Parking decision planning method and device, vehicle and computer readable storage medium

Abstract

The embodiment of the invention provides a parking decision planning method and device, a vehicle and a computer readable storage medium, and relates to the technical field of intelligent driving. According to the method, a plurality of collision circles can be added on the geometric topology rectangle of the vehicle, and the collision detection model is established; determining a current collision detection range of the vehicle according to the parking planning path and the vehicle state; predicting a collision risk area of the dynamic obstacle according to the motion trend of the dynamic obstacle; predicting a collision risk according to the collision risk area, the collision detection range and the collision detection model; and making a decision plan according to the collision risk. According to the method, the accuracy of parking scene recognition and collision detection can be improved, and the robustness and intelligent performance of automatic parking are improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and more specifically, to a parking decision planning method, device, vehicle and computer-readable storage medium. Background Art

[0002] At present, in the automatic parking scenario, after the intelligent vehicle perceives the surrounding environment and integrates the signals of various sensors, it will generate a grid map and give it to the decision-making planning module to generate a planning path. When dynamic obstacles such as other vehicles and pedestrians appear, if the perception of dynamic obstacles is inaccurate or the collision detection between dynamic obstacles and vehicles is inaccurate, it will interfere with the vehicle in the automatic parking work, causing the vehicle to brake abnormally or the automatic parking function to exit. Summary of the invention

[0003] The embodiments of the present application propose a parking decision planning method, device, vehicle and computer-readable storage medium to solve the above-mentioned technical problems.

[0004] In a first aspect, an embodiment of the present application provides a parking decision planning method, the method comprising: adding multiple collision circles to the geometric topological rectangle of the vehicle to establish a collision detection model; determining the current collision detection range of the vehicle according to the parking planning path and the vehicle status; predicting the collision risk area of ​​the dynamic obstacle according to the movement trend of the dynamic obstacle; predicting the collision risk according to the collision risk area, the collision detection range and the collision detection model; and making a decision plan according to the collision risk.

[0005] In a second aspect, an embodiment of the present application provides a parking decision planning device, which includes: a model building module, which is used to add multiple collision circles on the geometric topological rectangle of the vehicle to establish a collision detection model; a range determination module, which is used to determine the current collision detection range of the vehicle according to the parking planning path and the vehicle status; an area prediction module, which is used to predict the collision risk area of ​​the dynamic obstacle according to the movement trend of the dynamic obstacle; a risk prediction module, which is used to predict the collision risk according to the collision risk area, the collision detection range and the collision detection model; a decision planning module, which is used to make a decision plan according to the collision risk.

[0006] In a third aspect, an embodiment of the present application provides a vehicle, comprising: a memory and a processor, wherein an application is stored in the memory, and the application is used to enable the processor to execute the method provided by the embodiment of the present application when called by the processor.

[0007] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a program code stored thereon, wherein the program code is used to enable the processor to execute the method provided by the embodiment of the present application when called by the processor.

[0008] The parking decision planning method provided in the embodiment of the present application has the following technical effects: a new collision detection model is constructed based on the collision circle, the collision risk area of ​​the dynamic obstacle is predicted according to the movement trend of the dynamic obstacle, the collision risk is predicted according to the collision risk area, the collision detection range and the collision detection model, and a decision planning is made according to the collision risk, which can improve the accuracy of parking scene recognition and collision detection, so that when the vehicle encounters a dynamic obstacle during parking, it can make an intelligent decision based on the collision risk, enrich the application scenarios of the automatic parking function, improve the safety and intelligent performance of the automatic parking function, improve the robustness of the automatic parking planning, and improve the stability of the vehicle when there is an intrusion of a dynamic obstacle. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments and drawings obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0010] Figure 1 A schematic diagram showing a flow chart of a parking decision planning method provided by an embodiment of the present application is shown;

[0011] Figure 2 A schematic diagram of a collision detection model provided by an embodiment of the present application is shown;

[0012] Figure 3 A schematic diagram of a collision detection model provided by another embodiment of the present application is shown;

[0013] Figure 4 A schematic diagram showing different collision detection ranges provided by an embodiment of the present application is shown;

[0014] Figure 5 A schematic diagram showing a collision detection area provided by an embodiment of the present application is shown;

[0015] Figure 6 A schematic diagram showing a collision detection area provided by another embodiment of the present application is shown;

[0016] Figure 7 A schematic diagram showing a collision detection area provided by another embodiment of the present application is shown;

[0017] Figure 8 A schematic diagram showing a collision detection area provided by yet another embodiment of the present application is shown;

[0018] Fig. 9 A schematic diagram showing a collision detection area provided by another embodiment of the present application is shown;

[0019] Fig.10 A schematic diagram showing a collision detection area provided by yet another embodiment of the present application is shown;

[0020] Fig.11 A schematic diagram of the process of step S140 provided in an embodiment of the present application is shown;

