Automatic driving parking method and device based on path planning

Abstract

The application relates to the technical field of automobiles and provides an automatic driving parking method and device based on path planning, an electronic device and a storage medium. The method comprises the following steps: determining a target parking space for a vehicle and loading a garage map; according to environmental data of the garage map, taking the vehicle as a mass point, and generating a driving path from an initial position to a terminal position; predicting an obstacle according to external environmental information; when the obstacle is a pedestrian, performing collision prediction on the vehicle according to a motion trajectory and a safety distance; when the obstacle is another vehicle, performing collision prediction according to a driving path of the current vehicle and a driving speed and direction of the other vehicle; performing vehicle control according to a collision prediction result; and controlling the vehicle to reach the target parking space according to a real-time position after vehicle control. The application solves the problem in the prior art that accurate path planning cannot be performed based on a recommended parking space to realize automatic parking.

Description

Technical Field

[0001] This application relates to the field of automotive safety technology, and in particular to an autonomous driving parking method, device, electronic device, and readable storage medium based on path planning. Background Technology

[0002] With the rapid development of new energy vehicles, users are increasingly demanding autonomous driving functions, especially when parking in residential parking lots. Because drivers are often more relaxed as they approach home, and the time spent finding a suitable parking space can lead to a loss of focus, the frequency of minor collisions and scrapes in residential parking lots is increasing.

[0003] Currently, autonomous driving parking methods can be used to park in available spaces to reduce the occurrence of minor collisions. In this process, dynamic collision avoidance in autonomous driving often employs methods such as rolling window methods, neural network methods, and fuzzy logic algorithms. However, these dynamic collision avoidance methods often treat vehicles and obstacles as point masses, or only consider the vehicle's safe distance, simplifying the vehicle into a sphere. This simplification may lead to inaccurate collision predictions, causing the evaluated path to fail to reach its destination.

[0004] It is evident that the relevant technologies suffer from the inability to perform accurate route planning based on recommended parking spaces to achieve automatic parking. Summary of the Invention

[0005] In view of this, embodiments of this application provide an autonomous driving parking method, device, electronic device, and readable storage medium based on path planning, to solve the problem in related technologies that it is impossible to achieve automatic parking by accurately planning paths based on recommended parking spaces.

[0006] A first aspect of this application provides an autonomous driving parking method based on path planning, comprising:

[0007] Determine the target parking space for the vehicle, and based on the vehicle's initial position and the target parking space's corresponding endpoint position, perform a working environment initialization operation on the vehicle and load it into the garage map.

[0008] Based on the environmental data of the garage map, the vehicle is treated as a point mass, and a predetermined particle swarm optimization algorithm is used to perform global path planning for the vehicle to generate a driving path from the initial position to the final position.

[0009] During the process of controlling the vehicle to travel along the driving path, sensors are used to perceive the external environment information around the vehicle, and obstacles are predicted based on the external environment information, wherein the obstacles include pedestrians and / or other vehicles.

[0010] When the obstacle is a pedestrian, the pedestrian's trajectory is predicted, the safe distance between the vehicle and the pedestrian is determined, and a collision prediction is made for the vehicle based on the trajectory and the safe distance. The vehicle is then controlled based on the collision prediction result.

[0011] When the obstacle is another vehicle, the driving speed and direction of the other vehicle are obtained, collision prediction is performed based on the current vehicle's driving path and the driving speed and direction of the other vehicle, and vehicle control is performed based on the collision prediction result.

[0012] Based on the real-time location of the vehicle after control, the particle swarm optimization algorithm is used to re-plan the global path for the vehicle until the vehicle reaches the target parking space.

[0013] A second aspect of this application provides an automated parking device based on path planning, comprising:

[0014] The determination module is configured to determine the target parking space for the vehicle, perform a working environment initialization operation on the vehicle based on the initial position of the vehicle and the endpoint position corresponding to the target parking space, and load the garage map.

[0015] The path planning module is configured to, based on the environmental data of the garage map, treat the vehicle as a point mass and use a predetermined particle swarm optimization algorithm to perform global path planning for the vehicle, so as to generate a driving path from the initial position to the final position.

[0016] The prediction module is configured to, while controlling the vehicle to travel along the driving path, use sensors to perceive external environmental information around the vehicle and predict obstacles based on the external environmental information, wherein the obstacles include pedestrians and / or other vehicles.

[0017] The first control module is configured to predict the movement trajectory of the pedestrian when the obstacle is a pedestrian, determine the safe distance between the vehicle and the pedestrian, perform collision prediction on the vehicle based on the movement trajectory and the safe distance, and control the vehicle based on the collision prediction result.

[0018] The second control module is configured to, when the obstacle is another vehicle, acquire the driving speed and direction of the other vehicle, perform collision prediction based on the current vehicle's driving path and the driving speed and direction of the other vehicle, and perform vehicle control based on the collision prediction result.

[0019] The third control module is configured to re-plan the global path of the vehicle using the particle swarm optimization algorithm based on the real-time position of the vehicle after control, until the vehicle reaches the target parking space.

[0020] A third aspect of this application provides an electronic device, 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 steps of the above-described method.

[0021] A fourth aspect of this application provides a readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0022] The beneficial effects of the embodiments of this application compared with the prior art are:

[0023] By identifying a target parking space and loading it into a garage map, and using environmental data from the garage map, a driving path from the initial position to the destination position is generated, treating the vehicle as a point mass. During the vehicle's movement along this path, obstacles are predicted based on external environmental information. When the obstacle is a pedestrian, collision prediction is performed based on the pedestrian's trajectory and safe distance. When the obstacle is another vehicle, collision prediction is performed based on the current vehicle's driving path and the speed and direction of other vehicles. Vehicle control is then implemented based on the collision prediction results. Finally, based on the vehicle's real-time position after control, the vehicle is guided to the target parking space. This allows the vehicle to plan a parking path based on the recommended target parking space and autonomously drive to that space according to the planned path. Simultaneously, during the journey, the vehicle replans its path based on detection information. This effective combination of optimization and feedback mechanisms reduces computational load, ensures global convergence, and helps the vehicle avoid scrape accidents during parking, improving driving safety and user experience. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application, 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating an autonomous driving parking method based on path planning provided in an embodiment of this application;

[0026] Figure 2 This is a flowchart illustrating a particle swarm optimization algorithm provided in an embodiment of this application;

[0027] Figure 3 This is a schematic diagram of collision prediction when the obstacle is a pedestrian, provided in an embodiment of this application;

[0028] Figure 4 This is a schematic diagram of collision prediction that treats a vehicle as a cuboid, provided in an embodiment of this application;

[0029] Figure 5 This is a schematic diagram of the structure of an automated parking device based on path planning provided in an embodiment of this application;

[0030] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0031] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application can also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0032] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0033] Furthermore, it should be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes that element.

