Intelligent temporary parking method and device, vehicle and storage medium

Abstract

The application provides an intelligent temporary parking method and device, electronic equipment and storage medium, and belongs to the field of low-speed park intelligent parking; the method comprises the following steps: starting an automatic parking function and acquiring obstacle information in a sensing range; judging whether the path of a target vehicle according to a preset parking track to a specified temporary parking space is blocked according to the obstacle information; if yes, obtaining a plurality of parking sampling points according to the current sampling point of the target vehicle through the scatter sampling method, and planning a plurality of parking tracks by using a quintic polynomial curve; selecting one of the parking tracks as a target parking track through a prefabricated performance evaluation model; and selecting the parking spaces corresponding to the other suitable parking tracks as the space required for parallel parking according to the position of the target parking track in the time sequence. Through the application, the temporary parking point can be parked quickly according to the current surrounding environment information of the given temporary parking point, and the path planning for replacing the temporary parking point for re-parking in time.

Description

Technical Field

[0001] This invention belongs to the technical field of low-speed intelligent parking in parks, specifically relating to an intelligent temporary parking method, device, electronic equipment, and storage medium. Background Technology

[0002] In the structured roads of the low-speed park, autonomous vehicles for shuttle passengers can temporarily park on both sides of the road (for 5-10 minutes). Existing autonomous vehicles can pull over to park temporarily at given temporary parking spots, which can effectively solve the parking difficulties in the park and reduce the time required for shuttle passengers; the given temporary parking spots can be manually selected parking spots or parking spots marked on high-precision maps.

[0003] Currently, the intelligent temporary parking technology for existing autonomous vehicles typically faces two main problems when parking in low-speed parks: 1. Because temporary parking spots in low-speed parks are usually fixed points according to the park's planning; for example... Figure 1 As shown, when a static obstacle (such as a static obstacle vehicle) blocks the temporary parking point P, and dynamic obstacles (such as static obstacle vehicles) still exist in the surrounding environment, the vehicle's intelligent parking decision will choose to simply park and wait at the current location without any further action. This will prevent the vehicle from quickly achieving a complete parking near the temporary parking point P, thus affecting the traffic efficiency of the low-speed park. 2. If Figure 2 As shown, after a vehicle temporarily parks at temporary parking point P, the presence of static obstacles (such as static obstacle vehicles) at the current temporary parking point P will cause congestion in the lane if the temporary parking point is not changed, affecting the traffic efficiency of vehicles behind. However, the vehicle usually still needs to park and wait near the temporary parking point P for a certain period of time, but at this time it cannot intelligently change the parking point to make way, thus affecting the traffic efficiency of the low-speed park.

[0004] Therefore, it is particularly important to enable intelligent driving vehicles to quickly complete the parking process at a given temporary parking spot and plan the route to change parking spots in a timely manner when temporarily parking in low-speed parks, so as to effectively improve the traffic efficiency of low-speed parks. Summary of the Invention

[0005] To address at least one of the aforementioned technical problems, the present invention provides an intelligent temporary parking method, apparatus, electronic device, and storage medium.

[0006] In a first aspect, the invention provides an intelligent temporary parking method, comprising:

[0007] When the target vehicle enters the sensing range of the designated temporary parking space, the automatic parking function is activated and obstacle information within the sensing range is acquired; wherein, the obstacle information includes dynamic obstacles and static obstacles.

[0008] Based on the obstacle information, determine whether the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is obstructed;

[0009] If so, then based on the current sampling point of the target vehicle, several parking space sampling points are obtained by the point-scattering sampling method, and several parking trajectories are planned using a fifth-order polynomial curve.

[0010] One parking trajectory is selected from the aforementioned parking trajectories using a prefabricated performance evaluation model as the target parking trajectory; wherein, the prefabricated performance evaluation model includes comfort, safety, and efficiency.

[0011] When the target vehicle arrives at the target parking space according to the target parking trajectory, a suitable parking space corresponding to one of the other parking trajectories is selected as the space required for parallel parking based on the position of the target parking trajectory in the time sequence, so that the target vehicle stops correctly in the target parking space.

[0012] Preferably, the step of activating the automatic parking function and acquiring obstacle information within the perception range when the target vehicle enters the sensing range of the designated temporary parking space; wherein the obstacle information includes both dynamic and static obstacles, specifically includes:

[0013] The automatic parking function is activated when the target vehicle enters the sensing range of the designated temporary parking space.

[0014] The image acquisition device collects real-time environmental information about the target vehicle's surroundings; wherein the area corresponding to the perception range is a subset of the corresponding area of ​​the real-time surroundings.

[0015] The environmental information is screened to obtain obstacle information for both dynamic and static obstacles within the perception range.

[0016] Preferably, the specific steps of obtaining several parking space sampling points based on the current sampling point of the target vehicle using a point-scattering sampling method, and planning several parking trajectories using a fifth-order polynomial curve include:

[0017] Based on the Frenet coordinate system, with the road centerline as the reference line, the vertical axis represents the displacement along the road direction, and the horizontal axis represents the lateral displacement perpendicular to the road centerline, the current sampling point of the target vehicle in its current state is determined; wherein, the current sampling point includes the initial lateral position and the initial longitudinal position.

[0018] Based on the current sampling point, the end lateral position and end longitudinal position of several feasible berth states are obtained by the point sampling method.

[0019] The initial lateral and initial longitudinal positions are fitted with a fifth-order polynomial curve to several of the end lateral and end longitudinal positions to generate lateral and longitudinal trajectories, respectively.

[0020] The lateral and longitudinal trajectories are coupled according to a time series to generate several parking trajectories based on the Frenet coordinate system.

[0021] Preferably, the step of selecting a parking trajectory from the plurality of parking trajectories as the target parking trajectory using a prefabricated performance evaluation model; wherein, the prefabricated performance evaluation model includes comfort, safety, and efficiency, specifically includes:

[0022] The coordinate systems of the parking trajectories are transformed to generate the candidate parking trajectories based on the global coordinate system.

[0023] A loss function is constructed based on evaluation indicators that include at least comfort, safety, and efficiency to build a prefabricated performance evaluation model;

[0024] The cost of driving for the candidate parking trajectories is calculated based on the prefabricated performance evaluation model; wherein the cost includes comfort cost, safety cost, and driving efficiency cost.

[0025] The optimal parking value is selected from the estimated parking values, and the candidate parking trajectory corresponding to the optimal parking value is taken as the target parking trajectory.

[0026] Preferably, the prefabricated performance evaluation model is as follows:

[0027] J total =W s cost s +W l cost l ;

[0028] In the formula, W s W represents the cost weight of the longitudinal motion trajectory. l Cost represents the cost weight of the lateral movement trajectory. s The cost function represents the quality evaluation cost of longitudinal motion trajectory planning. l This represents the cost function for quality evaluation of lateral motion trajectory planning.

