Automatic parking longitudinal decision-making method, device and equipment, medium and vehicle

Abstract

The embodiment of the invention provides an automatic parking longitudinal decision-making method, device and equipment, a medium and a vehicle, and the method comprises the steps: in the automatic parking process, according to the own vehicle planning track of the vehicle and the own vehicle prediction track at the current moment, combining the semantic information of each obstacle in a real-time updated electronic map; determining a longitudinal collision distance between the vehicle and an obstacle; according to the planned trajectory of the vehicle and the semantic information of each obstacle in the electronic map, determining a transverse collision distance between the vehicle and the obstacle in combination with a vector projection mode; according to the longitudinal collision distance and the transverse collision distance, longitudinal decision making is conducted on the vehicle. By means of the method, during automatic parking decision planning, the semantic information of the obstacle is accurately recognized, simulation prediction is combined with the motion intention of the vehicle, the potential collision risk can be accurately predicted in advance, the speed of the vehicle can be adjusted more timely and more reasonably, avoidance measures can be taken, and the parking safety is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic parking, and particularly to an automatic parking longitudinal decision-making method, device, equipment, medium and vehicle. Background Art

[0002] Under the general trend of the intelligent development of automobiles, automatic parking technology has emerged, providing convenience for drivers. The existing automatic parking technology mainly deals with obstacles through collision detection algorithms, taking avoiding collisions with obstacles as a hard constraint for path planning algorithms, and then avoiding the collision risk of automatic parking paths through real-time updating of perceived environmental information and periodic replanning. However, the current understanding of the environment corresponding to automatic parking decision-making and planning is only based on the output of the current frame of perception fusion, and ignores the semantic information related to obstacles, and cannot make flexible decision-making and planning according to the real-time changing parking scenarios, resulting in the impact on the safety and parking efficiency of the automatic parking process and affecting the parking experience of users. Summary of the Invention

[0003] Embodiments of the present invention provide an automatic parking longitudinal decision-making method, device, equipment, medium and vehicle, which realizes more timely and reasonable adjustment of speed and taking avoidance measures during the parking process, effectively avoids the occurrence of collision accidents, and improves the safety of parking.

[0004] In a first aspect, an embodiment of the present invention provides an automatic parking longitudinal decision-making method, which includes:

[0005] During the automatic parking process, according to the vehicle's own planned trajectory and the vehicle's predicted trajectory at the current moment, combined with the semantic information of each obstacle in the real-time updated electronic map, determine the longitudinal collision distance between the vehicle and the obstacle;

[0006] According to the vehicle's own planned trajectory and the semantic information of each obstacle in the electronic map, combined with the vector projection method, determine the lateral collision distance between the vehicle and the obstacle;

[0007] According to the longitudinal collision distance and the lateral collision distance, make a longitudinal decision on the vehicle.

[0008] In a second aspect, an embodiment of the present invention provides an automatic parking longitudinal decision-making device, which includes:

[0009] A first determination module, configured to determine the longitudinal collision distance between the vehicle and the obstacle during the automatic parking process according to the vehicle's own planned trajectory and the vehicle's predicted trajectory at the current moment, combined with the semantic information of each obstacle in the real-time updated electronic map;

[0010] A second determination module, configured to determine a lateral collision distance between the vehicle and an obstacle according to the planned trajectory of the host vehicle and the semantic information of each obstacle in the electronic map, in combination with a vector projection method;

[0011] A longitudinal decision-making module, configured to perform longitudinal decision-making on the vehicle according to the longitudinal collision distance and the lateral collision distance.

[0012] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0013] At least one processor; and

[0014] A memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the automatic parking longitudinal decision-making method according to any embodiment of the present invention.

[0016] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, where the computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the automatic parking longitudinal decision-making method according to any embodiment of the present invention when executed by a processor.

[0017] In a fifth aspect, an embodiment of the present invention further provides a vehicle, where the vehicle includes the electronic device according to any embodiment of the present invention.

[0018] An embodiment of the present invention provides an automatic parking longitudinal decision-making method, device, equipment, medium and vehicle. The method includes: during automatic parking, determining a longitudinal collision distance between the vehicle and an obstacle according to the planned trajectory of the host vehicle and the predicted trajectory of the host vehicle at the current moment, in combination with the semantic information of each obstacle in the real-time updated electronic map; determining a lateral collision distance between the vehicle and an obstacle according to the planned trajectory of the host vehicle and the semantic information of each obstacle in the electronic map, in combination with a vector projection method; and performing longitudinal decision-making on the vehicle according to the longitudinal collision distance and the lateral collision distance. According to the above technical solution, when making an automatic parking decision and plan, the driving trajectory of the vehicle is considered based on multiple frames of history and future. By accurately identifying the semantic information of obstacles, including obstacle types, movement trends, etc., and combining the simulation prediction of the movement intention of the host vehicle, the potential collision risk can be accurately predicted in advance, and the accurate longitudinal collision distance and lateral collision risk level can be output. This enables the vehicle to adjust its speed and take avoidance measures more timely and reasonably during the parking process, effectively avoiding the occurrence of collision accidents and significantly improving the safety of parking.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. Description of the Drawings

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0021] Figure 1 It is a schematic flowchart of an automatic parking longitudinal decision-making method provided in Embodiment 1 of the present invention;

[0022] Figure 2 It is an interaction example diagram of automatic parking longitudinal decision-making in a dynamic obstacle scenario provided in Embodiment 1 of the present invention;

[0023] Figure 3 It is a schematic flowchart of another automatic parking longitudinal decision-making method provided in Embodiment 2 of the present invention;

[0024] Figure 4 It is an example diagram of the self-vehicle profile during the execution of an automatic parking longitudinal decision-making method provided in Embodiment 2 of the present invention;

[0025] Figure 5 It is a schematic structural diagram of an automatic parking longitudinal decision-making device provided in Embodiment 3 of the present invention;

[0026] Figure 6 It is a schematic structural diagram of an electronic device provided in Embodiment 4 of the present invention. Detailed Embodiments

[0027] In order to enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0028] It should be noted that the terms "first", "second", etc. in the description, claims and above-mentioned drawings of the present invention are used to distinguish similar objects and do not necessarily have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0029] It can be understood that under the general trend of the development of automotive intelligence, automatic parking technology has emerged, providing convenience for drivers. Existing automatic parking systems usually use sensors such as radar and cameras to detect obstacles around the vehicle, and then plan the parking path based on the geometric method of scatter sampling combined with arcs and straight lines or the hybrid A* search method that satisfies the kinematic constraints of the vehicle, and control the steering and speed of the vehicle to achieve automatic parking. The existing automatic parking technology mainly deals with obstacles through collision detection algorithms, taking avoiding collisions with obstacles as a hard constraint for the path planning algorithm, and then avoiding the collision risk of the automatic parking path by real-time updating the perceived environmental information and periodic replanning.

[0030] The current automatic parking decision-making and planning is based on the single-frame output of perception fusion for the understanding of the environment, without considering the predicted trajectories of dynamic obstacles and the simulation prediction of the motion intention of the host vehicle, lacking the interactive processing with the environment and the game intention with other pedestrians and vehicles, and unable to make speed decisions flexibly and confidently according to the real-time changing parking scenarios. For example, in a social parking lot with dynamic obstacles such as pedestrians suddenly crossing or vehicles moving, it is impossible to accurately predict their motion trajectories and intentions in advance, which may lead to sudden braking or even collisions. In a parking scenario where there is a following vehicle in the distance behind, due to insufficient parking efficiency, a scenario that could have been parked efficiently may be preempted by others.