[0034] The following describes in detail, with reference to the accompanying drawings, an autonomous driving parking method, apparatus, electronic device, and readable storage medium based on path planning, according to embodiments of this application.

[0035] Figure 1 This is a flowchart illustrating an autonomous driving parking method based on path planning, as provided in an embodiment of this application. Figure 1 The path planning-based autonomous driving parking method can be executed by the vehicle controller. For example... Figure 1 As shown, the method includes:

[0036] S101, determine the target parking space for the vehicle, perform working environment initialization operation on the vehicle based on the initial position of the vehicle and the end position corresponding to the target parking space, and load the garage map.

[0037] The target parking space is the intended location for the vehicle to be parked, as specified by the driver or the system. It can be a parking space or parking area for the safe parking of the vehicle.

[0038] The initial position of a vehicle refers to its position when the parking process begins, while the final position is the vehicle's target position during the parking process, i.e., the location of the target parking space.

[0039] Initializing the vehicle's working environment refers to the preparatory operations performed on the vehicle before the parking process begins. After the driver or parking system determines the target parking space, some operations need to be performed to put the vehicle in a suitable parking state.

[0040] Specific initialization operations for the working environment may include starting the vehicle, adjusting the seat and rearview mirrors, setting the parking dashboard, and adjusting the vehicle's posture. These operations enable the vehicle to better adapt to the environmental conditions during parking and ensure that the vehicle is ready to be driven into the target parking space smoothly.

[0041] A garage map is a system used to record and represent the distribution of parking spaces in a parking lot. It can be a two-dimensional or three-dimensional map, displaying information such as the location, type, and availability of different parking spaces.

[0042] By performing appropriate initialization operations on the working environment, parking efficiency and safety can be improved. These operations can be adjusted according to the driver's personal preferences and the requirements of the parking system to ensure that the vehicle is in optimal condition for a smooth parking process.

[0043] In this way, by identifying the target parking space, initializing the working environment for the vehicle, and loading the garage map, a more efficient parking process can be achieved. By accurately locating the target position, accurately executing the working environment initialization operation, and effectively utilizing the garage map, the efficiency and convenience of the parking system can be improved, reducing the parking inconvenience caused by traditional parking methods.

[0044] S102, based on the environmental data of the garage map, the vehicle is treated as a point mass, and a predetermined particle swarm optimization algorithm is used to perform global path planning for the vehicle to generate a driving path from the initial position to the destination position.

[0045] The environmental data of the garage map includes obstacles throughout the garage, driving roads available for vehicles, and the number and distribution of parking spaces.

[0046] In addition, garage maps may provide other environmental features, such as identifying parking areas, entrances and exits, emergency exits, emergency facilities, and navigation signs (such as arrows and signs) to guide vehicles to their destination.

[0047] It should be noted that the environmental data for the garage map can be obtained through LiDAR and cameras installed on the vehicle, or through other sensors with the same data collection function; no specific limitation is made here.

[0048] Treating a vehicle as a point mass is a simplified model for analyzing vehicle motion. In this model, the vehicle is idealized as a point without size or rotation, possessing only mass and position attributes. By treating the vehicle as a point mass, the point mass model can be used to study the vehicle's motion, mechanical behavior, and interactions with other objects. This simplified model reduces complexity, simplifies the problem, and helps us more easily apply physical laws and mathematical models to predict and analyze vehicle motion.

[0049] It should be noted that in some cases, such as when the vehicle is turning or the suspension system is affected, the actual shape and rotation of the vehicle may have a certain impact on the actual motion of the vehicle. Therefore, in some more precise and specific cases, more complex models and analysis methods may be needed to describe the motion of the vehicle.

[0050] Particle Swarm Optimization (PSO) is a collaborative optimization algorithm that searches and optimizes by continuously updating the positions and velocities of particles. Each particle remembers its own found optimal solution (local optimum), while the entire particle swarm shares the global optimum. During iteration, particles adjust their movement strategies by comparing their current position with their historical optimal solution and the global optimum, and based on their current velocity and position.

[0051] Thus, this step, based on the environmental data of the garage map, treats the vehicles as point masses and uses a pre-defined particle swarm optimization (PSO) algorithm to perform global path planning for the vehicles, generating driving paths. Through information exchange and sharing of the global optimal solution among the collaborative particles, the entire garage map is explored to find better path solutions. Furthermore, compared to other optimization algorithms, PSO is simple to implement, insensitive to initial value selection, and adaptable to different problems and environments. Its parallelism and interpretability also provide convenience and reliability for driving path planning.

[0052] S103, during the process of controlling the vehicle to travel along the driving path, the vehicle uses sensors to perceive the external environment information around the vehicle and predicts obstacles based on the external environment information, including pedestrians and / or other vehicles.

[0053] During the process of controlling the vehicle to travel along the planned driving path, sensors are used to perceive the external environmental information around the vehicle and predict obstacles. Sensors can include radar, cameras, lidar, etc. By perceiving the external environmental information, the system can predict and identify obstacles around the vehicle, which can be pedestrians, other vehicles, or other moving or stationary objects.

[0054] In addition, controlling the vehicle to automatically drive along the driving path to the target parking space can also avoid scrape and collision accidents caused by human error, thereby improving driving safety.

[0055] Pedestrians are individuals who walk on or around roads, including pedestrians, cyclists, scooter riders, etc. The ability to predict pedestrian situations helps vehicles avoid potential dangers caused by pedestrians, such as collision avoidance or proactive response strategies.

[0056] Other vehicles refer to other motor vehicles around the vehicle during its journey, such as cars and motorcycles. Predicting other vehicles can help the vehicle formulate appropriate driving strategies to avoid traffic accidents and congestion.

[0057] In this way, by using sensors to perceive information about the external environment around the vehicle and predict obstacles (including pedestrians and other vehicles), the vehicle can take timely measures to avoid collisions with pedestrians or other vehicles, thereby improving driving safety. In addition, accurate perception and prediction of the external environment helps the intelligent driving system make more accurate and efficient decisions and controls, thus achieving the goal of autonomous driving.

[0058] S104 When the obstacle is a pedestrian, the pedestrian's trajectory is predicted, the safe distance between the vehicle and the pedestrian is determined, and a collision prediction is made for the vehicle based on the trajectory and the safe distance. The vehicle is then controlled based on the collision prediction results.

[0059] When the obstacle is a pedestrian, the pedestrian's trajectory can be predicted by analyzing information such as the pedestrian's position, speed, and acceleration to infer the pedestrian's possible future movement path.

[0060] Determine a safe distance between vehicles and pedestrians to ensure sufficient space to avoid collisions. This safe distance can be calculated or set based on pedestrian speed, vehicle braking capacity, and system design standards.

