Narrow road passage method, device and vehicle

Abstract

The present application provides a method, device and vehicle for passing through a narrow road, the method comprising: if there is a passable target narrow road in front of the target vehicle, and the width of the target narrow road is greater than the minimum passable width and less than the sum of the minimum passable width and the preset width, then the alignment posture corresponding to the target narrow road is obtained, and the alignment posture is the posture of the target vehicle before entering the target narrow road; based on the current posture and alignment posture of the target vehicle, an initial path is obtained, and based on the initial path, the target vehicle is controlled to travel from the current posture to the alignment posture; after the target vehicle reaches the alignment posture, the target path corresponding to the target narrow road is obtained, and based on the target path, the target vehicle is controlled to pass through the target narrow road, and travel from the alignment posture to the target posture, and the target posture is the end posture of the target narrow road. Through the technical solution of the present application, it is possible to ensure that the vehicle can pass smoothly in the narrow road, so that the vehicle can pass through the narrow passage autonomously, saving the driver's travel time and reducing the possibility of vehicle scratches.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a narrow road passage method, device and vehicle. Background Art

[0002] Autonomous driving is a technology that uses advanced sensors and computing technologies to enable vehicles to navigate and operate autonomously without human intervention. Autonomous driving can provide people with a safer, more convenient and more efficient way to travel, while also having a profound impact on the entire transportation system and urban planning.

[0003] A narrow lane is a narrow passage that allows vehicles to pass. When a vehicle passes through a narrow lane, the distance between the two sides of the vehicle and the obstacles is very small, and it is easy for the vehicle to come into contact with the obstacles, causing the vehicle to break down.

[0004] When a vehicle supports an autonomous driving function and controls the vehicle to travel on a narrow road based on the autonomous driving function, how to ensure that the vehicle can pass through the narrow road smoothly? The relevant technology does not have a highly feasible solution for the vehicle to pass through the narrow road. Summary of the invention

[0005] The present application provides a narrow road passage method, the method comprising:

[0006] If there is a passable target narrow road in front of the target vehicle, and the width of the target narrow road is greater than the minimum passable width and less than the sum of the minimum passable width and the preset width, then the alignment posture corresponding to the target narrow road is obtained, and the alignment posture is the posture of the target vehicle before entering the target narrow road;

[0007] Acquire an initial path based on the current posture of the target vehicle and the alignment posture, and control the target vehicle to travel from the current posture to the alignment posture based on the initial path;

[0008] After the target vehicle reaches the alignment posture, a target path corresponding to the target narrow road is obtained, and the target vehicle is controlled to pass through the target narrow road based on the target path, and travels from the alignment posture to a target posture, where the target posture is the end posture of the target narrow road.

[0009] The present application provides a narrow passage device, the device comprising:

[0010] an acquisition module, configured to acquire, if there is a passable target narrow road in front of the target vehicle, and the width of the target narrow road is greater than the minimum passable width and less than the sum of the minimum passable width and a preset width, an alignment posture corresponding to the target narrow road, the alignment posture being the posture of the target vehicle before entering the target narrow road; and acquire an initial path based on the current posture of the target vehicle and the alignment posture;

[0011] A control module, configured to control the target vehicle to travel from the current posture to the aligned posture based on the initial path;

[0012] The acquisition module is further used to acquire a target path corresponding to the target narrow road after the target vehicle reaches the alignment posture;

[0013] The control module is further used to control the target vehicle to pass through the target narrow road based on the target path, and travel from the alignment posture to a target posture, where the target posture is the end posture of the target narrow road.

[0014] The present application provides a vehicle, comprising: a processor and a machine-readable storage medium, wherein the machine-readable storage medium stores machine-executable instructions that can be executed by the processor; the processor is used to execute the machine-executable instructions to implement the narrow lane passing method of the above-mentioned example of the present application.

[0015] It can be seen from the above technical solutions that in the embodiments of the present application, a highly feasible vehicle narrow lane passing solution is proposed, which can be applied in the assisted driving function or the automatic driving function, can ensure the smooth passage of vehicles in narrow lanes, enable vehicles to pass through narrow passages autonomously, save the passing time for drivers, and reduce the possibility of vehicle scratches. It can expand the passing capacity of vehicles in crowded scenes and improve the passing success rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments of the present application or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings of the embodiments of the present application.

[0017] Figure 1 It is a schematic flow chart of a narrow passage method in one embodiment of the present application;

[0018] Figure 2 is a schematic diagram of deploying multiple sensors on a target vehicle in one embodiment of the present application;

[0019] Figure 3 is a schematic diagram of an alignment process and a passage process in one embodiment of the present application;

[0020] Figure 4 It is a schematic flow chart of a narrow passage method in one embodiment of the present application;

[0021] Figure 5 is a schematic diagram of a centering path in one embodiment of the present application;

[0022] Fig. 6A is a schematic diagram of the minimum passable width of a straight narrow road in one embodiment of the present application;

[0023] Figure 6B It is a schematic diagram of the minimum passable width of a curved narrow road in one embodiment of the present application;

[0024] Figure 6C is a schematic diagram of the minimum passable width of a narrow corner road in one embodiment of the present application;

[0025] Fig. 7A is a schematic diagram of a basic non-adjustment condition in one embodiment of the present application;

[0026] Figure 7B is a schematic diagram of a high-order non-adjustment condition in one embodiment of the present application;

[0027] Fig. 8A This is a schematic diagram of selecting sampling points for posture alignment in one embodiment of the present application;

[0028] Figure 8B is a schematic diagram of a reachability judgment condition for a correct posture in one embodiment of the present application;

[0029] Fig. 9 is a schematic diagram of reaching the alignment posture in one embodiment of the present application;

[0030] Fig.10 is a schematic diagram of fixed curvature path planning in one embodiment of the present application;

[0031] Fig.11 It is a schematic structural diagram of a narrow passage device in one embodiment of the present application;

[0032] Fig.12 It is a hardware structure diagram of a vehicle in one embodiment of the present application. DETAILED DESCRIPTION

[0033] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, rather than limiting the present application. The singular forms of "a", "said" and "the" used in the present application and claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to any or all possible combinations of one or more associated listed items.

[0034] It should be understood that, although the terms first, second, third, etc. may be used to describe various information in the embodiments of the present application, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, in addition, the word "if" used may be interpreted as "at..." or "when..." or "in response to determination".

[0035] In the embodiment of the present application, a narrow road passing method is proposed, which can be applied to a vehicle (the vehicle is recorded as a target vehicle), and the target vehicle supports an assisted driving function or an automatic driving function. Figure 1 FIG. 1 is a flow chart of the narrow passage method, which may include:

[0036] Step 101: If there is a passable target narrow road in front of the target vehicle, and the width of the target narrow road is greater than the minimum passable width and less than the sum of the minimum passable width and a preset width, then obtain the alignment posture corresponding to the target narrow road, which is the posture of the target vehicle before entering the target narrow road.

[0037] Step 102: acquiring an initial path based on the current position of the target vehicle and the aligned position of the target vehicle, and controlling the target vehicle to travel from the current position to the aligned position based on the initial path.

[0038] Step 103: After the target vehicle reaches the aligned position, a target path corresponding to the target narrow road is obtained, and based on the target path, the target vehicle is controlled to pass through the target narrow road and travel from the aligned position to the target position; wherein the target position may be the end position of the target narrow road.

[0039] Exemplarily, the method for determining the minimum passable width may include, but is not limited to: if the target narrow road is a straight narrow road, the minimum passable width may be determined based on the vehicle width of the target vehicle and the configured safe driving distance. Alternatively, if the target narrow road is a curved narrow road, the minimum passable width may be determined based on the vehicle wheelbase of the target vehicle, the curvature of the curve of the target narrow road, the distance from the vehicle corner to the center of the front axle, the distance from the vehicle corner to the center of the rear axle, and the configured safe driving distance. Alternatively, if the target narrow road is a corner narrow road, the minimum passable width may be determined based on the minimum turning radius, the minimum turning width, the channel corner size, the channel width, and the difference between the turning radius of the vehicle corner and the safe distance; wherein the minimum turning width is determined based on the vehicle wheelbase, the maximum steering angle of the vehicle's front wheels, the distance from the vehicle corner to the center of the front axle, the distance from the vehicle corner to the center of the rear axle, and the safe driving distance.

