Control method for vehicle parking-out and vehicle
By acquiring obstacle location information and planning the journey in stages when the vehicle is parking, the safety issue of long-distance driving during automatic parking is solved, achieving safe and reliable long-distance parking and improving the user experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGZHOU XIAOPENG CONNECTIVITY TECH CO LTD
- Filing Date
- 2022-09-20
- Publication Date
- 2026-05-15
AI Technical Summary
In existing technologies, it is difficult for vehicles to safely and reliably plan long-distance driving when automatically parking, as they are easily interfered with by obstacles along the way, posing safety hazards.
Upon receiving a parking instruction, the vehicle travels a preset distance along the parking direction to obtain obstacle location information. Based on historical driving paths and obstacle location information, the type of parking path is determined, and a phased driving strategy is adopted, including first driving to the first target point and then driving to the second target point, to reasonably avoid obstacles.
It enables safe and reliable long-distance parking of vehicles, improves user experience, and effectively extends parking distance through a wider detection field of view and phased driving planning, ensuring driving safety and stability.
Smart Images
Figure CN115489517B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of autonomous driving technology, and in particular to a control method and vehicle for parking. Background Technology
[0002] With the development of autonomous driving technology, vehicles have automatic parking functions, which can help users solve parking problems.
[0003] In related technologies, when a vehicle automatically parks itself out of a parking space, it typically uses various radars and cameras installed on the vehicle to sense surrounding obstacles and plans its exit based on real-time detection results. However, this method can only park the vehicle a short distance away from the parking space. If the vehicle travels a long distance, it is easily affected by obstacles along the way, posing a safety hazard. Summary of the Invention
[0004] To address or partially address the problems existing in related technologies, this application provides a vehicle parking control method and a vehicle that can safely and reliably control the vehicle to automatically park out of a parking space and perform long-distance planned driving.
[0005] The first aspect of this application provides a method for controlling vehicle parking, comprising:
[0006] Upon receiving a parking instruction, the vehicle will travel a preset distance along the parking direction to obtain the location information of obstacles around the vehicle.
[0007] Based on the historical driving routes containing the current parking space obtained in advance, the type of parking exit route of the vehicle is determined according to a preset determination rule; wherein, the parking exit route includes one or more target points;
[0008] When the parking path is determined to be a multi-target path, according to the obstacle location information and a preset driving rule, the vehicle is first driven from its current location to the first target point, and then driven from the first target point to the second target point.
[0009] In one embodiment, the step of traveling the vehicle a preset distance along the parking direction includes:
[0010] The vehicle will travel a preset distance based on the distance between the vehicle and the guide line in the parking direction;
[0011] Based on the detection equipment on the vehicle, the location information of obstacles around the vehicle's current position is obtained.
[0012] In one embodiment, determining the type of parking exit path according to a preset determination rule based on pre-obtained historical driving paths containing the current parking space includes:
[0013] Generate a corresponding reference line based on the historical driving path;
[0014] Based on the curvature of the reference line, semantic segmentation is performed on the reference line to obtain the corresponding segmentation result;
[0015] When the segmentation result includes straight sections and curves, the positions of the first target point and the second target point on the reference line are determined respectively.
[0016] The type of parking exit path is determined based on the angle between the virtual pose at the second target point and the vehicle, as well as the lateral projection distance.
[0017] In one embodiment, determining the type of the corresponding parking path based on the angle between the virtual pose of the second target point and the vehicle, and the lateral projection distance, includes:
[0018] If the angle between the virtual pose at the second target point and the vehicle is 0°±5°, and the lateral projection distance is greater than or equal to a preset value, then the parking path type is a multi-target path with an S-shaped curve.
[0019] If the angle between the virtual pose at the second target point and the vehicle is 180°±5°, and the lateral projection distance is greater than or equal to a preset value, then the parking path type is a multi-target path with a U-shaped curve.
[0020] In one embodiment, after performing semantic segmentation on the reference line based on its curvature to obtain the corresponding segmentation result, the method further includes:
[0021] When the segmentation result is a straight road, the type of the parking exit path is determined to be a single-target point path with an L-shaped curve; or
[0022] The step of determining the type of parking exit path based on the angle between the virtual pose of the second target point and the vehicle, and the lateral projection distance, includes:
[0023] When the segmentation result includes straight roads and curves, if the angle between the virtual pose located at the second target point and the vehicle is 5° to 175°, or the lateral projection distance is less than a preset value, then the parking path is an L-shaped curve.
[0024] In one embodiment, the step of driving the vehicle from its current position to a first target point according to the obstacle location information and a preset driving rule includes:
[0025] When the simulated driving path of the vehicle from the current position to the first target point is a smooth curve and does not interfere with the position of the obstacle, the direction of the vehicle is adjusted to the specified direction and the vehicle continues to drive so that the vehicle can travel from the current position to the first target point.
[0026] In one embodiment, the step of driving the vehicle from its current position to a first target point according to the obstacle location information and a preset driving rule includes:
[0027] When the simulated driving path of the vehicle from the current position to the first target point cannot form a smooth curve, or when the formed smooth curve interferes with the position of the obstacle, the vehicle's direction is adjusted multiple times and the vehicle continues to drive so that the vehicle can travel from the current position to the first target point.