[0061] Collision prediction can be performed by predicting the trajectory of the pedestrian and determining the safe distance. Based on the predicted trajectory of the pedestrian and the motion state of the vehicle, it can be determined whether there is a potential collision risk. If the prediction results indicate that there is a possible collision, corresponding safety measures will be taken. If there is a collision risk, the vehicle control system can take measures such as emergency braking, lane changing or deceleration to maintain a safe distance and avoid collision with the pedestrian.

[0062] In this way, by predicting the movement trajectories of pedestrians and vehicles, determining safe distances, and implementing collision prediction and vehicle control, collisions between pedestrians and vehicles can be effectively prevented, the safety between pedestrians and vehicles can be improved, the risk of collisions can be reduced, and the safety and reliability of driving can be ensured.

[0063] S105: When the obstacle is another vehicle, obtain the driving speed and direction of the other vehicle, perform collision prediction based on the current vehicle's driving path and the driving speed and direction of the other vehicle, and perform vehicle control based on the collision prediction results.

[0064] When the obstacle is another vehicle, real-time speed and direction information of that vehicle is acquired via sensors or communication systems to understand its current motion state. Combining the current vehicle's path with the motion parameters of other vehicles, mathematical models and algorithms are used to predict potential collisions. This may include predicting the position and trajectory of other vehicles over a future period, and analyzing whether there are potential intersections or overlaps with the current vehicle's path.

[0065] Furthermore, based on the collision prediction results, appropriate control measures are taken to avoid collisions. For example, if the collision prediction indicates a potential collision risk, the vehicle control system can trigger actions such as emergency braking, lane changing, or deceleration to ensure that the vehicle maintains a safe distance from other vehicles and avoids a collision.

[0066] In this way, when the obstacle is another vehicle, by acquiring the motion information of the other vehicle, making collision predictions based on the current vehicle's driving path and the driving speed and direction of the other vehicles, and making vehicle control based on the collision prediction results, the risk of the vehicle colliding with other vehicles during autonomous driving can be effectively reduced.

[0067] S106: Based on the real-time position of the vehicle after control, the particle swarm optimization algorithm is used to re-plan the global path for the vehicle until the vehicle reaches the target parking space.

[0068] Based on the real-time location of the vehicle after control, the parking path is replanned using the particle swarm optimization algorithm to ensure that the path taken by the vehicle during the automatic driving process is the optimal path.

[0069] This embodiment determines the target parking space for the vehicle and loads it into the garage map. Based on the environmental data of the garage map, it generates a driving path from the initial position to the destination position, treating the vehicle as a point mass. While controlling the vehicle to travel along the driving path, it predicts obstacles based on external environmental information. When the obstacle is a pedestrian, it predicts a collision based on the pedestrian's trajectory and safe distance. When the obstacle is another vehicle, it predicts a collision based on the current vehicle's driving path and the speed and direction of other vehicles. It controls the vehicle based on the collision prediction results and, based on the vehicle's real-time position after control, guides it to the target parking space. This allows the vehicle to plan a parking path based on the recommended target parking space and autonomously drive to that space according to the planned path. Simultaneously, during the journey, the vehicle replans its path based on detection information. This effective combination of optimization and feedback mechanisms reduces computational load, ensures global convergence, and helps the vehicle avoid scrape accidents during parking, improving driving safety and user experience.

[0070] In some embodiments, determining a target parking space for a vehicle includes:

[0071] The system uses the preset license plate number in the property database and the vehicle's license plate number to determine whether the vehicle belongs to the owner. When the vehicle belongs to the owner, the system determines the owner's existing parking space as the target parking space for the vehicle. When the vehicle belongs to a visitor, the system receives visitor information entered through the visitor vehicle's user interface and determines the target parking space for the visitor vehicle based on the visitor information.

[0072] Specifically, the property database contains the license plate numbers of the owners' vehicles. By comparing the preset license plate numbers in the property database (such as the owners' license plate numbers) with the vehicle license plate numbers obtained by the garage gate camera, it is determined whether the vehicle belongs to the owner. If the vehicle's license plate number matches any owner's vehicle's license plate number, then the vehicle can be identified as the owner's vehicle.

[0073] After confirming that the vehicle belongs to the owner, the owner's existing parking space information can be obtained through the property database or other relevant record systems. Based on the owner's parking space allocation, the owner's existing parking space can be identified as the target parking space for the vehicle.

[0074] If the vehicle is not the owner's vehicle, i.e., a visitor's vehicle, the visitor's vehicle can access the parking lot's application through the visitor's vehicle's user interface to instruct the visitor to fill in visitor information. For example, the visitor can enter their own mobile phone number, the mobile phone number of the owner they are visiting, and the address information of the owner they are visiting into the application. The visitor's vehicle can then obtain a target parking space based on this visitor information.

[0075] Based on the matching results of parking preferences and visitor information, a suitable target parking space is determined for the vehicle. The target parking space should meet the vehicle's size requirements and be able to accommodate the parking method and stopping position described in the parking preferences.

[0076] This embodiment can determine whether a vehicle belongs to an owner by combining the preset license plate number in the property database with the vehicle's license plate number, and determine the owner's existing parking space as the target parking space based on the situation. For visitor vehicles, a parking application can be used to determine an available target parking space based on visitor information.

[0077] This allows the property database to develop different parking space recommendation strategies based on whether a vehicle belongs to an owner, thereby improving parking efficiency, simplifying management processes, enhancing user experience, and strengthening parking lot security.

[0078] Additionally, in some embodiments, determining a target parking space for a visitor's vehicle based on visitor information includes:

[0079] When the visitor information indicates that the visited owner already owns a parking space and that the parking space is vacant, the vacant parking space is determined to be the target parking space for the vehicle.

[0080] When the visitor information indicates that the visited owner does not own a parking space or that the owned parking space is already occupied, the parking preference information of the visitor's vehicle is obtained, wherein the parking preference information includes parking method and vehicle size;

[0081] Based on the parking method, vehicle size, and visitor information, the available parking space closest to the address of the visited homeowner is determined as the target parking space for the visitor's vehicle.

[0082] Specifically, based on the owner information indicated in the visitor information, it is checked whether the owner already owns a parking space and whether the owned parking space is vacant. If the owner's owned parking space is vacant, meaning no other vehicle is parked in the owner's owned parking space, then the vacant parking space can be used as the target parking space, and the visitor's vehicle can be parked in the owner's parking space. If the visitor information indicates that the owner does not own a parking space or that the owned parking space is occupied by other vehicles, then a recommendation needs to be made based on the owner's address information in the visitor information.