[0040] Exemplarily, the process of determining whether there is a passable target narrow road in front of the target vehicle may include: constructing an obstacle distance grid map based on observation data around the target vehicle, the observation data including data collected by a camera and / or data collected by a radar sensor. Determine a center path based on the obstacle distance grid map, the center path may include multiple path points, each path point is a path point with the largest distance from the obstacle determined based on the obstacle distance grid map. Obtain attribute information corresponding to each path point; wherein the attribute information includes the coordinates and orientation of the path point in the global coordinate system, the curvature of the curve corresponding to the path point, the distance from the path point to the center of the rear axle of the target vehicle along the path, and the distance from the path point to the obstacle. Based on the distance from each path point to the obstacle, if there is a road with a width greater than the minimum passable width and less than the sum of the minimum passable width and the preset width, it is determined that there is a passable target narrow road.

[0041] Exemplarily, before obtaining the alignment posture corresponding to the target narrow road, it is also possible to determine whether the target vehicle needs to adjust its posture before entering the target narrow road based on the current posture of the target vehicle; if so, the operation of obtaining the alignment posture corresponding to the target narrow road is performed. Among them, if the current posture does not meet the basic non-adjustment condition, and the current posture does not meet the high-order non-adjustment condition, it is determined that the target vehicle needs to adjust its posture before entering the target narrow road. Alternatively, if the current posture meets the basic non-adjustment condition, or the current posture meets the high-order non-adjustment condition, it is determined that the target vehicle does not need to adjust its posture before entering the target narrow road.

[0042] Exemplarily, if it is determined based on the current posture that the midpoint of the head of the target vehicle is located in the target fan-shaped area, the midpoint of the tail of the target vehicle is located in the target fan-shaped area, and the angle between the first line and the second line is less than the preset angle, then it is determined that the current posture meets the basic non-adjustment condition; otherwise, it is determined that the current posture does not meet the basic non-adjustment condition. Among them, the first line can be a line connecting the midpoint of the head and the starting point of the target narrow road, and the second line can be a line connecting the midpoint of the tail and the starting point of the target narrow road; the vertex position of the target fan-shaped area can be the starting point of the target narrow road, and the angle of the target fan-shaped area can be determined based on the width of the target narrow road, the vehicle width of the target vehicle, and the configured safe driving distance.

[0043] Exemplarily, multiple candidate poses are obtained from the detection arc corresponding to the current pose; if at least one candidate pose satisfies the basic non-adjustment condition, it is determined that the current pose satisfies the high-order non-adjustment condition; or, if all candidate poses do not satisfy the basic non-adjustment condition, it is determined that the current pose does not satisfy the high-order non-adjustment condition. The radius of the detection arc can be determined based on the angle Δψ and the distance Δd, the angle Δψ is the angle between the third line and the second line, and the distance Δd is the length of the second line; the third line can be the line between the midpoint of the head of the target vehicle and the midpoint of the tail of the target vehicle, and the second line can be the line between the midpoint of the tail of the target vehicle and the starting point of the target narrow road. The origin of the detection arc is located on the perpendicular line of the third line.

[0044] Exemplarily, obtaining the alignment posture corresponding to the target narrow road may include, but is not limited to: traversing multiple path points from the starting point of the target narrow road in the direction of the target narrow road. Based on the distance from each path point to the obstacle, a path point can be selected from multiple path points as a feature point. Obtain a fan-shaped detection area, the vertex position of the fan-shaped detection area can be the feature point, the starting direction is the direction of the feature point, and the ending direction is the direction of the starting point of the target narrow road. The fan-shaped detection area is divided vertically, and multiple sampling points are selected from the vertical division line. A sampling point is selected from the multiple sampling points as the target sampling point, and the alignment posture corresponding to the target narrow road is determined based on the posture of the target sampling point.

[0045] Exemplarily, when the target vehicle is at the target sampling point, at least one of the following conditions is satisfied: there is no obstacle directly in front of the target vehicle. The distance from the rear side of the target vehicle to the obstacle is greater than a first distance threshold. When there is an obstacle on the first side of the target vehicle, the distance from the target vehicle to the obstacle on that side is greater than a second distance threshold, and there is no obstacle in a designated area on the second side of the target vehicle. The first side may be the left side, and the second side may be the right side, or the first side may be the right side, and the second side may be the left side.

[0046] Exemplarily, obtaining the target path corresponding to the target narrow road may include, but is not limited to: obtaining a centered coarse path, which may include multiple path points. Obtaining multiple configured fixed curvature paths, each of which may include multiple path points. For each fixed curvature path, the path cost corresponding to the fixed curvature path is determined based on the closest distance from the path point on the fixed curvature path to the obstacle and the closest distance from the path point to the centered coarse path. Based on the path cost corresponding to each fixed curvature path, the fixed curvature path corresponding to the minimum path cost is determined as the target path corresponding to the target narrow road. Among them, the starting point of the target path is the path point corresponding to the current posture of the target vehicle.

[0047] Exemplarily, based on the shortest distance from the path point on the fixed curvature path to the obstacle and the shortest distance from the path point to the centered coarse path, determining the path cost corresponding to the fixed curvature path includes: using the following formula to determine the path cost corresponding to the fixed curvature path: Where J represents the path cost corresponding to the fixed curvature path, j represents the jth path point on the fixed curvature path, M represents the total number of path points, w1 and w2 represent the configured weight coefficients, represents the shortest distance from the jth path point to the obstacle, Represents the shortest distance from the jth path point to the centered coarse path.

[0048] It can be seen from the above technical solutions that in the embodiments of the present application, a highly feasible vehicle narrow lane passing solution is proposed, which can be applied in the assisted driving function or the automatic driving function, can ensure the smooth passage of vehicles in narrow lanes, enable vehicles to pass through narrow passages autonomously, save the passing time for drivers, and reduce the possibility of vehicle scratches. It can expand the passing capacity of vehicles in crowded scenes and improve the passing success rate.

[0049] The above technical solutions of the embodiments of the present application are described below in combination with specific application scenarios.

[0050] For vehicles that support assisted driving or autonomous driving functions (hereinafter referred to as target vehicles), target vehicles can be equipped with sensors such as cameras and ultrasonic radars. Figure 2 As shown in FIG. 1 , a schematic diagram of deploying multiple sensors on a target vehicle is shown, where V represents a camera and U represents an ultrasonic radar. Of course, this is just an example, and the target vehicle can also be deployed with sensors such as millimeter wave radar or lidar, and there is no restriction on this.

[0051] In order to adapt to various narrow road traffic scenarios to the greatest extent possible, so that the narrow road traffic method can be applied to assisted driving functions or automatic driving functions, the narrow road traffic method in the embodiment of the present application may include an alignment process and a traffic process. Figure 3 The figure shows a schematic diagram of the alignment process and the passage process.

[0052] Figure 3 (1) in the figure represents the alignment process. During the alignment process, it is necessary to search for a narrow road that the target vehicle can pass through, determine whether the target vehicle needs to adjust its posture before entering the narrow road, calculate the optimal alignment posture of the target vehicle entering the narrow road, and reach the alignment posture. Figure 3 (2) in the figure represents the passage process. During the passage process, path planning and servo control are required to control the target vehicle to pass through the narrow road safely.

[0053] See also Figure 4 As shown, the narrow road passage method proposed in the embodiment of the present application may include the following steps:

[0054] Step 401: Centered path planning. For example, based on data collected by a camera and / or data collected by a radar sensor (such as an ultrasonic radar), a centered path is planned in the channel in front of the target vehicle.

[0055] Step 402: Determine whether the narrow road is passable. For example, based on the center path and vehicle information, determine whether there is a passable narrow road ahead. If so, execute step 403; if not, end the process.

[0056] Step 403, posture adjustment judgment. For example, it is judged whether the target vehicle needs to adjust its posture by shifting gears before entering the narrow road. If so, step 404 can be executed; if not, step 406 can be executed.

[0057] Step 404: Calculate the alignment posture. For example, calculate the best alignment posture for the target vehicle to enter a narrow road.

[0058] Step 405: reaching the alignment posture. For example, through path planning and path following, the target vehicle is controlled to reach the alignment posture, that is, when the target vehicle reaches the position of the alignment posture, it is in the posture of the alignment posture.