[0021] Fig.12 A schematic diagram of a target collision circle set in a high obstacle collision detection model provided by an embodiment of the present application is shown;

[0022] Fig.13 A schematic diagram of a target collision circle set in a low obstacle collision detection model provided by an embodiment of the present application is shown;

[0023] Fig.14 A schematic diagram of a target collision circle set in a high obstacle collision detection model provided by another embodiment of the present application is shown;

[0024] Fig.15 A schematic diagram of a target collision circle set in a low obstacle collision detection model provided by another embodiment of the present application is shown;

[0025] Fig.16 A schematic diagram showing a target collision circle set in a high obstacle collision detection model provided by another embodiment of the present application is shown;

[0026] Fig.17 A schematic diagram showing a target collision circle set in a low obstacle collision detection model provided by another embodiment of the present application is shown;

[0027] Fig.18 A schematic diagram of a target collision circle set in a high obstacle collision detection model provided by yet another embodiment of the present application is shown;

[0028] Fig.19 A schematic diagram showing a target collision circle set in a low obstacle collision detection model provided by yet another embodiment of the present application is shown;

[0029] Fig. 20 A schematic diagram showing a target collision circle set in a high obstacle collision detection model provided by another embodiment of the present application is shown;

[0030] Fig.21 A schematic diagram showing a target collision circle set in a low obstacle collision detection model provided by another embodiment of the present application is shown;

[0031] Fig. 22 A schematic diagram showing a target collision circle set in a high obstacle collision detection model provided by yet another embodiment of the present application is shown;

[0032] Fig.23 A schematic diagram showing a target collision circle set in a low obstacle collision detection model provided by yet another embodiment of the present application is shown;

[0033] Fig.24 A schematic diagram of the process of step S143 provided in an embodiment of the present application is shown;

[0034] Fig.25 A schematic diagram of the process of step S145 provided in an embodiment of the present application is shown;

[0035] Fig.26 A schematic diagram showing the structure of a parking decision-making and planning device provided in one embodiment of the present application is shown;

[0036] Fig. 27 A schematic structural diagram of a vehicle provided in one embodiment of the present application is shown. DETAILED DESCRIPTION

[0037] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0038] The parking decision planning method provided in the embodiment of the present application can be applied to a parking decision planning device or a vehicle. In some embodiments, the parking decision planning device can be integrated in a vehicle. The vehicle can include but is not limited to a gasoline vehicle or an electric vehicle, and the electric vehicle can include but is not limited to a pure electric vehicle, a hybrid vehicle, or a fuel cell vehicle. Next, the parking decision planning method of the present application will be introduced by taking the vehicle as an example of the execution subject.

[0039] See also Figure 1 , Figure 1 FIG. 1 is a flow chart of a parking decision planning method provided by an embodiment of the present application. Figure 1 As shown, the parking decision planning method may include steps S110 to S150.

[0040] Step S110: adding multiple collision circles to the geometric topological rectangle of the vehicle to establish a collision detection model.

[0041] See also Figure 2 , Figure 2 FIG. 1 is a schematic diagram of a collision detection model provided by an embodiment of the present application. Figure 2As shown in , multiple collision circles (i.e., collision circle sets) can be added to the geometric topological rectangle of the vehicle to establish a collision detection model. Figure 2 As shown, the collision circle set in the embodiment of the present application may include a set of multiple circles with different radii and different centers.

[0042] Specifically, for obstacles of different heights, different collision circle sets can be added to the geometric topological rectangle of the vehicle to obtain collision detection models corresponding to obstacles of different heights. This application constructs different collision detection models for obstacles of different heights, which can improve the accuracy of subsequent collision detection.

[0043] In some embodiments, a height threshold may be predefined, and obstacles may be divided into high obstacles (i.e., obstacles with a height greater than the height threshold) and low obstacles (i.e., obstacles with a height not greater than the height threshold) according to the height of the obstacle and the height threshold. According to the difference in high and low obstacle detection, i.e., high obstacles cannot collide with all areas of the vehicle body, and low obstacles do not collide with the core area of ​​the vehicle, different collision circle sets may be selected to obtain collision detection models corresponding to high obstacles and low obstacles, respectively, to improve the accuracy of subsequent collision detection.

[0044] like Figure 2 As shown in FIG. 1 , for a high obstacle whose height is greater than the height threshold, a collision circle set located in all areas of the vehicle body can be added to the geometric topological rectangle of the vehicle to obtain a collision detection model corresponding to the high obstacle (such as Figure 2 The set of collision circles shown).

[0045] See also Figure 3 , Figure 3 FIG. 2 shows a schematic diagram of a collision detection model provided by another embodiment of the present application. Figure 3 As shown in FIG. 1 , for low obstacles whose height is not greater than the height threshold, a set of collision circles located in the core area of ​​the vehicle can be added to the geometric topological rectangle of the vehicle to obtain another collision detection model corresponding to low obstacles (such as Figure 3 The core area may include an area between a head area and a tail area of ​​the vehicle.