[0083] Based on the owner's address information and the availability of parking spaces, the system can find the nearest available parking space. In addition, the system can obtain the vehicle's parking preference information through the vehicle's user interface or other relevant methods. This parking preference information may include information such as parking method (e.g., parallel parking, side parking, etc.), vehicle size (e.g., length, width, height), and stopping position (e.g., in front, behind, angled parking, etc.).

[0084] The obtained parking preference information and visitor information are analyzed and matched, taking into account the visitor's vehicle size and parking method preferences and the actual situation of the parking lot, such as the available parking space types and the spacing between parking spaces, to determine the target parking space suitable for the visitor's vehicle from the vacant parking space closest to the owner's address.

[0085] This embodiment determines the target parking space for a vehicle based on visitor information and the owner's parking space availability. If the owner already has a vacant parking space, it is used as the target parking space. If the owner does not have a parking space or their parking space is occupied, a suitable target parking space is recommended for the visitor vehicle based on visitor information and the visitor vehicle's parking preference information. This allows the property management system to provide the most suitable parking space for visitor vehicles according to the actual situation, increasing the utilization rate of the parking lot and improving the user's parking experience.

[0086] In some embodiments, global path planning for the vehicle is performed using a predetermined particle swarm optimization algorithm, including:

[0087] A dimensional space is created, and a set of particles is randomly generated in the dimensional space as the initial solution for the path based on the initial position of the vehicle and the endpoint position corresponding to the target parking space. The path quality of each particle is evaluated according to a predefined fitness function. The optimal path of each particle is updated based on the particle swarm optimization algorithm by calculating the velocity and position of the particles. The global optimal path is updated based on the optimal paths of the particles, and the velocity and position of the particles are iteratively optimized until a stopping condition is reached. The global optimal path at the stopping condition is taken as the driving path for autonomous driving. The stopping condition includes reaching a preset number of iterations or the optimal value after multiple iterations being limited to a preset extreme value.

[0088] The following is combined with Figure 2 The workflow of a particle swarm optimization algorithm provided in this embodiment will be described, such as... Figure 2 As shown:

[0089] Specifically, based on the initial position of the vehicle and the final position of the target parking space, an appropriate dimensional space is created. The dimensional space can be represented as a two-dimensional plane, which contains the initial position of the vehicle and the final position of the target parking space. A set of particles is randomly generated in the dimensional space as the initial solution for the path. The position and velocity of each particle are initialized. Each particle represents a possible path, which consists of a defined sequence of coordinate points.

[0090] The quality of each particle's path is evaluated using a predefined fitness function, which calculates the particle's fitness value. The fitness function can take into account factors such as path length, smoothness, and collision avoidance to determine the quality of the path. Using a particle swarm optimization algorithm, the optimal path for each particle is updated based on its velocity and position. The updates to the particle's velocity and position can use formulas for acceleration and velocity to optimize the path search.

[0091] The global optimal path is updated based on the optimal path of each particle. The path of the particle with the best fitness value is taken as the global optimal solution, and the path and its fitness value are recorded. Through an iterative optimization process, the velocity and position of the particles and the global optimal path are continuously updated until a stopping condition is reached. The stopping condition can be reaching a predetermined number of iterations or reaching the convergence threshold of the fitness function. The global optimal path is used as the driving path for autonomous driving. This path is obtained through the particle swarm optimization algorithm and the optimal solution is obtained after multiple iterations. It can guide the vehicle to the target parking space while ensuring safety and efficiency.

[0092] As an example, suppose the position of the i-th particle in D-dimensional space is: xi = (x1, x2, ..., xD), where i = 1, 2, ..., N, and its velocity is: vi = (v1, v2, ..., vD), where i = 1, 2, ..., N. Based on the objective function, the fitness value is calculated, and thus P is derived. besti =P besti,1 ,...,P besti,d ,...,P besti,D With G besti =G besti,1 ,...,G besti,d ,...,G besti,D In each iteration, the population updates its velocity and position by following two extreme values:

[0093]

[0094] in:

[0095] c1 and c2 are learning factors for individuals and populations, respectively. k The inertial weight is used. Equation (2.2) defines the particle's flight speed.

[0096] The steps of the PSO algorithm are as follows:

[0097] Step 1: Initialize the population size N, the number of iterations K, and the size of the dimensional space D, and define the values ​​of w, c1, and c2;

[0098] Step 2: Randomly initialize x i v i P best and G best ;

[0099] Step 3: Iterate according to equations (1.1) and (2.2) to update the position of the particle;

[0100] Step 4: Evaluate the fitness value of each particle and update P. best and G best ;

[0101] Step 5: Proceed to Step 3 until K (preset number of) iterations have been performed, or until the accuracy of the optimal value is limited to a minimum value (preset extreme value) after multiple iterations. Then terminate the algorithm and output the optimal position G. best and fitness function value f(G) best ).

[0102] The objective function is: That is, the path length, P0 and P D The coordinates of the starting point and the target point in the path are represented by L(P), and the distance between each two adjacent path nodes is calculated. The sum of these values ​​gives the length of the path.

[0103] This embodiment uses the particle swarm optimization algorithm to generate and optimize autonomous driving paths in dimensional space, which can quickly obtain high-quality path solutions and improve the optimization effect and driving accuracy.

[0104] Furthermore, in some embodiments, when the obstacle is a pedestrian, the pedestrian's trajectory is predicted to determine a safe distance between the vehicle and the pedestrian. Based on the trajectory and the safe distance, a collision prediction is performed on the vehicle, and vehicle control is implemented based on the collision prediction results, including three scenarios:

[0105] In the first scenario, when pedestrians and vehicles are moving in opposite directions, the vehicle is controlled to stop at a safe distance and a voice prompt is issued to remind pedestrians to leave voluntarily.

[0106] Specifically, when pedestrians and vehicles are moving in opposite directions, to ensure safety, the vehicle's autonomous driving system controls the vehicle to stop at a safe distance from the pedestrians. It should be noted that the vehicle's sensors and algorithms can detect the pedestrians' positions and movements, and determine when to stop the vehicle based on the predetermined safe distance.

[0107] In addition, the vehicle can issue corresponding voice prompts through its built-in voice system or external speakers to remind pedestrians to move away from the vehicle. For example, the system can use voice prompts to remind pedestrians to be aware of the vehicle approaching or to ask pedestrians to move away from the vehicle to ensure safety.

[0108] This allows for the safety of pedestrians when they are moving in opposite directions from vehicles, and the vehicle can issue voice prompts to ensure that pedestrians move away from the front of the vehicle, reducing the potential risk of collision.

[0109] In the second scenario, when pedestrians and vehicles are moving in the same direction, the vehicle is controlled to travel at the same speed as the pedestrians, and a voice prompt is issued to remind the pedestrians to leave voluntarily.