[0059] Step 406: Centered path planning. For example, based on the data collected by the camera and / or the data collected by the radar sensor (such as ultrasonic radar), a centered path is planned in the channel in front of the target vehicle.

[0060] Step 407, judging whether the target vehicle has completely driven out of the narrow road. For example, judging whether the target vehicle has completely driven out of the narrow road. If so, the process ends. If not, executing step 408, continuing to execute the narrow road passing function.

[0061] Step 408: Travel path planning. For example, according to the relative position relationship between the target vehicle, the obstacle, and the center path, a drivable path is planned, and the target vehicle can travel along the path.

[0062] Step 409: Vehicle control. For example, the target vehicle is controlled in the horizontal and vertical directions according to the driving path.

[0063] The process from step 401 to step 409 is described below with reference to a specific example.

[0064] First, the center path planning of step 401.

[0065] For example, the following steps may be used to obtain the center path in the channel ahead of the target vehicle.

[0066] Step S11, obtaining observation data around the target vehicle, wherein the observation data may include but is not limited to data collected by a camera and / or data collected by a radar sensor, which is not limited.

[0067] Step S12: construct an obstacle distance grid map based on the observation data around the target vehicle. The obstacle distance grid map is used to represent the distance between the target vehicle and the obstacle. There is no restriction on the construction process of the obstacle distance grid map, as long as the obstacle distance grid map can be obtained based on the observation data.

[0068] For example, a grid map is a map composed of a series of grids, the size of the grid is the resolution of the map in the x / y direction, and each grid carries the attribute of whether the point is an obstacle. Each grid point in the obstacle distance grid map carries the information of the closest distance from the point to the obstacle grid point.

[0069] Step S13: determining a center path based on the obstacle distance grid map, where the center path includes a plurality of path points, each of which is a path point with a maximum distance from an obstacle determined based on the obstacle distance grid map.

[0070] Exemplarily, multiple path points can be selected on the driving path of the target vehicle. For each path point, the path point is the path point with the largest distance from the obstacle. For example, since each grid point in the obstacle distance grid map carries the information of the shortest distance from the point to the obstacle grid point, the path point with the largest distance from the obstacle can be determined based on the obstacle distance grid map. On this basis, the path composed of multiple path points (i.e., the path points with the largest distance from the obstacle) can be called a center path.

[0071] For example, the A* search algorithm can be used to search for multiple path points with the largest distance to obstacles, and the multiple path points can be combined into a central path. During the search process, it is necessary to be able to reflect the vehicle kinematic constraints in the obstacle distance grid map and maximize the path away from obstacles. There is no restriction on this search process.

[0072] See also Figure 5 As shown, it is a schematic diagram of the centered path. By planning the centered path, the channel with the largest width in front of the target vehicle can be determined, and a centered driving path can be formed in the center of the channel.

[0073] Step S14, obtaining attribute information corresponding to each path point in the centered path; wherein the attribute information includes the coordinates and orientation of the path point in the global coordinate system, the curve curvature corresponding to the path point, the distance from the path point along the path to the center of the rear axle of the target vehicle, and the distance from the path point to the obstacle.

[0074] Exemplarily, for each path point in the centered path, the path point can be in the form of (x, y, yaw, curv, s, obs_dis, narrow_flag). Among them, x and y represent the coordinates of the path point in the global coordinate system, and yaw represents the orientation of the path point in the global coordinate system. curv represents the curvature of the curve corresponding to the path point. s represents the distance from the path point to the center of the rear axle of the target vehicle along the path. obs_dis represents the distance from the path point to the obstacle. narrow_flag indicates whether the path point belongs to a narrow road. For example, if obs_dis meets the definition of narrow road width, narrow_flag is 1 ( Figure 5 If obs_dis does not meet the narrow path width definition, the narrow_flag value is 0 ( Figure 5 For whether obs_dis satisfies the narrow path width definition or obs_dis does not satisfy the narrow path width definition, please refer to the subsequent embodiments.

[0075] Second, determine whether the narrow road in step 402 is passable.

[0076] Based on the distance from each path point in the center path to the obstacle, if there is a road with a width greater than the minimum passable width and less than the sum of the minimum passable width and the preset width, it is determined that there is a passable narrow road (hereinafter referred to as a target narrow road), otherwise, it is determined that there is no passable target narrow road. Obviously, if there is a passable target narrow road in front of the target vehicle, the width of the target narrow road is greater than the minimum passable width, and the target narrow road is less than the sum of the minimum passable width and the preset width.

[0077] For example, a narrow lane can be wider than the minimum passable width d np and less than d np +1m channel, 1m represents the preset width, which can be configured empirically. Obviously, if there is a channel with a width greater than d np and less than d np +1m road, it is determined that there is a passable target narrow road, otherwise, it is determined that there is no passable target narrow road.

[0078] In a possible implementation, the minimum passable width may be determined using the following conditions:

[0079] Case 1: If the target narrow road is a straight narrow road, the minimum passable width can be determined based on the vehicle width of the target vehicle and the configured safe driving distance. For example, see Fig. 6A The figure shows the minimum passable width of a straight narrow road. The black solid line indicates the actual drivable boundary of the target narrow road. The minimum passable width d of the target narrow road is np It can be expressed as:np =W+2d sa Where W represents the vehicle width of the target vehicle, d sa It represents the safe driving distance of the vehicle and can be configured based on experience. Obviously, it can be based on the vehicle width W and the safe driving distance d sa Determine the minimum passable width d np .

[0080] Case 2: If the target narrow road is a curved narrow road, the minimum passable width can be determined based on the wheelbase of the target vehicle, the curve curvature of the target narrow road, the distance from the corner point of the vehicle to the center of the front axle, the distance from the corner point of the vehicle to the center of the rear axle, and the configured safe driving distance.

[0081] See also Figure 6B The figure shows the minimum passable width of a narrow curved road. The solid black line represents the actual edge of the target narrow road, and the dotted line represents the safe driving distance d. sa The minimum passable width d of the target narrow road np Can be:

[0082] Where L represents the wheelbase of the target vehicle, R represents the curvature of the target narrow road, R1 represents the turning radius of the front axle center of the target vehicle, R2 represents the turning radius of the rear axle center of the target vehicle, and R max Indicates the turning radius of the vehicle's outer corner, R min represents the turning radius of the corner point of the vehicle, d1 represents the distance from the corner point of the vehicle to the center of the front axle, d2 represents the distance from the corner point of the vehicle to the center of the rear axle, and d sa Indicates the safe driving distance of the vehicle, which can be configured based on experience. In summary, the safe driving distance d can be calculated based on the wheelbase L of the target vehicle, the curvature R of the target narrow road, the distance d1 from the vehicle corner to the center of the front axle, the distance d2 from the vehicle corner to the center of the rear axle, and the safe driving distance d sa , determine the minimum passable width d np , that is, the minimum passable width d of the narrow curved road np It can be determined by the road curvature and the vehicle parameters of the target vehicle.

[0083] Case 3: If the target narrow road is a corner narrow road, the minimum passable width can be determined based on the minimum turning radius, minimum turning width, channel corner size, channel width, the difference between the turning radius of the vehicle corner and the safe distance. Among them, the minimum turning width is determined based on the vehicle wheelbase, the maximum steering angle of the vehicle's front wheels, the distance from the vehicle corner to the center of the front axle, the distance from the vehicle corner to the center of the rear axle, and the safe driving distance.

[0084] See also Figure 6CThe figure shows the minimum passable width of the narrow road at the corner. The solid black line represents the actual feasible boundary of the target narrow road, and the dotted line represents the radius equal to the minimum turning radius R. v The minimum drivable boundary of the narrow turning lane, the outer side of which coincides with the black solid line. The minimum drivable width of the target narrow lane np Can be:

[0085] Among them, R v Indicates the minimum turning radius of the target vehicle, d min represents the minimum turning width of the target vehicle, θ represents the channel turning angle of the target narrow road (i.e. the turning angle of the road), and d represents the channel width of the target narrow road (i.e. the width of the channel at the entrance). R1 ​​represents the difference between the turning radius of the vehicle corner and the safety distance, such as the difference between the turning radius of the inner corner of the vehicle and the safety distance. R2 represents the sum of the turning radius of the outer corner of the vehicle and the safety distance, and the minimum passable width d is subtracted from the sum. np .