[0046] Step S120: Determine the current collision detection range of the vehicle according to the parking planning path and the vehicle status.

[0047] The parking planning path refers to the original parking path automatically planned by the planning module of the vehicle. The user can turn on the parking function of the vehicle by touching (for example, clicking) a parking function turn-on button on the vehicle display or a terminal connected to the vehicle.

[0048] After the parking function of the vehicle is turned on, the vehicle planning module can plan the original path (i.e., the parking planning path) based on the grid map, positioning information, vehicle status information and other information and vehicle dynamics principles. Among them, the main parameters of the grid map may include but are not limited to the drivable area information (freespace), obstacle segments (including high obstacle segments and low obstacle segments), the length and width of the parking space, the end position of automatic parking, etc. The main parameters of the positioning information include the horizontal and vertical coordinates of the vehicle, the curvature of the current positioning point, etc. The positioning information may include but is not limited to the vehicle's global navigation positioning information and inertial navigation positioning information. The vehicle status information may include but is not limited to the vehicle's current speed, acceleration, steering wheel angle, body heading angle, etc.

[0049] Specifically, the vehicle planning module can perform path search based on information such as grid maps, positioning information, vehicle status information and vehicle dynamics principles, using hybrid A* algorithm, RS (Reeds-Shepp) curve, geometric method and other algorithms to obtain an original path (i.e., parking planning path) from the starting point (i.e., the current position of the vehicle) to the end point of the parking space (i.e., the end point of automatic parking).

[0050] The vehicle can set different collision detection ranges according to the parking plan path and vehicle status. Figure 4 , Figure 4 FIG. 1 is a schematic diagram showing different collision detection ranges provided by an embodiment of the present application. Figure 4 As shown in , if the parking planning path is short, that is, the parking planning path is a short path, then the collision detection range is Figure 4 If the parking planning path is long, that is, the parking planning path is a long path, the collision detection range is Figure 4 The light gray area and the dark gray area in the diagram. Due to the differences in the horizontal and vertical lengths of the vehicle itself and the differences in horizontal and vertical driving, the horizontal and longitudinal safety distance settings are different, but the expansion of the collision circle is the same in all directions. From a safety perspective, to the same extent, the lateral safety distance is smaller than the longitudinal safety distance, so when judging the longitudinal safety distance, it is necessary to expand it in the longitudinal direction. In parking scenarios, when the parking planning path is longer, the vehicle has more room to accelerate, usually at a faster speed and with a higher risk, so its collision detection area should be expanded accordingly to enhance the sense of security when parking. The present application determines the current collision detection range of the vehicle based on the parking planning path and the vehicle status, which can improve the accuracy of subsequent collision detection and the safety of automatic parking.

[0051] Step S130: predicting a collision risk area of ​​the dynamic obstacle according to the movement trend of the dynamic obstacle.

[0052] The vehicle can sense the movement trend of the dynamic obstacle, and thus can predict the collision risk area of ​​the dynamic obstacle based on the movement trend of the dynamic obstacle. The movement trend of the dynamic obstacle may include the movement trajectory of the dynamic obstacle from the current moment to the moment when the vehicle is parked in the parking space. The movement trajectory may carry the speed, speed direction and predicted time of the dynamic obstacle. The collision risk area refers to the area where the dynamic obstacle may collide with the vehicle.

[0053] Dynamic obstacles, whether they are pedestrians traveling at low speed or two-wheeled or four-wheeled vehicles traveling at high speed, can be abstracted into rectangles in their bird's-eye view. The movement speed of dynamic obstacles can be obtained through multi-frame extraction. According to the speed size, speed direction and prediction time of dynamic obstacles, the following seven different collision risk areas can be obtained:

[0054] The first case: if the speed of the dynamic obstacle is extremely low (for example, less than 0.5 kph), the dynamic obstacle can be regarded as stationary, and the collision risk area can be a rectangular area abstracted from the dynamic obstacle.

[0055] Second case: Please refer to Figure 5 , Figure 5 A schematic diagram of a collision detection area provided by an embodiment of the present application is shown. Figure 5 In the equation, p0, p1, p2, and p3 are the original four corner points of the dynamic obstacle, and p00, p10, p30, and p40 are the predicted four corner points of the dynamic obstacle. Figure 5 As shown, p00 falls inside the triangle p0p10p30. In this case, the predicted collision area is p0p1p10p20p30p3p0.

[0056] The third case: please refer to Figure 6 , Figure 6 A schematic diagram of a collision detection area provided by another embodiment of the present application is shown. Figure 6 In the equation, p0, p1, p2, and p3 are the original four corner points of the dynamic obstacle, and p00, p10, p30, and p40 are the predicted four corner points of the dynamic obstacle. Figure 6 As shown, p10 falls into the interior of the triangle p1p00p20, and the predicted collision area is p0p1p2p20p30p00p0.

