Intelligent vehicle automatic parking path planning optimization method, system and device and storage medium

Abstract

The invention discloses an intelligent vehicle automatic parking path planning optimization method, system and device and a storage medium. The method comprises the following steps: planning a parking path of a vehicle from a current position to a target parking space by using a search algorithm; a virtual parking environment is constructed, a simulation test is carried out in the virtual parking environment, different parking scenes and environment changes are simulated, and whether a parking path is safe or not is verified; when the parking path is verified to be safe, the vehicle is controlled to be parked according to the parking path; and when the vehicle is parked, the environment change is monitored in real time through a sensor, and the parking path is dynamically adjusted by using a path-finding algorithm and reinforcement learning. An A * algorithm is used for carrying out preliminary path planning and optimization, the reliability of the algorithm is verified through simulation testing, path planning and control strategies are continuously optimized, environment changes are monitored in real time and paths are dynamically adjusted by fusing data of various sensors such as a laser radar, a camera and an ultrasonic radar, and it is ensured that a vehicle safely avoids obstacles.

Description

Technical Field

[0001] The present invention relates to the field of parking technology, and in particular to a method, system, device and storage medium for optimizing automatic parking path planning of an intelligent vehicle. Background Art

[0002] Traditional automatic parking systems have high requirements for parking spaces. For example, there cannot be obstacles such as telephone poles near the parking spaces. During the automatic parking process, if pedestrians or vehicles suddenly rush out from the rear, the system may not be able to respond in time, posing a safety hazard. For low obstacles (such as roadbeds, fire hydrants, etc.) or parking spaces with unclear markings, traditional automatic parking systems may not be able to accurately detect them. In rainy and snowy weather, the detection ability of traditional automatic parking systems will be affected due to unclear ground markings. Summary of the invention

[0003] The main purpose of the present invention is to propose a method, system, device and storage medium for optimizing the automatic parking path planning of an intelligent vehicle, aiming to enhance the safety of automatic parking and to identify obstacles that suddenly appear during parking.

[0004] To achieve the above-mentioned purpose, the present invention proposes an optimization method for automatic parking path planning of an intelligent vehicle, and the optimization method for automatic parking path planning of an intelligent vehicle comprises:

[0005] Use a search algorithm to plan a parking path from the vehicle's current position to the target parking space;

[0006] Constructing a virtual parking environment, performing simulation tests in the virtual parking environment, simulating different parking scenarios and environmental changes, and verifying whether the parking path is safe;

[0007] When the parking path is verified to be safe, controlling the vehicle to park according to the parking path;

[0008] When the vehicle is parking, environmental changes are monitored in real time through sensors, and the parking path is dynamically adjusted using a path-finding algorithm and reinforcement learning.

[0009] In one embodiment, before the step of using a search algorithm to plan a parking path from the current position of the vehicle to the target parking space, the step includes:

[0010] The environmental information of the vehicle parking is collected in real time by sensors, and a three-dimensional environmental model around the vehicle is created using a Kalman filter algorithm; the three-dimensional environmental model is used to construct a virtual parking environment.

[0011] In one embodiment, before the step of using a search algorithm to plan a parking path from the current position of the vehicle to the target parking space, the step includes:

[0012] Constructing a marker map according to the environmental information, and continuously updating the marker map through a particle filtering algorithm;

[0013] In the marked map, a machine learning model is used to analyze and match the target parking space that best meets the user's preferences based on the user's parking preferences.

[0014] In one embodiment, after the step of using a search algorithm to plan a parking path from the current position of the vehicle to the target parking space, the following steps are included:

[0015] When there are multiple vehicles parking at the same time in the parking lot, the multiple vehicles are controlled to share information through the V2V communication protocol, and the parking path of each vehicle is optimized by using game theory and multi-agent system algorithm.

[0016] In one embodiment, when the parking path is verified to be safe, the step of monitoring environmental changes in real time through sensors and dynamically adjusting the parking path using a path-finding algorithm and reinforcement learning includes:

[0017] A control method for predicting the vehicle is implemented by combining a PID controller on the vehicle and an algorithm model, and controlling the vehicle to park according to the parking path.