[0086] For example, for the minimum turning width d min For example, the minimum turning width d min It can be determined based on the vehicle wheelbase, the maximum steering angle of the vehicle's front wheels, the distance from the vehicle's corner point to the center of the front axle, the distance from the vehicle's corner point to the center of the rear axle, and the safe driving distance. For example, the minimum turning width d can be determined using the following formula: min : Where L represents the wheelbase of the target vehicle, R represents the maximum steering angle of the front wheels of the vehicle, R2 represents the turning radius of the rear axle center of the target vehicle, d1 represents the distance from the corner point of the vehicle to the center of the front axle, d2 represents the distance from the corner point of the vehicle to the center of the rear axle, and d sa Indicates the safe driving distance for the vehicle and can be configured based on experience.

[0087] Exemplarily, the above-mentioned parameter values ​​may be attributes of the target vehicle and attributes of the target narrow road. The attributes of the target vehicle and attributes of the target narrow road are both known values, and no limitation is imposed on them.

[0088] Third, the posture adjustment judgment of step 403 is made.

[0089] For example, based on the current posture of the target vehicle, it can be determined whether the target vehicle needs to adjust its posture before entering the target narrow road. If so, the alignment posture corresponding to the target narrow road is obtained, and then the target narrow road is entered based on the alignment posture. If not, it is not necessary to obtain the alignment posture corresponding to the target narrow road, that is, or, the target vehicle can safely enter the target narrow road without shifting or adjusting.

[0090] For example, it is possible to first determine whether the current posture satisfies the basic non-adjustment condition. If the basic non-adjustment condition is met, it can be determined that the target vehicle does not need to adjust its posture before entering the target narrow road. If the basic non-adjustment condition is not met, it is determined whether the current posture satisfies the high-order non-adjustment condition. If the high-order non-adjustment condition is met, it can be determined that the target vehicle does not need to adjust its posture before entering the target narrow road. If the high-order non-adjustment condition is not met, it can be determined that the target vehicle needs to adjust its posture before entering the target narrow road.

[0091] For example, see Fig. 7A As shown, it is a schematic diagram of the basic non-adjustment condition. If it is determined based on the current posture of the target vehicle that the midpoint of the head of the target vehicle is located in the target sector area, and the midpoint of the tail of the target vehicle is located in the target sector area, and the angle between the first connecting line and the second connecting line is less than the preset angle, then it is determined that the current posture meets the basic non-adjustment condition; otherwise, it is determined that the current posture does not meet the basic non-adjustment condition.

[0092] See also Fig. 7A As shown, the midpoint of the head of the target vehicle is a black solid dot, and the midpoint of the tail of the target vehicle is a black solid dot. The first connecting line is the connecting line between the midpoint of the head of the target vehicle and the starting point of the target narrow road, and the second connecting line is the connecting line between the midpoint of the tail of the target vehicle and the starting point of the target narrow road. The angle between the first connecting line and the second connecting line is γ, and the preset angle can be configured according to experience, such as 10 degrees.

[0093] Based on the constraint condition of "angle γ is less than 10 degrees", the setting of angle γ determines that the farther the initial position of the target vehicle is from the narrow road entrance, the larger the adjustment space of the target vehicle, and the greater the deviation between the initial posture of the target vehicle and the posture at the narrow road entrance can be; the closer the initial position of the target vehicle is to the narrow road entrance, the smaller the adjustment space of the target vehicle, and the smaller the deviation between the initial posture of the target vehicle and the posture at the narrow road entrance.

[0094] See also Fig. 7A As shown, the vertex position of the target sector area is the starting point of the target narrow road, and the angle of the target sector area is The midpoint of the target vehicle's head needs to be located in the target sector area, and the midpoint of the target vehicle's tail needs to be located in the target sector area. The vertex position and orientation of the target sector area are determined by the first narrow path point on the exploration path. It can be determined based on the narrow road width and the shape parameters of the target vehicle, such as the width of the target narrow road, the vehicle width of the target vehicle and the configured safe driving distance. For example, it can be determined using the following formula: d n represents the width of the target narrow road, W represents the vehicle width of the target vehicle, d sa Indicates safe driving distance.

[0095] For example, see Figure 7B As shown in FIG. 1 , it is a schematic diagram of the high-order non-adjustment condition. For the high-order non-adjustment condition, multiple candidate poses are obtained from the detection arc corresponding to the current pose of the target vehicle. If at least one candidate pose satisfies the basic non-adjustment condition, it is determined that the current pose satisfies the high-order non-adjustment condition; if all candidate poses do not satisfy the basic non-adjustment condition, it is determined that the current pose does not satisfy the high-order non-adjustment condition.

[0096] For example, starting from the current position of the target vehicle, traverse the points on the detection arc forward at intervals of 0.1m (0.1m is an empirical value and can be configured arbitrarily according to actual needs) as candidate poses. If the candidate pose currently traversed meets the basic non-adjustment condition, the traversal process is terminated to determine whether the current pose meets the high-order non-adjustment condition. If the candidate pose currently traversed does not meet the basic non-adjustment condition, continue to traverse the next point on the detection arc as a candidate pose, and so on, until the candidate pose currently traversed meets the basic non-adjustment condition, or all candidate poses on the detection arc do not meet the basic non-adjustment condition.

[0097] See also Figure 7B As shown, the radius R of the detection arc can be determined based on the angle Δψ and the distance Δd, the angle Δψ is the angle between the third line and the second line, the third line can be the line between the midpoint of the head of the target vehicle and the midpoint of the tail of the target vehicle, and the second line can be the line between the midpoint of the tail of the target vehicle and the starting point of the target narrow road. The distance Δd is the length of the second line, that is, the length between the midpoint of the tail of the target vehicle and the starting point of the target narrow road. The origin of the detection arc is located on the vertical line of the third line.

[0098] For example, the radius R of the detection arc can be determined by the following formula: At the same distance Δd, the larger the angle Δψ, the smaller the radius R; at the same angle Δψ, the smaller the distance Δd, the smaller the radius R.

[0099] Fourth, the alignment posture calculation in step 404 is performed.

[0100] For example, the alignment posture is the best position for the target vehicle to enter the narrow road. At this alignment posture, the target vehicle can directly enter the target narrow road without gear shifting, that is, it meets the convenience. At the same time, the alignment posture needs to meet the safety and accessibility conditions, that is, the alignment posture is collision-free and can be reached.

[0101] For example, in order to obtain the alignment posture corresponding to the target narrow road, the following steps can be used:

[0102] Step S21, traversing multiple path points from the starting point of the target narrow road in the direction of the target narrow road.

[0103] For example, see Fig. 8A The figure shows the selection of sampling points for the alignment posture. The black triangle represents the starting point of the target narrow road, that is, the first point in the path forward direction where narrow_flag = 1. Starting from the starting point of the target narrow road, multiple path points are traversed along the path in the direction of the target narrow road, for example, all path points within 2.0m (2.0m is an empirical value and can be arbitrarily configured according to actual needs) are traversed.

[0104] Step S22: After traversing multiple path points, based on the distance from each path point to the obstacle, one path point can be selected from the multiple path points as a feature point.

[0105] For example, for multiple path points traversed, if a path point meets any of the following conditions, then the path point is selected as a feature point, see Fig. 8A As shown in the figure, the black solid circles are the selected feature points.

[0106] 1. The distance between this path point and the obstacle is the smallest. 2. This path point is the first point on the path with the minimum distance to the obstacle within 0.5m before and after (0.5m is an empirical value and can be configured arbitrarily according to actual needs), that is, the distance between all path points and obstacles within 0.5m before this path point is constantly decreasing, and the distance between all path points and obstacles within 0.5m after this path point is constantly increasing, and this path point is the first one on the path with this feature.

[0107] Step S23, obtaining a sector-shaped detection area, wherein the vertex position of the sector-shaped detection area is the feature point, the starting direction of the sector-shaped detection area is the direction of the feature point (such as the yaw value of the feature point), and the ending direction of the sector-shaped detection area is the direction of the starting point of the target narrow road (such as the yaw value of the starting point).

[0108] Step S24: divide the sector-shaped detection area vertically, and select multiple sampling points from the vertical division line.