[0057] Fourth case: Please refer to Figure 7 , Figure 7 FIG. 2 is a schematic diagram showing a collision detection area provided by another embodiment of the present application. Figure 7 In the equation, p0, p1, p2, and p3 are the original four corner points of the dynamic obstacle, and p00, p10, p30, and p40 are the predicted four corner points of the dynamic obstacle. Figure 7As shown, p20 falls into the interior of the triangle p2p10p30, and the predicted collision area is p1p2p3p30p00p10p1.

[0058] Fifth case: Please refer to Figure 8 , Figure 8 FIG. 2 is a schematic diagram showing a collision detection area provided by another embodiment of the present application. Figure 8 In the equation, p0, p1, p2, and p3 are the original four corner points of the dynamic obstacle, and p00, p10, p30, and p40 are the predicted four corner points of the dynamic obstacle. Figure 8 As shown, p30 falls into the interior of the triangle p3p00p20, and the predicted collision area is p0p3p2p20p10p00p0.

[0059] Case 6: Please refer to Fig. 9 , Fig. 9 FIG. 2 is a schematic diagram showing a collision detection area provided by another embodiment of the present application. Fig. 9 In the equation, p0, p1, p2, and p3 are the original four corner points of the dynamic obstacle, and p00, p10, p30, and p40 are the predicted four corner points of the dynamic obstacle. Fig. 9 As shown, p00 is on the straight line p0p1, and the predicted collision area is p0p10p20p3p0.

[0060] Case 7: Please refer to Fig.10 , Fig.10 FIG. 2 is a schematic diagram showing a collision detection area provided by yet another embodiment of the present application. Fig.10 In the equation, p0, p1, p2, and p3 are the original four corner points of the dynamic obstacle, and p00, p10, p30, and p40 are the predicted four corner points of the dynamic obstacle. Fig.10 As shown, p00 is on the straight line p0p3, and the predicted collision area is p0p1p20p30p0.

[0061] The above 7 situations are all situations where dynamic obstacles may exist between two frames of the automatic parking planning module. Since the planning operation frequency is 100ms, dynamic obstacles cannot rush out of the original rectangular frame during this interval. Multi-frame sampling is similar to two-frame sampling, and so on.

[0062] Step S140: predicting collision risk according to the collision risk area, the collision detection range and the collision detection model.

[0063] See also Fig.11 , Fig.11 FIG. 2 shows a flow chart of step S140 provided in an embodiment of the present application. Fig.11 As shown, step S140 may include steps S141 to S145.

[0064] Step S141: According to the driving direction of the vehicle, a target collision circle set corresponding to the driving direction is determined from the collision detection model.

[0065] Among them, the vehicle's driving direction can be determined according to the parking planning path, and the vehicle's driving direction may include but is not limited to: turning to the left front, turning to the right front, turning to the left rear, turning to the right rear, driving straight ahead, and driving straight behind.

[0066] The target collision circle set may include a set of circles that need to be subjected to collision detection in the collision detection model. Different driving directions correspond to different target collision circle sets. The corresponding relationship between the driving direction and the target collision circle set may be preset.

[0067] Please also read Fig.12 and Fig.13 , Fig.12 A schematic diagram of a target collision circle set in a high obstacle collision detection model provided by an embodiment of the present application is shown. Fig.13 FIG. 1 is a schematic diagram showing a target collision circle set in a low obstacle collision detection model provided by an embodiment of the present application. Fig.12 and Fig.13 As shown, when the vehicle turns to the left front, the collision detection process only needs to consider Fig.12 and Fig.13 Whether the bold circle collides with the obstacle.

[0068] Please also read Fig.14 and Fig.15 , Fig.14 A schematic diagram of a target collision circle set in a high obstacle collision detection model provided by another embodiment of the present application is shown. Fig.15 FIG. 2 is a schematic diagram showing a target collision circle set in a low obstacle collision detection model provided by another embodiment of the present application. Fig.14 and Fig.15 As shown, when the vehicle turns to the right front, the collision detection process only needs to consider Fig.14 and Fig.15 Whether the bold circle collides with the obstacle.

[0069] Please also read Fig.16 and Fig.17 , Fig.16 FIG. 4 is a schematic diagram showing a target collision circle set in a high obstacle collision detection model provided by another embodiment of the present application. Fig.17 FIG. 2 is a schematic diagram showing a target collision circle set in a low obstacle collision detection model provided by another embodiment of the present application. Fig.16 and Fig.17As shown, when the vehicle turns to the left rear, the collision detection process only needs to consider Fig.16 and Fig.17 Whether the bold circle collides with the obstacle.

[0070] Please also read Fig.18 and Fig.19 , Fig.18 FIG. 4 is a schematic diagram showing a target collision circle set in a high obstacle collision detection model provided in another embodiment of the present application. Fig.19 FIG. 2 is a schematic diagram showing a target collision circle set in a low obstacle collision detection model provided by another embodiment of the present application. Fig.18 and Fig.19 As shown, when the vehicle turns to the right rear, the collision detection process only needs to consider Fig.18 and Fig.19 Whether the bold circle collides with the obstacle.

