WO2018223890A1

WO2018223890A1 - Automatic lane changing control method and device for vehicle

Info

Publication number: WO2018223890A1
Application number: PCT/CN2018/089261
Authority: WO
Inventors: 丁晨曦; 何彬; 章健勇
Original assignee: 蔚来汽车有限公司
Priority date: 2017-06-09
Filing date: 2018-05-31
Publication date: 2018-12-13
Also published as: CN109050533A; CN109050533B

Abstract

An automatic lane changing control method for a vehicle, comprising: determining whether a vehicle is in a lane changeable operating condition; in the case that the vehicle is in the lane changeable operating condition, prompting the driver to activate an automatic lane changing mode; and in response to activation of the automatic lane changing mode, executing the following steps: (a) selecting a target lane changing track from a predetermined lane changing track set according to road conditions and an estimated lane changing time; (b) determining a steering wheel rotation angle request according to a curvature change in the target lane changing track; and (c) controlling, according to the steering wheel rotation angle request, the vehicle to change lanes along the target lane changing track. The present invention also provides a corresponding automatic lane changing control device for a vehicle

Description

Vehicle automatic lane changing control method and device

Technical field

The present invention relates generally to the field of autonomous driving technology and, in particular, to a solution for vehicle adaptive control.

Background technique

At this stage, automotive control technology is moving in the direction of “smart cars”. On the basis of ordinary vehicles, more and more advanced sensors, controllers, actuators and other devices have been added to enable the vehicle to have intelligent environmental sensing capability, which can automatically analyze the safe and dangerous state of the vehicle, and make the vehicle according to people. Willingness to reach the destination, and ultimately achieve the purpose of replacing the person to operate.

Lane change is a common operation in the driving process of a car. At present, most of the most common lane change assist systems are installed with millimeter-wave radar on both sides of the vehicle to monitor the potential collision risk in front. The driver needs to monitor the vehicle behind the vehicle in real time and decide whether it can be safely changed. There are also blind spot detection systems that detect the rear of the vehicle through the vehicle's tail ultrasound system or the surround camera, alerting the driver to possible collision hazards. These lane change assist systems have not realized the automatic change of lanes in the true sense. In fact, it is still the driver's choice of the lane change timing and lane change trajectory. This does not reduce the burden on the driver, nor does it significantly reduce the probability of an accident. On the other hand, products that offer automatic lane change assistance on the market are often based on high-cost hardware configurations and large-scale data collection, which makes most consumers still unable to enjoy the convenience and safety of the vehicle's assisted driving system.

Therefore, it is desirable to design an automatic vehicle lane change scheme suitable for urban traffic scenarios that can be implemented at a lower cost.

Summary of the invention

Therefore, what is desired is to design an automatic driving scheme that can adapt to various road scenes while securing driving safety, especially to adapt to traffic intersections.

In view of this, the present invention provides a vehicle automatic lane change control method, which includes: determining whether the vehicle is in a changeable lane condition; prompting the driver to activate the automatic lane change mode when the vehicle is in a changeable working condition Performing the following steps in response to activation of the automatic lane change mode: (a) selecting a target lane change trajectory from a predetermined set of lane change trajectories according to road conditions and estimated lane change time; (b) based on the target lane change trajectory a curvature change to determine a steering wheel angle request; and (c) controlling the vehicle to change lanes along the target lane change trajectory according to the steering wheel angle request.

The method as described above, wherein each lane change trajectory in the set of lane change trajectories is fitted to a correspondence relationship between a longitudinal motion distance x and a lateral motion distance y of the vehicle in a Cartesian coordinate system y=f (x), wherein the x-axis is the forward direction of the vehicle, the y-axis is the vertical body direction, the coordinate origin indicates the position where the vehicle centroid is located at the start of the lane change, and the corresponding relationship curve is changed from the origin of the coordinate Start point.