[0109] For example, the sector-shaped detection area is divided vertically at intervals of 5 degrees (5 degrees is an empirical value and can be arbitrarily configured according to actual needs) to obtain multiple vertical division lines. Fig. 8A 4 longitudinal dividing lines are shown in FIG. For each longitudinal dividing line, a point is selected as a sampling point at every 0.5m interval (0.5m is an empirical value and can be configured arbitrarily according to actual needs) on the longitudinal dividing line. If the longest interval is 1.5m, 3 points can be selected as sampling points. In summary, a total of 12 points are selected as sampling points on the 4 longitudinal dividing lines, see Fig. 8A As shown, the black dashed circles represent the sampling points on the longitudinal dividing line.

[0110] Exemplarily, steps S21 to S24 are the process of generating alignment pose sampling points.

[0111] Step S25: Select a sampling point from the multiple sampling points as the target sampling point.

[0112] For example, since the positions of the sampling points are all facing the narrow road entrance, the convenience of the sampling points can be satisfied. At the same time, in order to ensure that the positions are safe and accessible, see Figure 8B As shown in the figure, it is a schematic diagram of the safety and accessibility judgment conditions of the correct posture. The obstacle distribution around the posture needs to meet Figure 8B When a sampling point is used as a target sampling point, a sampling point that satisfies the following conditions needs to be used as the target sampling point, that is, when the target vehicle is at the target sampling point, at least one of the following conditions is satisfied:

[0113] 1. There is no obstacle in front of the target vehicle. 2. The distance from the rear of the target vehicle to the obstacle is greater than the first distance threshold. Figure 8B As shown, the first distance threshold is d l 3. When there is an obstacle on the first side of the target vehicle, the distance from the target vehicle to the obstacle on that side is greater than the second distance threshold, and there is no obstacle in the designated area on the second side of the target vehicle. The first side is the left side, the second side is the right side, or the first side is the right side, and the second side is the left side. Figure 8B As shown, when there is an obstacle on the first side of the target vehicle, the distance from the target vehicle to the obstacle on that side is greater than the second distance threshold d s , there is no obstacle in the designated area on the second side of the target vehicle, and the designated area is based on the size parameter β and d a A dotted polygon.

[0114] Exemplarily, the first distance threshold d l , the second distance threshold d s , size parameter β, size parameter d a It can be obtained from the parameters of the target vehicle, such as the size, steering ability and safety distance, and there is no restriction on this. For example, d l =1.0m,d s =0.5m, β=45 degrees, d a =3.0m. During the alignment posture calculation process, the distance grid map and the separation axis method can be used to determine whether the obstacle meets the above conditions.

[0115] Step S26: determining the alignment posture corresponding to the target narrow road based on the posture of the target sampling point.

[0116] For example, the position of the target sampling point can be used as the alignment position corresponding to the target narrow road.

[0117] Exemplarily, step S25 to step S26 is a process of detecting the positive posture sampling points.

[0118] Fifth, regarding the arrival at the alignment posture in step 405 .

[0119] For example, based on the current position of the target vehicle and the alignment position of the target vehicle, an initial path can be obtained, and the target vehicle can be controlled to travel from the current position to the alignment position based on the initial path. When the target vehicle reaches the position of the alignment position, it is in the alignment position. For example, a P2P path planning method can be used (the P2P path planning method is a path planning algorithm that plans from one position to another, such as a hybrid A* method) to generate an initial path from the current position to the alignment position, and control the target vehicle to travel to the alignment position based on the initial path. See Fig. 9 The figure shows a schematic diagram of reaching the alignment position.

[0120] Sixth, centering path planning for step 406 .

[0121] In order to overcome the positioning error and control error in narrow channels, a two-stage real-time path planning method of "centering guidance + fine-tuning obstacle avoidance" is designed. In the "centering guidance" stage, a rough path is generated to guide the target vehicle to drive in the center of the narrow channel, so that the target vehicle is sufficiently away from obstacles on both sides, and the spatial safety margin of the target vehicle is improved. Since the centering path is planned and obtained in the obstacle distance grid map, the path points do not fully meet the vehicle kinematic constraints and curvature smoothness constraints, and the target vehicle cannot directly follow the path. Therefore, in the "fine-tuning obstacle avoidance" stage, considering the kinematic characteristics and ride comfort of the target vehicle and other factors, a fixed curvature path planning method is used to generate a driving path with fixed curvature, so as to achieve fine-tuning of the target vehicle's posture to avoid obstacles and follow the rough path. The combination of fixed curvature path and vehicle servo control can eliminate the safety hazards caused by path following errors in narrow scenes and improve vehicle driving safety.

[0122] For example, the following steps may be used to obtain the center path in the channel ahead of the target vehicle.

[0123] Step S31, obtaining a centered coarse path, the centered coarse path may include multiple path points, each path point is a path point with the largest distance from an obstacle, and the method for obtaining the centered coarse path refers to steps S11 to S13.

[0124] Step S32: Obtain fixed curvature paths corresponding to a plurality of fixed curvatures, each fixed curvature path including a plurality of path points. The fixed curvature path may also be referred to as an arc path.

[0125] See also Fig.10As shown in the figure, it is a schematic diagram of fixed curvature path planning. Multiple configured fixed curvatures can be obtained, if there are N fixed curvatures. For each fixed curvature, the fixed curvature path corresponding to the fixed curvature can be obtained, that is, the rear axle center of the target vehicle is used as the starting point of the fixed curvature path, and the fixed curvature path is determined based on the fixed curvature. N fixed curvatures correspond to N fixed curvature paths. The N fixed curvature paths are paths with equal spacing but the same length. The number of path points on each fixed curvature path is M.

[0126] Step S33: for each fixed curvature path, based on the shortest distance from a path point on the fixed curvature path to an obstacle and the shortest distance from the path point to the central coarse path, determine the path cost corresponding to the fixed curvature path.

[0127] For example, for each fixed curvature path, the path cost corresponding to the fixed curvature path can be determined using the following formula: J represents the path cost, j represents the jth path point on the fixed curvature path, M represents the total number of path points, w1 and w2 represent the configured weight coefficients, represents the shortest distance from the jth path point to the obstacle, Represents the shortest distance from the jth path point to the centered coarse path.

[0128] Step S34: Based on the path cost corresponding to each fixed curvature path, determine the fixed curvature path corresponding to the minimum path cost as the target path corresponding to the target narrow road.

[0129] For example, the target path corresponding to the target narrow road can be determined by the following formula: J i Indicates the path cost corresponding to the i-th fixed curvature path, where i ranges from 1 to N, meaning there are a total of N fixed curvature paths. min i J i It represents the fixed curvature path corresponding to the minimum path cost. The fixed curvature path corresponding to the minimum path cost is the fixed curvature path farthest from the obstacle and closest to the central coarse path, and can be used as the target path.

[0130] For example, for each fixed curvature path, when there is any point on the fixed curvature path that intersects with an obstacle, the fixed curvature path is directly eliminated and the fixed curvature path will not be used as the target path. For example, the starting point of the target path is the path point corresponding to the current posture of the target vehicle.

[0131] Seventh, regarding the judgment of exiting a narrow road, driving path planning, and vehicle control in steps 407 to 409.

[0132] Exemplarily, after obtaining the target path, the target vehicle can be controlled to pass through the target narrow road based on the target path, and travel from the alignment posture to the target posture, which is the best posture for the target vehicle to enter the target narrow road, and the target posture can be the end posture of the target narrow road.

[0133] In the process of controlling the target vehicle to pass through the target narrow road based on the target path, it is also possible to periodically determine whether the target vehicle has completely driven out of the target narrow road, such as determining whether the target vehicle has completely driven out of the target narrow road every 10 seconds. If so, the target vehicle successfully drives out of the target narrow road, and the process ends.

[0134] If not, the driving path is re-planned based on the current position of the target vehicle (i.e., the current position during driving). For example, a two-stage real-time path planning method of "centering guidance + fine-tuning obstacle avoidance" is adopted to re-plan the driving path of the target vehicle. This process can refer to steps S31-S34, so as to obtain the target path corresponding to the target narrow road, and continue to control the target vehicle to pass through the target narrow road based on the target path corresponding to the target narrow road, and so on, until the target vehicle has completely driven out of the target narrow road.