[0018] In one embodiment, the control method of the vehicle predicted by combining the PID controller on the vehicle and the algorithm model, and the step of controlling the vehicle to park according to the parking path, includes:

[0019] Using the PID controller and the MPC model on the vehicle to predict the control method of the vehicle, adjusting the direction and speed of the vehicle in real time, and controlling the vehicle to travel according to the parking path;

[0020] When the vehicle arrives in front of the target parking space, the vehicle is controlled to complete parking using fuzzy logic control and deep learning control models;

[0021] During the vehicle parking process, multi-sensor redundancy and Markov chain methods are used for safety monitoring and fault prediction.

[0022] The present invention further proposes an intelligent vehicle automatic parking path planning optimization system, the intelligent vehicle automatic parking path planning optimization system includes the intelligent vehicle automatic parking path planning optimization method as described above, the intelligent vehicle automatic parking path planning optimization system includes:

[0023] A path planning module is used to plan a parking path from the current position of the vehicle to the target parking space using a search algorithm;

[0024] A simulation module is used to construct a virtual parking environment, perform simulation tests in the virtual parking environment, simulate different parking scenarios and environmental changes, and verify whether the parking path is safe;

[0025] A control execution module, configured to control the vehicle to park according to the parking path when the parking path is verified to be safe;

[0026] The dynamic path adjustment module is used to monitor environmental changes in real time through sensors when the vehicle is parking, and dynamically adjust the parking path using a path-finding algorithm and reinforcement learning.

[0027] In one embodiment, the intelligent vehicle automatic parking path planning optimization system further includes:

[0028] An environment perception module is used to collect the environment information of the vehicle parking in real time through sensors, and use the Kalman filter algorithm to create a three-dimensional environment model around the vehicle; the three-dimensional environment model is used to construct a virtual parking environment;

[0029] A map generation and updating module, used to construct a marking map according to the environmental information, and continuously update the marking map through a particle filtering algorithm;

[0030] A user preference module, configured to analyze and match the target parking space that best meets the user's preference in the marked map according to the user's parking preference using a machine learning model;

[0031] The multi-vehicle cooperative communication module controls multiple vehicles to share information through the V2V communication protocol when there are multiple vehicles parked at the same time in the parking lot, and uses game theory and multi-agent system algorithm to optimize the parking path of each vehicle.

[0032] The present invention also proposes an intelligent vehicle automatic parking path planning optimization device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the above-mentioned intelligent vehicle automatic parking path planning optimization method.

[0033] The present invention also proposes a storage medium, which is a computer-readable storage medium. A computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the intelligent vehicle automatic parking path planning optimization method as described above are implemented.

[0034] The present invention relates to the field of parking technology, and specifically discloses a method, system, device and storage medium for automatic parking path planning optimization of intelligent vehicles. The method for automatic parking path planning optimization of intelligent vehicles includes: using a search algorithm to plan a parking path of a vehicle from a current position to a target parking space; constructing a virtual parking environment, performing simulation tests in the virtual parking environment, simulating different parking scenarios and environmental changes, and verifying whether the parking path is safe; when the parking path is verified to be safe, controlling the vehicle to park according to the parking path; when the vehicle is parked, monitoring environmental changes in real time through sensors, and dynamically adjusting the parking path using path-finding algorithms and reinforcement learning. Use the A* algorithm and nonlinear programming method to perform preliminary path planning and optimization to ensure that the path is the shortest and the energy consumption is the lowest, verify the reliability of the algorithm through simulation tests and continuously optimize the path planning and control strategies, and monitor environmental changes in real time and dynamically adjust the path by integrating data from multiple sensors such as laser radars, cameras and ultrasonic radars to ensure that the vehicle safely avoids obstacles. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0036] Figure 1 A schematic diagram of a flow chart of a first embodiment of a method for optimizing automatic parking path planning for an intelligent vehicle provided by the present invention;

[0037] Figure 2 A schematic diagram of a flow chart of a second embodiment of a method for optimizing automatic parking path planning for an intelligent vehicle provided by the present invention;

[0038] Figure 3 A schematic diagram of a flow chart of a third embodiment of a method for optimizing automatic parking path planning for an intelligent vehicle provided by the present invention;

[0039] Figure 4 A schematic diagram of the structure of an embodiment of the intelligent vehicle automatic parking path planning and optimization system provided by the present invention;

[0040] Figure 5 Another structural schematic diagram of an embodiment of the intelligent vehicle automatic parking path planning and optimization system provided by the present invention;

[0041] Figure 6 This is a schematic diagram of the structure of an intelligent vehicle automatic parking path planning optimization device for the hardware operating environment involved in the intelligent vehicle automatic parking path planning optimization method in the embodiment of the present application.