[0029] Preferably, the step of selecting suitable parking spaces corresponding to other parking trajectories as the space required for parallel parking, based on the position of the target parking trajectory in the time series when the target vehicle arrives at the target parking space according to the target parking trajectory, so that the target vehicle is parked correctly in the target parking space, specifically includes:

[0030] When the target vehicle arrives at the target parking space according to the target parking trajectory, it is determined whether the parking space of the target parking trajectory is at the first position of the time sequence;

[0031] If so, the parking space of the adjacent subsequent parking trajectory is used as the space required for parallel parking, so that the target vehicle can park upright by using the parking space of the subsequent parking trajectory, so that the target vehicle is parked upright in the target parking space.

[0032] If not, then the space required for parallel parking is the adjacent parking space of the previous parking trajectory, so that the target vehicle can park itself upright using the parking space of the previous parking trajectory, so that the target vehicle is parked upright in the target parking space.

[0033] Preferably, after the step of selecting suitable parking spaces corresponding to the other parking trajectories as the space required for parallel parking, based on the position of the target parking trajectory in the time series when the target vehicle arrives at the target parking space according to the target parking trajectory, so that the target vehicle is parked correctly in the target parking space, the method further includes:

[0034] When the target vehicle is in the target parking space, a decision is made based on whether the target vehicle is causing road congestion or whether to continue parking in the target parking space.

[0035] Preferably, the step of deciding whether to find a new parking space or continue parking in the target parking space when the target vehicle is in the target parking space specifically includes:

[0036] When the target vehicle is in the target parking space, determine whether the target vehicle is causing road congestion;

[0037] If so, then reset the parking trajectory planning of the new parking space near the target parking space so that the target vehicle is parked correctly in the new parking space;

[0038] If not, the target vehicle continues to park at the target parking space.

[0039] Secondly, the invention also provides an intelligent temporary parking system, comprising:

[0040] The acquisition module is used to activate the automatic parking function and acquire obstacle information within the perception range when the target vehicle enters the perception range of the designated temporary parking space; wherein, the obstacle information includes dynamic obstacles and static obstacles.

[0041] The prediction module is used to determine whether the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is blocked based on the obstacle information.

[0042] The planning module is used to obtain several parking space sampling points based on the current sampling points of the target vehicle by using a point-scattering sampling method if the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is blocked, and to plan several parking trajectories using a fifth-order polynomial curve.

[0043] The evaluation module is used to select one parking trajectory from the plurality of parking trajectories as the target parking trajectory through a prefabricated performance evaluation model; wherein, the prefabricated performance evaluation model includes comfort, safety, and efficiency.

[0044] The adjustment module is used to select, based on the position of the target parking trajectory in the time series, suitable parking spaces corresponding to other parking trajectories as the space required for parallel parking when the target vehicle arrives at the target parking space according to the target parking trajectory, so that the target vehicle stops correctly at the target parking space.

[0045] Preferably, the acquisition module includes:

[0046] The activation unit is used to activate the automatic parking function when the target vehicle enters the sensing range of the designated temporary parking space;

[0047] The acquisition unit is used to acquire real-time environmental information of the target vehicle's surroundings based on the image acquisition device; wherein the corresponding area of ​​the perception range belongs to a subset of the corresponding area of ​​the real-time surroundings.

[0048] The acquisition unit is used to screen the environmental information to obtain obstacle information of dynamic and static obstacles within the perception range.

[0049] Preferably, the planning module includes:

[0050] The determination unit is used to determine the current sampling point of the target vehicle in the current state based on the Frenet coordinate system, with the road centerline as the reference line, the vertical coordinate as the displacement along the road direction, and the horizontal coordinate as the lateral displacement perpendicular to the road centerline; wherein, the current sampling point includes the initial lateral position and the initial longitudinal position.

[0051] The sampling unit is used to obtain the end lateral position and end longitudinal position of several feasible berth states respectively by means of a point-scattering sampling method based on the current sampling point.

[0052] The fitting unit is used to fit the initial lateral position and the initial longitudinal position to several of the end lateral positions and end longitudinal positions respectively using a fifth-order polynomial curve to generate a lateral trajectory and a longitudinal trajectory.

[0053] The coupling unit is used to couple the lateral trajectory and the longitudinal trajectory according to the time sequence to generate several parking trajectories based on the Frenet coordinate system.

[0054] Preferably, the evaluation module includes:

[0055] A transformation unit is used to transform the coordinate system of the plurality of parking trajectories to generate the plurality of candidate parking trajectories based on the global coordinate system;

[0056] Construction unit, used to construct a loss function based on evaluation indicators including at least comfort, safety and efficiency to construct a prefabricated performance evaluation model;

[0057] The calculation unit is used to calculate the cost of driving corresponding to the plurality of candidate parking trajectories based on the pre-made performance evaluation model; wherein the cost includes comfort cost, safety cost, and driving efficiency cost.

[0058] The selection unit is used to select the optimal value from the value and take the candidate parking trajectory corresponding to the optimal value as the target parking trajectory.

[0059] Preferably, the adjustment module includes:

[0060] The first judgment unit is used to determine whether the parking space of the target parking trajectory is at the first position of the time sequence when the target vehicle travels to the target parking space according to the target parking trajectory.

[0061] The first parking alignment unit is used to use the parking space of the next adjacent parking trajectory as the space required for parallel parking if the parking space of the target parking trajectory is at the first position of the time sequence, so that the target vehicle can park and align itself using the parking space of the next parking trajectory, so that the target vehicle is parked and aligned at the target parking space.

[0062] The second parking alignment unit is used to use the parking space of the previous parking trajectory adjacent to the target parking trajectory as the space required for parallel parking if the parking space of the target parking trajectory is not in the first position of the time sequence, so that the target vehicle can park and align itself using the parking space of the previous parking trajectory, so that the target vehicle is parked in the target parking space.

[0063] Preferably, the intelligent temporary parking system further includes a congestion mitigation module, used to make a decision on whether the target vehicle is causing road congestion or to continue parking in the target parking space when the target vehicle is in the target parking space.

[0064] Preferably, the congestion mitigation module includes:

[0065] The second judgment unit is used to determine whether the target vehicle is causing road congestion when the target vehicle is in the target parking space.

[0066] A reset unit is used to reset the parking trajectory planning of a new parking space near the target parking space if the target vehicle causes road congestion, so that the target vehicle is parked correctly in the new parking space;

[0067] The continuation unit is used to ensure that the target vehicle continues to park in the target parking space if the target vehicle does not cause road obstruction.