[0110] Specifically, when pedestrians and vehicles move in the same direction, to ensure safety, the vehicle's automatic driving system controls the vehicle to maintain the same speed as the pedestrians, ensuring that the relative positions do not change drastically during the journey and reducing the potential risk of collision.

[0111] The vehicle can issue corresponding voice prompts through its built-in voice system or external speakers to remind pedestrians to move away from the vehicle.

[0112] This ensures the safety of pedestrians and vehicles when moving in the same direction. By making the vehicle travel at the same speed as the pedestrian and issuing voice prompts, pedestrians are alerted and reminded to move away from the vehicle, thereby reducing the potential risk of collision.

[0113] In the third scenario, when a pedestrian is to the side of the vehicle and at a safe distance, moving in the opposite or same direction as the vehicle, the vehicle will issue a voice prompt to remind the pedestrian to give way.

[0114] Specifically, when a pedestrian is to the side of a vehicle and maintains a safe distance from the vehicle while moving in the opposite or same direction, the possibility of a collision is relatively small because the pedestrian and the vehicle are not in the same driving lane and the vehicle's speed is generally greater than the pedestrian's walking speed. In this case, the vehicle can be controlled to issue a voice prompt to remind the pedestrian to actively avoid the vehicle.

[0115] In addition, when a pedestrian is detected ahead, the vehicle can predict the pedestrian's trajectory based on the pedestrian's movement direction and speed on the vehicle's display screen, and determine whether the pedestrian will enter the danger zone. If the pedestrian is outside the danger zone, the pedestrian can be marked with a safe color (green) on the vehicle's display screen. When the pedestrian enters the danger zone, the pedestrian image on the display screen can be marked as a red image, a danger warning sound will be emitted, and the vehicle will be controlled to take emergency braking measures.

[0116] This embodiment predicts the pedestrian's trajectory when the obstacle is a pedestrian, determines the safe distance between the vehicle and the pedestrian, predicts a collision based on the trajectory and safe distance, and controls the vehicle based on the collision prediction results. In cases where a collision is possible, different control strategies are adopted for the current vehicle to avoid collisions with pedestrians during autonomous driving, thereby improving the driving safety of autonomous driving and the safety of pedestrians.

[0117] Additionally, in some embodiments, when the obstacle is a pedestrian, the pedestrian's trajectory is predicted to determine a safe distance between the vehicle and the pedestrian. Based on the trajectory and the safe distance, a collision prediction is performed on the vehicle, and vehicle control is implemented based on the collision prediction result, including:

[0118] When a pedestrian is in front of a vehicle and the pedestrian's movement trajectory intersects the vehicle's movement direction, obtain the pedestrian's current position and the vehicle's current position, as well as the pedestrian's walking speed and the vehicle's driving speed.

[0119] Based on the current positions of pedestrians and vehicles, as well as the walking speed of pedestrians and the driving speed of vehicles, the system determines the distance between pedestrians and vehicles after a preset time. If the distance between pedestrians and vehicles is less than the safe distance, the system predicts that a collision will occur between the vehicle and the pedestrian and controls the vehicle to stop at the safe distance.

[0120] Specifically, when a pedestrian is in front of a vehicle and the pedestrian's trajectory intersects with the vehicle's direction of movement, the location information of the pedestrian and the vehicle can be obtained using sensors or other positioning devices. In addition, the pedestrian's speed can also be calculated using this information.

[0121] A common method for calculating collision risk is to use relative speed and distance. The current positions and speeds of pedestrians and vehicles are converted into relative positions and relative speeds. Then, their relative positions and relative speeds are compared to see if they meet the criteria for collision risk. If the criteria are met, it can be determined that a collision between the vehicle and the pedestrian is likely.

[0122] The following is combined with Figure 3 This application describes a collision prediction method where the obstacle is a pedestrian, as described in an embodiment of the present application. Figure 3As shown:

[0123] As an example, assume the safe distance between vehicles is d. f Let the vehicle's speed be v1 and the pedestrian's speed be v2. The vehicle moves along straight line AP, and the pedestrian moves along straight line BP. The intersection of their paths is point P. The current positions of the vehicle and pedestrian are A and B, respectively, and the distance between their current positions is r. Their distances from point P are a and b, respectively. Let the vehicle travel to point C along the planned path in time t (a preset time), and the pedestrian arrive at point D in the same time t. This is the angle between the directions of movement of the vehicle and the pedestrian. When the distance DC between the vehicle and the pedestrian is less than or equal to d... f If a collision is predicted, and the DC distance is x, the motion and set relationships between the vehicle and the pedestrian can be seen from the diagram as follows:

[0124]

[0125] Where d1 = v1t, d2 = v2t, let x = 0, or x = d f Substitute into the formula and solve for d.

[0126] If 0≤d1≤a, 0≤d1≤b, and 0≤d2≤a, 0≤d2≤b, it can be predicted that a collision will occur between a vehicle and a pedestrian.

[0127] When it is determined that a collision between a vehicle and a pedestrian will occur, the vehicle can be stopped at a safe distance by controlling the vehicle's braking system to ensure the safety of the pedestrian. This safe distance can be adjusted as needed to ensure sufficient safety space, and no specific limit is specified here.

[0128] This embodiment uses a preset algorithm to predict when a collision will occur between a vehicle and a pedestrian, and controls the vehicle to stop at a safe distance, ensuring the safety of pedestrians and vehicles when they cross paths, thereby reducing potential collision risks and improving the safety of autonomous driving.

[0129] In addition, in some embodiments, when the obstacle is another vehicle, the speed and direction of the other vehicle are obtained, collision prediction is performed based on the current vehicle's driving path and the speed and direction of the other vehicles, and vehicle control is performed based on the collision prediction results, including three scenarios:

[0130] In the first scenario, when other vehicles are traveling in front of the vehicle at a speed lower than the vehicle's current speed, the vehicle is controlled to travel at the same speed as the other vehicles.

[0131] Specifically, when other vehicles are traveling ahead at a speed lower than the current speed of the vehicle, a collision is inevitable after a certain period of time. In this case, controlling the vehicle to travel at the same speed as other vehicles can help avoid collisions with the vehicles ahead and improve driving safety.

[0132] The second scenario is when other vehicles are traveling in front of the vehicle and their speed is greater than the vehicle's current speed; or when other vehicles are traveling in opposite directions in different lanes, the vehicle is controlled to continue traveling at its current speed.

[0133] Specifically, when other vehicles are traveling in front of the current vehicle at a speed greater than the current vehicle's speed, or when other vehicles are traveling in opposite directions in different lanes, it can be determined that a collision will not occur between the two vehicles. The current vehicle can continue to maintain its current speed and keep a safe distance in accordance with road traffic rules.