The method as described above, wherein the corresponding relationship curve is a hyperbola or a convoluted line.

The method as described above, wherein the curvature change of each lane change trajectory is expressed as a curvature change function having a lateral motion distance y as a unique variable determined according to the correspondence relationship curve.

a method as described above, wherein the curvature change function is according to a formula

Determined from the correspondence curve, where y=f(x), C represents curvature,

Represents the second derivation of the y function,

Represents a derivation of the y function.

The method as described above, wherein determining the steering wheel angle request based on the curvature change of the target lane change trajectory comprises: detecting a real-time lateral movement distance of the vehicle; calculating a real-time curvature C of the target trajectory curve according to the curvature change function; The real-time curvature C adjusts the steering wheel angle request.

The method as described above, wherein detecting the real-time lateral movement distance of the vehicle comprises detecting a vertical distance change of the host vehicle and the reference lane line in real time according to a predetermined time interval.

The method as described above, wherein the lane change time is estimated based on one or more of a road condition, a driver's comfort level, and a driver's driving habit.

The method as described above, wherein determining whether the vehicle is in the changeable lane condition comprises: detecting whether there is a lane line indicating the changeable lane; estimating the vehicle separately from the vehicle in front of the lane, the vehicle in front of the target lane, and the vehicle behind the target lane The collision time between; and when the lane line indicating the changeable lane is detected and the estimated lane change time is less than the collision time, the vehicle is judged to be in the changeable lane condition.

The method as described above, wherein the estimating the collision time comprises: detecting respective vehicle speeds of the vehicle, the vehicle in front of the lane, the vehicle in front of the target lane, and the vehicle behind the target lane; detecting the vehicle in front of the vehicle and the vehicle in front of the lane, and the vehicle in front of the target lane a relative distance to the vehicle behind the target lane; and estimating the collision time based on the vehicle speed and the relative distance.

The method as described above, wherein determining whether the vehicle is in the changeable lane condition further comprises: determining a target lane change vehicle speed according to the road condition; and determining that the target lane change speed is within a range indicated by the road speed limit flag The car is in a changeable working condition.

The method as described above, further comprising enabling the driver to activate the automatic turning mode by turning a turn signal.

The method as described above, further comprising, in the case of a curved road, detecting a lane line curvature change and compensating for the steering wheel angle request in accordance with the lane line curvature change.

The method as described above, further comprising: detecting a vehicle speed of the preceding vehicle in response to activation of the automatic lane change mode, determining whether there is a collision risk with the preceding vehicle according to the vehicle speed of the preceding vehicle, and performing step a) to the risk of collision c).

In another aspect, the present invention also provides a vehicle automatic lane change control apparatus including a memory, a processor, and a computer program stored on the memory, wherein when the processor The vehicle automatic turning control device is caused to perform the method as described above when the computer program is executed.

DRAWINGS

The foregoing and other objects, features, and advantages of the invention will be apparent from

1 is a flow chart of a method of automatic lane change control of a vehicle according to an example of the present invention.

2a is a schematic illustration of a vehicle lane change trajectory in accordance with one example of the present invention.

Figure 2b illustrates the correspondence between the vehicle's forward distance and the curvature change of the lane change trajectory, in accordance with one embodiment of the present invention.

3 is a block diagram of a vehicle automatic lane change control device according to an example of the present invention.

detailed description

BRIEF DESCRIPTION OF THE DRAWINGS Exemplary embodiments of the present invention will now be described with reference to the drawings, wherein like reference numerals refer to the like. The examples described below are provided to enable those skilled in the art to understand the invention, and the examples are intended to be illustrative and not limiting. The illustrations of the various elements, components, modules, devices and device bodies in the figures are only illustrative of the existence of such elements, components, modules, devices, and device bodies, and also indicate the relative relationship between them, but are not intended to limit their specificity. Shape; the relationship of the steps in the flowchart is not limited to the order given, and can be adjusted according to the actual application without departing from the scope of protection of the present application.