[0031] In addition, the longitudinal decision-making and planning for automatic parking do not consider the semantics of obstacles. Longitudinal decision-making mainly controls the speed and acceleration of the vehicle during parking. The longitudinal decision-making of existing systems is relatively simple and fixed. Usually, it is adjusted based on a preset speed curve or a safety distance threshold, without fully considering the semantic information of obstacles. For example, in a parking space with limited space but where safe parking can be achieved through appropriate speed adjustment, the system may abandon parking due to overly conservative decision-making or experience multiple adjustment maneuvers to park, affecting the parking efficiency and user experience. When approaching an obstacle at a gear shift point on the path or when the path cuts into the parking space along the edge of the obstacle, although the planned path meets the hard constraints of collision detection, there will be a strong sense of oppression and panic in terms of user experience.

[0032] Therefore, an automatic parking longitudinal decision-making method is needed to solve the above problems.

[0033] Embodiment 1

[0034] Figure 1 As shown in the flowchart of an automatic parking longitudinal decision-making method provided in Embodiment 1 of the present invention, this method is applicable to the situation of making longitudinal decisions during the automatic parking process. This method can be executed by an automatic parking longitudinal decision-making device, which can be implemented in the form of hardware and / or software and is generally integrated in an electronic device.

[0035] As Figure 1 shown, an automatic parking longitudinal decision-making method provided in Embodiment 1 specifically may include the following steps:

[0036] S101. During the automatic parking process, based on the vehicle's own planned trajectory and the vehicle's predicted trajectory at the current moment, and in combination with the semantic information of each obstacle in the real-time updated electronic map, determine the longitudinal collision distance between the vehicle and the obstacle.

[0037] In this embodiment, the automatic parking longitudinal decision-making can specifically be understood as making control decisions on the speed and acceleration of the vehicle during parking. For the longitudinal decision-making during the vehicle's automatic parking process, it can be considered from two aspects: the longitudinal collision distance and the lateral collision distance between the vehicle and the obstacle. The longitudinal collision distance can specifically be understood as the distance between the vehicle and the obstacle in the longitudinal direction. The output of the longitudinal collision distance is one of the key safety features in the automatic parking system. Its purpose is to detect the longitudinal distance between the vehicle and the obstacle in front of or behind the vehicle in real time during the vehicle's travel along the planned trajectory, so as to enable the vehicle to make decisions such as decelerating and braking in a timely manner.

[0038] Different from the single-frame output that only considers environmental perception in the prior art, in this embodiment, the determination of the longitudinal collision distance is based on the predicted intention of the host vehicle, specifically including two parts: the longitudinal collision distance output based on the planned trajectory and the collision distance output based on kinematic derivation. Among them, the planned trajectory can be specifically understood as the driving trajectory of the vehicle during the parking process planned according to the starting point, ending point, and obstacle information of the vehicle parking, denoted as the host vehicle planned trajectory. The purpose of the collision distance output based on kinematic derivation is to accurately output the collision distance even when there is a deviation in the vehicle control following the planned trajectory, so as to determine the safety of the parking process. The predicted driving trajectory derived from kinematics is denoted as the host vehicle predicted trajectory.

[0039] In this embodiment, the semantic information of each obstacle in the environment is recorded in the electronic map, and the semantic information of the obstacle includes the position information, attributes, types, etc. of the obstacle. As the vehicle travels, the perception module will update the real-time perception information to the electronic map. After knowing the host vehicle planned trajectory and the host vehicle predicted trajectory of the vehicle, the outer envelope of the vehicle can be constructed, and the outer envelope of the vehicle can be fitted to the host vehicle planned trajectory and the host vehicle predicted trajectory. And according to the position information of each obstacle in the electronic map, it is judged whether there is an obstacle located in the host vehicle outer envelope fitted on the host vehicle planned trajectory or the host vehicle predicted trajectory. If there is an obstacle located in the host vehicle contour, it is necessary to determine the longitudinal collision distance between the vehicle and the obstacle. It can be understood that in this embodiment, the longitudinal collision distance between the vehicle and the obstacle will be determined based on the host vehicle planned trajectory. The longitudinal collision distance between the vehicle and the obstacle will also be determined based on the host vehicle predicted trajectory. When making a collision judgment, the evaluation will be based on these two longitudinal collision distances.

[0040] S102. According to the host vehicle planned trajectory and the semantic information of each obstacle in the electronic map, combined with the vector projection method, determine the lateral collision distance between the vehicle and the obstacle.

[0041] In this embodiment, for the longitudinal decision-making during the automatic parking process of the vehicle, in addition to considering from the longitudinal collision distance between the vehicle and the obstacle, it will also be judged based on the lateral collision distance. The lateral collision distance can be specifically understood as the distance between the vehicle and the obstacle in the lateral direction. The output of the lateral collision distance is also one of the key safety features in the automatic parking system. Its purpose is to detect the lateral distance between the left or right side of the vehicle and the obstacle in real time during the vehicle's driving along the planned trajectory, so as to enable the vehicle to make decisions such as decelerating and braking in a timely manner.

[0042] During the automatic parking process, in order to accurately calculate the lateral distance from the obstacle to the side of the vehicle, the method of vector projection is used. Specifically, for each planned trajectory point in the planned trajectory of the host vehicle, according to the vehicle pose at the planned trajectory point, the position information of the vehicle's central axis is determined. At the same time, the position information of each obstacle is obtained from the electronic map. According to the projection points of each obstacle to the vehicle's central axis and the position information of the obstacle, the lateral distance from each obstacle to the projection point of the vehicle's central axis is determined. Based on the lateral distance from each obstacle to the projection point of the vehicle's central axis and the width of the vehicle, the lateral collision distance from the obstacle to the vehicle can be determined.

[0043] S103. Make a longitudinal decision on the vehicle according to the longitudinal collision distance and the lateral collision distance.

[0044] In this embodiment, several risk level thresholds are set according to the longitudinal collision distance. Several risk level thresholds are set according to the lateral collision distance. By judging between which risk level thresholds the longitudinal collision distance is and between which risk level thresholds the lateral collision distance is, a longitudinal decision on the vehicle is determined. The longitudinal decision may include: not limiting the speed of the vehicle, limiting the speed of the vehicle, braking the vehicle, or even re-planning the driving trajectory.

[0045] In the above technical solution, when making an automatic parking decision and planning, the driving trajectory of the vehicle is considered based on multiple frames of history and future. By accurately identifying the semantic information of obstacles, including the type of obstacle, the movement trend, etc., and combining the simulation prediction of the host vehicle's movement intention, the potential collision risk can be accurately predicted in advance, and the accurate longitudinal collision distance and the lateral collision risk level can be output. This enables the vehicle to adjust the speed and take avoidance measures more timely and reasonably during the parking process, effectively avoiding the occurrence of collision accidents and significantly improving the safety of parking.

[0046] As an optional embodiment of the present invention, on the basis of the above embodiment, this optional embodiment can optimize the determination step of the host vehicle's predicted trajectory at the current moment, including:

[0047] a1) Determine the current turning radius of the vehicle according to the current steering wheel angle of the vehicle and the wheelbase of the vehicle.

[0048] In this embodiment, the possible driving curve of the host vehicle at the current steering wheel angle is predicted according to the pure tracking model. Let the current steering wheel angle be δ and the wheelbase be Wb, then the current turning radius of the vehicle

[0049] b1) Determine the host vehicle's predicted trajectory along the current steering wheel angle according to the current turning radius of the vehicle and the current pose of the vehicle, in combination with the pure tracking model.

