An intelligent obstacle avoidance method for lane centering

Abstract

The present invention provides an intelligent obstacle avoidance method for lane centering. This method uses sensing devices such as millimeter-wave radar and lidar to detect obstacles. The lateral and longitudinal distances of the obstacle relative to the vehicle are used to calculate the offset value the vehicle needs to avoid the obstacle. This offset is then applied to the lane centerline of the lane centering algorithm. This allows the LCC to maintain a certain degree of distance from the obstacle within the lane, reducing the driver's psychological stress.

Description

Technical Field

[0001] The present invention relates to the field of intelligent connected vehicles, and in particular to an intelligent obstacle avoidance method for lane centering. Background Art

[0002] The proportion of intelligent connected vehicles with L2 assisted driving functions is increasing. The lateral control principle of the L2 LCC (lane centering) function is to sense the left and right lane lines through the camera and output the lane line polynomial based on the vehicle coordinate system (for example, ), where the constant term C0 in the lane line cubic polynomial represents the lateral distance from the lane line to the center of the vehicle's rear axle. The polynomial coefficients of the left and right lane lines (C0, C1, C2, C3) are processed separately (filtering, etc.) and fused into the lane centerline. The control algorithm uses the fitted lane centerline as the tracking trajectory, calculates the specific lateral control amount, and sends it to the chassis to make the vehicle's center axis move along the lane centerline. However, since the definition of L2 autonomous driving function is still only assisted driving, the driver needs to take over the vehicle in some situations. Therefore, most current LCCs only support lane centering. However, this function cannot take lane avoidance measures for conditions such as fences on the side of the road, walls on the side of the road, and vehicles passing in the adjacent lane. The driver needs to actively intervene in the steering wheel to keep the vehicle at a safe distance from the obstacle in the lateral direction. When there is a fence in the lane where the vehicle is located or there are vehicles passing in the adjacent lane, the driver will feel tremendous psychological pressure if the vehicle still drives along the lane centerline.

[0003] Therefore, there is a need for an intelligent obstacle avoidance method for lane centering that can achieve active avoidance of the vehicle within the lane and improve driving safety and driving comfort. Summary of the Invention

[0004] In response to the deficiencies in the prior art, the purpose of the present invention is to provide an intelligent obstacle avoidance method for lane centering. After the method is applied to a vehicle, after the lane centering function is turned on, when there is a fence close to the side of the lane / there is a vehicle passing in the adjacent lane, etc., the vehicle can actively avoid it laterally within the lane, thereby reducing the psychological pressure on the driver.

[0005] In order to achieve the above technical effects, the present invention adopts the following technical solutions:

[0006] An intelligent obstacle avoidance method for lane centering includes the following steps:

[0007] Step S1: Obtain the left lane line polynomial, the right lane line polynomial, and the lane width based on the vehicle coordinate system; obtain the horizontal and vertical distances from the obstacle to the center of the vehicle, and determine whether there are obstacles on the left and right sides of the vehicle within the obstacle detection range;

[0008] Step S2: Calculate the lane centerline and the lateral drivable range within the lane by polynomial fusion of the left and right lane lines; calculate the lateral drivable range of the vehicle based on the obstacle by the lateral distance from the obstacle to the center of the vehicle;

[0009] Step S3: combining the lateral drivable range based on the obstacle and the lane centerline to obtain a driving centerline closest to the lane centerline within the lateral drivable range based on the obstacle, and calculating an offset distance of the driving centerline relative to the lane centerline;

[0010] Step S4: limiting the offset distance using the lateral drivable range within the lane calculated in step S2 to obtain a target offset distance;

[0011] Step S5: offsetting the lane centerline by the target offset distance to obtain the target driving centerline;

[0012] Step S6: The lane centering assist system (LCC) uses the target driving centerline as the target lane centering trajectory, calculates the lateral control command for the current vehicle along the trajectory, and sends it to the chassis;

[0013] Step S7: The vehicle implements an intelligent avoidance function based on lane centering.