[0028] In one embodiment, the step of driving the vehicle from the first target point to the second target point includes:
[0029] When the simulated driving path of the vehicle from the first target point to the second target point is a smooth curve and does not interfere with the position of the obstacle, the vehicle's direction is adjusted to the specified direction and the vehicle continues to drive so that the vehicle can travel from the first target point to the second target point.
[0030] In one embodiment, the step of driving the vehicle from the first target point to the second target point includes:
[0031] When the simulated driving path of the vehicle from the first target point to the second target point cannot form a smooth curve, or when the formed smooth curve interferes with the position of the obstacle, the vehicle's direction is adjusted multiple times and the vehicle continues to drive so that the vehicle can travel from the first target point to the second target point.
[0032] In one embodiment, the method further includes:
[0033] When the angle between the virtual pose of the vehicle and the first target point is less than 45°, if the vehicle travels a distance greater than a preset distance and the travel time is greater than a preset duration, the vehicle travels under the control of the preset planning rules.
[0034] In one embodiment, the parking command is triggered by an internal device of the vehicle or remotely triggered by an external device.
[0035] In one embodiment, the method further includes:
[0036] When the parking path is determined to be a single-target path, the vehicle is driven from its current position to the first target point according to the obstacle location information and a preset driving rule.
[0037] A second aspect of this application provides a vehicle comprising:
[0038] The detection module is used to travel a preset distance along the parking direction when a parking command is received, and to obtain the location information of obstacles around the vehicle.
[0039] The type determination module is used to determine the type of the parking exit path of the vehicle according to a preset determination rule based on the historical driving path containing the current parking space obtained in advance; wherein the parking exit path includes one or more target points;
[0040] The driving module is used to, when the parking path is determined to be a multi-target path, drive the vehicle from its current position to a first target point according to the obstacle location information and a preset driving rule, and then drive the vehicle from the first target point to a second target point.
[0041] A third aspect of this application provides a vehicle, comprising:
[0042] Processor; and
[0043] A memory that stores executable code, which, when executed by the processor, causes the processor to perform the method described above.
[0044] A fourth aspect of this application provides a computer-readable storage medium having executable code stored thereon, which, when executed by a processor of a vehicle, causes the processor to perform the method described above.
[0045] The technical solution provided in this application may include the following beneficial effects:
[0046] The vehicle parking control method of this application first moves the vehicle a short distance according to a preset distance, allowing the vehicle's detection equipment to have a wider detection field of view and more accurately acquire obstacle information around the vehicle. Then, it determines the type of parking path based on the historical driving path, and plans a driving strategy according to different types of parking paths. When the parking path is a multi-target point path, based on the historical driving path as a reference path and according to the detected obstacle position information, the vehicle is driven to the first target point in stages, and then to the second target point. This reasonably avoids obstacles while enabling the vehicle to reliably park over a long distance, effectively extending the vehicle's parking distance and improving the user experience.
[0047] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description
[0048] The above and other objects, features and advantages of this application will become more apparent from the more detailed description of exemplary embodiments thereof in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments thereof.
[0049] Figure 1 This is a schematic flowchart illustrating the vehicle parking control method according to an embodiment of this application;
[0050] Figure 2 This is a plan view of a parking lot in which the vehicle parking control method shown in the embodiments of this application is applied;
[0051] Figure 3 This is another schematic flowchart illustrating the vehicle parking control method shown in the embodiments of this application;
[0052] Figure 4 This is a schematic diagram of a single-target point path shown in an embodiment of this application;
[0053] Figure 5 This is a schematic diagram illustrating a multi-target point path according to an embodiment of this application;
[0054] Figure 6 This is a schematic diagram illustrating another multi-target point path in an embodiment of this application;
[0055] Figure 7 This is a schematic diagram of a smooth curve in an embodiment of this application;
[0056] Figure 8 This is a schematic diagram of a non-smooth curve in an embodiment of this application;
[0057] Figure 9 This is a schematic diagram of the vehicle structure shown in the embodiments of this application;
[0058] Figure 10 This is another structural schematic diagram of the vehicle shown in the embodiments of this application;
[0059] Figure 11 This is a schematic diagram of the vehicle structure shown in the embodiments of this application. Detailed Implementation
[0060] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to make this application more thorough and complete, and to fully convey the scope of this application to those skilled in the art.
[0061] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.
[0062] It should be understood that although the terms "first," "second," "third," etc., may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.
[0063] In related technologies, automatic parking technology for vehicles is limited to driving the vehicle to the vicinity of the parking space, and it is difficult to reliably park the vehicle over long distances.
[0064] To address the aforementioned issues, this application provides a vehicle parking control method that can safely and reliably control a vehicle to automatically park out of a parking space and perform long-distance planned driving.
[0065] The technical solutions of the embodiments of this application are described in detail below with reference to the accompanying drawings.
[0066] Figure 1 This is a schematic flowchart illustrating the vehicle parking control method shown in the embodiments of this application.