[0068] Thirdly, embodiments of this application provide a vehicle including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the intelligent temporary parking method as described in the first aspect.

[0069] Fourthly, embodiments of this application provide a storage medium storing a computer program that, when executed by a processor, implements the intelligent temporary parking method as described in the first aspect.

[0070] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0071] 1. When a designated temporary parking spot is blocked by static obstacles and there are still dynamic obstacles in the surrounding environment, a new parking spot can be selected by using planning methods such as point sampling, five-term polynomial curves and coupling. This allows vehicles to park near the designated temporary parking spot instead of waiting in the original road direction, which can effectively improve the efficiency of vehicle traffic and transportation in the entire park.

[0072] 2. If temporary parking at temporary parking spots causes lane congestion, new parking spots can be selected to allow passage by timely sampling, fifth-order polynomial curves, and coupling planning methods, which can effectively improve the overall vehicle traffic efficiency of the park.

[0073] 3. The algorithm employs point sampling, five-term polynomial curve and coupling planning methods, which have fast calculation speed and can select the optimal parking trajectory in a short time to realize the path and speed planning for parking on the side of the road. Furthermore, the previously calculated sampling points can be used to make decisions on the selection of new parking points to adjust the vehicle body to stop and complete parking. It can take into account both driving and parking state sampling. Attached Figure Description

[0074] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0075] Figure 1 This is a diagram illustrating one of the problems with existing intelligent temporary parking technologies.

[0076] Figure 2 This is a schematic diagram illustrating the second problem with existing intelligent temporary parking technology.

[0077] Figure 3 The flowchart is shown below for the intelligent temporary parking method provided in Embodiment 1 of the present invention;

[0078] Figure 4 This is a working diagram of the target vehicle finding a new parking space during parking, as provided in Embodiment 1 of the present invention.

[0079] Figure 5 This is a working diagram of the target vehicle completing the yielding process after temporary parking, as provided in Embodiment 1 of the present invention;

[0080] Figure 6 This is a block diagram of an intelligent temporary parking system corresponding to the method in Embodiment 1, provided in Embodiment 2 of the present invention;

[0081] Figure 7 This is a flowchart of step S205 in the intelligent temporary parking method provided in Embodiment 3 of the present invention;

[0082] Figure 8 This is a flowchart of S206 in the intelligent temporary parking method provided in Embodiment 3 of the present invention;

[0083] Figure 9 This is a block diagram of the intelligent temporary parking system structure corresponding to the method in Embodiment 3 provided in Embodiment 4 of the present invention;

[0084] Figure 10 This is a working diagram of the target vehicle completing the yielding process after temporary parking, as provided in Embodiment 5 of the present invention;

[0085] Figure 11 This is a block diagram of an intelligent temporary parking system corresponding to the method in Embodiment 5, provided in Embodiment 6 of the present invention;

[0086] Figure 12 This is a schematic diagram of the hardware structure of the vehicle provided in Embodiment 7 of the present invention.

[0087] Explanation of reference numerals in the attached figures:

[0088] 10-Acquisition Module, 11-Startup Unit, 12-Collection Unit, 13-Acquisition Unit;

[0089] 20-Prediction Module;

[0090] 30 - Planning module, 31 - Determining unit, 32 - Sampling unit, 33 - Fitting unit, 34 - Coupling unit;

[0091] 40 - Evaluation module, 41 - Conversion unit, 42 - Construction unit, 43 - Calculation unit, 44 - Optimal selection unit;

[0092] 50 - Adjustment module, 51 - First judgment unit, 52 - First stop adjustment unit, 53 - Second stop adjustment unit;

[0093] 60-Traffic congestion mitigation module, 61-Second judgment unit, 62-Reset unit, 63-Continuation unit;

[0094] 70-Pathway Module;

[0095] 80 - Bus, 81 - Processor, 82 - Memory, 83 - Communication interface. Detailed Implementation

[0096] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0097] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0098] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0099] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0100] Example 1

[0101] Specifically, Figure 3 The diagram shown is a flowchart of an intelligent temporary parking method provided in this embodiment.

[0102] like Figure 3 As shown, the intelligent temporary parking method of this embodiment includes the following steps:

[0103] S101, when the target vehicle enters the sensing range of the designated temporary parking space, the automatic parking function is activated and obstacle information within the sensing range is obtained; wherein, the obstacle information includes dynamic obstacles and static obstacles.

[0104] Specifically, in this embodiment, the temporary parking spaces in the low-speed park are equipped with sensing sensors with a certain range of sensing capabilities. The sensing information includes the temporary parking space itself and surrounding dynamic obstacles (usually moving vehicles, pedestrians, etc.) and static obstacles (usually parked vehicles, temporarily placed debris, etc.). When a target vehicle requiring temporary parking enters the sensing range of the designated temporary parking space, the target vehicle's intelligent parking system activates the automatic parking decision function. It uses the target vehicle's own camera to capture images of the surrounding environment and, with the help of the sensing range of the designated temporary parking space, collaboratively processes and acquires only obstacle information within the sensing range to reduce processing time.

[0105] Furthermore, step S101 of this embodiment specifically includes:

[0106] S1011, when the target vehicle enters the sensing range of the designated temporary parking space, the automatic parking function is activated.

[0107] Specifically, intelligent driving vehicles typically have intelligent parking systems that combine high-precision maps and positioning information to send a real-time signal to search for temporary parking spaces when they arrive at a designated temporary parking space. Additionally, the perception sensors at the temporary parking spaces in the low-speed zone also send real-time signals. When the target vehicle enters the range perceived by the perception sensors, the intelligent parking system receives the signal and starts up promptly.

[0108] S1012, The image acquisition device acquires real-time environmental information of the target vehicle's surroundings; wherein the area corresponding to the perception range is a subset of the corresponding area of ​​the real-time surroundings.

[0109] Specifically, when the intelligent parking system of the target vehicle is activated, the environmental image acquisition device on the target vehicle is turned on to collect information about the surrounding environment, acquiring environmental information within a certain range around the target vehicle. The acquisition range of the environmental image acquisition device is larger than the sensing range of the temporary parking space's sensors, meaning that the environmental acquisition range around the target vehicle includes the range sensed by the sensors. This improves the safety performance of subsequent parking decisions.

[0110] S1013, The environmental information is screened to obtain obstacle information of dynamic and static obstacles within the perception range.

[0111] Specifically, in this embodiment, since the collection range of the target vehicle's surrounding environment information is much larger than the sensing range of the temporary parking space's sensing sensor, not all of the target vehicle's surrounding environment information can be used. In order to reduce the extraction of subsequent effective information, it is necessary to screen the target vehicle's surrounding environment information and only need the environmental information between the target vehicle and the designated temporary parking space, that is, the relevant obstacle information within the sensing range. This relevant obstacle information includes dynamic obstacles (usually motor vehicles in motion, pedestrians, etc.) and static obstacles (usually motor vehicles in parking status, temporarily placed debris, etc.).