[0014] Preferably, in step S1, the left lane line polynomial, the right lane line polynomial and the lane width are acquired through a perception camera based on a vehicle coordinate system.

[0015] Preferably, in step S1, the horizontal and vertical distances from the obstacle to the center of the vehicle are acquired by millimeter wave radar or laser radar.

[0016] Preferably, in step S2, if the lateral distance of the left obstacle relative to the left edge of the vehicle body is greater than the lateral distance of the right obstacle relative to the right edge of the vehicle body, it means that the distance between the left and right obstacles is sufficient for the vehicle to pass safely, and the vehicle can drive normally within the drivable range based on the obstacles; if the lateral distance of the left obstacle relative to the left edge of the vehicle body is less than or equal to the lateral distance of the right obstacle relative to the right edge of the vehicle body, it means that the distance between the left and right obstacles is not sufficient for the vehicle to pass at a safe distance, and the vehicle needs to drive centered between the two obstacles.

[0017] Preferably, if it is detected that there are no obstacles on both the left and right sides, the lateral distances of the obstacles on the left and right sides are preset values, and the lateral drivable range of the vehicle based on the obstacles is also a fixed value.

[0018] Preferably, in step S3, if the lane centerline is within the vehicle drivable range based on the obstacle, the offset distance of the driving centerline relative to the lane centerline is 0.

[0019] Preferably, the method is applicable to L2 level lane centering assistance function.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] 1. The intelligent obstacle avoidance method for lane centering provided by the present invention enables a vehicle with lateral-related functions (such as LCC and TJA) enabled in Level 2 autonomous driving to automatically maintain a safe distance from obstacles within its lane when obstacles exist or are about to appear on the left or right side of the vehicle (for example, a fence in the lane in which the vehicle is located, a truck passing in an adjacent lane, etc.), thereby greatly reducing the psychological stress on the driver in such conditions.

[0022] 2. The intelligent obstacle avoidance method for lane centering provided by this invention has low sensor requirements. It only requires the carrier vehicle to have lane centering capabilities and the necessary configuration: a single camera to detect lane markings and a few sensors to detect surrounding obstacles. It also requires very low computing power. This reduces implementation costs for automakers and provides a better user experience for drivers. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0024] Figure 1 Flowchart of the intelligent obstacle avoidance method for lane centering of the present invention;

[0025] Figure 2 Schematic diagram of vehicle coordinate system and direction;

[0026] Figure 3 Schematic diagram of obstacle detection in step S1 of the embodiment;

[0027] Figure 4 Schematic diagram of the drivable range (light green area) within the lane in the embodiment;

[0028] Figure 5 A schematic diagram of the lateral drivable range of a vehicle based on obstacles with no obstacles on both sides;

[0029] Figure 6 A schematic diagram of the lateral drivable range of a vehicle with an obstacle on the left side based on the obstacle;

[0030] Figure 7 A schematic diagram of the lateral driving range of a vehicle based on obstacles when the distance between the left and right obstacles is sufficient for the vehicle to pass safely;

[0031] Figure 8A schematic diagram of the vehicle's lateral drivable range based on obstacles when the distance between the left and right obstacles is not sufficient for the vehicle to maintain a safe distance through;

[0032] Figure 9 This is a schematic diagram showing the lane centerline on the left side of the obstacle's drivable range;

[0033] Figure 10 This is a schematic diagram showing the lane centerline on the right side of the obstacle's drivable range;

[0034] Figure 11 This is a schematic diagram of the lane centerline within the drivable range of obstacles;

[0035] Figure 12 This is a flowchart of the steps of the intelligent obstacle avoidance method for lane centering of the present invention. DETAILED DESCRIPTION

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Generally, the components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations.

[0037] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for protection, but merely represents selected embodiments of the present application. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments in the present application without creative work are within the scope of protection of the present application.