As mentioned in the background art, there is currently no cost-effective and efficient vehicle automatic lane change control scheme. The present invention provides a vehicle lane change assisting scheme capable of realizing automatic lane change control of a vehicle in an urban traffic scene without requiring high-end assisted navigation configurations such as high-precision maps and GPS, and large-scale data processing.

The vehicle assisted driving scheme provided by the present invention can replace the driver's responsibility to control the vehicle for a period of time, especially if a lane change is required. According to the method of the present invention, the vehicle can have both vertical and horizontal automatic control, but the driver still needs to monitor the driving activity, that is, observe the surrounding situation, and the driver can share control with the vehicle, and must be on standby at the vehicle. When you exit the automatic control, you can always pick it up.

1 is a flow chart of a method of automatic turning control of a vehicle according to an example of the present invention. As shown in FIG. 1, the method first includes determining in step 11 whether the vehicle is in a changeable lane condition. In the context of the present invention, a turntable condition refers to the condition of the road in which the vehicle can be transferred from one lane to another.

In some embodiments, determining whether the vehicle is in a changeable lane condition includes detecting whether there is a lane line indicating a changeable lane. In an implementation, one or more front cameras disposed on the vehicle body may be used to identify whether the lane line of the lane indicates a changeable lane position according to traffic regulations, thereby determining whether the vehicle is in an area where lane change is permitted. In addition, determining whether the vehicle is in a convertible condition may further include determining whether a collision may occur between the vehicle and the vehicle in which it is located and the vehicle in the target lane.

In some embodiments, determining whether a collision is likely to occur includes estimating the time required for the vehicle to change lanes on the one hand, and estimating the vehicle between the vehicle and the vehicle in front of the lane, the vehicle in front of the target lane, and the vehicle behind the target lane. Possible collision time. Generally, it can be judged that the lane change has no collision risk if the estimated lane change time is less than the possible collision time, and that the current vehicle is in a turnable condition when the lane line indicating the changeable lane is detected at the same time. .

In terms of estimated lane change time, one or more of road conditions, driver comfort, and driver's driving habits may be based. For example, the time required for lane change may be first estimated according to different road widths and degrees of straightness, which may be a time range. It is also possible to further consider the driver's habits and comfort in this time range interval. For example, it can be observed by the driver's driving habits for a long time that the driver prefers faster driving, and his own operation style when changing lanes is also biased, so that a relatively reasonable range can be selected. Short lane change time. In addition, considering the comfort of the driver and the passengers in the car, the estimated lane change time can also be adjusted, for example, the lane change time is not too short, so that the passengers in the car feel the violent shaking and cause discomfort or cause hurt.

In terms of estimating the risk of collision, in some embodiments, the vehicle, the vehicle in front of the lane, and the vehicle in front of the lane can be detected by fusing information from various sensors such as a front camera system, a front millimeter wave radar, a lateral millimeter wave radar, and the like. The respective vehicle speeds of the vehicle in front of the target lane and the vehicle behind the target lane, and the relative distance between the vehicle and the vehicle in front of the lane, the vehicle in front of the target lane, and the vehicle behind the target lane. Based on the detected vehicle speed and relative distance, the collision time between the vehicle and the vehicle in front of the lane, the vehicle in front of the target lane, and the vehicle behind the target lane can be respectively determined.

Further, in order to accurately determine whether the vehicle is in a changeable working condition, the target lane changing speed can also be determined according to the road condition, for example, according to parameters such as the speed of the front and rear vehicles on the target lane and the relative distance from the vehicle. The goal is to change lane speed. For example, the speed limit flag of the road is identified by the front camera, and the vehicle is judged to be in the changeable road condition if the determined target lane change speed is within the range indicated by the road speed limit indicator. The determined target lane change speed can also be a certain range of vehicle speed. As long as the vehicle speed range and the road speed limit intersect, it can also be judged that the vehicle is in the changeable lane condition.

