Early warning protection method and device for automatic parking assistance

Abstract

The invention provides an early warning protection method and device for automatic parking assistance, and relates to the technical field of combined control of vehicle subsystems of different types or different functions. The method comprises the following steps: generating a planned path trajectory; determining a preview point according to the speed; predicting whether an obstacle point exists in the expansion model between the current target path point and the preview point, issuing first-level early warning when the obstacle point exists, and controlling the speed to be reduced to be not higher than a first speed; predicting and generating a kinematic path through the vehicle kinematic model; predicting whether an obstacle point exists in the vehicle expansion model on the kinematic path or not, and if yes, issuing second-level early warning and controlling the speed to be reduced to be not higher than a second speed; and when an obstacle point exists in the expansion model at present, the obstacle point is outside the body model and is in the current advancing direction, issuing three-level early warning and controlling to enter an emergency stop state. Accuracy and safety of the automatic parking function are guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of joint control of different types or different functions of vehicle subsystems, in particular to the technical field of predicting or avoiding possible or imminent collisions and automatic operation for parking, and more particularly to a pre-warning protection method and device for automatic parking assistance. BACKGROUND

[0002] The automated parking assistance (APA) system belongs to the category of advanced driving assistance systems (ADAS), and the APA system detects the position of obstacles through sensor fusion and algorithm control, plans a parking path, and automatically controls the steering wheel to complete the parking operation. The system mainly relies on the perception of the vehicle environment, the identification and judgment of the parking space, and the dynamic path planning to achieve automatic parking. Compared with other ADAS functions, the APA function is more suitable for low-speed scenarios.

[0003] As a popular automatic assistance driving function, APA has derived a series of pre-warning protection functions. At present, the pre-warning protection of APA depends largely on the hardware performance of the ultrasonic probe. When the user experiences this function and is not focused, it may cause a poor driving experience due to the disappearance or diversion of ultrasonic noise. SUMMARY

[0004] To solve the above problems in the prior art, in a first aspect, embodiments of the present application provide a method for pre-warning protection of automatic parking assistance, the method comprising: generating a planning path trajectory according to an initial pose of a vehicle and information of a target parking space, the planning path trajectory comprising a plurality of planning path points; determining a preview point farther than a current target path point on the planning path trajectory according to a real-time speed of the vehicle; predicting whether there is an obstacle point within a vehicle inflation model when the vehicle travels to a point between the current target path point and the preview point, and in the case that there is an obstacle point, issuing a first-level pre-warning to a driver of the vehicle and controlling the speed of the vehicle to be reduced to be not higher than a first speed; predicting a kinematic path in a future predetermined time period by a vehicle kinematic model according to a real-time position, a real-time speed and a traveling direction of the vehicle; predicting whether there is an obstacle point within the vehicle inflation model when the vehicle travels on the kinematic path, and in the case that there is an obstacle point, issuing a second-level pre-warning to the driver of the vehicle and controlling the speed of the vehicle to be reduced to be not higher than a second speed, wherein the second speed is lower than the first speed; in the case that there is an obstacle point within the vehicle inflation model at present and the obstacle point is outside a vehicle body model and in a current traveling direction, issuing a third-level pre-warning to the driver of the vehicle and controlling the vehicle to enter an emergency stop state.

[0005] In some embodiments, the method further comprises: in response to controlling the vehicle to enter the emergency stop state, triggering re-generation of the planning path trajectory.

[0006] In some embodiments, the method further comprises: in the case that the driver takes over the vehicle or receives the re-generated planning path trajectory, releasing the emergency stop state of the vehicle.

[0007] In some embodiments, after controlling the vehicle to enter the emergency stop state, the method further comprises: predicting a re-predicted kinematic path in the future predetermined time period by the vehicle kinematic model according to a real-time position, a real-time speed and a traveling direction of the vehicle; predicting whether there is an obstacle point within the vehicle body model when the vehicle travels on the re-predicted kinematic path; detecting whether a distance between the vehicle and the obstacle point increases according to the real-time position of the vehicle; in the case that there is no obstacle point within the vehicle body model and the distance between the vehicle and the obstacle point increases, releasing the emergency stop state of the vehicle.