[0038] like Figure 1 As shown, this embodiment provides an intelligent obstacle avoidance method for lane centering. The method detects obstacles (including fences, curbs, vehicles, etc.) through sensing equipment such as millimeter wave radar / lidar, and uses the lateral and longitudinal distances of the obstacles relative to the vehicle to fit and calculate the offset value that the vehicle needs to move away from the obstacles and compensate it to the lane centerline of the lane centering algorithm. This allows the LCC to achieve a certain degree of distance from the obstacles within the lane, reducing the psychological pressure on the driver. The method specifically includes the following steps: (All formulas / distances in the following steps are based on the following Figure 2 The center of the rear axle of the vehicle shown is used as the coordinate axis, and the body axis is divided into x and y axes, with positive on the left and negative on the right, and positive on the front and negative on the rear)

[0039] The parameters in this embodiment are defined as follows:

[0040] car_width: vehicle width;

[0041] dl_min: The minimum lateral distance between the vehicle body edge and the lane line;

[0042] do_min: The minimum lateral distance required between the edge of the vehicle and the obstacle;

[0043] RL_dis: Right Line distance, the lateral distance between the right lane line and the right edge of the vehicle body;

[0044] LL_dis: Left Line distance, the lateral distance between the left lane line and the left edge of the vehicle body;

[0045] RObs_dx: Right Obstacledx, the distance of the right obstacle relative to the vehicle's x-axis;

[0046] LObs_dx: Left Obstacledx, the distance of the left obstacle relative to the vehicle's x-axis;

[0047] RO_dis: Right Obstacledistance, the lateral distance of the right obstacle relative to the right edge of the vehicle body;

[0048] LO_dis: Left Obstacledistance, the lateral distance of the left obstacle relative to the left edge of the vehicle body;

[0049] The cubic polynomial of the left lane line:

[0050] The cubic polynomial of the right lane line:

[0051] Lane centerline cubic polynomial:

[0052] Note: The constant term C0 in the lane line cubic polynomial represents the lateral distance from the lane line to the center of the vehicle's rear axle.

[0053] The intelligent obstacle avoidance method for lane centering provided in this embodiment must meet three key conditions: the vehicle remains within the lane, avoids left and right obstacles, and drives close to the lane center. Prioritizing the vehicle to remain within the lane is the top priority, followed by avoiding obstacles while meeting this condition, and finally, driving as close to the lane center as possible while meeting the first two conditions. The method provided in this embodiment first simultaneously satisfies the two lower-priority conditions—avoiding left and right obstacles and driving close to the lane center—and then constrains these results using the highest-priority condition of maintaining the vehicle within the lane, achieving a logically clear avoidance strategy. The method for avoiding left and right obstacles includes obtaining the lateral and longitudinal distances of the obstacles in the vehicle coordinate system, screening valid obstacles based on a set obstacle detection range, and finally calculating the possible range for avoiding obstacles. The step of driving the vehicle close to the lane center includes selecting the location closest to the lane centerline as the driving centerline within the possible range for avoiding obstacles. The final steps to keep the vehicle in the lane include: calculating the polynomial coefficients of the lane centerline and the lane width using the left and right lane line polynomials in the vehicle coordinate system; then calculating the drivable range within the lane based on the lane centerline; and finally, combining the previous steps to meet the requirements of avoiding left and right obstacles and driving close to the lane center as much as possible within the drivable range within the lane to obtain the final driving centerline. Different from the avoidance strategy of autonomous driving at level L3 and above, this strategy is applicable to the lane centering assist function at level L2. The overall method specifically includes the following steps:

[0054] Step S1: Obtain the left lane line polynomial, the right lane line polynomial and the lane width based on the vehicle coordinate system through the perception camera; Figure 3 As shown, the millimeter wave / lidar radar obtains the horizontal and vertical distances from the obstacle to the center of the vehicle, and detects and determines whether there are obstacles within a certain range on the left and right sides of the vehicle: (Taking the right side as an example, the obstacle detection range is generally within 30m in the longitudinal direction and 0.8 to 2.25m in the lateral direction. Figure 3 The orange part is an obstacle)

[0055] (1) If any, take the lateral distance to the nearest longitudinal obstacle:

[0056] RObs_dx = lateral distance to the right obstacle

[0057] LObs_dx = lateral distance to the left obstacle

[0058] (2) If it does not exist:

[0059] RObs_dx = 5 (default value)

[0060] LObs_dx = 5 (default value)

[0061] Step S2: Calculate the lane centerline and the lateral drivable range within the lane by polynomial fusion of the left and right lane lines; calculate the lateral drivable range of the vehicle based on the obstacle by using the lateral distance from the obstacle to the center of the vehicle obtained in step S1.