[0135] In the process of controlling the target vehicle to pass through the target narrow road based on the target path, the target vehicle may be controlled in the lateral direction and / or in the longitudinal direction, and there is no limitation on this.

[0136] In the related art, in the process of controlling the target vehicle to pass through the target narrow road, the lateral and longitudinal deviations from the center of the rear axle of the target vehicle to the path point can be calculated to generate control quantities such as the front wheel steering angle and the accelerator pedal opening to achieve path following. However, this process depends on positioning information. Under normal circumstances, there will be an error of about 0.1m to 0.2m in positioning. In narrow scenes, the positioning error superimposed on the control error will affect the safety of the target vehicle. In response to the above problems, in this embodiment, instead of adopting the method of "calculating the lateral and longitudinal deviations from the center of the rear axle of the target vehicle to the path point, generating control quantities such as the front wheel steering angle and the accelerator pedal opening to achieve path following", a vehicle servo control method is used to achieve path following. The vehicle servo control method eliminates the influence of positioning error and avoids the superposition of positioning error and control error.

[0137] For example, the vehicle servo control method does not need to calculate the lateral and longitudinal deviations from the rear axle center of the target vehicle to the path point. In the lateral control of the target vehicle, the vehicle servo control method directly calculates the front wheel steering angle corresponding to the target vehicle using the curvature value of the fixed curvature path. The calculation formula is as follows: δ = tan -1(Lκ). δ represents the front wheel steering angle corresponding to the target vehicle, L represents the wheelbase of the target vehicle, and κ represents the curvature value of the fixed curvature path (i.e., the target path). In the longitudinal control of the target vehicle, the vehicle servo control method enables the target vehicle to maintain a constant speed.

[0138] Referring to the contents of the sixth point and the seventh point, the target vehicle can enter the passage stage when the target vehicle does not need to adjust its posture or the target vehicle reaches the alignment posture. For the passage stage, the present application designs a driving scheme that combines a two-stage real-time path planning method with a vehicle servo control method. Among them, the two-stage real-time path planning method is a method of centering guidance + fine-tuning obstacle avoidance. Centering guidance is the process of generating a coarse path, see step S31, and fine-tuning obstacle avoidance is to generate a driving path with a fixed curvature based on the coarse path, see steps S32-S34. Among them, the vehicle servo control method is a method of directly calculating the front wheel steering angle corresponding to the target vehicle using the curvature value of the fixed curvature path. This scheme can better cope with the uncertainty of perceived information in narrow scenes, improve the passability of the target vehicle, eliminate the influence of path following errors, and improve the driving safety of the target vehicle in crowded scenes.

[0139] It can be seen from the above technical solutions that in the embodiments of the present application, a highly feasible vehicle narrow lane passage solution is proposed, which can be applied to the assisted driving function or the automatic driving function, and can ensure the smooth passage of the vehicle in the narrow lane, so that the vehicle can pass through the narrow passage autonomously, saving the driver's passage time and reducing the possibility of the vehicle being scratched. The vehicle's traffic capacity in crowded scenes can be expanded and the traffic success rate can be improved. If applied to the assisted driving function, the vehicle can autonomously pass through the narrow passage that is difficult for the driver to manually operate the vehicle through, saving the driver's passage time and reducing the possibility of the vehicle being scratched. If applied to the automatic driving function (such as automatic valet parking, etc.), the vehicle's traffic capacity in scenes such as crowded parking lots can be expanded, the success rate of autonomous parking and pick-up of the vehicle can be improved, and the time of passengers can be saved.

[0140] Through the two-stage traffic solution of "alignment + passage", the alignment stage is used to search for the narrow road that can be passed in front of the vehicle, determine whether the vehicle can directly enter the narrow road, calculate the vehicle's preparation posture for entering the narrow road (i.e., the best alignment posture), and reach the preparation posture, etc., see the relevant contents of step 401, step 402, step 403, step 404, and step 405. The passage stage is used to guide and control the vehicle to safely pass through the narrow road through methods such as path planning and vehicle servo control, see the relevant contents of step 406, step 407, step 408, and step 409. When this solution is applied to the automatic driving function, it can improve the vehicle's passing ability in crowded scenes. When this solution is applied to the assisted driving function, it can assist the driver to pass through narrow passages.

[0141] In the path planning process of the traffic phase, a two-stage real-time path planning method of "centering guidance + fine-tuning obstacle avoidance" can be adopted. In the "centering guidance" stage, a rough path is generated to guide the vehicle to drive in the center of the narrow road, so that the vehicle is away from obstacles on both sides and the spatial safety margin of the vehicle is improved. For the centering guidance process, see step S31. In the "fine-tuning obstacle avoidance" stage, factors such as vehicle kinematic characteristics and ride comfort are considered to generate a driving path with a fixed curvature, and the vehicle posture is fine-tuned to avoid obstacles and follow the rough path. For the process of fine-tuning obstacle avoidance, see steps S32-step S34. By adopting the path planning method of "centering guidance + fine-tuning obstacle avoidance", the uncertainty of perception information in crowded scenes can be better dealt with, the influence of path following errors and positioning errors can be eliminated, and the vehicle's passing ability and driving safety in narrow scenes can be improved.

[0142] With regard to the "searching for a passable narrow road in front of the vehicle" in the alignment stage, i.e., the judgment of passability of the narrow road in step 402, a method for quickly judging whether a narrow road is passable is proposed. The characteristic parameters such as the position, length, width and curvature of the narrow road in front of the vehicle are obtained through the designed center path planning method. By establishing a mathematical relationship between the minimum passable width of the narrow road and the characteristic parameters of the narrow road and the characteristic parameters of the vehicle, a quick judgment can be made as to whether the narrow road is passable before the vehicle enters the narrow road.

[0143] With regard to the "determination of whether the vehicle can directly enter the narrow road", i.e., the posture adjustment judgment in step 403, in the alignment stage, a method for quickly judging whether the vehicle needs to adjust its posture is proposed. With regard to whether the vehicle can directly enter the narrow road without changing gears, basic non-adjustment conditions and high-order non-adjustment conditions are established using information such as the width and orientation of the narrow road entrance and the relative position relationship between the narrow road and the vehicle. Based on the basic non-adjustment conditions and the high-order non-adjustment conditions, a quick judgment of whether the vehicle needs to adjust its posture before entering the narrow road is achieved.

[0144] With regard to the "calculation of the vehicle's preparatory posture for entering a narrow road" in the alignment stage, i.e., the alignment posture calculation in step 404, a calculation method for the alignment posture for narrow road passage is proposed, including "generating alignment posture sampling points" and "alignment posture sampling point detection" to ensure the convenience, safety and accessibility of the alignment posture.

[0145] Based on the same application concept as the above method, a narrow passage device is proposed in the embodiment of the present application, see Fig.11 FIG. 1 is a schematic diagram of the structure of the device, and the device may include:

[0146] The acquisition module 1111 is used for acquiring the alignment posture corresponding to the target narrow road if there is a passable target narrow road in front of the target vehicle, and the width of the target narrow road is greater than the minimum passable width and less than the sum of the minimum passable width and the preset width, wherein the alignment posture is the posture of the target vehicle before entering the target narrow road; acquiring an initial path based on the current posture of the target vehicle and the alignment posture; the control module 1112 is used for controlling the target vehicle to travel from the current posture to the alignment posture based on the initial path; the acquisition module 1111 is also used for acquiring the target path corresponding to the target narrow road after the target vehicle reaches the alignment posture; the control module 1112 is also used for controlling the target vehicle to pass through the target narrow road based on the target path, and travel from the alignment posture to the target posture; wherein the target posture is the end posture of the target narrow road.

[0147] Exemplarily, the acquisition module 1111 is also used to acquire the minimum passable width. When acquiring the minimum passable width, the acquisition module 1111 is specifically used to: if the target narrow road is a straight narrow road, then based on the vehicle width of the target vehicle and the configured safe driving distance, determine the minimum passable width; or, if the target narrow road is a curved narrow road, then based on the vehicle wheelbase of the target vehicle, the curve curvature of the target narrow road, the distance from the vehicle corner point to the center of the front axle, the distance from the vehicle corner point to the center of the rear axle, and the configured safe driving distance, determine the minimum passable width; or, if the target narrow road is a corner narrow road, based on the minimum turning radius, the minimum turning width, the channel corner size, the channel width, and the difference between the turning radius of the vehicle corner point and the safe distance, determine the minimum passable width; wherein, the minimum turning width is determined based on the vehicle wheelbase, the maximum steering angle of the vehicle's front wheels, the distance from the vehicle corner point to the center of the front axle, the distance from the vehicle corner point to the center of the rear axle, and the safe driving distance.