[0050] In this embodiment, the pure tracking model uses a bicycle model to simulate the kinematic behavior of the vehicle. The current pose of the vehicle includes the current position and the current heading angle of the vehicle, which is equivalent to giving the traveling direction of the vehicle. Given the traveling direction of the vehicle and the current turning radius of the vehicle, combined with the pure tracking model, a kinematic driving curve of the vehicle along the current steering wheel angle is constructed, denoted as the ego-vehicle prediction trajectory of the vehicle along the current steering wheel angle.

[0051] The above technical solution specifies how to determine the ego-vehicle prediction trajectory of the vehicle at the current moment, providing basic data for predicting the risk of collision between the ego-vehicle and obstacles based on the ego-vehicle prediction trajectory.

[0052] As another alternative embodiment of the present invention, on the basis of the above embodiment, the method can be optimized to further include:

[0053] If the obstacle is a dynamic obstacle, then according to the ego-vehicle planned trajectory of the vehicle and the obstacle prediction trajectory of the obstacle, the predicted collision time between the vehicle and the obstacle is determined, and longitudinal decision-making is performed on the vehicle according to the predicted collision time.

[0054] In this embodiment, during the automatic parking process, if there are dynamic obstacles, in order to ensure the safety of parking, the potential collision risk with dynamic obstacles can be monitored and processed in real time. For example, dynamic obstacles include pedestrians, vehicles, etc. For the collision detection of dynamic obstacles, interference judgment needs to be performed according to the ego-vehicle planned trajectory and the obstacle prediction trajectory. Traverse the time stamps of the ego-vehicle trajectory to obtain the planned position of the vehicle and the predicted position of the obstacle corresponding to each time stamp, and construct envelopes of the two at their respective positions. By judging whether the envelopes overlap, the predicted collision time between the vehicle and the obstacle is judged. Furthermore, longitudinal decision-making is performed on the vehicle according to the predicted collision time.

[0055] Exemplarily, Figure 2 is an interaction example diagram of longitudinal decision-making for automatic parking in a dynamic obstacle scenario provided by Embodiment 1 of the present invention. As Figure 2 shown, the execution subject of this embodiment can be considered as a parking controller. The planning module can provide the ego-vehicle planned trajectory for the parking controller. The parking controller can also request information about dynamic obstacles from the perception module. The perception module will feedback the current state of the dynamic obstacles to the trajectory prediction module. The trajectory prediction module generates an obstacle prediction trajectory according to the current state of the dynamic obstacles and returns the obstacle prediction trajectory to the parking controller. The parking controller sends the ego-vehicle planned trajectory and the obstacle prediction trajectory to the collision detection module. The collision detection module executes the collision detection algorithm and returns the collision detection result to the parking controller. The parking controller adjusts the planned trajectory or triggers a brake stop according to the collision detection result.

[0056] The above technical solution takes into account the predicted trajectory of dynamic obstacles and the simulated prediction of the motion intention of the host vehicle, realizes the interactive processing with the environment and the game intention with other pedestrians and vehicles, can make speed decisions flexibly and confidently according to the parking scenario that changes in real time, and improves the safety of automatic parking.

[0057] As a specific implementation, the step of determining the predicted collision time between the vehicle and the obstacle according to the planned trajectory of the host vehicle and the predicted trajectory of the obstacle can be optimized, including:

[0058] a2) For each timestamp corresponding to the planned trajectory of the host vehicle, obtain the planned position information of the host vehicle at the timestamp and the predicted position information of the obstacle at the timestamp.

[0059] Specifically, traverse the timestamps of the host vehicle trajectory. For each timestamp corresponding to the planned trajectory of the host vehicle, obtain the planned position information of the host vehicle at the timestamp and the predicted position information of the obstacle at the timestamp.

[0060] b2) Construct the host vehicle envelope of the vehicle at the position corresponding to the planned position information of the host vehicle, and construct the obstacle envelope of the obstacle at the position corresponding to the predicted position information of the obstacle.

[0061] In this embodiment, a convex polygon of the vehicle is constructed at the position corresponding to the planned position information of the host vehicle, denoted as the host vehicle envelope. A convex polygon of the obstacle is constructed at the position corresponding to the predicted position information of the obstacle, denoted as the obstacle envelope.

[0062] It should be noted that when constructing the envelope of the vehicle or the obstacle, the size of the envelope expansion can be set according to actual needs. Exemplarily, for static obstacles such as walls, columns or building edges, since such obstacles are static, a relatively small expansion range can be set for such obstacles, that is, the constructed obstacle envelope can be relatively small. For dynamic obstacles such as pedestrians and non-motor vehicles, since such obstacles are dynamic and there is a relatively high risk of loss in case of collision, therefore, a relatively large collision range can be set for such obstacles, that is, the constructed obstacle envelope can be relatively large.

[0063] c2) Determine whether there is an overlap between the host vehicle envelope and the obstacle envelope.

[0064] Specifically, determine whether there is an overlap between the host vehicle envelope and the obstacle envelope to further determine whether the vehicle and the obstacle will collide. If there is no overlap between the host vehicle envelope and the obstacle envelope, the vehicle and the obstacle will not collide; if there is an overlap between the host vehicle envelope and the obstacle envelope, the vehicle and the obstacle will collide.

[0065] d2) If there is an overlap, determine the predicted collision time between the vehicle and the obstacle according to the time stamp corresponding to the existing overlap.

[0066] In this embodiment, if the ego vehicle includes an overlap with the obstacle envelope, it means that the vehicle will collide with the obstacle. Further, according to the time stamp corresponding to the existing overlap and the time stamp corresponding to the current moment, the difference between the two is calculated to determine the predicted collision time when the vehicle collides with the obstacle.

[0067] Exemplarily, assume that the predicted position of the dynamic obstacle at time t is (xo(t), yo(t)), and the predicted position of the ego vehicle at time t is (xe(t), ye(t)). A convex polygon of the two is constructed to determine whether there is an overlap. If there is an overlap, the predicted collision time is output.

[0068] The above technical solution specifies the steps of how to determine whether there is a collision risk between the vehicle and the obstacle according to the planned trajectory of the ego vehicle and the predicted trajectory of the obstacle, and how to determine the predicted collision time if there is a collision risk, providing a basis for subsequent longitudinal decision-making of the vehicle.

[0069] As a specific implementation, the steps of longitudinal decision-making for the vehicle according to the predicted collision time can be optimized, including:

[0070] a3) If the predicted collision time is greater than the first time threshold, control the vehicle to accelerate.

[0071] In this embodiment, the predicted collision time is compared with the set first time threshold and second time threshold to make a longitudinal decision for the vehicle. The first time threshold and the second time threshold are set according to the actual situation, and the first time threshold is greater than the second time threshold. Specifically, if the predicted collision time is greater than the first time threshold, it means that the predicted collision time is long, and the ego vehicle can appropriately accelerate to preempt passing to reduce the intersection time with the dynamic obstacle.

[0072] b3) If the predicted collision time is greater than the second time threshold and less than or equal to the first time threshold, control the vehicle to decelerate.

[0073] Specifically, if the predicted collision time is greater than the second time threshold and the predicted collision time is less than or equal to the first time threshold, it means that the predicted collision time is close, and the ego vehicle should appropriately decelerate and slow down to increase the safety distance.

[0074] c3) If the predicted collision time is less than or equal to the second time threshold, control the vehicle to brake to a stop and wait for the obstacle to leave before continuing the parking operation.

[0075] Specifically, if the predicted collision time is less than or equal to the second time threshold, it means that the predicted collision time is very close. The host vehicle needs to immediately brake and enter the pause state, and wait for the dynamic obstacle to leave before continuing the parking operation.

[0076] The above technical solution specifies how to make longitudinal decisions for the vehicle based on the predicted collision time, innovatively combines prediction information in the decision-making and planning process of automatic parking, can interact with the real-time changing parking scenario and play a game with other moving vehicles or pedestrians, and realizes a more confident and safe parking experience.