[0042] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components in a certain specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0045] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features limited to "first" and "second" may explicitly or implicitly include at least one of the features. In addition, if "and / or" or "and / or" appears in the full text, its meaning includes three parallel solutions. Taking "A and / or B" as an example, it includes solution A, solution B, or solutions that satisfy both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in this field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0046] An intelligent vehicle is a comprehensive system that integrates environmental perception, planning and decision-making, and multi-level assisted driving. It uses a variety of technologies such as computers, modern sensors, information fusion, communications, artificial intelligence, and automatic control, and is a typical high-tech complex. Intelligent vehicles add advanced sensors (such as radars, cameras, etc.), controllers, actuators, and other devices to general vehicles to achieve information exchange with people, vehicles, roads, etc., so that the vehicle has intelligent environmental perception capabilities, can automatically analyze the safety and danger status of vehicle driving, and reach the destination according to people's wishes, ultimately achieving the purpose of replacing human operation.

[0047] Automatic parking refers to a technology that allows a car to park automatically without manual control. It usually includes environmental perception, parking space detection and recognition, etc. Automatic parking can be divided into semi-automatic parking and fully automatic parking according to the degree of automation. Semi-automatic parking requires the driver to control the speed of the vehicle, while the system is responsible for controlling the steering; fully automatic parking is more intelligent and can automatically control all operations such as speed, steering, acceleration and deceleration. The driver only needs to monitor inside or outside the car.

[0048] Traditional automatic parking systems have high requirements for parking spaces. For example, there cannot be obstacles such as telephone poles near the parking spaces. During the automatic parking process, if pedestrians or vehicles suddenly rush out from the rear, the system may not be able to respond in time, posing a safety hazard. For low obstacles (such as roadbeds, fire hydrants, etc.) or parking spaces with unclear markings, traditional automatic parking systems may not be able to accurately detect them. In rainy and snowy weather, the detection ability of traditional automatic parking systems will be affected due to unclear ground markings.

[0049] In summary, it is necessary to propose a path planning optimization method for automatic parking of intelligent vehicles to solve the above problems.

[0050] The main solution of this application is: using a search algorithm to plan a parking path for a vehicle from its current position to a target parking space; constructing a virtual parking environment, performing simulation tests in the virtual parking environment, simulating different parking scenarios and environmental changes, and verifying whether the parking path is safe; when the parking path is verified to be safe, controlling the vehicle to park according to the parking path; while the vehicle is parking, monitoring environmental changes in real time through sensors, and dynamically adjusting the parking path using a path-finding algorithm and reinforcement learning.

[0051] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an intelligent vehicle automatic parking path planning optimization device capable of realizing the above functions, etc. The following takes the parking controller as an example to illustrate this embodiment and the following embodiments.

[0052] Based on this, the present application embodiment provides a method for optimizing the automatic parking path planning of an intelligent vehicle, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the intelligent vehicle automatic parking path planning optimization method of the present application.

[0053] In this embodiment, the intelligent vehicle automatic parking path planning optimization method includes steps S10 to S40.

[0054] Step S10: Using a search algorithm to plan a parking path for the vehicle from the current position to the target parking space.

[0055] It should be noted that the A* algorithm is an algorithm that finds the lowest passing cost for a path with multiple nodes on a graph plane. The A* algorithm, full name A* search algorithm, calculates the evaluation function value of each node and selects the node with the smallest evaluation to expand until the target node is found or all nodes are searched. The search algorithm is the A* algorithm, which uses the A* algorithm to plan the optimal path from the current position of the vehicle to the target parking space in a static environment. The path will take into account the shortest distance, obstacle avoidance and lane restrictions of the parking lot, evaluate the energy consumption of the initially planned path, and use nonlinear programming methods to optimize the path so that the vehicle consumes the least energy during driving while maintaining a reasonable driving time.

[0056] It can be understood that the A* algorithm is used to plan the optimal path from the current position of the vehicle to the target parking space, as shown in the formula:

[0057] f(n)=g(n)+h(n)

[0058] Where f(n) is the estimated cost of the best path from the initial node to the target node through node n, g(n) is the actual cost from the initial node to node n, and h(n) is the estimated cost of the best path from node n to the target node.

[0059] Specifically, the energy consumption of the initially planned path is evaluated, and the path is optimized using a nonlinear programming method to minimize the energy consumption of the vehicle during driving while maintaining a reasonable driving time.