[0134] The third scenario is when other vehicles are in front of your vehicle and are entering the parking space, stop your vehicle outside the safe area.

[0135] Specifically, when other vehicles are in front of the vehicle and are entering a parking space, the vehicle's autonomous driving system should control the vehicle to stop outside the safe area and wait for the vehicle in front to finish parking. After that, the vehicle can restart the autonomous driving system to continue driving, thereby ensuring that other vehicles can park smoothly while avoiding collisions with other vehicles or causing traffic chaos.

[0136] This embodiment obtains the speed and direction of other vehicles when the obstacle is another vehicle, and performs collision prediction based on the current vehicle's driving path and the speed and direction of other vehicles. In cases where a collision is possible, different control strategies are adopted for the current vehicle for different collision situations, thereby avoiding collisions between the current vehicle and other vehicles during autonomous driving and improving the driving safety of autonomous driving.

[0137] In some embodiments, when the obstacle is another vehicle, the driving speed and direction of the other vehicle are obtained, collision prediction is performed based on the current vehicle's driving path and the driving speed and direction of the other vehicle, and vehicle control is performed based on the collision prediction results, including:

[0138] When other vehicles are in front of the vehicle and their driving directions intersect with the vehicle's driving direction, obtain the vehicle's first driving speed and the other vehicles' second driving speed, and treat the vehicle as a first cuboid and the other vehicles as a second cuboid;

[0139] Based on the first and second driving speeds, determine whether each edge of the first cuboid intersects with the six faces of the second cuboid in the plane within the target time. If an intersection is found, determine whether the intersection point is on other vehicles to determine whether the two vehicles will collide within the target time. If a collision is determined, control the vehicles to stop outside the safe area.

[0140] The following is combined with Figure 4 This application describes a collision prediction method that treats a vehicle as a cuboid, as exemplified by the present application. Figure 4 As shown:

[0141] Specifically, the prediction of whether a collision will occur is based on treating the vehicle as a cuboid. This prediction calculates whether each edge of the autonomous vehicle intersects with the obstacle vehicle; if an intersection is predicted, a collision will happen. Similarly, if two vehicles are approximated as cuboids, a collision is considered to occur if the two cuboids overlap.

[0142] As an example, suppose the speeds of two vehicles are v1 (first driving speed), v2 (second driving speed), and v1 (second driving speed), respectively. After detecting an obstacle vehicle, calculate any time T (target time) and determine whether the two vehicles intersect at time T. This involves solving for whether each edge of the autonomous vehicle at this position intersects with the six faces of the obstacle vehicle, thus determining whether a collision has occurred. The determination method is as follows:

[0143] Determine whether a line segment intersects the plane containing the surface;

[0144]

[0145] Where P is a point on one of the planes of the obstacle vehicle, and line segment AB is one of the twelve edges of the current vehicle. Let P be a vector pointing to A. Let be the normal vector on the plane of other vehicles; if the above formula is satisfied, then and The directions are opposite, meaning line segment AB intersects the plane; if they intersect, find the intersection point and continue; otherwise, end.

[0146]

[0147] The first equation represents the equation of the obstacle plane, and the second equation represents the equation of line segment AB. Solve for t and Q(t); if Q(t) has a solution, then continue with the following judgment.

[0148] Determine if the intersection point is on other vehicles; if so, they intersect; otherwise, they do not intersect.

[0149] Assuming the four vertices of the plane of other vehicles are M, N, T, and R, the intersection point lies on other vehicles if the following formula is satisfied:

[0150]

[0151] If a collision is predicted between two vehicles, the current vehicle is stopped at a safe distance to allow the obstacle vehicle to pass first.

[0152] This embodiment treats the vehicle as a first cuboid and other vehicles as second cuboids, and performs collision prediction by calculating whether the cuboids overlap. This helps to improve the accuracy of collision prediction while enhancing the safety of the autonomous driving process.

[0153] All of the above-mentioned optional technical solutions can be combined in any way to form the optional embodiments of this application, and will not be described in detail here.

[0154] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the process of the embodiments of this application.

[0155] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0156] Figure 5 This application provides an embodiment of an automated driving parking device based on path planning, such as... Figure 5 As shown, the device includes:

[0157] The determination module 501 is configured to determine the target parking space for the vehicle, perform working environment initialization operations on the vehicle based on the initial position of the vehicle and the end position corresponding to the target parking space, and load the garage map.

[0158] The path planning module 502 is configured to perform global path planning for the vehicle, treating the vehicle as a point mass, based on the environmental data of the garage map, and using a predetermined particle swarm algorithm to generate a driving path from the initial position to the destination position.

[0159] The prediction module 503 is configured to use sensors to perceive external environmental information around the vehicle while controlling the vehicle to travel along the driving path, and to predict obstacles based on the external environmental information, wherein obstacles include pedestrians and / or other vehicles.

[0160] The first control module 504 is configured to predict the movement trajectory of a pedestrian when the obstacle is a pedestrian, determine the safe distance between the vehicle and the pedestrian, predict a collision with the vehicle based on the movement trajectory and the safe distance, and control the vehicle based on the collision prediction result.

[0161] The second control module 505 is configured to, when the obstacle is another vehicle, acquire the driving speed and direction of the other vehicle, perform collision prediction based on the current vehicle's driving path and the driving speed and direction of the other vehicle, and perform vehicle control based on the collision prediction results.

[0162] The third control module 506 is configured to re-plan the global path of the vehicle using the particle swarm optimization algorithm based on the real-time position of the vehicle after control, until the vehicle reaches the target parking space.

[0163] In some embodiments, the determining module 501 is further configured to determine whether a vehicle is an owner's vehicle by using a preset license plate number in the property database and the vehicle's license plate number; when the vehicle is an owner's vehicle, determine an existing parking space of the owner as the target parking space for the vehicle; when the vehicle is a visitor's vehicle, receive visitor information input through the visitor's vehicle's user interface, and determine the target parking space for the visitor's vehicle based on the visitor information.

[0164] In some embodiments, the determining module 501 is further configured to: when the visitor information indicates that the visited owner owns a parking space and the parking space is vacant, determine the vacant parking space as the target parking space for the vehicle; when the visitor information indicates that the visited owner does not own a parking space or that the owned parking space is already occupied, obtain the parking preference information of the visitor vehicle, wherein the parking preference information includes parking method and vehicle size; and determine the vacant parking space closest to the address information of the visited owner as the target parking space for the visitor vehicle based on the parking method, vehicle size and the visitor information.