[0148] Exemplarily, when the acquisition module 1111 determines that there is a passable target narrow road in front of the target vehicle, it is specifically used to: construct an obstacle distance grid map based on observation data around the target vehicle, wherein the observation data includes data collected by a camera and / or data collected by a radar sensor; determine a center path based on the obstacle distance grid map, and the center path includes multiple path points, each path point is a path point with the largest distance from the obstacle determined based on the obstacle distance grid map; obtain attribute information corresponding to each path point; wherein the attribute information includes the coordinates and orientation of the path point in the global coordinate system, the curve curvature corresponding to the path point, the distance from the path point along the path to the center of the rear axle of the target vehicle, and the distance from the path point to the obstacle; based on the distance from each path point to the obstacle, if there is a road with a width greater than the minimum passable width and less than the sum of the minimum passable width and a preset width, it is determined that there is a passable target narrow road.

[0149] Exemplarily, the acquisition module 1111 is also used to determine whether the target vehicle needs to adjust its posture before entering the target narrow road based on the current posture of the target vehicle before acquiring the alignment posture corresponding to the target narrow road; if so, perform an operation of acquiring the alignment posture corresponding to the target narrow road; wherein, if the current posture does not satisfy the basic non-adjustment condition and the current posture does not satisfy the high-order non-adjustment condition, it is determined that the target vehicle needs to adjust its posture before entering the target narrow road; or, if the current posture satisfies the basic non-adjustment condition or the current posture satisfies the high-order non-adjustment condition, it is determined that the target vehicle does not need to adjust its posture before entering the target narrow road.

[0150] Exemplarily, the acquisition module 1111 determines whether the target vehicle needs to adjust its posture before entering the target narrow road based on the current posture of the target vehicle, and is specifically used for: if it is determined based on the current posture that the midpoint of the head of the target vehicle is located in the target fan-shaped area, the midpoint of the tail of the target vehicle is located in the target fan-shaped area, and the angle between the first line and the second line is less than a preset angle, then it is determined that the current posture meets the basic no-adjustment condition; otherwise, it is determined that the current posture does not meet the basic no-adjustment condition; wherein the first line is a line connecting the midpoint of the head and the starting point of the target narrow road, and the second line is a line connecting the midpoint of the tail and the starting point of the target narrow road; the vertex position of the target fan-shaped area is the starting point of the target narrow road, and the angle of the target fan-shaped area is determined based on the width of the target narrow road, the vehicle width of the target vehicle and the configured safe driving distance.

[0151] Exemplarily, when the acquisition module 1111 determines whether the target vehicle needs to adjust its posture before entering the target narrow road based on the current posture of the target vehicle, it is specifically used to: acquire multiple candidate postures from the detection arc corresponding to the current posture; if at least one candidate posture satisfies the basic non-adjustment condition, determine that the current posture satisfies the high-order non-adjustment condition; or, if all candidate postures do not satisfy the basic non-adjustment condition, determine that the current posture does not satisfy the high-order non-adjustment condition; wherein the radius of the detection arc is determined based on the angle Δψ and the distance Δd, the angle Δψ is the angle between the third line and the second line, and the distance Δd is the length of the second line; the third line is the line between the midpoint of the head of the target vehicle and the midpoint of the tail of the target vehicle, and the second line is the line between the midpoint of the tail of the target vehicle and the starting point of the target narrow road; the origin of the detection arc is located on the perpendicular line of the third line.

[0152] Exemplarily, when the acquisition module 1111 acquires the alignment posture corresponding to the target narrow road, it is specifically used to: traverse multiple path points from the starting point of the target narrow road in the direction of the target narrow road; select feature points from the multiple path points based on the distance from each path point to the obstacle; acquire a fan-shaped detection area, the vertex position of the fan-shaped detection area is the feature point, the starting direction is the direction of the feature point, and the ending direction is the direction of the starting point of the target narrow road; divide the fan-shaped detection area vertically, and select multiple sampling points from the vertical division line; select a sampling point from the multiple sampling points as the target sampling point, and determine the alignment posture corresponding to the target narrow road based on the posture of the target sampling point.

[0153] Exemplarily, when the target vehicle is at the target sampling point, the following conditions are satisfied: there is no obstacle directly in front of the target vehicle; the distance from the rear side of the target vehicle to the obstacle is greater than a first distance threshold; when there is an obstacle on the first side of the target vehicle, the distance from the target vehicle to the obstacle on that side is greater than a second distance threshold, and there is no obstacle in a designated area on the second side of the target vehicle; the first side is the left side and the second side is the right side, or the first side is the right side and the second side is the left side.

[0154] Exemplarily, when the acquisition module 1111 acquires the target path corresponding to the target narrow road, it is specifically used to: acquire a centered rough path, wherein the centered rough path includes multiple path points; acquire fixed curvature paths corresponding to multiple configured fixed curvatures, wherein each fixed curvature path includes multiple path points; for each fixed curvature path, based on the closest distance from the path point on the fixed curvature path to the obstacle and the closest distance from the path point to the centered rough path, determine the path cost corresponding to the fixed curvature path; based on the path cost corresponding to each fixed curvature path, determine the fixed curvature path corresponding to the minimum path cost as the target path corresponding to the target narrow road; wherein the starting point of the target path is the path point corresponding to the current posture of the target vehicle.

[0155] Exemplarily, the acquisition module 1111 determines the path cost corresponding to the fixed curvature path based on the shortest distance from the path point on the fixed curvature path to the obstacle and the shortest distance from the path point to the centered coarse path by using the following formula to determine the path cost corresponding to the fixed curvature path:

[0156]

[0157] Where J represents the path cost corresponding to the fixed curvature path, j represents the jth path point on the fixed curvature path, M represents the total number of path points, w1 and w2 represent the configured weight coefficients, represents the shortest distance from the jth path point to the obstacle, represents the shortest distance from the jth path point to the centered coarse path.

[0158] Based on the same application concept as the above method, a vehicle is proposed in the embodiment of the present application, see Fig.12 As shown, the vehicle includes: a processor 1211 and a machine-readable storage medium 1212, the machine-readable storage medium 1212 stores machine-executable instructions that can be executed by the processor 1211; the processor 1211 is used to execute the machine-executable instructions to implement the narrow lane passage method disclosed in the above example of this application.

[0159] Based on the same application concept as the above method, an embodiment of the present application also provides a machine-readable storage medium, on which a number of computer instructions are stored. When the computer instructions are executed by a processor, the narrow lane passage method disclosed in the above example of the present application can be implemented.

[0160] The above-mentioned machine-readable storage medium may be any electronic, magnetic, optical or other physical storage device, which may contain or store information, such as executable instructions, data, etc. For example, the machine-readable storage medium may be: RAM (Radom Access Memory), volatile memory, non-volatile memory, flash memory, storage drive (such as hard disk drive), solid state drive, any type of storage disk (such as CD, DVD, etc.), or similar storage medium, or a combination thereof.

[0161] The systems, devices, modules or units described in the above embodiments may be implemented by a computer entity or by a product having a certain function. A typical implementation device is a computer, and the specific form of the computer may be a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email transceiver device, a game console, a tablet computer, a wearable device or a combination of any of these devices.

[0162] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0163] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the embodiments of the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0164] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0165] Moreover, these computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0166] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0167] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A narrow passage method, characterized in that: The method comprises: If there is a passable target narrow road in front of the target vehicle, and the width of the target narrow road is greater than the minimum passable width and less than the sum of the minimum passable width and the preset width, then the alignment posture corresponding to the target narrow road is obtained, and the alignment posture is the posture of the target vehicle before entering the target narrow road; Acquire an initial path based on the current posture of the target vehicle and the aligned posture, and control the target vehicle to travel from the current posture to the aligned posture based on the initial path; After the target vehicle reaches the alignment posture, a target path corresponding to the target narrow road is acquired, and based on the target path, the target vehicle is controlled to pass through the target narrow road and travel from the alignment posture to a target posture, where the target posture is the end posture of the target narrow road; Wherein, the obtaining of the alignment posture corresponding to the target narrow road comprises: traversing a plurality of path points from the starting point of the target narrow road in the direction of the target narrow road; selecting a feature point from the plurality of path points based on the distance from each path point to the obstacle; obtaining a sector-shaped detection area, wherein the vertex position of the sector-shaped detection area is the feature point, the starting direction is the direction of the feature point, and the ending direction is the direction of the starting point of the target narrow road; dividing the sector-shaped detection area longitudinally, and selecting a plurality of sampling points from the longitudinal dividing line; selecting a sampling point from the plurality of sampling points as the target sampling point, and determining the alignment posture corresponding to the target narrow road based on the posture of the target sampling point.