[0112] S102, determine whether the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is blocked based on the obstacle information.

[0113] Specifically, in this embodiment, when the intelligent parking system of the target vehicle enters the range sensed by the designated temporary parking space, the intelligent parking system automatically generates a parking trajectory for the target vehicle to drive to the designated temporary parking space, i.e., a preset parking trajectory. Based on this preset parking trajectory, combined with the obstacle information obtained in the previous step, it can be determined whether there are obstacles on the preset parking trajectory. If there are obstacles, it means that parking cannot be performed according to the preset parking trajectory; if there are no obstacles, parking can be performed directly according to the preset parking trajectory.

[0114] S103, if so, then based on the current sampling point of the target vehicle, a number of parking space sampling points are obtained by the point-scattering sampling method, and a number of parking trajectories are planned using a fifth-order polynomial curve.

[0115] Specifically, such as Figure 4 As shown, when there is an obstacle in the designated temporary parking space for the target vehicle, a new parking space search decision is made. In order to improve the computing power and speed of the decision, a point sampling method combined with a fifth-order polynomial curve method is used to quickly and accurately plan the parking trajectory of the parking points near the designated temporary parking space according to the specific situation.

[0116] Furthermore, step S103 of this embodiment specifically includes:

[0117] S1031, Based on the Frenet coordinate system, with the road centerline as the reference line, the vertical coordinate is the displacement along the road direction, and the horizontal coordinate is the lateral displacement perpendicular to the road centerline, determine the current sampling point of the target vehicle in its current state; wherein, the current sampling point includes the initial lateral position and the initial longitudinal position.

[0118] S1032, Based on the current sampling point, obtain the end lateral position and end longitudinal position of several feasible berth states respectively by the point sampling method;

[0119] S1033, The initial lateral position and the initial longitudinal position are fitted with a fifth-order polynomial curve to a number of the end lateral positions and end longitudinal positions to generate a lateral trajectory and a longitudinal trajectory, respectively.

[0120] S1034, the lateral trajectory and longitudinal trajectory are coupled according to the time series to generate several parking trajectories based on the Frenet coordinate system.

[0121] Specifically, in combination Figure 4 As shown, taking the road centerline where the temporary parking space is located as the reference line, the vertical coordinate S represents the displacement along the road direction, and the horizontal coordinate D represents the lateral displacement perpendicular to the road centerline. Let the coordinates of the temporary parking point P be (s1, d1). Therefore, the nearest feasible parking area obtained through environmental perception search is the right side of the current vehicle, as shown below. Figure 4 As shown in the dashed box, sampling is performed on the right side. The longitudinal position s1 and lateral position d1 of point P are used as reference values. The longitudinal positions of each sampling point are [s1+2.0, s1+4.0, s1+6.0, s1+8.0], i.e., the longitudinal sampling interval Δs = 2.0m. The lateral position of each sampling point is d1. The four circles A, B, C, and D in the sampling area Q in the figure are the final state sampling points, with velocity and acceleration of 0. The current position of the target vehicle is point E (s0, d0). The current velocity and acceleration are used as the initial state of the vehicle. The initial and final states of the horizontal and vertical axes are connected by a fifth-order polynomial curve. Then, the horizontal and vertical trajectories are coupled to plan four corresponding parking trajectories, which include vehicle path and speed information, such as... Figure 4 As shown by the black curve in the middle.

[0122] S104, select one parking trajectory from the plurality of parking trajectories as the target parking trajectory through a prefabricated performance evaluation model; wherein, the prefabricated performance evaluation model includes comfort, safety and efficiency.

[0123] Specifically, a pre-designed performance evaluation model is used to evaluate the collision-free parking trajectory planned by the intelligent parking system in order to select the optimal target parking trajectory. The evaluation factors of the pre-designed performance evaluation model in this embodiment include, but are not limited to, comfort, safety, and efficiency.

[0124] Furthermore, step S104 of this embodiment specifically includes:

[0125] S1041, the coordinate system of the plurality of parking trajectories is transformed to generate the plurality of candidate parking trajectories based on the global coordinate system.

[0126] Specifically, trajectory planning based on the Frenet coordinate system will output separate horizontal and vertical trajectories. However, the final output of the reference motion trajectory for autonomous vehicles needs to be directly applicable to the execution module. Therefore, the trajectory planning results obtained in the Frenet coordinate system need to be output in the global coordinate system.

[0127] S1042, construct a loss function based on evaluation indicators including at least comfort, safety and efficiency to construct a prefabricated performance evaluation model.

[0128] The prefabricated performance evaluation model is as follows:

[0129] J total =W s cos t s +W l cos t l ;

[0130] In the formula, W s W represents the cost weight of the longitudinal motion trajectory. l Cost represents the cost weight of the lateral movement trajectory. s The cost function represents the quality evaluation cost of longitudinal motion trajectory planning. l This represents the cost function for quality evaluation of lateral motion trajectory planning.

[0131] Specifically, in this embodiment, the quality evaluation cost function for longitudinal motion trajectory planning and the quality evaluation cost function for lateral motion trajectory planning are obtained through longitudinal trajectory solving and lateral trajectory solving, respectively; the specific calculations are as follows:

[0132] 1. Solving the longitudinal trajectory

[0133] The corresponding trajectory duration T = [t0, t1], and the initial longitudinal state of the vehicle at time t0 is: The longitudinal state of the vehicle at the end time t1 is The fifth-degree polynomial relation is as follows, where t∈T:

[0134] s(t)=c s0 +c s1 t+c s2 t 2 +c s3 t 3 +c s4 t 4 +c s5 t 5 Taking the second derivative of the formula, we get:

[0135]

[0136]

[0137] The initial and final states S0 and S1, calculated using the above formula, can be expressed as matrix equations as follows:

[0138]

[0139] In the above matrix equation, M s C is a constant matrix. s Let be the polynomial coefficient matrix, and let the initial and final state matrix be... The polynomial coefficient matrix Cs = Ms⁻¹Ps can be obtained from the above matrix equation. The coefficient matrix C of the matrix equation needs to be solved again after each reprogramming trigger. s The longitudinal motion trajectory within the corresponding trajectory duration (T) can be described as follows:

[0140]

[0141] The third derivative of the longitudinal trajectory s(t) with respect to time t The jerk describes the first derivative of longitudinal acceleration, and its magnitude can be used to describe the comfort level of an autonomous vehicle during its motion. The overall comfort level over a corresponding trajectory duration T is represented by J. T (s(t)) represents the quality of the trajectory, and its calculation formula is as follows:

[0142]

[0143] The quality assessment cost function for longitudinal motion trajectory planning can be abstracted into the following function:

[0144]

[0145] In the formula, g(T) is the duration corresponding to the longitudinal motion trajectory, i.e., g(T) = T, which is used to describe the trajectory planning efficiency; S(i) is the trajectory length of the longitudinal motion trajectory along the reference line direction, which is used to describe the trajectory behavior. It tends to select a longer trajectory because a longer trajectory can avoid obstacles at a greater distance, improving comfort; s(t1) is the longitudinal position of the final state sampling of the longitudinal motion trajectory, and s(target) is the target longitudinal position. When following another vehicle, it is the value obtained by subtracting the longitudinal position of the preceding vehicle. In the parking scenario, it is the longitudinal position of the parking point, (s(t1) - s(target)). 2 Used to describe the final state position cost of the longitudinal motion trajectory; W is the comfort weighting coefficient. t For trajectory efficiency weighting coefficients; For trajectory behavior weighting coefficients; The final state position cost weight coefficient of the trajectory.

[0146] 2. Solving the lateral trajectory

[0147] In each trajectory planning cycle, after the longitudinal trajectory planning is completed, each longitudinal trajectory has a corresponding time consumption and trajectory length. The lateral trajectory is then solved based on the longitudinal trajectory length. The final state constraints of the lateral trajectory include lateral position, velocity, and acceleration. Combining the vehicle's current lateral position, velocity, and acceleration, there are a total of six constraints. The relationship between the corresponding trajectory length S and lateral displacement L is solved using a fifth-order polynomial function.

[0148] The corresponding trajectory duration S = [s0, s1], and the vehicle's lateral state corresponding to the initial longitudinal position s0 is: The lateral state of the vehicle at the end time s1 is The fifth-degree polynomial relation is as follows, where s∈S:

[0149] Taking the second derivative of this expression with respect to s

[0150]

[0151] After calculating the initial and final states L0 and L1 using the above formulas, they can be expressed as matrix equations as follows:

[0152]

[0153] In the formula M l C is a constant matrix. l Let be the polynomial coefficient matrix. Let the initial and final state matrix be... The polynomial coefficient matrix C can be obtained from the above matrix equation. l =M l -1 Pl After each replanning trigger, the coefficient matrix C of the matrix equation needs to be solved again. l The longitudinal motion trajectory within the corresponding trajectory length (S) can be described as follows:

[0154]

[0155] The lateral trajectory l(s) is given by the third derivative with respect to the trajectory length s. This describes the third derivative of the lateral displacement with respect to the longitudinal displacement. Its magnitude can be used to describe the severity of the lateral displacement change of the vehicle's position during autonomous vehicle movement, and is therefore used for trajectory comfort evaluation. The overall comfort value within the corresponding trajectory length S is represented by J. S (l(s)) represents the following formula:

[0156] The quality assessment cost function for lateral motion trajectory planning can be abstracted into the following function:

[0157]

[0158] In the formula, g(s) is the trajectory length corresponding to the lateral motion trajectory, i.e., g(s) = s, which is used to describe the trajectory driving efficiency; L(i) is the longitudinal position of the lateral motion trajectory along the final state sampling, which is used to describe the trajectory behavior, L(i) = l(s1), which is used to determine the degree of deviation of the sampling point from the reference center line. W is the comfort weighting coefficient. s This is the weighting coefficient for trajectory driving efficiency; For trajectory behavior weighting coefficients.

[0159] 3. Horizontal and vertical trajectory coupling

[0160] Each planning cycle will generate a series of trajectories, with each trajectory corresponding to a travel time of Ti and T. i =[T min ,ΔT,T max ], T min T max These represent the shortest and longest trajectory durations, respectively, with a sampling interval of ΔT. After sampling the end position s1 of the longitudinal trajectory, several sets of longitudinal motion trajectories can be obtained:

[0161] S bundle =Ω(T) i ,s0,s1),T i =[T min ,ΔT,T max ]

[0162] In the formula, Ω(t,s0,s1) is the set of all longitudinal motion trajectories, corresponding to the duration T. i The longitudinal trajectory motion state within is S i =Ω(T) i ,s0,s1). Lateral sampling is performed based on the longitudinal motion trajectory, and each longitudinal motion trajectory S i Samples are taken from several lateral end positions, and the set of lateral motion trajectories is solved:

[0163] L bundle =Ψ(S i ,l0,l1),S i ∈S boundle ,

[0164] In the formula, Ψ(S) i (l0, l1) represents the corresponding longitudinal motion trajectory S i The set of lateral motion trajectories is used to achieve lateral and longitudinal trajectory coupling. The total motion trajectory obtained by coupling the lateral and longitudinal trajectories is J. total The calculation is as follows:

[0165] J total =W s cost s +W l cost l .

[0166] S1043, calculate the cost of driving corresponding to the several candidate parking trajectories based on the prefabricated performance evaluation model; wherein, the cost includes comfort cost, safety cost, and driving efficiency cost.

[0167] S1044, Select the optimal value from the value and use the candidate parking trajectory corresponding to the optimal value as the target parking trajectory.

[0168] S105, when the target vehicle arrives at the target parking space according to the target parking trajectory, select a suitable parking space corresponding to one of the other parking trajectories as the space required for parallel parking based on the position of the target parking trajectory in the time sequence, so that the target vehicle stops upright in the target parking space.

[0169] Specifically, after driving to the end of the parking trajectory, the next sampling point after the end of the trajectory is used as the new temporary parking point. If the vehicle is not parked straight, a parallel parking operation is performed to adjust the vehicle's posture and reach the new parking point.

[0170] Furthermore, step S105 of this embodiment specifically includes:

[0171] S1051, when the target vehicle arrives at the target parking space according to the target parking trajectory, determine whether the parking space of the target parking trajectory is at the first position of the time sequence.

[0172] S1052, if so, then the parking space of the adjacent subsequent parking trajectory is used as the space required for parallel parking, so that the target vehicle can park upright by using the parking space of the subsequent parking trajectory, so that the target vehicle is upright in the target parking space.

[0173] Specifically, in combination Figure 4 As shown, if the end point of the parking trajectory is point A, then the point before it is used as a temporary parking point to adjust the vehicle's attitude; if the end point of the parking trajectory is point B, then point A is used as a temporary parking point. Before planning the parking trajectory, it will be checked whether there is a risk of collision with the obstacle behind when the vehicle is at point A. If there is, then point C is selected as a temporary parking point and the parking trajectory is planned.