[0008] In some embodiments, the pre-warning sensitivity is adjustable according to one or more of the following: a boundary expansion size of the vehicle inflation model compared to the vehicle body model; a number of path points spaced between the preview point and the current target path point; a time length of the predetermined time period.

[0009] In some embodiments, the method further includes predicting whether an obstacle point exists within the vehicle expansion model when the vehicle passes through the plurality of planned path points, and sending a prompt message to the vehicle driver if an obstacle point exists.

[0010] In some embodiments, the execution frequency of generating the kinematic path within a predetermined future time period through prediction by the vehicle kinematic model is greater than or equal to 30 Hz and less than or equal to 100 Hz.

[0011] In some embodiments, the real-time speed of the vehicle is calculated based on pulse values ​​measured by wheel speed pulse meters disposed on rear wheels of the vehicle.

[0012] In some embodiments, the initial position of the vehicle is determined based on the position, speed, and direction of travel of the vehicle when the automatic parking assistance function is activated.

[0013] In a second aspect, an embodiment of the present invention provides a warning protection device for automatic parking assistance, the device comprising a memory and a processor, the memory storing a computer program, which implements the warning protection method for automatic parking assistance described in any of the above embodiments when the computer program is executed by the processor.

[0014] The automatic parking assistance warning protection method and device proposed in the embodiment of the present invention ensure the accuracy of function triggering by combining vehicle speed with real-time positioning and estimating the vehicle posture at a certain frequency using the vehicle kinematic model; the kinematic model is combined with the planned collision model posture estimation to ensure the accuracy of the vehicle's emergency stop function and optimize the operating experience of the APA function when the function is triggered; combined with the vehicle kinematic model, obstacle collision prediction is added to the vehicle, which promotes the continuity of the APA function and ensures the normal deployment of the function; multi-party verification is performed through predictions of multiple models to ensure the safety of passengers in the automatic parking state. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The above and other objects, features and advantages of the embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings, in which several embodiments of the present invention are shown by way of example and not limitation, in which:

[0016] Figure 1 A flowchart of a warning protection method for automatic parking assistance according to an embodiment of the present invention is shown.

[0017] In the drawings, the same or corresponding reference numerals denote the same or corresponding parts. DETAILED DESCRIPTION

[0018] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way.

[0019] In one aspect, an embodiment of the present invention provides a warning protection method for automatic parking assistance. Figure 1 , which shows a flow chart of a warning protection method for automatic parking assistance according to an embodiment of the present invention. The method includes steps S101-S106.

[0020] In step S101, a planned path trajectory is generated based on the vehicle's initial posture and information about the target parking space. The planned path trajectory includes a plurality of planned path points. The initial posture includes the vehicle's initial position and posture. As one embodiment of the present invention, the vehicle's initial posture is determined based on the vehicle's position, speed, and direction of travel when the automatic parking assistance function is activated. For example, the vehicle's speed when the automatic parking assistance function is activated can be calculated based on the pulse value measured by a wheel speed pulse meter installed on the vehicle's rear wheels. The direction of travel can be obtained based on gear information (e.g., N, P, R, D gear).

[0021] The planned path trajectory can be generated according to any suitable path planning algorithm that exists or will appear in the future.

[0022] In step S102, a preview point that is farther than the current target path point is determined on the planned path trajectory according to the real-time speed of the vehicle.

[0023] As one embodiment of the present invention, the real-time speed of the vehicle is calculated based on the pulse value measured by a wheel speed pulse meter provided on the rear wheel of the vehicle.

[0024] The current target pathpoint is the next pathpoint the vehicle is currently tracking along the planned path. Preview points are calculated backwards from the current target pathpoint along the planned path. For example, if the vehicle is traveling at a high speed, the preview point can be set farther away from the current target pathpoint, meaning that there are more pathpoints between the preview point and the target pathpoint. In other words, on the planned pathpoint, preview points are pathpoints traversed backwards based on vehicle speed. Faster speeds lead to more preview points. For example, a preview point can be 8-15 (inclusive) pathpoints away from the current target pathpoint.