[0077] Embodiment 2

[0078] Figure 3 As shown in the flowchart of another automatic parking longitudinal decision-making method provided in Embodiment 2 of the present invention, this embodiment is a further optimization of the above embodiment. In this embodiment, the determination of the longitudinal collision distance between the vehicle and the obstacle is further optimized by "combining the semantic information of each obstacle in the electronic map according to the self-vehicle planned trajectory of the vehicle and the self-vehicle predicted trajectory at the current moment", and the determination of the lateral collision distance between the vehicle and the obstacle is optimized by "combining the self-vehicle planned trajectory and the semantic information of each obstacle in the electronic map, and combining the vector projection method", and the determination of making longitudinal decisions for the vehicle is optimized by "according to the longitudinal collision distance and the lateral collision distance".

[0079] As Figure 3 shown, Embodiment 2 of the present invention provides an automatic parking longitudinal decision-making method, which specifically includes the following steps:

[0080] S201. According to the vehicle poses at each planned trajectory point on the self-vehicle planned trajectory at a preset step interval, and the gear of the vehicle, fit the self-vehicle contour of the vehicle to each planned trajectory point of the target self-vehicle planned trajectory.

[0081] In this embodiment, the target self-vehicle planned trajectory is included in the self-vehicle planned trajectory. The minimum envelope of the vehicle can be constructed based on the key vertices of the vehicle as the self-vehicle contour of the vehicle. The gear of the vehicle indicates whether the vehicle is in the forward direction or the reverse direction at this time. If the gear of the vehicle is in the forward direction, the collision distance in the forward direction needs to be output; if the gear of the vehicle is in the reverse direction, the collision distance in the reverse direction needs to be output. The target self-vehicle planned trajectory can be intercepted from the self-vehicle planned trajectory according to the gear of the vehicle. The vehicle pose includes the coordinate position of the vehicle and the heading angle of the vehicle at this coordinate position. The set step size can be set according to the actual situation. After constructing the self-vehicle contour, taking the set step interval as a planned trajectory point, the self-vehicle contour is planned and deduced on the target self-vehicle planned trajectory, which is equivalent to fitting the self-vehicle contour of the vehicle to each planned trajectory point in the target planned trajectory according to the vehicle pose.

[0082] As a specific implementation, the step of optimizing the vehicle pose and the gear of the vehicle at each planned trajectory point on the planned trajectory of the host vehicle at a preset step length and fitting the host vehicle contour of the vehicle to each planned trajectory point of the target host vehicle planned trajectory includes:

[0083] a3) Construct the host vehicle contour of the vehicle.

[0084] This step is used to construct the minimum bounding envelope of the vehicle based on the key vertices of the vehicle as the host vehicle contour of the vehicle. Preferably, the host vehicle contour can adopt an octagonal bounding box, and the vertices of the octagonal bounding box are calibrated according to the installation positions of the ultrasonic radars on the vehicle to obtain the minimum vehicle octagonal contour as the host vehicle contour. Exemplarily, Figure 4 FIG. is an example diagram of the host vehicle contour during the execution of an automatic parking longitudinal decision-making method provided in Embodiment 2 of the present invention. As Figure 4 shown, the dots on the vehicle edge in the figure represent the installation positions of the 12-channel ultrasonic radars on the vehicle. Taking these installation positions as the vertices of the octagon, an octagonal bounding box can be obtained as the host vehicle contour.

[0085] b3) Determine the target host vehicle planned trajectory according to the planned trajectory of the host vehicle and the gear of the vehicle.

[0086] In this embodiment, the gear of the vehicle indicates whether the vehicle is in the forward direction or the reverse direction at this time. If the gear of the vehicle is in the forward direction, the collision distance in the forward direction needs to be output; if the gear of the vehicle is in the reverse direction, the collision distance in the reverse direction needs to be output. Based on this, it can be determined whether the vehicle is moving forward or backward according to the gear of the vehicle, and then it can be determined whether the target host vehicle planned trajectory is from the current position forward to the end point or from the current position backward to the end point.

[0087] c3) Fit the host vehicle contour of the vehicle to each planned trajectory point in the target planned trajectory according to the vehicle poses at each planned trajectory point on the target host vehicle planned trajectory at an interval of the set step length.

[0088] In this embodiment, the vehicle pose includes the coordinate position of the vehicle and the heading angle of the vehicle at this coordinate position. The set step length can be set according to the actual situation. For example, the set step length can be set to 10 cm, and no specific limitation is made here. After constructing the host vehicle contour, taking the interval of the set step length as a planned trajectory point, the host vehicle contour is predicted and deduced on the target host vehicle planned trajectory, which is equivalent to fitting the host vehicle contour of the vehicle to each predicted trajectory point in the target predicted trajectory according to the vehicle pose.

[0089] How the above technical solution deduces the host vehicle contour of the vehicle on the planned trajectory of the host vehicle provides a basis for determining the collision risk with obstacles based on the host vehicle contour.

[0090] S202. Determine whether there is an obstacle within the self-vehicle contour fitted on the target self-vehicle planned trajectory according to the position information of each obstacle in the electronic map.

[0091] In this embodiment, after determining the self-vehicle contour fitted on the target self-vehicle planned trajectory, it is possible to determine whether there is an obstacle within the self-vehicle contour according to the position information of each obstacle in the electronic map. If there is no obstacle within the self-vehicle contour, there is no need to further determine the longitudinal collision distance between the vehicle and the obstacle; if there is an obstacle within the self-vehicle contour, it is necessary to further determine the longitudinal collision distance between the vehicle and the obstacle.

[0092] S203. If so, determine the first longitudinal collision distance between the vehicle and the obstacle.

[0093] Specifically, if there is an obstacle within the self-vehicle contour, it is necessary to determine the longitudinal distance between the vehicle and the obstacle, and record this distance as the first longitudinal collision distance. Determine the distance between the current position of the vehicle and the obstacle in terms of several set step lengths, and then multiply it by the specific length of each set step length to calculate the longitudinal collision distance between the vehicle and the obstacle.

[0094] Exemplarily, assume that the gear of the vehicle is in the forward direction, the set step length is 10 cm, and there is no obstacle within the self-vehicle contour within 3 planned trajectory points (i.e., within 30 cm) from the current position forward. There is an obstacle within the self-vehicle contour at the 4th planned trajectory point from the current position forward (i.e., at 40 cm), then the longitudinal collision distance between the vehicle and the obstacle is determined to be 40 cm.

[0095] S204. According to the vehicle poses at each prediction trajectory point on the self-vehicle prediction trajectory at a preset step length interval, and the gear of the vehicle, fit the self-vehicle contour of the vehicle to each prediction trajectory point of the target self-vehicle prediction trajectory.

[0096] In this embodiment, the target self-vehicle prediction trajectory is included in the self-vehicle prediction trajectory. Based on the key vertices of the vehicle, the minimum outer envelope of the vehicle can be constructed as the self-vehicle contour of the vehicle. The gear of the vehicle indicates whether the vehicle is in the forward direction or the reverse direction at this time. If the gear of the vehicle is in the forward direction, the collision distance in the forward direction needs to be output; if the gear of the vehicle is in the reverse direction, the collision distance in the reverse direction needs to be output. The target self-vehicle prediction trajectory can be intercepted from the self-vehicle prediction trajectory according to the gear of the vehicle. The vehicle pose includes the coordinate position of the vehicle and the heading angle of the vehicle at this coordinate position. The set step length can be set according to the actual situation. After constructing the self-vehicle contour, taking the set step length interval as a prediction trajectory point, predicting and deriving the self-vehicle contour on the target self-vehicle prediction trajectory is equivalent to fitting the self-vehicle contour of the vehicle to each prediction trajectory point in the target prediction trajectory according to the vehicle pose.