[0067] See Figure 1 An embodiment of this application illustrates a vehicle parking control method, comprising:
[0068] S110, upon receiving a parking instruction, will drive the vehicle a preset distance along the parking direction and obtain the location information of obstacles around the vehicle.
[0069] The system involves a vehicle parked in its current parking space. Upon receiving a parking exit command, the vehicle activates its automatic parking function, allowing it to autonomously exit from the current space to its final parking destination. Optionally, the parking exit command can be triggered by either internal vehicle devices or remotely via external devices. For example, a user can trigger the parking exit command from inside the vehicle by controlling its internal devices; or, a user can send the command from outside the vehicle via an external device such as a linked remote control device or an application on a user terminal; or, the command can be sent to the corresponding vehicle via a remote server—there are no restrictions on this.
[0070] It is understood that when a vehicle is parked in the current parking space, there may be obstacles on the left and / or right sides of the vehicle. These obstacles could be other vehicles, walls, or pillars. Since vehicles are generally equipped with detection equipment such as radar and cameras, these obstacles can partially obstruct the detection equipment, thus affecting the monitoring field of view. Therefore, the vehicle can be driven a preset distance in the parking exit direction to avoid obstructing the vehicle's detection equipment. The parking exit direction can be the direction of exiting the current parking space. It should be noted that the parking space type applicable to this application's method can be... Figure 2 The perpendicular parking space shown indicates that the current parking space where the vehicle is located is a perpendicular parking space, which is different from the lateral parking space used in related technologies. Furthermore, the preset distance can be 0.5 meters to 2 meters. By moving the vehicle a short distance, causing part of the vehicle body to move out of the current parking space, the detection equipment can obtain a wider detection field of view to monitor the position information of surrounding obstacles.
[0071] like Figure 2 As shown, the location information of obstacles around the vehicle includes the locations of obstacles to the left, right, and in the parking direction. By determining the location information of each obstacle, it is easy to monitor the distance between the vehicle and the obstacles in real time during the vehicle's movement, thereby preventing the vehicle from scraping against obstacles during parking.
[0072] S120, based on the historical driving path including the current parking space obtained in advance, determine the type of the vehicle's parking exit path according to the preset judgment rules; wherein, the parking exit path type includes one or more target points.
[0073] Historical driving routes can be obtained using relevant technologies. For example, within the current parking lot, historical driving routes can be acquired and stored in the vehicle through vehicle learning. These routes include a portion or the entire driving route from a preset location in the parking lot, such as the parking lot entrance, to the current parking space. Historical driving routes can also be obtained by requesting from a cloud server; there are no restrictions on the source. It should be noted that within the current parking lot, driving routes whose endpoint is the current parking space, or whose driving route passes through the current parking space, can all be used as historical driving routes for the current vehicle to exit. It can be understood that the number of historical driving routes in the same parking lot can be greater than or equal to one. When the number of historical driving routes is greater than one, optionally, in this step, the historical driving route with the shortest driving distance, starting from the vehicle's current parking space and ending at the location selected by the user, can be determined as the reference route for the exit. Alternatively, a historical driving route directly selected by the user can also be used as the reference route for the exit. It should be noted that the parking exit path refers to the path a vehicle takes from the current parking space to the parking exit destination. The parking exit path may be a part of the historical driving path, that is, not the entire historical driving path. The parking exit destination is located on the historical driving path and may or may not be a location selected by the user.
[0074] Furthermore, the type of parking exit path can be preset, including single-target point paths and multi-target point paths. A single-target point path refers to a parking exit path with only one target point as the parking exit endpoint; a multi-target point path refers to a parking exit path with multiple segments and corresponding target points. For example, a multi-target point path includes two or more target points, each of which is the endpoint of its corresponding segment, and the last target point is the parking exit endpoint of the entire path. It can be understood that corresponding planning and driving strategies can be adopted based on the different types of parking exit paths.
[0075] In this step, based on the determined historical driving path, the type of the current parking path can be identified through preset judgment rules. For example, when there is only one target point, the parking path type is a single-target-point path. When there is more than one target point, the parking path type is a multi-target-point path. It should be noted that the number of target points mentioned here refers to the actual number of target points that are ultimately available for driving, not the number of target points derived during the calculation process.
[0076] S130, when the parking path is determined to be a multi-target point path, according to the obstacle location information and the preset driving rules, the vehicle is first driven from the current position to the first target point, and then driven from the first target point to the second target point.
[0077] In this step, when the parking exit path is determined to be a multi-target point path, the entire parking exit path can be divided into multiple segments based on target points at different locations. This allows for the planning of driving strategies based on different segments, enabling autonomous driving in different segments according to corresponding preset driving rules.
[0078] Taking a parking path with two target points as an example, the parking path includes a first target point and a second target point. The first target point is located closer to the vehicle, and the second target point is located farther away. According to preset driving rules, the segment from the vehicle's current position to the first target point is the first segment of the parking path, and the segment from the first target point to the second target point is the second segment of the parking path. Optionally, different preset driving rules can be used to plan driving for different segments, without limitation. Compared to the traditional strategy of planning driving with a single destination as the target, this method of planning driving for different target points is more conducive to overcoming obstacles faced by different road segments and ensuring driving safety. At the same time, it can also perform long-distance parking, that is, not only moving the vehicle from the current parking space to the vicinity of the parking space, but also moving the vehicle from the current parking space to the second target point farther away from the parking space.