[0025] In step S103, it is predicted whether there is an obstacle point within the vehicle expansion model when the vehicle reaches a point between the current target path point and the preview point. If an obstacle point is present, a first-level warning is issued to the vehicle driver and the vehicle speed is controlled to be reduced to no more than a first speed. The first-level warning may, for example, remind the driver to prepare to brake.

[0026] As an embodiment of the present invention, the first speed may be a predefined fixed value. As an example only, the first speed may be set to a value within a range of 3 m / s or more and 5 m / s or less, such as 4 m / s.

[0027] As another embodiment of the present invention, the first speed may be dynamically set according to the distance between the obstacle and the vehicle at that time.

[0028] The vehicle expansion model is generated by setting expansion parameters based on the vehicle's dimensions. Intuitively, the vehicle expansion model's outline is larger than the vehicle's main body model's outline, and the area it covers forms a warning zone. For example, the boundary of the vehicle expansion model can extend beyond the boundary of the vehicle's main body model by a distance between 10 cm and 20 cm. As another example, the vehicle expansion model can be drawn on a map based on the rear axle center position inferred from wheel speed positioning, vehicle parameters, and expansion parameters.

[0029] In step S104, a kinematic path for a predetermined future time period is predicted and generated using the vehicle kinematic model based on the vehicle's real-time position, speed, and direction of travel. In one embodiment of the present invention, the frequency of predicting and generating the kinematic path for a predetermined future time period using the vehicle kinematic model is greater than or equal to 30 Hz and less than or equal to 100 Hz.

[0030] Generally speaking, a vehicle's kinematic model calculates its path within a predetermined future timeframe based on the vehicle's real-time position, speed, and direction. Unlike path planning when the parking function is activated, the kinematic model does not need to consider the orientation of the target parking space.

[0031] As an example only, the predetermined time period may be set to 600ms, 800ms, 1000ms, etc.

[0032] In step S105, the presence of an obstacle point within the vehicle expansion model while the vehicle is traveling along the kinematic path is predicted. If an obstacle point is present, a secondary warning is issued to the driver and the vehicle's speed is controlled to be no higher than a second speed, where the second speed is lower than the first speed. The secondary warning may, for example, prompt the driver to immediately apply the brakes. For example, the second speed may be set within a range of 0.8 m / s to 2.5 m / s, for example, 1.5 m / s.

[0033] In step S106, based on the vehicle's real-time position, speed, and direction of travel, if an obstacle point is currently present within the vehicle's expansion model and is both outside the vehicle's main body model and in the current direction of travel, a Level 3 warning is issued to the driver, and the vehicle is controlled to enter an Emergency Stop (ESTOP) state. The Level 3 warning can notify the driver that the emergency braking function has been activated.

[0034] As an example, determining whether an obstacle is in the current traveling direction may be determined based on the positional relationship of the obstacle point relative to the vehicle body model.

[0035] The emergency stop state brings the entire vehicle to an abrupt halt. For example, lateral control sends the current steering wheel angle, while longitudinal control sends a deceleration command of -5 to bring the vehicle to a standstill. The ESTOP state can be sent from the decision-making layer to the regulatory control layer via the relevant message middleware.

[0036] As an embodiment of the present invention, the method may further include: in response to controlling the vehicle to enter an emergency stop state, triggering regeneration of the planned path trajectory.

[0037] As an embodiment of the present invention, the method may further include: releasing the emergency stop state of the vehicle when the driver takes over the vehicle or receives a regenerated planned path trajectory.

[0038] As one embodiment of the present invention, after controlling the vehicle to enter an emergency stop state, the method may further include: generating a re-predicted kinematic path for a predetermined future time period using a vehicle kinematic model based on the vehicle's real-time position, real-time speed, and direction of travel; predicting whether an obstacle point exists within the vehicle's body model when the vehicle is traveling on the re-predicted kinematic path; detecting whether the distance between the vehicle and the obstacle point increases based on the vehicle's real-time position; and if the obstacle point does not exist within the vehicle's body model and the distance between the vehicle and the obstacle point increases, releasing (or canceling) the vehicle's emergency stop state. Otherwise, maintaining the ESTOP state ensures vehicle safety while the APA function is in operation.