[0071] Please also read Fig. 20 and Fig.21 , Fig. 20 A schematic diagram of a target collision circle set in a high obstacle collision detection model provided by another embodiment of the present application is shown. Fig.21 FIG. 2 is a schematic diagram showing a target collision circle set in a low obstacle collision detection model provided by another embodiment of the present application. Fig. 20 and Fig.21 As shown, when the vehicle is moving straight ahead, the collision detection process only needs to consider Fig. 20 and Fig.21 Whether the bold circle collides with the obstacle.

[0072] Please also read Fig. 22 and Fig.23 , Fig. 22 FIG. 4 is a schematic diagram showing a target collision circle set in a high obstacle collision detection model provided by yet another embodiment of the present application. Fig.23 FIG. 2 is a schematic diagram showing a target collision circle set in a low obstacle collision detection model provided by yet another embodiment of the present application. Fig. 22 and Fig.23 As shown, when the vehicle is moving straight ahead, the collision detection process only needs to consider Fig. 22 and Fig.23 Whether the bold circle in the figure below collides with the obstacle.

[0073] Step S142: determining a point set of a collision boundary of the collision risk area according to the time it takes for the vehicle to complete the planned parking path and the speed of the dynamic obstacle.

[0074] According to the difference between long and short paths, the time it takes for the vehicle to complete each long and short path can be calculated, which is recorded as time t. That is, the time t for the vehicle to complete the parking planning path can be calculated. According to this time t and the speed (vector) v of the dynamic obstacle output by the prediction module, the product v*t of speed v and time t is calculated. v*t is the point set of the collision boundary of the collision risk area (such as Figures 5 to 8 The length of the bold solid line in the figure) is used to obtain the point set of the final collision rectangular boundary of the collision risk area.

[0075] Step S143: determining the minimum distance between the collision risk area and the collision detection range according to the point set and the target collision circle set.

[0076] According to the vehicle's motion mode, calculate the distance between the point set of the collision boundary of the collision risk area and the center of each circle in the target collision circle set, and then subtract the radius of each circle from the distance to take the minimum distance, which is DISTMin. For details, please refer to Fig.24 , Fig.24 FIG. 4 shows a flow chart of step S143 provided in an embodiment of the present application. Fig.24 As shown, step S143 may include steps S1431 to S1433.

[0077] Step S1431: Calculate the distances between each point set of the collision boundary of the collision risk area and each center of the target collision circle set to obtain a plurality of first distances, each first distance corresponding to a collision circle.

[0078] Step S1432: subtract the radius of the corresponding collision circle from the multiple first distances to obtain multiple second distances.

[0079] For each first distance, the second distance corresponding to the first distance may be obtained by subtracting the radius of the collision circle corresponding to the first distance from the first distance.

[0080] Step S1433: Acquire a minimum distance among a plurality of second distances as the minimum distance between the collision risk area and the collision detection range.

[0081] After obtaining the multiple second distances, the sizes of the multiple second distances may be compared, and the smallest distance therein may be used as the minimum distance DISTMin between the collision risk area and the collision detection range.

[0082] Step S144: Determine the collision risk state according to the minimum distance.

[0083] In the embodiment of the present application, the collision risk state may include a safe state, a deceleration state, and a parking state. As shown in Table 1, a corresponding threshold interval can be predefined for each collision risk state, and different threshold intervals correspond to different collision risk states. In Table 1, DISTMin represents the minimum distance, D1 and D2 represent different thresholds, D1 is greater than D2, and the specific values ​​of D1 and D2 can be predefined according to actual needs.

[0084] Table 1

[0085]

[0086]

[0087] The target threshold interval to which the minimum distance belongs can be determined; and the collision risk state corresponding to the target is determined to be the collision risk state between the vehicle and the dynamic obstacle. Taking Table 1 as an example, if the target threshold interval to which the minimum distance belongs is DISTMin≥D1, the collision risk state between the vehicle and the dynamic obstacle is a safe state. If the target threshold interval to which the minimum distance belongs is D1>DISTMin>D2, the collision risk state between the vehicle and the dynamic obstacle is a deceleration state. If the target threshold interval to which the minimum distance belongs is DISTMin≤D2, the collision risk state between the vehicle and the dynamic obstacle is a parking state.

[0088] Step S145: updating the collision risk state according to the minimum distance of multiple consecutive frames.

[0089] Since front-end perception is difficult to achieve centimeter-level accuracy in parking scenarios, this application processes the minimum distance DISTmin to obtain a more accurate minimum distance change trend, and updates the collision risk status based on the minimum distance change trend to avoid abnormal braking, deceleration or acceleration of the vehicle due to misjudgment, thereby improving the user experience of automatic parking.

[0090] See also Fig.25 , Fig.25 FIG. 4 shows a flow chart of step S145 provided in an embodiment of the present application. Fig.25 As shown, the change trend determination method (ie, step S145) may include steps S1451 to S1455.