Those skilled in the art will appreciate that the determination of the interchangeable operating conditions in the context of the present invention is not limited to the circumstances of the above examples. For example, in the case where the vehicle is not detected within a certain range, it may not be necessary to judge the risk of the collision, but it may be judged whether or not the vehicle is in the changeable lane condition only by determining whether the vehicle is in the changeable area. Accordingly, the lane change time can be estimated more from the driver's driving habits and comfort.

Further, if the environmental condition is satisfied, that is, the vehicle is in a road condition capable of lane change, the driver may be prompted to activate the automatic lane change mode in step 12, and it is determined in step 13 whether the automatic lane change mode is activated. . In some examples, the driver can be enabled to activate the automatic turn mode by turning the turn signal and to determine that the automatic lane change mode can be entered in response to activation of the turn signal. In still other examples, the driver may activate the automatic lane change mode by setting a special mode start/switch button.

Steps 14 through 16 can be performed with the automatic lane change mode activated. This may also include several situations, such as whether to perform a lane change immediately after the driver activates the automatic lane change mode or to make an autonomous decision after observing a specific time for the road condition, and determine whether to perform steps 14 to 16. For example, after the automatic lane changing mode is activated, it is possible to detect the vehicle speed of the preceding vehicle, determine whether there is a collision risk with the preceding vehicle according to the vehicle speed of the preceding vehicle, and in case of a collision risk or only the vehicle speed of the preceding vehicle is lower than a certain speed. Steps 14 through 16 are performed only if the limits are not necessarily dangerous.

As shown in FIG. 1, the target lane change trajectory is selected from the predetermined set of lane change trajectories according to the road condition and the estimated lane change time in step 14. In the implementation, the road shape in the city can be classified in advance to determine the corresponding lane change trajectories. Further, the lane change trajectory in the same road shape can also be refined according to different lane change times. For example, some drivers may prefer a faster lane change experience during the lane change process, while others may prefer a relatively gentle lane change process, which may correspond to different range of lane change trajectories. In general, the predetermined driving trajectory should conform to the actual road conditions on the one hand and the actual driving behavior on the other hand.

A set of lane change trajectories can be obtained by pre-analysis, statistics and acquisition of these lane change trajectories. In order to be able to conveniently select a target lane change trajectory from the set, in the present invention, each lane change trajectory is correspondingly stored corresponding to both the road condition and the lane change time for extraction in the actual lane change process. Thus, the automatic lane changing scheme provided by the present invention can widely adapt to different traffic conditions in the city, and at the same time, enables the driver to obtain the best driving experience at the same time. In contrast, most of the current mass-changing lane-assisting systems use only fixed lane-change trajectory models, regardless of trajectory planning under different road conditions.

In some embodiments, each lane change trajectory can be fitted to a correspondence curve y=f(x) between the longitudinal motion distance x and the lateral motion distance y of the vehicle in a Cartesian coordinate system, where the x-axis is The direction of the car is forward, the y-axis is the direction of the vertical body, and the origin of the coordinate indicates the position of the center of mass of the vehicle at the start of the lane change, and the corresponding relationship curve can be the starting point of the lane change, which is convenient for calculation. Figure 2a is a schematic illustration of such a lane change trajectory.

The corresponding relationship curve representing the lane change trajectory may be a hyperbola or a convoluted line. Those skilled in the art can understand that in the actual driving process, the gyroscopic line and the hyperbola are more in line with the driver's driving behavior habits with respect to the circular lane change trajectory.