[0039] As one embodiment of the present invention, the warning sensitivity is adjustable based on one or more of the following: the boundary expansion size of the vehicle expansion model compared to the vehicle body model; the number of path points between the preview point and the current target path point; and the length of the predetermined time period. By way of example, by setting the aforementioned values, two, three, or more levels of sensitivity can be achieved. When set to high sensitivity, the warning time is advanced, leaving more braking distance, but it is also more prone to false warnings. Conversely, when set to low sensitivity, the warning time is delayed, requiring more urgent braking, which can reduce the interference caused by false warnings.

[0040] For example, for the boundary expansion size of the vehicle expansion model, you can set three levels of sensitivity, such as setting the boundary expansion size to 10cm, 20cm, and 30cm. You can also set four levels of sensitivity, such as setting the boundary expansion size to 15cm, 20cm, 25cm, and 30cm.

[0041] For the number of path points between the preview point and the current target path point, you can set a secondary sensitivity, for example, setting the number of path points to 10 or 15 respectively, or a tertiary sensitivity, for example, setting the number of path points to 8, 12, or 16 respectively.

[0042] The length of the predetermined time period can be set to two levels of sensitivity, for example, the predetermined time period can be set to 500ms and 1500ms respectively, or to four levels of sensitivity, for example, the predetermined time period can be set to 500ms, 1000ms, 1500ms, and 2000ms respectively.

[0043] In addition to the first, second, and third level warnings described above, as an embodiment of the present invention, the method may also include: predicting whether there are obstacle points within the vehicle expansion model when the vehicle passes through multiple planned path points, and sending prompt information to the vehicle driver if there are obstacle points.

[0044] On the other hand, an embodiment of the present invention provides a warning protection device for automatic parking assistance, which includes a memory and a processor. The memory stores a computer program, which implements the warning protection method for automatic parking assistance described in any of the above embodiments when the computer program is executed by the processor.

[0045] The warning protection method and device for automatic parking assistance proposed in the embodiments of the present invention are more suitable for the APA function than the existing general parking distance control method due to its combination with the regulation and control algorithm in the APA function. The warning protection technology for automatic parking assistance proposed in the embodiments of the present invention adds speed limit warnings, brake warnings, and emergency stop warnings to the parking process in different scenarios according to the three-level warning information, and adds a state recovery interface when the function is triggered by an obstacle. It calculates the vehicle collision risk based on the wheel speed pulse, vehicle motion posture, steering wheel angle, etc. combined with the re-planned path points, and determines whether the vehicle emergency stop state is triggered / released, thereby optimizing the automatic parking function experience, and can provide good assistance and bottom-up protection for APA.

[0046] The automatic parking assistance warning protection method and device proposed in the embodiments of the present invention jointly construct a collision risk assessment model based on the automatic parking function through modules such as a path planning module, a kinematic model, a vehicle expansion model, and a vehicle body model.

[0047] The warning protection method and device for automatic parking assistance proposed in the embodiments of the present invention realize parking emergency braking and secondary judgment recovery functions. This function has a significant performance optimization for ensuring the safety of vehicles when parking in parking spaces and the continuity of parking functions in complex scenarios, especially dead-end roads and horizontal space parking spaces.

[0048] The automatic parking assistance warning protection method and device proposed in the embodiment of the present invention ensures the accuracy of function triggering by combining the vehicle wheel speed pulse meter with real-time positioning and estimating the vehicle posture with the vehicle kinematic model at a certain frequency (for example, a frequency greater than or equal to 30 Hz and less than or equal to 100 Hz); the kinematic model is combined with the planned collision model posture estimation to ensure the accuracy of the vehicle's emergency stop function and optimize the operating experience of the APA function when the function is triggered; combined with the vehicle kinematic model, obstacle collision prediction is added to the vehicle, which promotes the continuity of the APA function and ensures the normal deployment of the function; multi-party verification is performed through predictions of multiple models to ensure the safety of passengers in the automatic parking state.