[0060] Optionally, the A* algorithm is applied for path planning while considering energy consumption optimization to achieve efficient and energy-saving path planning. In the preliminary path planning process, the A* algorithm not only considers the shortest distance from the initial node to the target node, but also estimates the optimal path cost from node n to the target node through the heuristic method h(n). Compared with the traditional path planning method, the planned path is more optimized, which not only considers the lowest energy consumption but also maintains a reasonable driving time. In addition, this step also incorporates practical factors such as obstacle avoidance and parking lot lane restrictions, so that the planned path is more in line with the actual application scenario, improving the feasibility and safety of the path.

[0061] Step S20: construct a virtual parking environment, perform simulation tests in the virtual parking environment, simulate different parking scenarios and environmental changes, and verify whether the parking path is safe.

[0062] It should be noted that simulation tests are carried out in a virtual parking environment to simulate different parking scenarios and environmental changes, verify the reliability and stability of the path planning algorithm, and identify potential problems and optimization opportunities by analyzing the simulation test data, continuously adjust and optimize the path planning and control strategies, and improve the overall performance of the system.

[0063] Specifically, verify the reliability and stability of the path planning algorithm to ensure that it can perform well in practical applications. Through this step, perform simulation tests to verify the reliability of the algorithm, and use the test data for continuous optimization to improve system performance.

[0064] It is understandable that after the vehicle parking path planning is completed, when there are multiple vehicles parking, the parking controller also needs to perform the following steps:

[0065] Step S21: When there are multiple vehicles parked at the same time in the parking lot, control the multiple vehicles to share information through the V2V communication protocol, and use game theory and multi-agent system algorithm to optimize the parking path of each vehicle.

[0066] It should be noted that when there are multiple vehicles parked at the same time in a parking lot, the vehicles share location information, speed information and parking intentions through the V2V communication protocol; game theory and multi-agent system algorithms are used to optimize path planning between multiple vehicles to avoid conflicts between vehicles.

[0067] Specifically, game theory and multi-agent system algorithms are applied to optimize multi-vehicle path planning. Game theory optimizes the overall parking efficiency by modeling and analyzing the decisions and actions between vehicles, reducing waiting time and resource waste, and improving overall parking efficiency.

[0068] It can be understood that the V2V (Vehicle to Vehicle) communication protocol, that is, the communication protocol between vehicles, refers to the protocol for vehicles to communicate and share information through wireless communication technology. The V2V communication protocol allows vehicles to directly exchange information, such as key data such as position, speed, acceleration, etc., so as to achieve the coordinated driving of vehicles and the optimization of intelligent transportation systems. Through the V2V protocol, vehicles can perceive the status of surrounding vehicles in real time, warn of potential collision risks in advance, and realize functions such as coordinated lane change and coordinated acceleration and deceleration, thereby improving road safety and traffic efficiency. Game theory, also known as game theory, countermeasure theory or game theory, mainly studies the interaction between formulated incentive structures, as well as mathematical theories and methods of phenomena with struggle or competition nature. Multi-agent system (MARL) algorithm, the MARL algorithm aims to enable each agent to learn how to maximize its own rewards while maintaining its contribution to the global reward maximization. When there are multiple vehicles parking at the same time in the parking lot, the vehicles share position information, speed information and parking intentions through the V2V communication protocol. Using game theory and multi-agent system algorithms, the path planning between multiple vehicles is optimized to avoid conflicts between vehicles and improve parking efficiency.

[0069] Step S30: When the parking path is verified to be safe, controlling the vehicle to park according to the parking path.

[0070] Step S40: When the vehicle is parking, the environment changes are monitored in real time by sensors, and the parking path is dynamically adjusted using a path-finding algorithm and reinforcement learning.

[0071] It is understandable that the vehicle's sensors continuously monitor changes in the surrounding environment during driving, such as the sudden appearance of pedestrians or other moving vehicles; if changes in the environment are detected, the D*Lite algorithm and reinforcement learning-based methods are immediately used to dynamically adjust the path. The D*Lite algorithm is a variant of the A* algorithm and is particularly suitable for path planning in dynamic environments. By dynamically adjusting the path, it ensures that the vehicle can safely avoid obstacles and successfully arrive at the target parking space. The introduction of the D*Lite algorithm and reinforcement learning methods enables the vehicle to adapt to dynamic environmental changes in real time and ensure safe obstacle avoidance.

