selected as the preliminary following target; otherwise, the front object in the adjacent lane (130) is selected as the preliminary following target.

6. The method according to any one of claims 1 to 5, wherein the step (S4) includes comparing the detection value of the preliminary following target in the host trajectory (120) with the detection value in the adjacent lane (130). If the larger detection value of the two is greater than or equal to a set detection value threshold, then the preliminary following target is determined as the final following target, where the detection value is determined as the number of times the preliminary following target is continuously determined to be a vehicle after a fixed time interval.

7. The method according to any one of claims 1 to 5, further comprising: After determining the final following target, automatically adjusting the speed of the host vehicle (100) according to the distance and relative speed between the final following target and the host vehicle (100).

8. A processor module for a vehicle with an adaptive cruise function, the processor module being configured to: Receive detection signals of front objects of the host vehicle (100) in the host lane (110) and the adjacent lane (130), and determine whether each front object is a vehicle; Detect whether the host vehicle (100) is performing a lane change behavior by detecting the lateral deviation degree of the host vehicle (100) relative to the host lane (110); When the host vehicle (100) is performing a lane change behavior, detect the distance between the front object of the host vehicle (100) in the host lane (110) and the adjacent lane (130) to which the host vehicle (100) is changing lanes and the host vehicle (100), and the coincidence rate between the front object in the host lane (110) and the host trajectory of the host vehicle (100), and select the front object in the host lane (110) or the adjacent lane (130) as the preliminary following target, where the host trajectory is the driving trajectory in a future period predicted based on the current driving action of the host vehicle (100); Determine whether the preliminary following target is a vehicle based on the determination result of whether each front object is a vehicle. When it is determined that the preliminary following target is a vehicle, determine the preliminary following target as the final following target; and Repeat the above steps at fixed time intervals until the host vehicle (100) has completely moved into the adjacent lane (130).

9. A system for selecting a following target during a lane change in an adaptive cruise process, comprising: The processor module according to claim 8; A sensor module configured to obtain information including the position of the host vehicle (100) in the host lane (110), the distances and relative positions between the host vehicle (100) and the front objects in the host lane (110) and the adjacent lane (130), and images of the front objects, and to send the obtained information to the processor module; And An adaptive cruise control module configured to receive from the sensor module the distance and relative speed between the final following target determined by the processor module and the host vehicle (100), and to automatically adjust the speed of the host vehicle (100) according to the distance and relative speed.

10. A machine-readable storage medium storing executable instructions that, when executed by a processor module, implement the method according to any one of claims 1 to 7.

Images