[0097] As a specific implementation, the step of optimizing the vehicle pose and the gear of the vehicle at each planned trajectory point with a preset step length on the predicted planned trajectory and fitting the predicted contour of the vehicle to each planned trajectory point of the target predicted planned trajectory includes:

[0098] a4) Construct the ego contour of the vehicle.

[0099] This step is used to construct the minimum bounding envelope of the vehicle based on the key vertices of the vehicle as the ego contour of the vehicle. Preferably, the ego contour can adopt an octagonal bounding box, and the vertices of the octagonal bounding box are calibrated according to the installation positions of the ultrasonic radars on the vehicle to obtain the minimum octagonal contour of the vehicle as the ego contour.

[0100] b4) Determine the target predicted planned trajectory according to the predicted planned trajectory and the gear of the vehicle.

[0101] In this embodiment, the gear of the vehicle indicates whether the vehicle is in the forward direction or the reverse direction at this time. If the gear of the vehicle is in the forward direction, the collision distance in the forward direction needs to be output; if the gear of the vehicle is in the reverse direction, the collision distance in the reverse direction needs to be output. Based on this, it can be determined whether the vehicle is moving forward or backward according to the gear of the vehicle, and then it can be determined whether the target ego prediction trajectory is from the current position forward to the end point or from the current position backward to the end point.

[0102] c4) Fit the predicted contour of the vehicle to each planned trajectory point in the target planned trajectory according to the vehicle pose at each planned trajectory point with a set step length in the target predicted planned trajectory.

[0103] In this embodiment, the vehicle pose includes the coordinate position of the vehicle and the heading angle of the vehicle at this coordinate position. The set step length can be set according to the actual situation. For example, the set step length can be 10 cm, and no specific limitation is made here. After constructing the ego contour, taking the set step length as a predicted trajectory point, the ego contour of the vehicle is predicted and deduced on the target ego prediction trajectory, which is equivalent to fitting the ego contour of the vehicle to each predicted trajectory point in the target predicted trajectory according to the vehicle pose.

[0104] The above technical solution specifies how to deduce the ego contour of the vehicle on the ego prediction trajectory, providing a basis for determining the collision risk with obstacles based on the ego contour subsequently.

[0105] S205. According to the position information of each obstacle in the electronic map, determine whether there is an obstacle located in the ego contour fitted on the target ego prediction trajectory.

[0106] In this embodiment, after determining the self-vehicle contour fitted on the predicted trajectory of the target self-vehicle, it is possible to determine whether there is an obstacle within the self-vehicle contour based on the position information of each obstacle in the electronic map. If there is no obstacle within the self-vehicle contour, there is no need to further determine the longitudinal collision distance between the vehicle and the obstacle; if there is an obstacle within the self-vehicle contour, it is necessary to further determine the longitudinal collision distance between the vehicle and the obstacle.

[0107] S206. If so, determine the second longitudinal collision distance between the vehicle and the obstacle.

[0108] Specifically, if there is an obstacle within the self-vehicle contour, it is necessary to determine the longitudinal distance between the vehicle and the obstacle, and record this distance as the second longitudinal collision distance. Determine the distance between the current position of the vehicle and the obstacle in several set steps, and then multiply it by the specific length of each set step to calculate the second longitudinal collision distance between the vehicle and the obstacle.

[0109] Exemplarily, assume that the gear of the vehicle is in the forward direction, the set step is 10 cm, and there is no obstacle within the self-vehicle contour within 3 predicted trajectory points (i.e., within 30 cm) from the current position forward. There is an obstacle within the self-vehicle contour at the 4th predicted trajectory point from the current position forward, that is, there is an obstacle within the self-vehicle contour at 40 cm, then the second longitudinal collision distance between the vehicle and the obstacle is determined to be 40 cm.

[0110] By executing steps S204 - S206, it can be ensured that the collision distance can still be accurately output in the case of deviation of vehicle control following the planned trajectory, so as to ensure the safety of the parking process.

[0111] S207. For each planned trajectory point in the self-vehicle planned trajectory, determine the position information of the vehicle's central axis according to the vehicle pose at the planned trajectory point.

[0112] During the automatic parking process, in order to reduce the collision risk between the vehicle and the obstacle, in addition to considering the longitudinal collision distance between the vehicle and the obstacle, it is also necessary to consider the lateral collision distance between the vehicle and the obstacle. In order to accurately calculate the lateral distance from the obstacle to the side of the vehicle, the method of vector projection is used in this embodiment. The vehicle pose includes the coordinate position, heading angle, etc. of the vehicle. Specifically, for each planned trajectory point in the self-vehicle planned trajectory, the vehicle pose at each planned trajectory point is known, then the coordinates of the front center point and the rear center point of the vehicle can be obtained. Further, by subtracting the coordinates of the rear center point from the coordinates of the front center point, a vector representation of the vehicle's central axis can be obtained, which is denoted as the position information of the central axis.

[0113] Exemplarily, the coordinates of the front center point of the vehicle are represented as A(ax, ay), and the coordinates of the rear center point are represented as B(bx, by), then the central axis vector representation is

[0114] S208. Obtain the position information of each obstacle within the set range of the planned trajectory points from the electronic map.

[0115] In this embodiment, the set range can be set according to the actual situation and is not specifically limited here. Considering that the environment where the vehicle is located may contain multiple obstacle points, but only the obstacles near the planned trajectory points may collide with the vehicle. Therefore, traverse the poses of each planned trajectory point, and after thinning each planned trajectory point, obtain the position information of the obstacles within the set range of each planned trajectory point.

[0116] S209. Determine the lateral distance from each obstacle to the projection point on the vehicle's central axis according to the position information of the vehicle's central axis and the position information of each obstacle.

[0117] In this embodiment, let the coordinates of the obstacle point on the electronic map be P(px, py), and the central axis vector be The vector from the front center point A of the vehicle to the obstacle point P is expressed as The calculation parameter k represents the projection position of the obstacle on the vehicle's central axis,

[0118] According to the parameter k, calculate the projection point Q(qx, qy)=(ax + k·abx, ay + k·aby) of the obstacle point on the central axis, and calculate the Euclidean distance from the projection point to the obstacle point to obtain the lateral distance

[0119] S210. Determine the minimum lateral distance from the lateral distances from each obstacle to the projection point on the vehicle's central axis.

[0120] Specifically, for each planned trajectory point, after calculating the lateral distances from each obstacle within its set range to the projection point on the vehicle's central axis, the minimum lateral distance can be determined therefrom. The minimum lateral distance can be expressed as: min{d}.

[0121] S211. Determine the lateral collision distance between the vehicle and the obstacle according to the minimum lateral distance and the width of the vehicle.

[0122] Continuing from the above description, let the vehicle width be W, then the minimum lateral collision distance between the vehicle and the obstacle is Take this minimum lateral collision distance as the lateral collision distance between the vehicle and the obstacle.

[0123] 212. If both the first longitudinal collision distance and the second longitudinal collision distance of the vehicle are greater than the first longitudinal distance threshold, and the lateral collision distance is greater than the first lateral distance threshold, then control the vehicle speed to remain unchanged.

[0124] In this embodiment, according to the longitudinal collision distance, two risk level thresholds are set, denoted as the first longitudinal distance threshold and the second longitudinal distance threshold respectively, and the first longitudinal distance threshold is greater than the second longitudinal distance threshold. According to the lateral collision distance, two risk level thresholds are set, denoted as the first lateral distance threshold and the second lateral distance threshold respectively, and the first lateral distance threshold is greater than the second lateral collision distance threshold. Each collision distance threshold can be set according to the actual situation.