[0165] In some embodiments, the path planning module 502 is further configured to create a dimensional space, randomly generate a set of particles in the dimensional space as the initial solution of the path based on the initial position of the vehicle and the endpoint position corresponding to the target parking space; evaluate the path quality of each particle according to a predefined fitness function, update the optimal path of each particle according to the particle swarm optimization algorithm by calculating the velocity and position of the particles; update the global optimal path according to the optimal path of the particles, and iteratively optimize the velocity and position of the particles until a stopping condition is reached, and use the global optimal path at the stopping condition as the driving path of autonomous driving, wherein the stopping condition includes the number of iterations reaching a preset number, or the optimal value after multiple iterations being limited to a preset extreme value.

[0166] In some embodiments, the second control module 505 is further configured to: predict a collision between the vehicle and the pedestrian when the pedestrian and the vehicle are moving in opposite directions; control the vehicle to stop at a safe distance and issue a voice prompt message to prompt the pedestrian to move away; predict a collision between the vehicle and the pedestrian when the pedestrian and the vehicle are moving in the same direction and the vehicle's speed is greater than the pedestrian's walking speed; control the vehicle to travel at the same speed as the pedestrian and issue a voice prompt message to prompt the pedestrian to move away; and predict that a collision will not occur between the vehicle and the pedestrian when the pedestrian is to the side of the vehicle and at a safe distance from the vehicle, and issue a voice prompt message to prompt the pedestrian to avoid the collision.

[0167] In some embodiments, the second control module 505 is further configured to, when a pedestrian is in front of a vehicle and the pedestrian's movement trajectory intersects with the vehicle's movement direction, obtain the current position of the pedestrian and the current position of the vehicle, as well as the pedestrian's walking speed and the vehicle's driving speed; determine the distance between the pedestrian and the vehicle after a preset time based on the current position of the pedestrian and the current position of the vehicle, as well as the pedestrian's walking speed and the vehicle's driving speed; and if the distance between the pedestrian and the vehicle is less than a safe distance, predict that a collision will occur between the vehicle and the pedestrian, and control the vehicle to stop outside the safe distance.

[0168] In some embodiments, the third control module 506 is further configured to predict a collision between the vehicle and other vehicles when other vehicles are traveling in front of the vehicle at a speed less than the vehicle's current speed, and control the vehicle to travel at the same speed as other vehicles; when other vehicles are traveling in front of the vehicle at a speed greater than the vehicle's current speed; or when other vehicles are traveling in opposite directions in different lanes, predict that a collision will not occur between the vehicle and other vehicles, and control the vehicle to continue traveling at its current speed; and when other vehicles are in front of the vehicle and are entering a parking space, predict a collision will occur between the vehicle and other vehicles, and control the vehicle to stop outside the safe area.

[0169] In some embodiments, the third control module 506 is further configured to, when other vehicles are in front of the vehicle and the driving directions of the other vehicles intersect with the driving direction of the vehicle, acquire the first driving speed of the vehicle and the second driving speed of the other vehicles, and regard the vehicle as a first cuboid and the other vehicles as a second cuboid; based on the first driving speed and the second driving speed, determine whether each edge of the first cuboid and the six faces of the second cuboid intersect in the plane within a target time; if an intersection is determined to exist, determine whether the intersection point is on the other vehicle, thereby determining whether the two vehicles will collide within the target time; if a collision is determined to occur, control the vehicle to stop outside the safe area.

[0170] The apparatus provided in this application embodiment can implement all the method steps of the above method embodiments and achieve the same technical effect, which will not be repeated here.

[0171] Figure 6 This is a schematic diagram of the electronic device 6 provided in an embodiment of this application. Figure 6 As shown, the electronic device 6 of this embodiment includes a processor 601, a memory 602, and a computer program 603 stored in the memory 602 and executable on the processor 601. When the processor 601 executes the computer program 603, it implements the steps in the various method embodiments described above. Alternatively, when the processor 601 executes the computer program 603, it implements the functions of each module / unit in the various device embodiments described above.

[0172] Electronic device 6 can be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 6 may include, but is not limited to, processor 601 and memory 602. Those skilled in the art will understand that... Figure 6 This is merely an example of electronic device 6 and does not constitute a limitation on electronic device 6. It may include more or fewer components than shown, or different components.

[0173] The processor 601 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0174] The memory 602 can be an internal storage unit of the electronic device 6, such as a hard disk or RAM of the electronic device 6. The memory 602 can also be an external storage device of the electronic device 6, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc., equipped on the electronic device 6. The memory 602 can also include both internal and external storage units of the electronic device 6. The memory 602 is used to store computer programs and other programs and data required by the electronic device.

[0175] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0176] If integrated modules / units are implemented as software functional units and sold or used as independent products, they can be stored in a readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The readable storage medium may include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0177] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions 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 this application, and should all be included within the protection scope of this application.

Claims

1. An autonomous driving parking method based on path planning, characterized in that, include: Determine the target parking space for the vehicle, and based on the vehicle's initial position and the target parking space's corresponding endpoint position, perform a working environment initialization operation on the vehicle and load it into the garage map. Based on the environmental data of the garage map, the vehicle is treated as a point mass, and a predetermined particle swarm optimization algorithm is used to perform global path planning for the vehicle to generate a driving path from the initial position to the final position. During the process of controlling the vehicle to travel along the driving path, sensors are used to perceive the external environment information around the vehicle, and obstacles are predicted based on the external environment information, wherein the obstacles include pedestrians and / or other vehicles. When the obstacle is a pedestrian, the pedestrian's trajectory is predicted, the safe distance between the vehicle and the pedestrian is determined, and a collision prediction is made for the vehicle based on the trajectory and the safe distance. The vehicle is then controlled based on the collision prediction result. When the obstacle is another vehicle, the driving speed and direction of the other vehicle are obtained, collision prediction is performed based on the current vehicle's driving path and the driving speed and direction of the other vehicle, and vehicle control is performed based on the collision prediction result. Based on the real-time location of the vehicle after control, the particle swarm algorithm is used to re-plan the global path for the vehicle until the vehicle reaches the target parking space. The process of determining a target parking space for a vehicle includes: when the vehicle is a visitor vehicle, receiving visitor information input through the user interface of the visitor vehicle; when the visitor information indicates that the visited owner owns a parking space and the parking space is vacant, determining the vacant parking space as the target parking space for the vehicle; when the visitor information indicates that the visited owner does not own a parking space or that the owned parking space is already occupied, obtaining the parking preference information of the visitor vehicle, wherein the parking preference information includes parking method and vehicle size; and determining the vacant parking space closest to the visited owner's address as the target parking space for the visitor vehicle based on the parking method, vehicle size, and visitor information.

2. The autonomous driving parking method based on path planning according to claim 1, wherein determining the target parking space for the vehicle includes: The vehicle's license plate number is compared with the preset license plate number in the property database to determine whether the vehicle belongs to the owner. When the vehicle is owned by the owner, the owner's existing parking space is identified as the target parking space for the vehicle.