Accordingly, in order to accurately select the lane change trajectory, in addition to detecting the road condition, it is necessary to estimate the required lane change time in the automatic lane change scheme of the present invention. As already described above, the required lane change time can be estimated based on one or more of road conditions, driver comfort, and driver's driving habits. For example, the time required for lane change may be first estimated according to different road widths and degrees of straightness, which may be a time range. It is also possible to further consider the driver's habits and comfort in this time range interval. For example, it can be observed by the driver's driving habits for a long time that the driver prefers faster driving, and his own operation style when changing lanes is also biased, so that a relatively reasonable range can be selected. Short lane change time. In addition, considering the comfort of the driver and the passengers in the car, the estimated lane change time can also be adjusted, for example, the lane change time is not too short, so that the passengers in the car feel the violent shaking and cause discomfort or cause hurt.

In step 15, the steering wheel angle request is determined based on the curvature change of the selected target lane change trajectory.

In some embodiments, the change in curvature of each lane change trajectory can be represented as a curvature change function with the lateral motion distance y of the vehicle as a unique variable. This curvature change function can be determined, for example, according to a correspondence curve y=f(x) between the lateral movement distance y of the vehicle and the longitudinal movement distance x. Thus, according to this one-to-one correspondence, the curvature change of the lane change trajectory can be determined by detecting only the lateral movement distance of the vehicle during the lane changing process, thereby further providing the steering wheel angle request. Figure 2b illustrates the correspondence between the vehicle's forward distance and the curvature change of the lane change trajectory, in accordance with one embodiment of the present invention. Figure 2b can also be considered to indicate the change in curvature of the lane change trajectory over time if the vehicle is a uniform speed change lane.

In some embodiments, the vertical distance change of the host vehicle and the reference lane line may be detected in real time at predetermined time intervals to determine the y value. According to the above correspondence curve, although it is possible to obtain a curvature change by obtaining the vehicle forward distance, the detection of the vehicle forward distance requires real-time positioning. The solution for determining the curvature change by detecting only the lateral movement distance will greatly reduce the sensor configuration requirements compared to the solution for obtaining the vehicle advance distance by real-time positioning.

In some embodiments, the curvature change function with the lateral motion distance y as a unique variable may be according to a formula

It is determined from the corresponding relationship curve representing the lane change trajectory. As mentioned before, y=f(x), C means curvature,

Represents the second derivation of the y function,

Represents a derivation of the y function.

In the case of a curved road, it is also possible to detect the lane line curvature change by the front camera and compensate for the steering wheel angle request according to the lane line curvature change.

Finally, in step 16, the vehicle is controlled to change lanes along the target lane change trajectory according to the steering wheel angle request. Thus, the driver can completely not participate in the control of the vehicle during the entire lane change process.

The vehicle automatic lane change control method provided by the present invention can be applied to a rich urban traffic scene, so that the driver can realize fully automatic driving when changing lanes, and is not limited to whether the vehicle has a high-end navigation configuration. At the same time, the method of the invention can make the vehicle only equipped with some commonly used sensor devices, a complex data processing system that does not require massive data, and does not need to monitor the actual motion trajectory of the vehicle in real time to drive along the target trajectory, which can greatly Reducing the cost of data processing reduces the threshold in both hardware and software, and contributes to the widespread implementation of automatic lane change schemes in the market, thereby improving driving convenience, comfort and safety.

3 is a block diagram of a vehicle automatic turning control device according to an example of the present invention. As shown in FIG. 3, the vehicle automatic turning control device 300 includes a memory 31 and a processor 33 in which computer programs are stored on the memory 31, and these computer programs, when executed by the processor 33, can cause the vehicle automatic turning control device to execute The vehicle automatic turning control method according to the present invention as described above.

The vehicle automatic turning control device 300 can be implemented separately or integrated in the electronic control unit ECU of the vehicle. In the latter case, device 300 can be implemented by sharing a processor and memory in the ECU.

It should be noted that the above specific embodiments are merely illustrative of the technical solutions of the present invention and are not limited thereto. While the invention has been described in detail herein with reference to the preferred embodiments of the embodiments of the invention All are covered by the scope of the claimed invention.