[0125] Specifically, if the first longitudinal collision distance of the vehicle is greater than the first longitudinal distance threshold and the second longitudinal collision distance is greater than the first longitudinal distance threshold, it indicates that the vehicle is relatively far from the obstacle in the longitudinal direction. And if the lateral collision distance is greater than the first lateral distance threshold, it indicates that the vehicle is relatively far from the obstacle in the lateral direction. Based on the above description, it can be determined that there is no risk currently, the vehicle is not speed-limited, and the vehicle speed can be controlled to remain unchanged.

[0126] S213. If the first longitudinal collision distance or the second longitudinal collision distance of the vehicle is greater than the second longitudinal distance threshold and less than or equal to the first longitudinal distance threshold, or the lateral collision distance is greater than the second lateral distance threshold and less than or equal to the first lateral distance threshold, then control the vehicle to decelerate.

[0127] Specifically, if the first longitudinal collision distance of the vehicle is greater than the second longitudinal distance threshold and less than or equal to the first longitudinal distance threshold, or the second longitudinal collision distance is greater than the second longitudinal distance threshold and less than or equal to the first longitudinal distance threshold, it indicates that the vehicle is relatively close to the obstacle in the longitudinal direction. If the lateral collision distance is greater than the second lateral distance threshold and less than or equal to the first lateral distance threshold, it indicates that the vehicle is relatively close to the obstacle in the lateral direction. When any of the situations described in this step above occurs, the vehicle is in the medium and low risk collision level, and the vehicle needs to be speed-limited.

[0128] S214. If the first longitudinal collision distance or the second longitudinal collision distance of the vehicle is less than or equal to the second longitudinal distance threshold, or the lateral collision distance is less than or equal to the second lateral distance threshold, then control the vehicle to brake to a stop.

[0129] Specifically, if the first longitudinal collision distance of the vehicle is less than or equal to the second longitudinal distance threshold, or the second longitudinal collision distance is less than or equal to the second longitudinal distance threshold, it indicates that the vehicle is very close to the obstacle in the longitudinal direction. If the lateral collision distance is less than or equal to the second lateral distance threshold, it indicates that the vehicle is very close to the obstacle in the lateral direction. When any of the situations described in this step above occurs, the vehicle is in the high risk collision level, and it is necessary to immediately brake to a stop and re-plan and adjust the driving trajectory.

[0130] The above technical solution specifies how to determine the longitudinal collision distance and the lateral collision distance between the vehicle and the obstacle, and how to make longitudinal decisions for the vehicle based on the longitudinal collision distance and the lateral collision distance. During the decision-making and planning process of automatic parking, predictive information is innovatively combined, enabling interaction with the real-time changing parking scenario and gaming with other moving vehicles and pedestrians to achieve a more confident and safe parking experience. By accurately identifying the semantic information of the obstacle, including the obstacle type, movement trend, etc., and combining the simulation and prediction of the movement intention of the vehicle itself, the system can accurately predict potential collision risks in advance and output the accurate longitudinal collision distance and the lateral collision risk level. This enables the vehicle to adjust its speed and take avoidance measures more timely and reasonably during the parking process, effectively avoiding the occurrence of collision accidents and significantly improving the safety of parking. Moreover, this longitudinal decision-making method has strong adaptability and flexibility, and can make full use of the space in the curb scenario. It meets the parking needs of different users in different scenarios, has obvious advantages over traditional systems in terms of the adaptability of the parking scenario, and broadens the application scope of automatic parking technology.

[0131] As an alternative embodiment of the embodiment of the present invention, on the basis of the above embodiment, this alternative embodiment can optimize that the method further includes:

[0132] When the vehicle parks in the garage in place, if the lateral collision distance is greater than the second lateral distance threshold and less than the third lateral distance threshold, the folding of the vehicle's rearview mirror is triggered.

[0133] In this embodiment, the medium and low risk levels can be further divided between the first lateral distance threshold and the second collision distance, and the division threshold is denoted as the third collision distance threshold. The third lateral distance threshold is less than the first lateral distance threshold and greater than the second lateral distance threshold. If the lateral collision distance is less than the first lateral distance threshold and greater than the third lateral distance threshold, the current is a low risk collision level; if the lateral collision distance is less than the third lateral distance threshold and greater than the second lateral distance threshold, the current is a medium risk collision level. This step is only judged when the vehicle is in the last step of parking in the garage, that is, when the vehicle parks in the garage in place, if the lateral collision distance is greater than the second lateral distance threshold and less than the third lateral distance threshold, the folding of the vehicle's rearview mirror is triggered, which can ensure the safe parking of the vehicle.

[0134] The above technical solution can support the request for folding the rearview mirror and make full use of the space in the curb scenario.

[0135] It can be understood that in this embodiment, the three-dimensional semantic information of the electronic map and moving objects are innovatively utilized, combined with the simulation and prediction of the motion intention of the host vehicle, to achieve the output of the predicted longitudinal collision distance, lateral collision risk level and collision semantic output during the parking process, to make a more flexible and confident speed decision, to support the folding request of the rearview mirror, to avoid obstacles of dynamic obstacles and to utilize the space of extreme scenarios, ensuring the safety and efficiency of the automatic parking process.

[0136] Embodiment III

[0137] Figure 5 FIG. is a schematic structural diagram of an automatic parking longitudinal decision-making device provided in Embodiment III of the present invention. The device is applicable to the situation of making longitudinal decisions during the automatic parking process. The automatic parking longitudinal decision-making device can be implemented in the form of hardware and / or software and is generally integrated in an electronic device. As Figure 5 shown, the device includes: a first determination module 31, a second determination module 32, and a longitudinal decision-making module 33, wherein,

[0138] The first determination module 31 is configured to determine the longitudinal collision distance between the vehicle and an obstacle according to the self-vehicle planned trajectory of the vehicle and the self-vehicle predicted trajectory at the current moment, in combination with the semantic information of each obstacle in the electronically mapped real-time update.

[0139] The second determination module 32 is configured to determine the lateral collision distance between the vehicle and an obstacle according to the self-vehicle planned trajectory and the semantic information of each obstacle in the electronic map, in combination with a vector projection method.

[0140] The longitudinal decision-making module 33 is configured to make a longitudinal decision on the vehicle according to the longitudinal collision distance and the lateral collision distance.

[0141] In the above technical solution, when making an automatic parking decision and planning, the driving trajectory of the vehicle is considered based on multiple frames of history and future. By accurately identifying the semantic information of obstacles, including obstacle types, motion trends, etc., and combining the simulation and prediction of the motion intention of the host vehicle, potential collision risks can be accurately predicted in advance, and accurate longitudinal collision distances and lateral collision risk levels can be output. This enables the vehicle to adjust its speed and take avoidance measures more timely and reasonably during the parking process, effectively avoiding the occurrence of collision accidents and significantly improving the safety of parking.