3. The autonomous driving parking method based on path planning according to claim 1, characterized in that, The step of using a predetermined particle swarm optimization algorithm to perform global path planning for the vehicle includes: Create a dimensional space, and randomly generate a group of population particles in the dimensional space as the initial solution for the path based on the initial position of the vehicle and the endpoint position corresponding to the target parking space; The path quality of each particle is evaluated according to a predefined fitness function, and the optimal path of each particle is updated according to the particle swarm algorithm by calculating the velocity and position of the particles. The global optimal path is updated based on the optimal path of the particle, and the velocity and position of the particle are iteratively optimized until a stopping condition is reached. The global optimal path at the point where the stopping condition is reached is used as the driving path for autonomous driving. The stopping condition includes the number of iterations reaching a preset number, or the optimal value after multiple iterations being limited to a preset extreme value.

4. The autonomous driving parking method based on path planning according to claim 1, characterized in that, When the obstacle is a pedestrian, the process includes predicting the pedestrian's trajectory, determining a safe distance between the vehicle and the pedestrian, predicting a collision with the vehicle based on the trajectory and the safe distance, and controlling the vehicle based on the collision prediction result. When the pedestrian and the vehicle are moving in opposite directions, it is predicted that the vehicle and the pedestrian will collide. The vehicle is controlled to stop at the safe distance and a voice prompt is issued to prompt the pedestrian to leave the area. When the pedestrian and the vehicle are moving in the same direction and the vehicle's speed is greater than the pedestrian's walking speed, it is predicted that the vehicle and the pedestrian will collide. The vehicle is then controlled to travel at the same speed as the pedestrian, and the voice prompt message is issued to prompt the pedestrian to leave the area. When the pedestrian is to the side of the vehicle and at a safe distance, moving in the opposite direction or in the same direction as the vehicle, it is predicted that the vehicle and the pedestrian will not collide. The vehicle is then controlled to issue the voice prompt, which is used to prompt the pedestrian to actively avoid the collision.

5. The autonomous driving parking method based on path planning according to claim 1, characterized in that, When the obstacle is a pedestrian, the process includes predicting the pedestrian's trajectory, determining a safe distance between the vehicle and the pedestrian, predicting a collision with the vehicle based on the trajectory and the safe distance, and controlling the vehicle based on the collision prediction result. When the pedestrian is in front of the vehicle and the pedestrian's movement trajectory intersects the vehicle's movement direction, obtain the pedestrian's current position and the vehicle's current position, as well as the pedestrian's walking speed and the vehicle's driving speed. Based on the current positions of the pedestrian and the vehicle, as well as the pedestrian's walking speed and the vehicle's driving speed, the distance between the pedestrian and the vehicle after a preset time is determined. If the distance between the pedestrian and the vehicle is less than the safe distance, a collision between the vehicle and the pedestrian is predicted, and the vehicle is controlled to stop outside the safe distance.

6. The autonomous driving parking method based on path planning according to claim 1, characterized in that, When the obstacle is another vehicle, the system acquires the speed and direction of that other vehicle, performs collision prediction based on the current vehicle's path and the speed and direction of the other vehicle, and performs vehicle control based on the collision prediction result, including: When another vehicle is traveling in front of the vehicle and its speed is less than the vehicle's current speed, it is predicted that the vehicle and the other vehicle will collide, and the vehicle is controlled to travel at the same speed as the other vehicle. When the other vehicle is traveling in front of the vehicle and its speed is greater than the vehicle's current speed; or when the other vehicle is traveling in the opposite direction to the vehicle in a different lane, it is predicted that the vehicle and the other vehicle will not collide, and the vehicle is controlled to continue traveling at its current speed. When another vehicle is in front of the vehicle and is entering the parking space, it is predicted that the vehicle and the other vehicle will collide, and the vehicle is controlled to stop outside the safe area.

7. The autonomous driving parking method based on path planning according to claim 1, characterized in that, When the obstacle is another vehicle, the system acquires the speed and direction of that other vehicle, performs collision prediction based on the current vehicle's path and the speed and direction of the other vehicle, and performs vehicle control based on the collision prediction result, including: When the other vehicle is in front of the vehicle and the driving direction of the other vehicle intersects with the driving direction of the vehicle, the first driving speed of the vehicle and the second driving speed of the other vehicle are obtained, and the vehicle is regarded as a first cuboid and the other vehicle is regarded as a second cuboid. Based on the first driving speed and the second driving speed, determine whether each edge of the first cuboid and the six faces of the second cuboid intersect in their respective planes within the target time. If the intersection is determined to exist, determine whether the intersection point is on the other vehicle, thereby determining whether the two vehicles will collide within the target time. If it is determined that the two vehicles will collide, control the vehicles to stop outside the safe area.

8. An automated parking device based on path planning, characterized in that, include: The determination module is configured to determine the target parking space for the vehicle, perform a working environment initialization operation on the vehicle based on the initial position of the vehicle and the endpoint position corresponding to the target parking space, and load the garage map. The path planning module is configured to, based on the environmental data of the garage map, treat the vehicle as a point mass and use a predetermined particle swarm optimization algorithm to perform global path planning for the vehicle, so as to generate a driving path from the initial position to the final position. The prediction module is configured to, while controlling the vehicle to travel along the driving path, use sensors to perceive external environmental information around the vehicle and predict obstacles based on the external environmental information, wherein the obstacles include pedestrians and / or other vehicles. The first control module is configured to predict the movement trajectory of the pedestrian when the obstacle is a pedestrian, determine the safe distance between the vehicle and the pedestrian, perform collision prediction on the vehicle based on the movement trajectory and the safe distance, and control the vehicle based on the collision prediction result. The second control module is configured to, when the obstacle is another vehicle, acquire the driving speed and direction of the other vehicle, perform collision prediction based on the current vehicle's driving path and the driving speed and direction of the other vehicle, and perform vehicle control based on the collision prediction result. The third control module is configured to re-plan the global path of the vehicle using the particle swarm algorithm based on the real-time position of the vehicle after control, until the vehicle reaches the target parking space. The determining module is further configured to: when the vehicle is a visitor vehicle, receive visitor information input through the user interface of the visitor vehicle; when the visitor information indicates that the visited owner owns a parking space and the parking space is vacant, determine the vacant parking space as the target parking space for the vehicle; when the visitor information indicates that the visited owner does not own a parking space or that the owned parking space is already occupied, obtain the parking preference information of the visitor vehicle, wherein the parking preference information includes parking method and vehicle size; and determine the vacant parking space closest to the visited owner's address as the target parking space for the visitor vehicle based on the parking method, vehicle size, and visitor information.

9. A readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images