[0091] Step S1451: construct a distance series according to the minimum distance of multiple consecutive frames.

[0092] The vehicle can lock the minimum distance of multiple consecutive fixed frames (recorded as n) to obtain the distance sequence DISTmin[n]. In each planning cycle, the first element in the distance sequence is removed and a new minimum distance is added to obtain a new distance sequence.

[0093] Step S1452: Calculate the average values ​​of the distance series of multiple consecutive frames respectively.

[0094] The average values ​​of multiple (e.g., 5) different distance series DISTmin[n] obtained from multiple consecutive frames (e.g., 5 frames) are calculated respectively to obtain the average values ​​of the multiple distance series. After that, the sizes of the multiple (e.g., 5) average values ​​can be compared one by one.

[0095] Step S1453: When the average value of the distance series of multiple frames gradually decreases, it is determined that the collision risk increases, and when the average value decreases by a deceleration threshold or a parking threshold, the collision risk state is updated to a deceleration state or a parking state.

[0096] If the order of the size of multiple (such as 5) average values ​​satisfies the time sequence, that is, the average value at the later time is smaller than the average value at the earlier time, that is, the multiple average values ​​are getting smaller and smaller in the time sequence, it can be determined that the current collision risk is increasing. When the average value increases to the deceleration threshold / parking threshold, the collision risk state can be updated to the deceleration state / parking state, and a deceleration control instruction / parking control instruction is sent. Before the average value decreases the deceleration threshold or the parking threshold, the collision risk state is not updated. Among them, " / " means "or".

[0097] Step S1454: when the average value of the distance series of multiple frames gradually increases, it is determined that the collision risk is reduced, and when the average value increases to a safety threshold, the collision risk state is updated to a safe state.

[0098] If the size of multiple (such as 5) average values ​​is exactly opposite to the time sequence, that is, the average value at a later time is larger than the average value at an earlier time, that is, the multiple average values ​​are getting larger in time sequence, it can be determined that the current collision risk is reduced, and when the average value increases to the safety threshold, the collision risk state is updated to a safe state. Before the average value increases to the safety threshold, the collision risk state is not updated.

[0099] Step S1455: When the average value of the distance series of multiple frames fluctuates, the collision risk state is not updated.

[0100] If the sizes of multiple (such as 5) average values ​​have no obvious relationship with the time series, for example, multiple (such as 5) average values ​​fluctuate, it can be determined that the current collision risk remains unchanged and the collision risk state does not need to be updated.

[0101] Based on steps S1451 to S1455, the present application can avoid to a great extent the problem of dynamic obstacle positioning jumping and forcing the vehicle to stop due to inaccurate perception, thereby greatly improving the accuracy of collision detection and the safety of automatic parking.

[0102] Step S150: making a decision plan according to the collision risk.

[0103] Assuming that the motion state of the dynamic obstacle does not change, the vehicle can determine the parking planning path based on the current waypoint extending the preset length as the target path; obtain the collision risk status of each waypoint on the target path; when there is a parking state in the collision risk state of each waypoint, re-plan the parking path, and regard the collision risk area of ​​the dynamic obstacle as a non-driving area during re-planning. Among them, the non-driving area refers to an area where the vehicle cannot travel, which can also be called an unavailable area or unavailable space. The present application performs re-planning when there is a parking risk, which can improve the continuity and intelligent experience of the parking process and avoid the problem of needing to stop every time a dynamic obstacle appears for re-planning.

[0104] For example, when performing dynamic replanning, the A* algorithm can be used to search for a path away from obstacles. During the search, the collision risk area of ​​the dynamic obstacle is regarded as the unavailable space of the grid map.

[0105] If the vehicle has been forced to stop by a dynamic obstacle, it can stop and wait, and send instructions to the state machine, the parking function enters pause and prompts the driver that an obstacle is passing by. If the pedestrian moves away, the original parking planning path is restored. If the original parking planning path triggers the above re-planning judgment, re-planning is performed. If the dynamic obstacle is stationary, after waiting for more than the pre-defined waiting time, the dynamic obstacle is converted into a static obstacle, re-planning is performed, and the process returns to step S110.

[0106] Steps S110 to S150 have the following technical effects: a new collision detection model is constructed based on the collision circle, the collision risk area of ​​the dynamic obstacle is predicted according to the movement trend of the dynamic obstacle, the collision risk is predicted according to the collision risk area, the collision detection range and the collision detection model, and a decision plan is made according to the collision risk, which can improve the accuracy of parking scene recognition and collision detection, so that when the vehicle encounters a dynamic obstacle during parking, it can make an intelligent decision based on the collision risk, enrich the application scenarios of the automatic parking function, improve the safety and intelligent performance of the automatic parking function, improve the robustness of the automatic parking planning, and improve the stability of the vehicle when a dynamic obstacle intrudes.