[0062] (1) Figure 4 As shown in the light green area, the lateral driving range within the lane is: [RL_dis, LL_dis] (x-axis: lane centerline, y-axis: vehicle y-axis), where:

[0063] Lane width = Cl0-Cr0

[0064]

[0065] RL_dis=-LL_dis

[0066] Among them, RL_dis (Right Line distance) refers to the lateral distance between the right lane line and the right edge of the vehicle body; LL_dis (Left Line distance) refers to the lateral distance between the left lane line and the left edge of the vehicle body, car_width refers to the vehicle width, and dl_min refers to the minimum lateral distance that needs to be maintained between the edge of the vehicle body and the lane line.

[0067] (2) Figures 5 to 8 As shown in the figure, the lateral travel range of the vehicle's center axis based on the obstacle is: [RO_dis, LO_dis] (coordinate system: vehicle coordinate system, RO_dis is the lateral distance of the right obstacle relative to the right edge of the vehicle body, and LO_dis is the lateral distance of the left obstacle relative to the left edge of the vehicle body)

[0068] ① When there are no obstacles on both sides of the vehicle, the lateral travel range of the vehicle's center axis based on the obstacles is Figure 5 The light brown area shows:

[0069]

[0070] ② When there is an obstacle on one side of the vehicle (taking the left side as an example), the lateral travel range of the vehicle's center axis based on the obstacle is Figure 6 The light brown area shows:

[0071]

[0072] ③ When there are obstacles on both sides of the vehicle, the lateral travel range of the vehicle's center axis based on the obstacles is:

[0073]

[0074] 1) If Figure 7 As shown, if LO_dis > RO_dis, it indicates that the distance between the obstacles on the left and right sides is sufficient for the vehicle to pass safely, and the vehicle can drive normally within the drivable range [RO_dis, LO_dis] based on the obstacles.

[0075] 2) As Figure 8 shown, if LO_dis <= RO_dis, it indicates that the distance between the obstacles on the left and right sides is not sufficient for the vehicle to maintain a safe distance to pass. At this time, the vehicle needs to drive in the middle between the two obstacles centrally.

[0076] Step S3: Combine the lateral drivable range based on the obstacles and the lane center line to obtain the driving center line 1 closest to the lane center line within the lateral drivable range based on the obstacles, and calculate the offset distance d1 of the driving center line 1 relative to the lane center line.

[0077] (1) As Figure 9 shown, if the lane center line is on the left side of the drivable range of the obstacles, LO_dis < C0, then d1 = LO_dis - C0;

[0078] (2) As Figure 10 shown, if the lane center line is on the right side of the drivable range of the obstacles, RO_dis > C0, then d1 = RO_dis - C0;

[0079] (3) As Figure 11 shown, if the lane center line is within the drivable range of the obstacles, RO_dis <= C0 <= LO_dis, then d1 = 0.

[0080] Where, C0 represents the lateral distance from the lane line to the center of the rear axle of the vehicle.

[0081] If the obstacle detection accuracy is not sufficient to support stable avoidance, the offset distance d1 in the present invention can also be given different fixed values according to the presence or absence of the obstacles:

[0082] 1. Appropriately give a fixed value: To achieve a fixed offset distance within the lane (for example, when there is an obstacle on the left side, d1 = -0.35)

[0083] 2. Give a very large fixed value: To achieve driving at a distance of d l _min from the lane line on the other side where the obstacle exists (for example, when there is an obstacle on the left side, d1 = -5).

[0084] Step S4: Limit the offset distance d1 through the lateral drivable range within the lane to obtain the target offset distance d2.