[0079] As can be seen from this example, the vehicle parking control method of this application first moves the vehicle a short distance according to a preset distance, allowing the vehicle's detection equipment to have a wider detection field of view and more accurately obtain obstacle information around the vehicle. Then, it determines the type of parking path based on the historical driving path, and thus plans a driving strategy according to different types of parking paths. When the parking path is a multi-target point path, based on the historical driving path as a reference path and according to the detected obstacle position information, the vehicle is driven to the first target point in stages, and then to the second target point. This reasonably avoids obstacles while enabling the vehicle to reliably park over a long distance, effectively extending the vehicle's parking distance and improving the user experience.
[0080] Figure 3 This is a schematic flowchart illustrating the vehicle parking control method shown in the embodiments of this application.
[0081] See Figure 3 An embodiment of this application illustrates a vehicle parking control method, comprising:
[0082] S210, upon receiving a parking instruction, when an obstacle is detected on at least one side of the vehicle, the vehicle travels a preset distance based on the distance between the vehicle and the guide line in the parking direction; and the vehicle obtains the location information of obstacles around the vehicle's current position based on the detection equipment on the vehicle.
[0083] In this step, please refer to the following: Figure 3 and Figure 4 The guide line can be the center line of the road preceding the parking direction. It can be understood that, based on the historical driving path passing through the current parking space, the guide line can also be based on the center line of the road along the historical driving path, i.e., the reference line mentioned below. Before a clear driving strategy is planned, to avoid vehicles traveling excessive distances and for safety reasons, such as... Figure 3 As shown, using the guide lines as a reference, and based on the vehicle's orientation within the parking space, the distance H between the edge of the vehicle's front or rear and the guide lines should be greater than 0 to prevent vehicles from exceeding the guide lines. Figure 4 As shown, the preset distance L for the vehicle to travel is 0.5 meters to 2 meters, provided that it does not exceed the guide line.
[0084] The detection equipment on the vehicle can include various types of radar and camera devices located on the vehicle body, such as ultrasonic radar, millimeter-wave radar, surround-view cameras, etc., without limitation. After the vehicle has moved a preset distance, the detection equipment can accurately obtain the location information of obstacles located in different directions of the vehicle, and determine the distance between each obstacle and the vehicle body based on the obstacle location information.
[0085] S220: Generate a corresponding reference line based on the historical driving path; perform semantic segmentation on the reference line based on its curvature to obtain the corresponding segmentation result; determine the type of the corresponding parking exit path based on the segmentation result, and determine the first target point or the first target point and the second target point corresponding to the parking exit path.
[0086] Based on the determined historical driving path, a corresponding reference line can be drawn as the reference line for the parking exit path of the corresponding road segment. Generally, the historical driving path consists of straight sections and / or curves; similarly, the reference line contains straight lines and / or curves. Based on the curvature at different positions on the reference line, the corresponding road segment on the reference line can be classified as a straight section or a curve. Optionally, when the curvature of the line segment is less than a preset curvature value, it indicates that the corresponding road segment is a straight section; otherwise, it is a curve. For example, the preset curvature value can be 0.05 to 0.1, such as 0.05, 0.07, 0.09, or 0.1.
[0087] It is understandable that semantic segmentation of the reference line using curvature can determine the segmentation result of the reference line containing the parking exit path. For example, the segmentation result may show that the reference line includes straight sections; or, the segmentation result may show that the reference line includes both straight sections and curves, with at least one curve. It should be noted that the parking exit path refers to a relatively long route taken by the vehicle from the current parking space, not necessarily the entire historical driving path. Therefore, the segmentation result of the reference line only needs to correspond to the segment containing the parking exit path, meaning that segmentation of the complete historical driving path is not necessary.
[0088] In one implementation, when the segmentation result includes a straight path, the berthing path type is a single-target point path. Further, as... Figure 4 As shown, the parking exit path can be a single-target-point path with an L-shaped curve, meaning the shape of the vehicle's path from its current position to the first target point is similar to an L-shape. Further, when the parking exit path is a single-target-point path, it contains only one target point, which is the first target point. The first target point is set on a straight section as the parking exit endpoint. Optionally, to accurately obtain the position of the first target point on the single-target path, in one embodiment, the projection point of the current parking space on the reference line is used as the initial point; starting from the initial point, the first target point is set on a straight section as the parking exit endpoint according to a first preset adjustment distance. That is, the distance between the initial point S0 and the first target point S1 is the first preset adjustment distance A.
[0089] In one embodiment, when the segmentation result includes straight sections and curves, a first sample target point can be designated on the straight section closest to the current parking space, and a second sample target point can be designated on the curve connected to the straight section. The first sample target point can be located at the end of the straight section (the end furthest from the current parking space), and the second sample target point is located at the end of the curve, i.e., the exit point of the curve. It should be noted that the first sample target point is not necessarily the first target point, and the second sample target point is not necessarily the second target point; furthermore, the parking exit path may or may not have a second target point, but it will always have a first target point. Specific details will be provided later.