[0107] See also Fig.26 , Fig.26 FIG. 1 is a schematic diagram showing the structure of a parking decision planning device provided by an embodiment of the present application. Fig.26As shown, the parking decision device 100 may include: a model building module 110, a range determination module 120, an area prediction module 130, a risk prediction module 140, and a decision planning module 150. The model building module 110 is used to add multiple collision circles to the geometric topological rectangle of the vehicle to establish a collision detection model. The range determination module 120 is used to determine the current collision detection range of the vehicle according to the parking planning path and the vehicle state. The area prediction module 130 is used to predict the collision risk area of ​​the dynamic obstacle according to the movement trend of the dynamic obstacle. The risk prediction module 140 is used to predict the collision risk according to the collision risk area, the collision detection range and the collision detection model. The decision planning module 150 is used to make a decision plan based on the collision risk.

[0108] In some embodiments, the model building module 110 is further used to: for obstacles of different heights, add different collision circle sets to the geometric topological rectangle of the vehicle to obtain collision detection models corresponding to obstacles of different heights.

[0109] In some embodiments, the model building module 110 is also used to: for obstacles with a height greater than a height threshold, add a set of collision circles located in all areas of the vehicle body to the vehicle's geometric topological rectangle to obtain a collision detection model; for obstacles with a height not greater than the height threshold, add a set of collision circles located in the core area of ​​the vehicle to the vehicle's geometric topological rectangle to obtain another collision detection model, wherein the core area includes the area between the head area and the tail area of ​​the vehicle.

[0110] In some embodiments, the risk prediction module 140 is also used to: determine a target collision circle set corresponding to the driving direction from a collision detection model according to the driving direction of the vehicle; determine a point set of the collision boundary of the collision risk area according to the time it takes for the vehicle to complete the planned parking path and the speed of the dynamic obstacle; determine a minimum distance between the collision risk area and the collision detection range according to the point set and the target collision circle set; determine a collision risk state according to the minimum distance; and update the collision risk state according to the minimum distance in multiple consecutive frames.

[0111] In some embodiments, the risk prediction module 140 is further used to: determine the target threshold interval to which the minimum distance belongs, different threshold intervals correspond to different collision risk states; determine the collision risk state corresponding to the target as the collision risk state between the vehicle and the dynamic obstacle.

[0112] In some embodiments, the risk prediction module 140 is also used to: construct a distance series based on the minimum distance of multiple consecutive frames; respectively calculate the average values ​​of the distance series of multiple consecutive frames; when the average value of the distance series of multiple frames gradually decreases, determine that the collision risk increases, and update the collision risk state to a deceleration state or a parking state when the average value decreases by a deceleration threshold or a parking threshold; when the average value of the distance series of multiple frames gradually increases, determine that the collision risk decreases, and when the average value increases to a safety threshold, update the collision risk state to a safe state; when the average value of the distance series of multiple frames fluctuates, the collision risk state is not updated.

[0113] In some embodiments, the decision-making planning module 150 is used to: determine a parking planned path that extends a preset length based on a current waypoint as a target path; obtain the collision risk status of each waypoint on the target path; and re-plan the parking path when a parking state exists in the collision risk status of each waypoint, and during the re-planning, the collision risk area of ​​the dynamic obstacle is regarded as a non-driving area.

[0114] Those skilled in the art can clearly understand that the above device provided in the embodiment of the present application can implement the method provided in the embodiment of the present application. The specific working process of the above-described device and module can refer to the corresponding process of the method in the embodiment of the present application, which will not be repeated here.

[0115] In the embodiments provided in the present application, the coupling, direct coupling or communication connection between the modules shown or discussed may be indirect coupling or communication coupling through some interfaces, devices or modules, and may be electrical, mechanical or other forms, and the embodiments of the present application do not impose specific limitations on this.

[0116] In addition, each functional module in the embodiment of the present application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules.

[0117] See also Fig. 27 , Fig. 27 FIG. 1 shows a schematic diagram of the structure of a vehicle provided by an embodiment of the present application. Fig. 27 As shown, the vehicle 200 may include a memory 210 and a processor 220, wherein the memory 210 stores an application program, and the application program is configured to enable the processor 220 to execute the method provided in the embodiment of the present application when called by the processor 220.

[0118] The processor 220 may include one or more processing cores. The processor 220 uses various interfaces and lines to connect various parts of the entire vehicle 200, and is used to run or execute instructions, programs, code sets or instruction sets stored in the memory 210, and call to run or execute data stored in the memory 210, perform various functions of the vehicle 200 and process data.

[0119] The processor 220 can be implemented in at least one of the following hardware forms: digital signal processing (DSP), field programmable gate array (FPGA), and programmable logic array (PLA). The processor 220 can integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem is used to process wireless communications. It is understandable that the above-mentioned modem may not be integrated into the processor 220, but may be implemented separately through a communication chip.