[0085] (1) If d1 < RL_dis, then d2 = RL_dis;

[0086] (2) If d1>LL_dis, then d2=LL_dis;

[0087] (3) If RL_dis<=d1<=LL_dis, then d2=d1.

[0088] Step S5: offset the lane centerline by the target offset distance d2 to obtain the driving centerline 2.

[0089] Driving center line 2:

[0090] Step S6: LCC uses driving centerline 2 as the target trajectory centered in the lane, calculates the lateral control instructions for the current vehicle traveling along this trajectory, and sends them to the chassis.

[0091] Step S7: The vehicle implements an intelligent avoidance function based on lane centering.

[0092] The above describes the specific embodiments of the present invention. Based on the above description, relevant personnel can make various changes and modifications without departing from the scope of the technical concept of this invention.

Claims

1. An intelligent obstacle avoidance method for lane centering, characterized in that: The following steps are involved: Step S1: Obtain the left lane line polynomial, the right lane line polynomial and the lane width based on the vehicle coordinate system; Obtain the horizontal and vertical distances from the obstacle to the center of the vehicle, and determine whether there are obstacles on the left and right sides of the vehicle within the obstacle detection range; Step S2: Calculate the lane centerline and the lateral drivable range within the lane by polynomial fusion of the left and right lane lines; calculate the lateral drivable range of the vehicle based on the obstacle by the lateral distance from the obstacle to the center of the vehicle; Step S3: combining the lateral drivable range based on the obstacle and the lane centerline to obtain a driving centerline closest to the lane centerline within the lateral drivable range based on the obstacle, and calculating an offset distance of the driving centerline relative to the lane centerline; Step S4: limiting the offset distance using the lateral drivable range within the lane calculated in step S2 to obtain a target offset distance; Step S5: offsetting the lane centerline by the target offset distance to obtain the target driving centerline; Step S6: The lane centering assist system (LCC) uses the target driving centerline as the target trajectory for lane centering, calculates the lateral control instructions for the current vehicle to travel along the target trajectory, and issues them to the chassis. Step S7: The vehicle implements an intelligent avoidance function based on lane centering; All distances described above are based on the center of the vehicle's rear axle as the coordinate axis, the vehicle's travel direction as the x-axis, and the direction perpendicular to the vehicle's travel direction as the y-axis. Left is positive, right is negative, and front is positive and rear is negative.

2. The intelligent obstacle avoidance method for lane centering according to claim 1, characterized in that: In step S1, the left lane line polynomial, the right lane line polynomial and the lane width are acquired through a perception camera based on a vehicle coordinate system.

3. The intelligent obstacle avoidance method for lane centering according to claim 1, characterized in that: In step S1, the horizontal and vertical distances from the obstacle to the center of the vehicle are obtained by millimeter wave radar or laser radar.

4. The intelligent obstacle avoidance method for lane centering according to claim 1, characterized in that: In step S2, if the lateral distance of the left obstacle relative to the left edge of the vehicle body is greater than the lateral distance of the right obstacle relative to the right edge of the vehicle body, it means that the distance between the left and right obstacles is sufficient for the vehicle to pass safely, and the vehicle can drive normally within the drivable range based on the obstacles; if the lateral distance of the left obstacle relative to the left edge of the vehicle body is less than or equal to the lateral distance of the right obstacle relative to the right edge of the vehicle body, it means that the distance between the left and right obstacles is not sufficient for the vehicle to pass at a safe distance, and the vehicle needs to drive centered between the two obstacles.

5. The intelligent obstacle avoidance method for lane centering according to claim 1, characterized in that: If it is detected that there are no obstacles on the left and right sides, the lateral distances of the obstacles on the left and right sides are preset values, and the lateral drivable range of the vehicle based on the obstacles is also a fixed value.

6. The intelligent obstacle avoidance method for lane centering according to claim 1, characterized in that: In step S3 , if the lane centerline is within the vehicle's drivable range based on the obstacle, the offset distance of the driving centerline relative to the lane centerline is 0.

7. The intelligent obstacle avoidance method for lane centering according to claim 1, characterized in that: The method is applicable to L2 level lane centering assistance function.

Images