[0072] It should be noted that the reinforcement learning-based method is a machine learning technology that focuses on how the agent optimizes its behavior strategy through trial and error learning in the environment in order to obtain the maximum long-term cumulative reward. When the vehicle is driving along the initially planned path, it continuously monitors changes in the surrounding environment, including detecting sudden pedestrians or other moving vehicles. If changes in the environment are detected, the D*Lite algorithm and the reinforcement learning-based method are immediately used to dynamically adjust the path to ensure that the vehicle can safely avoid obstacles and successfully arrive at the target parking space.

[0073] In this embodiment, the A* algorithm and nonlinear programming method are used for preliminary path planning and optimization to ensure the shortest path with the lowest energy consumption. The reliability of the algorithm is verified through simulation tests, and the path planning and control strategies are continuously optimized. By integrating data from multiple sensors such as lidar, cameras, and ultrasonic radar, environmental changes are monitored in real time and the path is dynamically adjusted to ensure that the vehicle safely avoids obstacles.

[0074] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 2 , Figure 2 A flow chart of the second embodiment of the method for optimizing the automatic parking path planning of an intelligent vehicle of the present application is provided. In the steps, the method for optimizing the automatic parking path planning of an intelligent vehicle includes:

[0075] Step S201: Collecting the parking environment information of the vehicle in real time through sensors, and using the Kalman filter algorithm to create a three-dimensional environment model around the vehicle; the three-dimensional environment model is used to construct a virtual parking environment.

[0076] It should be noted that various sensors on the vehicle, including lidar, cameras and ultrasonic radar, are activated to obtain real-time environmental information in the parking lot, including obstacles, lane lines, parking spaces, etc. The data obtained from different sensors are fused and processed, and the Kalman filter algorithm is used to create a three-dimensional environmental model around the vehicle to ensure the accuracy and completeness of the environmental information.

[0077] Specifically, the laser radar, camera and ultrasonic radar are activated to obtain real-time environmental data in the parking lot, including obstacle distance, lane line position and parking space vacancy status. The Kalman filter algorithm is used to fuse different sensor data to create a three-dimensional environmental model around the vehicle:

[0078] x k =Fx k-1 +Bu k +w k-1

[0079] In the formula, x k is the state estimate, F is the state transfer matrix, B is the control input matrix, u k is the control input, w k-1 It is process noise; verify the fused environmental data to ensure the accuracy and completeness of the data.

[0080] It can be understood that the Kalman filter algorithm is an algorithm that uses the linear system state equation to optimally estimate the system state through the system input and output observation data. It assumes that the system state is linear and that both the process noise and the observation noise are Gaussian distributed. The Kalman filter algorithm is introduced to fuse multi-sensor data, improve the accuracy and robustness of the environmental model, and verify the fused environmental data in real time to ensure the reliability of subsequent steps. Compared with traditional environmental perception and data acquisition methods, the Kalman filter algorithm fuses the data of laser radar, camera and ultrasonic radar to create a more accurate and real-time three-dimensional environment model around the vehicle. Among them, the application of the state transfer matrix F and the control input matrix B enables the system to better understand and predict changes in the vehicle's surrounding environment, thereby improving the accuracy and integrity of the data, which not only enhances the system's ability to adapt to environmental changes, but also provides more reliable basic data support for subsequent steps.

[0081] Step S202: construct a marker map according to the environmental information, and continuously update the marker map through a particle filtering algorithm.

[0082] It should be noted that SLAM technology is used to construct the identification map of the parking lot based on sensor data, identifying the locations of parking spaces, passages and obstacles. During vehicle driving, changes in the environment, including new obstacles or moving vehicles, are detected in real time. The current identification map is continuously updated through a particle filtering algorithm to keep the information up to date.

[0083] Specifically, SLAM (Simultaneous Localization and Mapping) technology, that is, simultaneous positioning and map building technology, is a technology that builds maps and locates simultaneously in an unknown environment in real time. SLAM technology mainly uses sensors to collect environmental information, and then integrates this information through algorithms to determine the location of the device in an unknown environment and build an environmental map. Using SLAM technology, the identification map of the parking lot is constructed based on sensor data, and the particle filter algorithm is used to update the map. The particle filter algorithm approximates the posterior probability distribution through a group of random samples (particles), identifies the location of parking spaces, passages and obstacles, establishes the topological structure of the parking environment, detects changes in the environment in real time, such as new obstacles or moving vehicles, and updates the current identification map through the particle filter algorithm to keep the information up to date. Through this step, the identification map is constructed and updated using SLAM technology and particle filter algorithm to reflect changes in the parking environment in real time.