2. The method according to claim 1, characterized in that The method for determining the minimum passable width includes: If the target narrow road is a straight narrow road, the minimum passable width is determined based on the vehicle width of the target vehicle and the configured safe driving distance; or, If the target narrow road is a curved narrow road, the minimum passable width is determined based on the wheelbase of the target vehicle, the curvature of the curve of the target narrow road, the distance from the corner of the vehicle to the center of the front axle, the distance from the corner of the vehicle to the center of the rear axle, and the configured safe driving distance; or, If the target narrow road is a corner narrow road, the minimum passable width is determined based on the minimum turning radius, the minimum turning width, the channel corner size, the channel width, the difference between the turning radius of the vehicle corner point and the safety distance; wherein the minimum turning width is determined based on the vehicle wheelbase, the maximum steering angle of the vehicle's front wheels, the distance from the vehicle corner point to the center of the front axle, the distance from the vehicle corner point to the center of the rear axle, and the safe driving distance.

3. The method according to claim 1 or 2, characterized in that: The process of determining that there is a passable target narrow road in front of the target vehicle includes: Constructing an obstacle distance grid map based on observation data around the target vehicle, wherein the observation data includes data collected by a camera and / or data collected by a radar sensor; Determine a center path based on the obstacle distance grid map, wherein the center path includes a plurality of path points, each of which is a path point with a maximum distance from an obstacle determined based on the obstacle distance grid map; Acquire attribute information corresponding to each path point; wherein the attribute information includes the coordinates and orientation of the path point in the global coordinate system, the curvature of the curve corresponding to the path point, the distance from the path point along the path to the center of the rear axle of the target vehicle, and the distance from the path point to the obstacle; Based on the distance from each path point to the obstacle, if there is a road whose width is greater than the minimum passable width and less than the sum of the minimum passable width and the preset width, it is determined that there is a passable target narrow road.

4. The method according to claim 1, characterized in that Before acquiring the alignment posture corresponding to the target narrow road, the method further includes: determining whether the posture of the target vehicle needs to be adjusted before entering the target narrow road based on the current posture of the target vehicle; if so, performing an operation of acquiring the alignment posture corresponding to the target narrow road; If the current posture does not satisfy the basic non-adjustment condition and the current posture does not satisfy the high-order non-adjustment condition, it is determined that the posture of the target vehicle needs to be adjusted before entering the target narrow road; Alternatively, if the current posture satisfies the basic non-adjustment condition, or the current posture satisfies the high-order non-adjustment condition, it is determined that the posture does not need to be adjusted before the target vehicle enters the target narrow road.

5. The method according to claim 4, characterized in that If it is determined based on the current posture that the midpoint of the head of the target vehicle is located in the target sector area, the midpoint of the tail of the target vehicle is located in the target sector area, and the angle between the first connecting line and the second connecting line is less than a preset angle, then it is determined that the current posture meets the basic non-adjustment condition; otherwise, it is determined that the current posture does not meet the basic non-adjustment condition; Among them, the first connecting line is a connecting line between the midpoint of the head and the starting point of the target narrow road, and the second connecting line is a connecting line between the midpoint of the tail and the starting point of the target narrow road; the vertex position of the target fan-shaped area is the starting point of the target narrow road, and the angle of the target fan-shaped area is determined based on the width of the target narrow road, the vehicle width of the target vehicle and the configured safe driving distance.

6. The method according to claim 4 or 5, characterized in that: Acquire multiple candidate poses from the detection arc corresponding to the current pose; if at least one candidate pose satisfies the basic non-adjustment condition, determine that the current pose satisfies the high-order non-adjustment condition; Alternatively, if all candidate postures do not satisfy the basic non-adjustment condition, determining that the current posture does not satisfy the high-order non-adjustment condition; The radius of the detection arc is determined based on the angle Δψ and the distance Δd, the angle Δψ is the angle between the third connecting line and the second connecting line, and the distance Δd is the length of the second connecting line; the third connecting line is a connecting line between the midpoint of the head of the target vehicle and the midpoint of the tail of the target vehicle, and the second connecting line is a connecting line between the midpoint of the tail of the target vehicle and the starting point of the target narrow road; The origin of the detection arc is located on the perpendicular line of the third connecting line.

7. The method according to claim 1, characterized in that When the target vehicle is at the target sampling point, the following conditions are met: There is no obstacle in front of the target vehicle; The distance from the rear side of the target vehicle to the obstacle is greater than a first distance threshold; When there is an obstacle on the first side of the target vehicle, the distance from the target vehicle to the obstacle on the side is greater than a second distance threshold, and there is no obstacle in the designated area on the second side of the target vehicle; The first side is the left side and the second side is the right side, or the first side is the right side and the second side is the left side.

8. The method according to claim 1, characterized in that The obtaining of the target path corresponding to the target narrow road includes: Acquire a centered coarse path, wherein the centered coarse path includes a plurality of path points; Acquire multiple configured fixed curvature paths, each fixed curvature path including multiple path points; For each fixed curvature path, the path cost corresponding to the fixed curvature path is determined based on the shortest distance from the path point on the fixed curvature path to the obstacle and the shortest distance from the path point to the centered coarse path; based on the path cost corresponding to each fixed curvature path, the fixed curvature path corresponding to the minimum path cost is determined as the target path corresponding to the target narrow road; wherein the starting point of the target path is the path point corresponding to the current posture of the target vehicle.

9. The method according to claim 8, characterized in that The determining the path cost corresponding to the fixed curvature path based on the shortest distance from the path point on the fixed curvature path to the obstacle and the shortest distance from the path point to the central coarse path includes: The path cost corresponding to the fixed curvature path is determined using the following formula: J represents the path cost corresponding to the fixed curvature path, j represents the jth path point on the fixed curvature path, M represents the total number of path points, w1 and w2 represent the configured weight coefficients, represents the shortest distance from the jth path point to the obstacle, represents the shortest distance from the jth path point to the centered coarse path.

10. A narrow passage device, characterized in that: The device comprises: an acquisition module, configured to acquire, if there is a passable target narrow road in front of the target vehicle, and the width of the target narrow road is greater than the minimum passable width and less than the sum of the minimum passable width and a preset width, an alignment posture corresponding to the target narrow road, the alignment posture being the posture of the target vehicle before entering the target narrow road; and acquire an initial path based on the current posture of the target vehicle and the alignment posture; A control module, configured to control the target vehicle to travel from the current posture to the aligned posture based on the initial path; The acquisition module is further used to acquire a target path corresponding to the target narrow road after the target vehicle reaches the alignment posture; The control module is further used to control the target vehicle to pass through the target narrow road based on the target path, and drive from the alignment posture to a target posture, where the target posture is the end posture of the target narrow road; Among them, when the acquisition module acquires the alignment posture corresponding to the target narrow road, it is specifically used to: traverse multiple path points from the starting point of the target narrow road in the direction of the target narrow road; select feature points from the multiple path points based on the distance from each path point to the obstacle; acquire a fan-shaped detection area, the vertex position of the fan-shaped detection area is the feature point, the starting direction is the direction of the feature point, and the ending direction is the direction of the starting point of the target narrow road; divide the fan-shaped detection area vertically, and select multiple sampling points from the vertical division line; select a sampling point from the multiple sampling points as the target sampling point, and determine the alignment posture corresponding to the target narrow road based on the posture of the target sampling point.

11. A vehicle, characterized in that: include: a processor and a machine-readable storage medium storing machine-executable instructions executable by the processor; The processor is used to execute machine executable instructions to implement the method described in any one of claims 1-9.

Images