[0090] To accurately determine the type of parking path when the segmentation result includes both straight and curved sections, in one embodiment, the positions of the first and second sample target points on the reference line are adjusted according to a preset adjustment method to determine their respective positions. In a specific embodiment, the projection point of the current parking space on the reference line is used as the initial point; starting from the initial point, the position of the first sample target point is adjusted according to a first preset adjustment distance to obtain the position of the first target point on the reference line; then, the position of the second sample target point is adjusted according to a second preset adjustment distance to obtain the position of the second target point on the reference line. That is, as... Figure 5As shown, the distance between the first target point S1 and the initial point S0 on the reference line is the first preset adjustment distance A, and the distance between the second target point S2 and the first target point S1 on the reference line is the second preset adjustment distance B. For example, A can be 5 to 10 meters, and B can be 15 to 30 meters. It should be noted that this distance is not the straight-line distance between the two points, but rather the length of the reference line between them. By adjusting the positions of the first and second sample target points on the reference line, the positions of the first and second target points can be obtained. Based on this, it is necessary to further determine the type of berthing path according to the following scheme.
[0091] In one embodiment, the type of parking exit path is determined based on the angle between the virtual pose at the second target point and the vehicle, and the lateral projection distance. For example... Figure 5 and Figure 6 As shown, in a specific implementation, if the angle between the virtual pose at the second target point and the vehicle in the current parking space is 0°±5°, and the lateral projection distance is greater than or equal to a preset value, then the parking path is an S-shaped curve; if the angle between the virtual pose at the second target point and the vehicle in the current parking space is 180°±5°, and the lateral projection distance is greater than or equal to a preset value, then the parking path is a U-shaped curve. Wherein, as... Figure 5 and Figure 6 As shown, the lateral projection distance D refers to the projected distance from the rear axle center of the current vehicle location to the straight line containing the second target point. Wherein, Figure 5 The angle between the virtual pose of the vehicle and the second target point is approximately 180°. Figure 6 The angle between the virtual pose of the vehicle and the second target point is approximately 0°.
[0092] In one specific implementation, when the segmentation result includes straight sections and curves, if the angle between the virtual pose at the second target point and the vehicle in the current parking space is 5° to 175°, or the lateral projection distance is less than a preset value, then the parking path is... Figure 4 The L-shaped curve is shown. In other words, although the segmentation result of the reference line includes both the straightaway and the curve, when the aforementioned conditions are met, the parking path type is still a single-target point path, that is, it only contains the first target point, and the second target point needs to be discarded to ensure the safety of autonomous driving.
[0093] S230, when the parking path type is a multi-target point path, according to the obstacle location information and the preset driving rules, the vehicle is first driven from the current position to the first target point, and then the vehicle is driven from the first target point to the second target point.
[0094] Based on the identification result of step S220, the parking path type may be a multi-target point path with a U-shaped bend or an S-shaped bend, both of which include a first target point and a second target point. The first and second road segments can use the same preset driving rules.
[0095] In one embodiment, when the simulated driving path of the vehicle from its current position to the first target point is a smooth curve and does not interfere with the position of obstacles, the vehicle's direction is adjusted to a designated direction and driving continues, so that the vehicle travels from its current position to the first target point. Wherein, for example... Figure 7 As shown, a smooth curve is a curve formed by connecting a straight line and an arc. The starting point of the straight line is connected to the center of the rear axle of the vehicle, and the ending point of the straight line is connected to the starting point of the arc. The ending point of the arc is connected to the first target point. According to the geometric sampling method in related technologies, when any smooth curve can directly connect the vehicle and the first target point, or when the error of the connected smooth curve is within a preset error range, and the smooth curve does not interfere with surrounding obstacles, the vehicle's wheels only need to be adjusted to the corresponding designated direction once by the steering wheel. Without needing to adjust the direction multiple times, the vehicle can use a smooth curve as the planned simulated driving path to directly reach the first target point from its current position. The preset error range means that the angle β between the virtual vehicle pose at the end of the smooth curve and the virtual vehicle pose at the first target point is less than 15°. That is, even if the end of the smooth curve is not the first target point, but is still near the first target point, it is still within an acceptable range.
[0096] The opposite of the above, such as Figure 8 As shown, in one embodiment, when the simulated driving path of the vehicle from its current position to the first target point cannot form a smooth curve, or when the formed smooth curve interferes with the position of the obstacle, the vehicle's direction is adjusted multiple times and the vehicle continues to drive so that it can travel from its current position to the first target point. In other words, when the wheels cannot directly exit using the above method due to surrounding obstacles or narrow space, a geometric parking maneuver can be used. This involves adjusting the vehicle's direction multiple times while it is moving, allowing the vehicle to successfully reach the first target point.
[0097] Specifically, when executing the geometric parking maneuver, as the vehicle moves, multiple smooth curves with different radii are continuously and synchronously drawn in real time to gradually reduce the angle between the end of the curve and the first target point until the angle error is within a preset error range and does not interfere with the position of obstacles. The wheel direction is then adjusted synchronously according to the corresponding smooth curves to drive until the first target point is reached. In other words, in the current geometric parking maneuver, the vehicle will drive according to the real-time changing smooth curves, and the driving direction of the parking path will be adjusted in real time.