Claims

A vehicle automatic lane change control method includes:

Determine whether the vehicle is in a changeable working condition;

Prompt the driver to activate the automatic lane changing mode when the vehicle is in a changeable working condition;

In response to the activation of the automatic lane change mode, the following steps are performed:

(a) selecting a target lane change trajectory from a predetermined set of lane change trajectories according to road conditions and estimated lane change time;

(b) determining a steering wheel angle request based on a change in curvature of the target lane change trajectory;

(c) controlling the vehicle to perform lane change along the target lane change trajectory according to the steering wheel angle request.
The method of claim 1 wherein each of the lane change trajectories in the set of lane change trajectories is fitted to a correspondence between a longitudinal motion distance x and a lateral motion distance y of the vehicle in a Cartesian coordinate system y=f(x), wherein the x-axis is the forward direction of the vehicle, the y-axis is the vertical body direction, the coordinate origin indicates the position of the vehicle centroid at the start of the lane change, and the corresponding relationship curve is the coordinate origin Change the starting point of the lane.
The method of claim 2 wherein said corresponding relationship curve is a hyperbola or a convoluted line.
The method of claim 2, wherein the curvature change of each lane change trajectory is represented as a curvature change function determined by the corresponding relationship curve with the lateral motion distance y as a unique variable.
The method of claim 4, wherein the curvature change function is according to a formula
Determined from the correspondence curve, where y=f(x), C represents curvature,
Represents the second derivation of the y function,
Represents a derivation of the y function.
The method of claim 4, wherein determining the steering wheel angle request based on the change in curvature of the target lane change trajectory comprises:

Detecting the real-time lateral movement distance of the vehicle;

Calculating a real-time curvature C of the target trajectory curve according to the curvature change function;

The steering wheel angle request is adjusted based on the real-time curvature C.
The method of claim 6, wherein detecting the real-time lateral movement distance of the vehicle comprises detecting a vertical distance change of the host vehicle and the reference lane line in real time in accordance with a predetermined time interval.
The method of claim 1 wherein the lane change time is estimated based on one or more of a road condition, a driver's comfort level, and a driver's driving habits.
The method of claim 1 wherein determining whether the vehicle is in a changeable condition comprises:

Detecting whether there is a lane line indicating that the lane can be changed;

Estimating the collision time between the vehicle and the vehicle in front of the lane, the vehicle in front of the target lane, and the vehicle behind the target lane;

When the lane line indicating the changeable lane is detected and the estimated lane change time is less than the collision time, it is judged that the vehicle is in the changeable lane condition.
The method of claim 9 wherein estimating the collision time comprises:

Detecting the respective vehicle speeds of the vehicle, the vehicle in front of the lane, the vehicle in front of the target lane, and the vehicle behind the target lane;

Detecting the relative distance between the vehicle and the vehicle in front of the lane, the vehicle in front of the target lane, and the vehicle behind the target lane;

The collision time is estimated based on the vehicle speed and the relative distance.
The method of claim 9 wherein determining whether the vehicle is in a changeable operating condition further comprises:

Judging the target lane change speed according to the road condition;

When the target lane change speed is within the range indicated by the road speed limit flag, it is determined that the vehicle is in the changeable lane condition.
The method of claim 1 further comprising enabling a driver to activate said automatic turning mode by turning a turn signal.
The method of claim 1 further comprising, in the case of a curvilinear road, detecting lane line curvature changes and compensating for said steering wheel angle request based on lane line curvature changes.
The method of claim 1, further comprising detecting a vehicle speed of the preceding vehicle in response to activation of the automatic lane change mode, determining whether there is a collision risk with the preceding vehicle based on the vehicle speed of the preceding vehicle, and performing the step in the event of a collision risk a) to c).
A vehicle automatic lane change control device including a memory, a processor, and a computer program stored on the memory, wherein when the computer program is executed on the processor The vehicle automatic turning control device performs the method according to any one of claims 1-14.