[0049] For illustrative purposes, the foregoing description of the embodiments of the present invention has been given, which is not exhaustive nor intended to limit the present invention to disclosed exact forms. It will be appreciated by those skilled in the art that various changes may be made without departing from the scope of the present invention, and that elements therein may be replaced with equivalents. In addition, without departing from the basic scope of the present invention, many modifications may be made so that specific situations or materials are adapted to the teachings of the present invention. Therefore, the present invention is not intended to be limited to the specific embodiments disclosed as the best mode for realizing the present invention, and the present invention will include all embodiments within the scope of the appended claims.

Claims

1. A warning protection method for automatic parking assistance, characterized in that: The method comprises: Generate a planned path trajectory based on the initial position of the vehicle and information about the target parking space, wherein the planned path trajectory includes a plurality of planned path points; Determining a preview point on the planned path trajectory that is farther than the current target path point based on the real-time speed of the vehicle; predicting whether an obstacle point exists within the vehicle expansion model when the vehicle travels to a point between the current target path point and the preview point, and issuing a first-level warning to the vehicle driver and controlling the vehicle speed to be no higher than a first speed if an obstacle point exists; Based on the vehicle's real-time position, real-time speed and direction of travel, the vehicle kinematic model is used to predict and generate a kinematic path within a predetermined time period in the future; predicting whether an obstacle point exists within the vehicle expansion model when the vehicle travels on the kinematic path, and issuing a secondary warning to the vehicle driver if the obstacle point exists, and controlling the vehicle speed to be reduced to no higher than a second speed, wherein the second speed is lower than the first speed; Based on the vehicle's real-time position, real-time speed and direction of travel, if there is an obstacle point within the vehicle's expansion model and the obstacle point is outside the vehicle's main body model and in the current direction of travel, a third-level warning is issued to the vehicle driver and the vehicle is controlled to enter an emergency stop state.

2. The automatic parking assist warning protection method according to claim 1, characterized in that: The method further comprises: In response to controlling the vehicle to enter an emergency stop state, regeneration of the planned path trajectory is triggered.

3. The automatic parking assist warning protection method according to claim 2, characterized in that: The method further comprises: When the driver takes over the vehicle or receives a regenerated planned path trajectory, the vehicle's emergency stop state is released.

4. The automatic parking assist warning protection method according to claim 1, characterized in that: After controlling the vehicle to enter an emergency stop state, the method further includes: Based on the real-time position, real-time speed and direction of travel of the vehicle, a re-predicted kinematic path within a predetermined time period in the future is generated through the vehicle kinematic model prediction; Predicting whether there are obstacle points within the vehicle body model when the vehicle travels on the re-predicted kinematic path; According to the real-time position of the vehicle, detect whether the distance between the vehicle and the obstacle point increases; When there is no obstacle point within the vehicle body model and the distance between the vehicle and the obstacle point increases, the emergency stop state of the vehicle is released.

5. The automatic parking assistance warning protection method according to any one of claims 1 to 4, characterized in that: Alert sensitivity is adjustable based on one or more of the following: The boundary expansion size of the vehicle expansion model compared to the vehicle body model; The number of path points between the preview point and the current target path point; The length of the predetermined time period.

6. The automatic parking assistance warning protection method according to any one of claims 1 to 4, characterized in that: The method further comprises: It is predicted whether there is an obstacle point within the vehicle expansion model when the vehicle passes through the plurality of planned path points, and if the obstacle point exists, prompt information is sent to the vehicle driver.

7. The automatic parking assistance warning protection method according to any one of claims 1 to 4, characterized in that: The execution frequency of generating a kinematic path within a predetermined future time period through prediction by the vehicle kinematic model is greater than or equal to 30 Hz and less than or equal to 100 Hz.

8. The automatic parking assistance warning protection method according to any one of claims 1 to 4, characterized in that: The real-time speed of the vehicle is calculated based on the pulse value measured by the wheel speed pulse meter installed on the rear wheel of the vehicle.

9. The automatic parking assistance warning protection method according to any one of claims 1 to 4, characterized in that: The initial position of the vehicle is determined based on the position, speed and direction of travel of the vehicle when the automatic parking assistance function is activated.

10. A warning protection device for automatic parking assistance, characterized in that: The device includes a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the warning protection method for automatic parking assistance according to any one of claims 1 to 9 is implemented.

Images