[0098] Similarly, once the vehicle reaches the first target point, the same geometric sampling method is used to connect the first and second target points. If a smooth curve of the above type can be found that directly connects the first and second target points, or if the angle between the connected points is within a preset error range, the vehicle can adjust its wheels to the corresponding designated direction and drive directly from the first target point to the second target point, thus completing a long-distance parking maneuver. Conversely, if the geometric sampling method cannot be used when the vehicle departs from the first target point, a geometric rubbing-in-the-loop method is used until the second target point is reached.
[0099] S240, when the parking path type is a single target point path, the vehicle is driven from the current position to the first target point according to the obstacle position information and the preset driving rules.
[0100] It is understandable that either step S240 or step S230 can be performed.
[0101] Similarly, in this step, the geometric sampling method or geometric rubbing method in step S230 can be used to drive the vehicle from the current position to the first target point, which will not be elaborated here.
[0102] It is understood that in the above steps, the vehicle can successively execute the corresponding results during the strategy planning process, enabling the vehicle to continue driving. To effectively improve driving efficiency, in one embodiment, when the angle between the vehicle and the virtual pose of the first target point is less than 45°, if the vehicle's continuous driving distance is greater than a preset driving distance (e.g., 8 to 15 meters) and the driving time is greater than a preset duration (e.g., 2 to 5 seconds), the vehicle's driving is switched to a preset planning rule. That is, during the execution of steps S210 to S240, if the aforementioned conditions are met, the vehicle's driving control method can be switched. The preset planning rule can be a method based on lattice polynomial curve generation in related technologies. This method is suitable for curves with small curvature and relatively wide spaces, i.e., suitable for driving paths that tend to be smooth and far from obstacles. Conversely, when the angle between the vehicle and the virtual pose of the first target point is detected to be greater than 45°, or the vehicle's continuous driving distance is less than or equal to the preset driving distance, or the driving time is less than the preset duration, the control method is not switched, or the changed method is switched back to the control method of this application.
[0103] As can be seen from this example, the vehicle parking control method of this application can accurately determine the type of parking path according to the preset judgment logic after moving the vehicle body out of the preset distance, and thus execute the preset driving rules accordingly. That is, it can satisfy short-distance parking for single-target point paths, as well as long-distance parking for multi-target point paths, flexibly cope with various parking scenarios, and ensure the safety and stability of the parking process, meet user needs, and improve user experience.
[0104] Corresponding to the aforementioned application function implementation method embodiments, this application also provides a vehicle and corresponding embodiments.
[0105] Figure 9 This is a schematic diagram of the vehicle structure shown in the embodiments of this application.
[0106] See Figure 9 The vehicle illustrated in this application includes a detection module 310, a type determination module 320, and a driving module 330, wherein:
[0107] The detection module 310 is used to move the vehicle a preset distance along the parking direction when it receives a parking instruction to obtain the location information of obstacles around the vehicle.
[0108] The type determination module 320 is used to determine the type of the vehicle's parking exit path according to a preset determination rule based on the historical driving path containing the current parking space obtained in advance; wherein, the parking exit path type includes one or more target points.
[0109] The driving module 330 is used to, when the parking path is determined to be a multi-target path, drive the vehicle from its current position to the first target point according to the obstacle location information and the preset driving rules, and then drive the vehicle from the first target point to the second target point.
[0110] Further, see Figure 10 The vehicle also includes a command receiving module 340, which receives parking commands. These commands are triggered by internal vehicle devices or remotely by external devices. The driving module 330, after receiving the parking command from the information receiving module, drives the vehicle a preset distance based on the distance between the vehicle and the guide line in the parking direction. The detection module 310 uses detection devices on the vehicle to acquire information about the positions of obstacles around the vehicle's current location.
[0111] The type determination module 320 includes a segmentation submodule 321 and a recognition submodule 322. The segmentation submodule 321 is used to generate a corresponding reference line based on the historical driving path; and to perform semantic segmentation on the reference line based on its curvature to obtain the segmentation result of the reference line.
[0112] The identification submodule 322 is used to determine the positions of the first target point and the second target point on the reference line when the segmentation result includes straight roads and curves; and to determine the type of the corresponding parking path based on the angle between the virtual pose of the second target point and the vehicle and the lateral projection distance. The parking path types include single-target point paths and multi-target point paths; single-target point paths include L-shaped curves, and multi-target point paths include U-shaped curves and S-shaped curves. Specifically, if the angle between the virtual pose of the second target point and the vehicle is 0°±5°, and the lateral projection distance is greater than or equal to a preset value, then the parking path type is a multi-target path with an S-shaped curve; if the angle between the virtual pose of the second target point and the vehicle is 180°±5°, and the lateral projection distance is greater than or equal to a preset value, then the parking path type is a multi-target path with a U-shaped curve. When the segmentation result includes both straight sections and curves, if the angle between the virtual pose at the second target point and the vehicle is between 5° and 175°, or the lateral projection distance is less than a preset value, then the parking path type is a single-target path with an L-shaped curve. When the segmentation result is a straight section, the parking path type is a single-target path with an L-shaped curve.