[0174] S106, when the target vehicle is in the target parking space, a decision is made based on whether the target vehicle is causing road congestion or whether to continue parking in the target parking space.

[0175] Specifically, such as Figure 5 As shown, the presence of static obstacles around the current location causes lane congestion. However, the target vehicle still needs to temporarily stop and wait near point P for a certain period of time. In order not to affect the traffic efficiency of vehicles behind, a yielding decision needs to be made.

[0176] Furthermore, step S106 of this embodiment specifically includes:

[0177] S1061, when the target vehicle is in the target parking space, determine whether the target vehicle is causing road congestion.

[0178] S1062, if so, then reset the parking trajectory planning of the new parking space near the target parking space so that the target vehicle is parked correctly in the new parking space.

[0179] Specifically, in combination Figure 5As shown, a new temporary parking point is found by using point sampling. The current temporary parking point P is the target vehicle's coordinate point E (s0, d0), and the coordinates of the static obstacle center point F are (s2, d2). Similarly, point sampling is performed on the left side, but the longitudinal position s2 and the lateral position d2 of point F are used as reference values. The longitudinal positions of each final state sampling point are [s2+3.0, s2+5.0, s2+7.0, s2+9.0]. For safety reasons, the first sampling point is modified to s2+3.0. The longitudinal sampling interval remains Δs = 2.0m, and the lateral position of the final state sampling point is d2. Similarly, the initial and final states of the horizontal and vertical axes are connected by a fifth-order polynomial curve, and then the horizontal and vertical trajectories are coupled to plan four corresponding parking trajectories, including vehicle path and speed information.

[0180] In summary, the automatic parking function is activated and obstacle information within the perception range is acquired. Based on the obstacle information, it is determined whether the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is obstructed. If so, several parking space sampling points are obtained using a point-sampling method based on the target vehicle's current sampling point, and several parking trajectories are planned using a fifth-order polynomial curve. One parking trajectory is selected as the target parking trajectory through a pre-built performance evaluation model. Based on the position of the target parking trajectory in the time series, suitable parking spaces corresponding to several other parking trajectories are selected as the space required for parallel parking. Through the above steps, the path planning for stopping at a given temporary parking point and timely changing to a new temporary parking point can be quickly completed based on the current surrounding environment information of the temporary parking point, improving the traffic efficiency of the park.

[0181] Example 2

[0182] This embodiment provides a structural block diagram of a system corresponding to the method described in Embodiment 1. Figure 6 This is a structural block diagram of the intelligent temporary parking system according to this embodiment, as follows: Figure 6 As shown, the system includes:

[0183] The acquisition module 10 is used to activate the automatic parking function and acquire obstacle information within the perception range when the target vehicle enters the perception range of the designated temporary parking space; wherein, the obstacle information includes dynamic obstacles and static obstacles.

[0184] The prediction module 20 is used to determine whether the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is blocked based on the obstacle information.

[0185] The planning module 30 is used to obtain several parking space sampling points by means of a point-scattering sampling method based on the current sampling point of the target vehicle if the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is blocked, and to plan several parking trajectories using a fifth-order polynomial curve.

[0186] Evaluation module 40 is used to select one parking trajectory from the plurality of parking trajectories as the target parking trajectory through a prefabricated performance evaluation model; wherein, the prefabricated performance evaluation model includes comfort, safety and efficiency.

[0187] The adjustment module 50 is used to select, based on the position of the target parking trajectory in the time series, suitable parking spaces corresponding to other parking trajectories as the space required for parallel parking when the target vehicle arrives at the target parking space according to the target parking trajectory, so that the target vehicle stops correctly at the target parking space.

[0188] The congestion mitigation module 60 is used to make a decision on whether to find a new parking space or continue parking in the target parking space when the target vehicle is in the target parking space, based on whether the target vehicle is causing road congestion.

[0189] Furthermore, the acquisition module 10 includes:

[0190] The activation unit 11 is used to activate the automatic parking function when the target vehicle enters the sensing range of the designated temporary parking space.

[0191] The acquisition unit 12 is used to acquire real-time environmental information of the target vehicle's surroundings according to the image acquisition device; wherein the corresponding area of ​​the perception range belongs to a subset of the corresponding area of ​​the real-time surroundings.

[0192] The acquisition unit 13 is used to screen the environmental information to obtain obstacle information of dynamic and static obstacles within the perception range.

[0193] Furthermore, the planning module 30 includes:

[0194] The determining unit 31 is used to determine the current sampling point of the target vehicle in the current state based on the Frenet coordinate system, with the road centerline as the reference line, the vertical coordinate as the displacement along the road direction, and the horizontal coordinate as the lateral displacement perpendicular to the road centerline; wherein, the current sampling point includes the initial lateral position and the initial longitudinal position.

[0195] Sampling unit 32 is used to obtain the end lateral position and end longitudinal position of several feasible berth states respectively by means of a point-sampling method based on the current sampling point.

[0196] Fitting unit 33 is used to fit the initial lateral position and the initial longitudinal position to several of the end lateral positions and end longitudinal positions respectively through a fifth-order polynomial curve to generate a lateral trajectory and a longitudinal trajectory.

[0197] The coupling unit 34 is used to couple the lateral trajectory and the longitudinal trajectory according to the time sequence to generate several parking trajectories based on the Frenet coordinate system.

[0198] Furthermore, the evaluation module 40 includes:

[0199] The conversion unit 41 is used to convert the coordinate system of the plurality of parking trajectories to generate the plurality of candidate parking trajectories based on the global coordinate system.

[0200] Construction unit 42 is used to construct a prefabricated performance evaluation model by building a loss function based on evaluation indicators including at least comfort, safety and efficiency.

[0201] The calculation unit 43 is used to calculate the cost of driving corresponding to the plurality of candidate parking trajectories based on the prefabricated performance evaluation model; wherein the cost includes comfort cost, safety cost, and driving efficiency cost.

[0202] The selection unit 44 is used to select the optimal value from the value and take the candidate parking trajectory corresponding to the optimal value as the target parking trajectory.

[0203] Furthermore, the adjustment module 50 includes:

[0204] The first judgment unit 51 is used to determine whether the parking space of the target parking trajectory is at the first position of the time sequence when the target vehicle travels to the target parking space according to the target parking trajectory.

[0205] The first parking alignment unit 52 is used to use the parking space of the next adjacent parking trajectory as the space required for parallel parking if the parking space of the target parking trajectory is at the first position of the time sequence, so that the target vehicle can park and align itself by using the parking space of the next parking trajectory, so that the target vehicle is parked and aligned at the target parking space.

[0206] Furthermore, the congestion mitigation module 60 includes:

[0207] The second judgment unit 61 determines whether the target vehicle is causing road congestion when the target vehicle is in the target parking space.