[0142] Optionally, the first determination module 31 includes:

[0143] The first fitting unit is configured to fit the self-vehicle contour of the vehicle to each planned trajectory point of the target self-vehicle planned trajectory according to the vehicle poses at each planned trajectory point on the self-vehicle planned trajectory at a preset step interval and the gear of the vehicle, where the target self-vehicle planned trajectory is included in the self-vehicle planned trajectory;

[0144] The first judgment unit is configured to judge whether there is an obstacle located in the self-vehicle contour fitted on the target self-vehicle planned trajectory according to the position information of each obstacle in the electronic map;

[0145] The first determination unit is configured to, if any, determine the first longitudinal collision distance between the vehicle and the obstacle;

[0146] The second fitting unit is configured to fit the self-vehicle contour of the vehicle to each prediction trajectory point of the target self-vehicle prediction trajectory according to the vehicle poses at each prediction trajectory point on the self-vehicle prediction trajectory at a preset step interval and the gear of the vehicle, where the target self-vehicle prediction trajectory is included in the self-vehicle prediction trajectory;

[0147] The second judgment unit is configured to judge whether there is an obstacle located in the self-vehicle contour fitted on the target self-vehicle prediction trajectory according to the position information of each obstacle in the electronic map;

[0148] The second determination unit is configured to, if any, determine the second longitudinal collision distance between the vehicle and the obstacle.

[0149] Optionally, the first fitting unit is specifically configured to:

[0150] Construct the self-vehicle contour of the vehicle;

[0151] Determine the target self-vehicle planned trajectory according to the self-vehicle planned trajectory and the gear of the vehicle;

[0152] Fit the self-vehicle contour of the vehicle to each planned trajectory point in the target planned trajectory according to the vehicle poses at each planned trajectory point on the target self-vehicle planned trajectory at a set step interval.

[0153] Optionally, the device further includes a trajectory prediction module, configured to:

[0154] Determine the current turning radius of the vehicle according to the current steering wheel angle of the vehicle and the wheelbase of the vehicle;

[0155] Determine the self-vehicle prediction trajectory of the vehicle along the current steering wheel angle in combination with the pure tracking model according to the current turning radius of the vehicle and the current pose of the vehicle.

[0156] Optionally, the second determination module 32 is specifically configured to:

[0157] For each planned trajectory point in the self-vehicle planned trajectory, determine the position information of the vehicle central axis according to the vehicle pose at the planned trajectory point;

[0158] Obtain the position information of each obstacle within a set range of the planned trajectory point from the electronic map;

[0159] Determine the lateral distance from each obstacle to the projection point on the vehicle central axis according to the position information of the vehicle central axis and the position information of each obstacle.

[0160] The minimum lateral distance is determined from the lateral distances of the projection points of each obstacle to the vehicle's central axis.

[0161] Based on the minimum lateral distance and the width of the vehicle, the lateral collision distance between the vehicle and the obstacle is determined.

[0162] Optionally, the longitudinal decision-making module 33 is specifically configured to:

[0163] If both the first longitudinal collision distance and the second longitudinal collision distance of the vehicle are greater than the first longitudinal distance threshold, and the lateral collision distance is greater than the first lateral distance threshold, then keep the vehicle speed unchanged.

[0164] If the first longitudinal distance or the second longitudinal collision distance of the vehicle is greater than the second longitudinal distance threshold and less than or equal to the first longitudinal distance threshold, or the lateral collision distance is greater than the second lateral distance threshold and less than or equal to the first lateral distance threshold, then control the vehicle to decelerate.

[0165] If the first longitudinal distance or the second longitudinal collision distance of the vehicle is less than or equal to the second longitudinal distance threshold, or the lateral collision distance is less than or equal to the second lateral distance threshold, then control the vehicle to stop.

[0166] The first longitudinal distance threshold is greater than the second longitudinal distance threshold, and the first lateral distance threshold is greater than the second lateral collision distance threshold.

[0167] Optionally, the longitudinal decision-making module 33 is further configured to:

[0168] When the vehicle parks in the garage in place, if the lateral collision distance is greater than the second distance threshold and less than the third collision distance, then trigger the folding of the vehicle's rearview mirror, where the third collision distance is greater than the second collision distance and less than the first collision distance.

[0169] Optionally, the device further includes a dynamic decision-making module, which is used to:

[0170] If the obstacle is a dynamic obstacle, then based on the vehicle's own planned trajectory and the obstacle's predicted trajectory, determine the predicted collision time between the vehicle and the obstacle, and make a longitudinal decision on the vehicle according to the predicted collision time.

[0171] Optionally, the dynamic decision-making module includes a time calculation unit, which is used to:

[0172] For each timestamp corresponding to the vehicle's own planned trajectory, obtain the vehicle's own planned position information at the timestamp and the obstacle's predicted position information at the timestamp.

[0173] Construct the ego vehicle envelope of the vehicle at the position corresponding to the ego vehicle planning position information, and construct the obstacle envelope of the obstacle at the position corresponding to the obstacle prediction position information;

[0174] Determine whether there is an overlap between the ego vehicle envelope and the obstacle envelope;

[0175] If there is an overlap, determine the predicted collision time between the vehicle and the obstacle according to the time stamp corresponding to the existence of the overlap.

[0176] Optionally, the dynamic decision-making module further includes a dynamic decision-making unit for:

[0177] If the predicted collision time is greater than the first time threshold, control the vehicle to accelerate;

[0178] If the predicted collision time is greater than the second time threshold and less than or equal to the first time threshold, control the vehicle to decelerate;

[0179] If the predicted collision time is less than or equal to the second time threshold, control the vehicle to brake and stop and wait for the obstacle to leave and then continue the parking operation;

[0180] The first time threshold is greater than the second time threshold.

[0181] The automatic parking longitudinal decision-making device provided by the embodiments of the present invention can execute the automatic parking longitudinal decision-making method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0182] Embodiment 4

[0183] Figure 6 It is a schematic structural diagram of an electronic device provided by Embodiment 4 of the present invention. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processors, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are only examples and are not intended to limit the implementation of the present invention described herein and / or claimed.

[0184] As Figure 6As shown, the electronic device 40 includes at least one processor 41 and a memory communicatively connected to the at least one processor 41, such as a read-only memory (ROM) 42, a random access memory (RAM) 43, etc. The memory stores a computer program executable by the at least one processor. The processor 41 can execute various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 42 or the computer program loaded from the storage unit 48 into the random access memory (RAM) 43. In the RAM 43, various programs and data required for the operation of the electronic device 40 can also be stored. The processor 41, the ROM 42, and the RAM 43 are connected to each other via a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0185] Multiple components in the electronic device 40 are connected to the I / O interface 45, including: an input unit 46, such as a keyboard, a mouse, etc.; an output unit 47, such as various types of displays, speakers, etc.; a storage unit 48, such as a disk, an optical disc, etc.; and a communication unit 49, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0186] The processor 41 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 41 include but are not limited to a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 41 executes the various methods and processes described above, such as the automatic parking longitudinal decision-making method.

[0187] In some embodiments, the automatic parking longitudinal decision-making method can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as the storage unit 48. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 40 via the ROM 42 and / or the communication unit 49. When the computer program is loaded into the RAM 43 and executed by the processor 41, one or more steps of the automatic parking longitudinal decision-making method described above can be executed. Alternatively, in other embodiments, the processor 41 can be configured to execute the automatic parking longitudinal decision-making method by any other appropriate means (e.g., by means of firmware).

[0188] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that receives data and instructions from a storage system, at least one input device, and at least one output device, and transmits the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0189] The computer programs for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, a special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer programs can be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine, or entirely on the remote machine or server.

[0190] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the foregoing.

[0191] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0192] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0193] The computing system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0194] An embodiment of the present invention also provides a vehicle, including the electronic device provided in any embodiment of the present invention.

[0195] In the process of implementing the computer program product, computer program code for performing the operations of the present disclosure may be written in one or more programming languages or combinations thereof. The above-mentioned programming languages include, but are not limited to, object-oriented programming languages - such as Java, Smalltalk, C++, and also include conventional procedural programming languages - such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network - including a local area network (LAN) or a wide area network (WAN) - or, alternatively, may be connected to an external computer (e.g., by using an Internet service provider to connect through the Internet).