[0113] The driving module 330 also includes a path planning module 331 and a path driving module 332. When the identification submodule 322 determines that the parking path is a multi-target point path, the path planning module 331 first uses a geometric sampling method to obtain a smooth curve from the current position to the first target point. When the simulated driving path from the current position to the first target point is a smooth curve and does not interfere with the position of the obstacle, the path driving module 332 adjusts the direction of the vehicle to the specified direction and continues to drive so that the vehicle can drive from the current position to the first target point. When the path planning module 331 determines that the simulated driving path from the current position to the first target point cannot form a smooth curve, or the formed smooth curve interferes with the position of the obstacle, the path planning module 331 also uses a geometric kneading method to draw multiple smooth curves synchronously with different radii in real time as the vehicle moves, in order to reduce the angle between the end of the curve and the first target point, until the angle error is within the preset error range and does not interfere with the position of the obstacle. Then, the path driving module 332 adjusts the direction of the vehicle multiple times according to the corresponding smooth curve and continues to drive so that the vehicle can drive from the current position to the first target point. Similarly, the path planning module 331 and the path driving module 332 use the same method to drive the vehicle from the first target point to the second target point, which will not be elaborated here.
[0114] When the identification submodule 322 determines that the parking path is a single-target point path, the driving module 330 drives the vehicle from its current position to the first target point according to the obstacle location information and preset driving rules. Specifically, the same method described above is executed by the path planning module 331 and the path driving module 332 in the driving module 330, which will not be elaborated here.
[0115] The vehicle in this application also includes a switching module 350, which is used to switch to the preset planning rule to control the vehicle's driving when the angle between the virtual pose of the vehicle and the first target point is less than 45°, and the vehicle's continuous driving distance is greater than the preset driving distance and the driving time is greater than the preset time.
[0116] In summary, the vehicle described in this application, depending on the parking path, can perform long-distance parking to the second target point in stages for multi-target point paths while ensuring safety and stability; for single-target point paths, it can also quickly perform short-distance parking to the first target point. The vehicle described in this application can flexibly handle various parking scenarios and meet user needs.
[0117] Regarding the modules in the above embodiments, the specific ways in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated further here.
[0118] Figure 11 This is a schematic diagram of the vehicle structure shown in the embodiments of this application.
[0119] See Figure 11 The vehicle 1000 includes a memory 1010 and a processor 1020.
[0120] The processor 1020 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.
[0121] Memory 1010 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. ROM may store static data or instructions required by processor 1020 or other modules of the computer. Permanent storage devices may be read-write storage devices. Permanent storage devices may be non-volatile storage devices that retain stored instructions and data even when the computer is powered off. In some embodiments, permanent storage devices use mass storage devices (e.g., magnetic or optical disks, flash memory) as permanent storage devices. In other embodiments, permanent storage devices may be removable storage devices (e.g., floppy disks, optical drives). System memory may be a read-write storage device or a volatile read-write storage device, such as dynamic random access memory. System memory may store some or all of the instructions and data required by the processor during operation. Furthermore, memory 1010 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (e.g., DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, the memory 1010 may include a removable storage device that is readable and / or writable, such as a laser disc (CD), a read-only digital multifunction optical disc (e.g., DVD-ROM, dual-layer DVD-ROM), a read-only Blu-ray disc, a high-density optical disc, a flash memory card (e.g., SD card, mini SD card, Micro-SD card, etc.), a magnetic floppy disk, etc. Computer-readable storage media do not contain carrier waves or transient electronic signals transmitted wirelessly or via wired connections.
[0122] The memory 1010 stores executable code, which, when processed by the processor 1020, can cause the processor 1020 to execute part or all of the methods described above.
[0123] Furthermore, the method according to this application can also be implemented as a computer program or computer program product, which includes computer program code instructions for performing some or all of the steps in the method described above.
[0124] Alternatively, this application may be implemented as a computer-readable storage medium (or a non-transitory machine-readable storage medium or a machine-readable storage medium) storing executable code (or computer program or computer instruction code) thereon, which, when executed by a processor of an electronic device (or server, etc.), causes the processor to perform part or all of the steps of the methods described above according to this application.
[0125] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.
Claims
1. A method for controlling vehicle parking, characterized in that, include: Upon receiving a parking instruction, the vehicle will travel a preset distance along the parking direction to obtain the location information of obstacles around the vehicle. Based on the historical driving paths containing the current parking space obtained in advance, the type of the parking exit path of the vehicle is determined according to a preset judgment rule; wherein, based on the historical driving paths, a corresponding reference line is generated; based on the curvature of the reference line, semantic segmentation is performed on the reference line to obtain the corresponding segmentation result; based on the segmentation result, the type of the corresponding parking exit path is determined; the parking exit path includes one or more target points; When the parking path is determined to be a multi-target path, according to the obstacle location information and a preset driving rule, the vehicle is first driven from its current location to the first target point, and then driven from the first target point to the second target point.