[0120] The memory 210 may include a random access memory (RAM) or a read-only memory (ROM). The memory 210 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 210 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function, instructions for implementing the above-mentioned various method embodiments, etc. The data storage area may store data created by the vehicle 200 during use, etc.

[0121] An embodiment of the present application also provides a computer-readable storage medium, on which a program code is stored. The program code is configured to execute the method provided in the embodiment of the present application when called by a processor.

[0122] The computer-readable storage medium may be an electronic memory such as a flash memory, an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a hard disk or a ROM.

[0123] In some embodiments, the computer readable storage medium includes a non-volatile computer readable medium (Non-Transitory Computer-Readable Storage Medium, referred to as Non-TCRSM). The computer readable storage medium has a storage space for the program code that executes any method step in the above method. These program codes can be read from or written into one or more computer program products. The program code can be compressed in an appropriate form.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A parking decision planning method, characterized in that: include: Add multiple collision circles to the geometric topological rectangle of the vehicle to establish a collision detection model; Determine the current collision detection range of the vehicle based on the parking planning path and vehicle status; Predicting a collision risk area of ​​the dynamic obstacle based on the movement trend of the dynamic obstacle; predicting a collision risk according to the collision risk area, the collision detection range, and the collision detection model; Based on the collision risk, decision planning is made.

2. The method according to claim 1, characterized in that: The method of adding a plurality of collision circles to the geometric topological rectangle of the vehicle and establishing a collision detection model includes: For obstacles of different heights, different collision circle sets are added to the geometric topological rectangle of the vehicle to obtain collision detection models corresponding to obstacles of different heights.

3. The method according to claim 2, characterized in that For obstacles of different heights, different collision circle sets are added to the geometric topological rectangle of the vehicle to obtain collision detection models corresponding to obstacles of different heights, including: For obstacles with a height greater than a height threshold, a collision circle set located in all areas of the vehicle body is added to the geometric topological rectangle of the vehicle to obtain a collision detection model; For obstacles whose height is not greater than the height threshold, a set of collision circles located in the core area of ​​the vehicle is added to the geometric topological rectangle of the vehicle to obtain another collision detection model, where the core area includes an area between the head area and the tail area of ​​the vehicle.

4. The method according to claim 1, characterized in that The predicting of the collision risk according to the collision risk area, the collision detection range and the collision detection model includes: According to the driving direction of the vehicle, a target collision circle set corresponding to the driving direction is determined from the collision detection model; Determining a point set of a collision boundary of the collision risk area according to the time it takes for the vehicle to complete the planned parking path and the speed of the dynamic obstacle; Determining a minimum distance between the collision risk area and the collision detection range according to the point set and the target collision circle set; determining a collision risk state according to the minimum distance; The collision risk state is updated according to the minimum distance in the consecutive multiple frames.

5. The method according to claim 4, characterized in that The predicting the collision risk according to the minimum distance includes: Determine a target threshold interval to which the minimum distance belongs, where different threshold intervals correspond to different collision risk states; A collision risk state corresponding to the target is determined to be a collision risk state between the vehicle and the dynamic obstacle.

6. The method according to claim 4 or 5, characterized in that: The updating of the collision risk state according to the minimum distance of the continuous multiple frames includes: Constructing a distance series according to the minimum distances of the continuous multiple frames; Calculate the average values ​​of the distance series of multiple consecutive frames respectively; When the average value of the distance series of multiple frames gradually decreases, it is determined that the collision risk increases, and when the average value decreases by a deceleration threshold or a parking threshold, the collision risk state is updated to a deceleration state or a parking state; When the average value of the distance series of multiple frames gradually increases, it is determined that the collision risk is reduced, and when the average value increases to a safety threshold, the collision risk state is updated to a safe state; When the average value of the distance series of multiple frames fluctuates, the collision risk state is not updated.

7. The method according to claim 1, characterized in that The making of decision planning according to the collision risk includes: Determine a parking planning path based on the current waypoint and extending a preset length as a target path; Obtaining the collision risk status of each waypoint on the target path; When a parking state exists in the collision risk state of each waypoint, the parking path is replanned, and the collision risk area of ​​the dynamic obstacle is regarded as a non-driving area during the replanning.

8. A parking decision-making and planning device, characterized in that: include: A model building module is used to add multiple collision circles to the geometric topological rectangle of the vehicle to establish a collision detection model; A range determination module is used to determine the current collision detection range of the vehicle based on the parking planning path and vehicle status; An area prediction module, used to predict the collision risk area of ​​the dynamic obstacle according to the movement trend of the dynamic obstacle; A risk prediction module, used to predict the collision risk according to the collision risk area, the collision detection range and the collision detection model; The decision-making planning module is used to make a decision plan according to the collision risk.

9. A vehicle, characterized in that: include: A memory and a processor, wherein an application is stored in the memory, and the application is used to enable the processor to execute the method according to any one of claims 1 to 7 when called by the processor.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, and when the program codes are called by a processor, the processor executes the method according to any one of claims 1 to 7.

Images