[0208] The reset unit 62 is used to reset the parking trajectory planning of a new parking space near the target parking space if the target vehicle causes road congestion, so that the target vehicle can be parked correctly in the new parking space.

[0209] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0210] Example 3

[0211] Specifically, this embodiment provides a flowchart of an intelligent temporary parking method. The difference between this embodiment and the method in Embodiment 1 lies in:

[0212] 1. The specific process steps of step S205 in this embodiment are different from those of step S105 in embodiment 1, such as... Figure 7 As shown, the specific steps of step S205 in this embodiment include:

[0213] S2051, when the target vehicle arrives at the target parking space according to the target parking trajectory, determine whether the parking space of the target parking trajectory is at the first position of the time sequence.

[0214] S2052, if not, then the parking space of the adjacent previous parking trajectory is used as the space required for parallel parking, so that the target vehicle can park upright using the parking space of the previous parking trajectory, so that the target vehicle is parked upright in the target parking space.

[0215] 2. The specific process steps of step S206 in this embodiment are different from those of step S106 in embodiment 1, such as... Figure 8 As shown, the specific steps of step S206 in this embodiment include:

[0216] S2061, when the target vehicle is in the target parking space, determine whether the target vehicle is causing road congestion;

[0217] S2062, if not, the target vehicle continues to park at the target parking space.

[0218] Example 4

[0219] This embodiment provides a structural block diagram of a system corresponding to the method described in Embodiment 3. Figure 9 This is a structural block diagram of the intelligent temporary parking system according to this embodiment, as follows: Figure 9 As shown, the system differs from the system in Example 2 in that:

[0220] 1. The adjustment module of the system in this embodiment has a different function from the adjustment module of the system in embodiment 2: The adjustment module 50 in this embodiment includes:

[0221] The first judgment unit 51 is used to determine whether the parking space of the target parking trajectory is at the first position of the time sequence when the target vehicle travels to the target parking space according to the target parking trajectory.

[0222] The second parking alignment unit 53 is used to use the parking space of the adjacent previous parking trajectory as the space required for parallel parking if the parking space of the target parking trajectory is not in the first position of the time sequence, so that the target vehicle can park and align itself using the parking space of the previous parking trajectory, so that the target vehicle is parked in the target parking space.

[0223] 2. The specific functions of the congestion mitigation module in this embodiment differ from those in the system of embodiment 2: The congestion mitigation module 60 in this embodiment includes:

[0224] The second judgment unit 61 is used to determine whether the target vehicle causes road congestion when the target vehicle is in the target parking space.

[0225] The continuation unit 63 is used to ensure that if the target vehicle does not cause road obstruction, the target vehicle continues to park in the target parking space.

[0226] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0227] Example 5

[0228] Specifically, Figure 10 The diagram shown is a flowchart of an intelligent temporary parking method provided in this embodiment.

[0229] like Figure 10 As shown, the intelligent temporary parking method of this embodiment includes the following steps:

[0230] S301, when the target vehicle enters the sensing range of the designated temporary parking space, the automatic parking function is activated and obstacle information within the sensing range is obtained; wherein, the obstacle information includes dynamic obstacles and static obstacles.

[0231] S302, determine whether the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is blocked based on the obstacle information.

[0232] S303, if not, proceed directly to the designated temporary parking space according to the preset parking trajectory.

[0233] Example 6

[0234] This embodiment provides a structural block diagram of a system corresponding to the method described in Embodiment 5. Figure 11 This is a structural block diagram of the intelligent temporary parking system according to this embodiment, as follows: Figure 11 As shown, the system includes:

[0235] The acquisition module 10 is used to activate the automatic parking function and acquire obstacle information within the perception range when the target vehicle enters the perception range of the designated temporary parking space; wherein, the obstacle information includes dynamic obstacles and static obstacles.

[0236] The prediction module 20 is used to determine, based on the obstacle information, whether the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is obstructed.

[0237] The path module 70 is used to proceed to the designated temporary parking space according to the preset parking trajectory if the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is not obstructed.

[0238] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0239] Example 7

[0240] Combination Figure 12 The intelligent temporary parking method described can be implemented by vehicles. Figure 12 This is a schematic diagram of the hardware structure of a vehicle according to this embodiment.

[0241] The vehicle may include a processor 81 and a memory 82 storing computer program instructions.

[0242] Specifically, the processor 81 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement this application.

[0243] The memory 82 may include a mass storage device for data or instructions. For example, and not limitingly, the memory 82 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), flash memory, an optical disk drive, a magneto-optical disk drive, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 82 may include removable or non-removable (or fixed) media. Where appropriate, the memory 82 may be internal or external to a data processing device. In a particular embodiment, the memory 82 is non-volatile memory. In a particular embodiment, the memory 82 includes read-only memory (ROM) and random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), an electrically alterable read-only memory (EAROM), or flash memory, or a combination of two or more of these. Where appropriate, the RAM can be Static Random-Access Memory (SRAM) or Dynamic Random-Access Memory (DRAM). DRAM can be Fast Page Mode Dynamic Random-Access Memory (FPMDRAM), Extended Data Out Dynamic Random-Access Memory (EDODRAM), Synchronous Dynamic Random-Access Memory (SDRAM), etc.

[0244] The memory 82 can be used to store or cache various data files that need to be processed and / or communicated, as well as possible computer program instructions executed by the processor 81.

[0245] The processor 81 reads and executes the computer program instructions stored in the memory 82 to implement the intelligent temporary parking methods of embodiments 1, 3, and 5 described above.

[0246] In some embodiments, the vehicle may further include a communication interface 83 and a bus 80. For example, Figure 12 As shown, the processor 81, memory 82, and communication interface 83 are connected through bus 80 and complete communication with each other.

[0247] The communication interface 83 is used to enable communication between the various modules, devices, units, and / or equipment in this application. The communication interface 83 can also enable data communication with other components such as external devices, image / data acquisition devices, databases, external storage, and image / data processing workstations.

[0248] Bus 80 includes hardware, software, or both, that couples components of a device together. Bus 80 includes, but is not limited to, at least one of the following: data bus, address bus, control bus, expansion bus, and local bus. For example, and not as a limitation, bus 80 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 80 may include one or more buses. Although this application describes and illustrates a specific bus, this application considers any suitable bus or interconnection.

[0249] The vehicle can access the intelligent temporary parking system and execute the intelligent temporary parking methods in embodiments 1, 3, and 5 of this document.

[0250] In addition, in conjunction with the intelligent temporary parking methods in embodiments 1, 3, and 5 above, this application can provide a storage medium for implementation. This storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement the intelligent temporary parking methods of embodiments 1, 3, and 5 above.