2. The method according to claim 1, characterized in that, The step of driving the vehicle a preset distance along the parking direction includes: The vehicle will travel a preset distance based on the distance between the vehicle and the guide line in the parking direction; Based on the detection equipment on the vehicle, the location information of obstacles around the vehicle's current position is obtained.
3. The method according to claim 1, characterized in that, The step of determining the type of parking exit path based on pre-obtained historical driving paths containing the current parking space and according to preset judgment rules includes: Generate a corresponding reference line based on the historical driving path; Based on the curvature of the reference line, semantic segmentation is performed on the reference line to obtain the corresponding segmentation result; When the segmentation result includes straight sections and curves, the positions of the first target point and the second target point on the reference line are determined respectively. The type of parking exit path is determined based on the angle between the virtual pose at the second target point and the vehicle, as well as the lateral projection distance.
4. The method according to claim 3, characterized in that, The step of determining the type of parking exit path based on the angle between the virtual pose of the second target point and the vehicle, and the lateral projection distance, includes: If the angle between the virtual pose at the second target point and the vehicle is 0°±5°, and the lateral projection distance is greater than or equal to a preset value, then the parking path type is a multi-target path with an S-shaped curve. If the angle between the virtual pose at the second target point and the vehicle is 180°±5°, and the lateral projection distance is greater than or equal to a preset value, then the parking path type is a multi-target path with a U-shaped curve.
5. The method according to claim 3, characterized in that: After performing semantic segmentation on the reference line based on its curvature to obtain the corresponding segmentation result, the method further includes: When the segmentation result is a straight road, the type of the parking exit path is determined to be a single-target point path with an L-shaped curve; or The step of determining the type of parking exit path based on the angle between the virtual pose of the second target point and the vehicle, and the lateral projection distance, includes: When the segmentation result includes straight roads and curves, if the angle between the virtual pose located at the second target point and the vehicle is 5°~175°, or the lateral projection distance is less than a preset value, then the parking path is an L-shaped curve.
6. The method according to claim 1, characterized in that, The step of driving the vehicle from its current position to the first target point according to the obstacle location information and a preset driving rule includes: When the simulated driving path of the vehicle from the current position to the first target point is a smooth curve and does not interfere with the position of the obstacle, the direction of the vehicle is adjusted to the specified direction and the vehicle continues to drive so that the vehicle can travel from the current position to the first target point.
7. The method according to claim 1, characterized in that, The step of driving the vehicle from its current position to the first target point according to the obstacle location information and a preset driving rule includes: When the simulated driving path of the vehicle from the current position to the first target point cannot form a smooth curve, or when the formed smooth curve interferes with the position of the obstacle, the vehicle's direction is adjusted multiple times and the vehicle continues to drive so that the vehicle can travel from the current position to the first target point.
8. The method according to claim 6 or 7, characterized in that, The step of driving the vehicle from the first target point to the second target point includes: When the simulated driving path of the vehicle from the first target point to the second target point is a smooth curve and does not interfere with the position of the obstacle, the vehicle's direction is adjusted to the specified direction and the vehicle continues to drive so that the vehicle can travel from the first target point to the second target point.
9. The method according to claim 6 or 7, characterized in that, The step of driving the vehicle from the first target point to the second target point includes: When the simulated driving path of the vehicle from the first target point to the second target point cannot form a smooth curve, or when the formed smooth curve interferes with the position of the obstacle, the vehicle's direction is adjusted multiple times and the vehicle continues to drive so that the vehicle can travel from the first target point to the second target point.
10. The method according to claim 1, characterized in that, The method further includes: When the angle between the virtual pose of the vehicle and the first target point is less than 45°, if the vehicle travels a distance greater than a preset distance and the travel time is greater than a preset duration, the vehicle travels under the control of the preset planning rules.
11. The method according to claim 1, characterized in that, The parking command is triggered by either an internal device of the vehicle or a remote device.
12. The method according to any one of claims 1 to 7 or 10 to 11, characterized in that, The method further includes: When the parking path is determined to be a single-target path, the vehicle is driven from its current position to the first target point according to the obstacle location information and a preset driving rule.
13. A vehicle, characterized in that, include: The detection module is used to travel a preset distance along the parking direction when a parking command is received, and to obtain the location information of obstacles around the vehicle. The type determination module is used to determine the type of the parking exit path of the vehicle according to a preset determination rule based on the historical driving path containing the current parking space obtained in advance; wherein, the parking exit path includes one or more target points; a corresponding reference line is generated based on the historical driving path; the reference line is semantically segmented based on the curvature of the reference line to obtain the corresponding segmentation result; and the type of the corresponding parking exit path is determined based on the segmentation result. The driving module is used to, when the parking path is determined to be a multi-target path, drive the vehicle from its current position to a first target point according to the obstacle location information and a preset driving rule, and then drive the vehicle from the first target point to a second target point.
14. A vehicle, characterized in that, include: processor; as well as A memory having executable code stored thereon, which, when executed by the processor, causes the processor to perform the method as described in any one of claims 1-12.
15. A computer-readable storage medium having executable code stored thereon, which, when executed by a processor of a vehicle, causes the processor to perform the method as described in any one of claims 1-12.