[0196] It should be understood that various forms of the processes shown above may be used, steps may be reordered, added or deleted. For example, the steps recited in the present invention may be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is imposed herein.

[0197] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An automatic parking longitudinal decision method, characterized in that: include: During the automatic parking process, the longitudinal collision distance between the vehicle and the obstacle is determined based on the planned trajectory of the vehicle and the predicted trajectory of the vehicle at the current moment, combined with the semantic information of each obstacle in the real-time updated electronic map; Determine the lateral collision distance between the vehicle and the obstacle based on the planned trajectory of the vehicle and the semantic information of each obstacle in the electronic map in combination with a vector projection method; A longitudinal decision is made for the vehicle according to the longitudinal collision distance and the lateral collision distance.

2. The method according to claim 1, characterized in that The method of determining the longitudinal collision distance between the vehicle and the obstacle based on the planned trajectory of the vehicle and the predicted trajectory of the vehicle at the current moment, combined with the semantic information of each obstacle in the electronic map, includes: According to the vehicle posture at each planned trajectory point on the planned trajectory of the self-vehicle at intervals of a preset step length, and the gear position of the vehicle, the self-vehicle contour of the vehicle is fitted to each planned trajectory point of the target self-vehicle planned trajectory, and the target self-vehicle planned trajectory is included in the self-vehicle planned trajectory; According to the position information of each obstacle in the electronic map, determining whether there is an obstacle located in the ego vehicle contour fitted on the target ego vehicle planning trajectory; If so, determining a first longitudinal collision distance between the vehicle and the obstacle; According to the vehicle posture at each predicted trajectory point on the predicted trajectory of the ego vehicle at intervals of a preset step length, and the gear position of the vehicle, the ego vehicle profile of the vehicle is fitted to each predicted trajectory point of the target ego vehicle predicted trajectory, and the target ego vehicle predicted trajectory is included in the ego vehicle predicted trajectory; According to the position information of each obstacle in the electronic map, determining whether there is an obstacle located in the ego vehicle profile fitted on the target ego vehicle predicted trajectory; If so, a second longitudinal collision distance between the vehicle and the obstacle is determined.

3. The method according to claim 2, characterized in that The step of fitting the vehicle contour to each planned trajectory point of the target self-vehicle planned trajectory according to the vehicle posture at each planned trajectory point with a preset step length interval on the self-vehicle planned trajectory and the gear position of the vehicle comprises: constructing an ego-vehicle profile of the vehicle; Determining the target self-vehicle planned trajectory according to the self-vehicle planned trajectory and the gear position of the vehicle; According to the vehicle posture at each planning trajectory point with a set step length in the target ego-vehicle planning trajectory, the ego-vehicle contour of the vehicle is fitted to each planning trajectory point in the target planning trajectory.

4. The method according to claim 1, characterized in that: The step of determining the predicted trajectory of the vehicle at the current moment includes: Determining a current turning radius of the vehicle according to a current steering wheel angle of the vehicle and a wheelbase of the vehicle; According to the current turning radius of the vehicle and the current posture of the vehicle, in combination with a pure tracking model, a predicted self-vehicle trajectory of the vehicle along the current steering wheel angle is determined.

5. The method according to claim 1, characterized in that The determining of the lateral collision distance between the vehicle and the obstacle according to the planned trajectory of the vehicle and the semantic information of each obstacle in the electronic map in combination with a vector projection method includes: For each planned trajectory point in the planned trajectory of the self-vehicle, determining the position information of the center axis of the vehicle according to the vehicle posture at the planned trajectory point; Obtaining the location information of each obstacle within the set range of the planned trajectory point from the electronic map; Determine the lateral distance from each obstacle to the projection point of the center axis of the vehicle according to the position information of the center axis of the vehicle and the position information of each obstacle; Determine a minimum lateral distance from the lateral distances of the projection points of each obstacle to the center axis of the vehicle; A lateral collision distance between the vehicle and an obstacle is determined according to the minimum lateral distance and the width of the vehicle.

6. The method according to claim 1, characterized in that The making a longitudinal decision on the vehicle according to the longitudinal collision distance and the lateral collision distance includes: If the first longitudinal collision distance and the second longitudinal collision distance of the vehicle are both greater than the first longitudinal distance threshold, and the lateral collision distance is greater than the first lateral distance threshold, controlling the vehicle speed to remain unchanged; If the first longitudinal distance of the vehicle or the second longitudinal collision distance is greater than the second longitudinal distance threshold and less than or equal to the first longitudinal distance threshold, or the lateral collision distance is greater than the second lateral distance threshold and less than or equal to the first lateral distance threshold, controlling the vehicle to decelerate; If the first longitudinal distance of the vehicle or the second longitudinal collision distance is less than or equal to a second longitudinal distance threshold, or the lateral collision distance is less than or equal to a second lateral distance threshold, controlling the vehicle to stop; The first longitudinal distance threshold is greater than the second longitudinal distance threshold, and the first lateral distance threshold is greater than the second lateral collision distance threshold.

7. The method according to claim 6, characterized in that Also includes: When the vehicle is parked in place, if the lateral collision distance is greater than a second distance threshold and less than a third collision distance, the rearview mirror of the vehicle is triggered to fold, and the third collision distance is greater than the second collision distance and less than the first collision distance.

8. The method according to claim 1, characterized in that Also includes: If the obstacle is a dynamic obstacle, the estimated collision time between the vehicle and the obstacle is determined according to the planned trajectory of the vehicle and the predicted trajectory of the obstacle, and a longitudinal decision is made for the vehicle according to the estimated collision time.

9. The method according to claim 8, characterized in that The determining the estimated collision time between the vehicle and the obstacle according to the planned trajectory of the vehicle and the predicted trajectory of the obstacle includes: For each timestamp corresponding to the planned trajectory of the ego vehicle, obtaining the planned position information of the ego vehicle at the timestamp and the predicted obstacle position information of the obstacle at the timestamp; Constructing an ego vehicle envelope of the vehicle at a position corresponding to the ego vehicle planned position information, and constructing an obstacle envelope of the obstacle at a position corresponding to the obstacle predicted position information; Determine whether the vehicle envelope overlaps the obstacle envelope; If there is an overlap, the estimated collision time between the vehicle and the obstacle is determined according to the timestamp corresponding to the overlap.

10. The method according to claim 8, characterized in that The making a longitudinal decision on the vehicle according to the estimated collision time includes: If the estimated collision time is greater than a first time threshold, controlling the vehicle to accelerate; If the estimated collision time is greater than the second time threshold and less than or equal to the first time threshold, controlling the vehicle to decelerate; If the estimated collision time is less than or equal to a second time threshold, the vehicle is controlled to stop and the parking operation is continued after the obstacle leaves; The first time threshold is greater than the second time threshold.

11. An automatic parking longitudinal decision-making device, characterized in that: include: A first determination module is used to determine the longitudinal collision distance between the vehicle and the obstacle during the automatic parking process according to the planned trajectory of the vehicle and the predicted trajectory of the vehicle at the current moment, combined with the semantic information of each obstacle in the electronic map updated in real time; A second determination module is used to determine the lateral collision distance between the vehicle and the obstacle according to the planned trajectory of the vehicle and the semantic information of each obstacle in the electronic map in combination with a vector projection method; The longitudinal decision module is used to make a longitudinal decision for the vehicle according to the longitudinal collision distance and the lateral collision distance.

12. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the automatic parking longitudinal decision method according to any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the automatic parking longitudinal decision method according to any one of claims 1 to 10 when executed.

14. A vehicle, characterized in that: The invention comprises implementing the electronic device as claimed in any one of claims 12.

Images