[0251] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0252] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent temporary parking method, characterized in that, include: When the target vehicle enters the sensing range of the designated temporary parking space, the automatic parking function is activated and obstacle information within the sensing range is acquired; wherein, the obstacle information includes dynamic obstacles and static obstacles; Based on the obstacle information, determine whether the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is obstructed; If so, then based on the current sampling point of the target vehicle, several parking space sampling points are obtained by the point-scattering sampling method, and several parking trajectories are planned using a fifth-order polynomial curve. One parking trajectory is selected from the aforementioned parking trajectories using a prefabricated performance evaluation model as the target parking trajectory; wherein, the prefabricated performance evaluation model includes comfort, safety, and efficiency. When the target vehicle arrives at the target parking space according to the target parking trajectory, a suitable parking space corresponding to one of the other parking trajectories is selected as the space required for parallel parking based on the position of the target parking trajectory in the time sequence, so that the target vehicle stops correctly in the target parking space; When the target vehicle enters the sensing range of the designated temporary parking space, the automatic parking function is activated and obstacle information within the sensing range is acquired; wherein, the obstacle information includes both dynamic and static obstacles, the specific steps include: The automatic parking function is activated when the target vehicle enters the sensing range of the designated temporary parking space. The image acquisition device collects real-time environmental information about the target vehicle's surroundings; wherein the area corresponding to the perception range is a subset of the corresponding area of ​​the real-time surroundings. The environmental information is screened to obtain obstacle information for both dynamic and static obstacles within the perception range.

2. The intelligent temporary parking method according to claim 1, characterized in that, The specific steps of obtaining several parking space sampling points based on the current sampling point of the target vehicle using a point-scattering sampling method, and planning several parking trajectories using a fifth-order polynomial curve include: Based on the Frenet coordinate system, with the road centerline as the reference line, the vertical axis represents the displacement along the road direction, and the horizontal axis represents the lateral displacement perpendicular to the road centerline, the current sampling point of the target vehicle in its current state is determined; wherein, the current sampling point includes the initial lateral position and the initial longitudinal position. Based on the current sampling point, the end lateral position and end longitudinal position of several feasible berth states are obtained by the point sampling method. The initial lateral and initial longitudinal positions are fitted with a fifth-order polynomial curve to several of the end lateral and end longitudinal positions to generate lateral and longitudinal trajectories, respectively. The lateral and longitudinal trajectories are coupled according to a time series to generate several parking trajectories based on the Frenet coordinate system.

3. The intelligent temporary parking method according to claim 1, characterized in that, The step of selecting a parking trajectory from the plurality of parking trajectories as the target parking trajectory using a prefabricated performance evaluation model, wherein the prefabricated performance evaluation model includes comfort, safety, and efficiency, specifically includes: The coordinate systems of the aforementioned parking trajectories are transformed to generate several candidate parking trajectories based on the global coordinate system. A loss function is constructed based on evaluation indicators that include at least comfort, safety, and efficiency to build a prefabricated performance evaluation model; The cost of driving for the candidate parking trajectories is calculated based on the prefabricated performance evaluation model; wherein the cost includes comfort cost, safety cost, and driving efficiency cost. The optimal parking value is selected from the estimated parking values, and the candidate parking trajectory corresponding to the optimal parking value is taken as the target parking trajectory.

4. The intelligent temporary parking method according to claim 1 or 3, characterized in that, The prefabricated performance evaluation model is as follows: ； In the formula, This represents the value weight of the longitudinal motion trajectory. This represents the value weight of the lateral movement trajectory. This represents the cost function for evaluating the quality of longitudinal motion trajectory planning. This represents the cost function for quality evaluation of lateral motion trajectory planning.

5. The intelligent temporary parking method according to claim 1, characterized in that, The step of selecting suitable parking spaces corresponding to other parking trajectories as the space required for parallel parking, based on the position of the target parking trajectory in the time series, when the target vehicle arrives at the target parking space according to the target parking trajectory, specifically includes: When the target vehicle arrives at the target parking space according to the target parking trajectory, it is determined whether the parking space of the target parking trajectory is at the first position of the time sequence; If so, the parking space of the adjacent subsequent parking trajectory is used as the space required for parallel parking, so that the target vehicle can park upright by using the parking space of the subsequent parking trajectory, so that the target vehicle is parked upright in the target parking space. If not, then the space required for parallel parking is the adjacent parking space of the previous parking trajectory, so that the target vehicle can park itself upright using the parking space of the previous parking trajectory, so that the target vehicle is parked upright in the target parking space.

6. The intelligent temporary parking method according to claim 1, characterized in that, After the step of selecting suitable parking spaces corresponding to other parking trajectories as the space required for parallel parking, based on the position of the target parking trajectory in the time series when the target vehicle arrives at the target parking space according to the target parking trajectory, so that the target vehicle is parked correctly in the target parking space, the method further includes: When the target vehicle is in the target parking space, a decision is made based on whether the target vehicle is causing road congestion or whether to continue parking in the target parking space.

7. An intelligent temporary parking system, characterized in that, include: The acquisition module is used to activate the automatic parking function and acquire obstacle information within the perception range when the target vehicle enters the perception range of the designated temporary parking space; wherein, the obstacle information includes dynamic obstacles and static obstacles; The prediction module is used to determine whether the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is blocked based on the obstacle information. The planning module is used to obtain several parking space sampling points based on the current sampling points of the target vehicle by using a point-scattering sampling method if the path of the target vehicle to the designated temporary parking space according to the preset parking trajectory is blocked, and to plan several parking trajectories using a fifth-order polynomial curve. The evaluation module is used to select one parking trajectory from the plurality of parking trajectories as the target parking trajectory through a prefabricated performance evaluation model; wherein, the prefabricated performance evaluation model includes comfort, safety, and efficiency. The adjustment module is used to select, based on the position of the target parking trajectory in the time sequence, suitable parking spaces corresponding to other parking trajectories as the space required for parallel parking when the target vehicle arrives at the target parking space according to the target parking trajectory, so that the target vehicle stops correctly at the target parking space; The acquisition module includes: The activation unit is used to activate the automatic parking function when the target vehicle enters the sensing range of the designated temporary parking space; The acquisition unit is used to acquire real-time environmental information of the target vehicle's surroundings based on the image acquisition device; wherein the corresponding area of ​​the perception range belongs to a subset of the corresponding area of ​​the real-time surroundings. The acquisition unit is used to screen the environmental information to obtain obstacle information of dynamic and static obstacles within the perception range.

8. A vehicle comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the intelligent temporary parking method as described in any one of claims 1 to 6.

9. A storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the intelligent temporary parking method as described in any one of claims 1 to 6.

Images