[0084] It is understandable that the identification map is a high-precision map. The so-called high-precision map is actually a thematic map that serves the autonomous driving system in contrast to ordinary navigation electronic maps. The absolute position accuracy of the high-precision map is close to 1m, and the relative position accuracy is at the centimeter level, which can reach 10-20cm. It can accurately and comprehensively characterize road features and requires higher real-time performance, which is the most significant feature of the high-precision map.

[0085] Step S203: In the marked map, a machine learning model is used to analyze and match the target parking space that best meets the user's preferences based on the user's parking preferences.

[0086] It is understandable that when planning the route, the user's parking preferences are fully considered, such as being close to the exit, avoiding the wall, etc., and the machine learning model is used to analyze and match the parking location that best meets the user's preferences.

[0087] Specifically, a machine learning model refers to a mathematical model that can make predictions or inferences based on data. It automatically completes tasks such as classification, regression, clustering, generation, etc. by learning patterns in the data.

[0088] In this embodiment, an accurate three-dimensional environmental model is created by integrating data from multiple sensors such as lidar, camera, and ultrasonic radar, and an identification map is constructed based on the environmental information of the sensor. The identification map is continuously updated through a particle filtering algorithm. In the identification map, the A* algorithm and nonlinear programming method are used for preliminary path planning and optimization. At the same time, the user's parking preferences are taken into consideration and the parking location that best meets the user's needs is matched.

[0089] Based on the first embodiment and / or the second embodiment of the present application, in the third embodiment of the present application, the same or similar contents as those of the above-mentioned first and second embodiments can refer to the above introduction, and will not be repeated later. Figure 3 , Figure 3 A flow chart of the third embodiment of the method for optimizing automatic parking path planning for intelligent vehicles in this application.

[0090] Step S300: predicting the control method of the vehicle by combining the PID controller on the vehicle and the algorithm model, and controlling the vehicle to park according to the parking path.

[0091] It can be understood that the vehicle PID controller is an important tool used in vehicle control systems. It is based on the PID (proportional-integral-differential) control algorithm. By continuously calculating and adjusting the control quantity of the system, the vehicle can accurately and stably reach the desired trajectory or state.

[0092] Step S300 includes steps S301 to S303:

[0093] Step S301: Use the PID controller and MPC model on the vehicle to predict the control method of the vehicle, adjust the direction and speed of the vehicle in real time, and control the vehicle to travel according to the parking path.

[0094] It should be noted that the vehicle uses the PID controller and MPC model predictive control method to adjust the direction and speed in real time according to the planned path:

[0095]

[0096] Where u(t) is the control input, e(t) is the error, and K p , K i , K d They are proportional, integral, and differential coefficients respectively.

[0097] It is understandable that MPC (Model Predictive Control) is an advanced control strategy that combines prediction models, rolling optimization and feedback correction mechanisms to achieve optimal control of complex industrial processes. The basic principle of MPC is to use the prediction model to predict the future output state of the system based on the current system state and possible future inputs at each sampling moment, and determine the current control action by solving the optimal control problem in a finite time domain. Combining PID controller and MPC model predictive control method, accurate vehicle control and parking positioning can be achieved.

[0098] Step S302: When the vehicle arrives in front of the target parking space, the vehicle is controlled to complete parking using fuzzy logic control and deep learning control models.

[0099] It should be noted that when the vehicle approaches the target parking space, it switches to a low-speed precision control strategy, using fuzzy logic control and deep learning control models to enable the vehicle to accurately position and complete parking in a narrow space.

[0100] It is understandable that fuzzy logic control is a control method based on fuzzy logic theory. Fuzzy logic control is a science based on multi-valued logic, using fuzzy set methods to study fuzzy thinking, language forms and their laws. Fuzzy logic control uses fuzzy set theory, fuzzy language variables and fuzzy logic reasoning to imitate the uncertainty concept judgment and reasoning thinking of the human brain to reason and control the description system with unknown or uncertain models, as well as the control objects with strong nonlinearity and large lag. The deep learning control model aims to achieve intelligent control of complex systems through deep neural networks. The deep learning control model uses the powerful data processing and pattern recognition capabilities of deep neural networks (such as convolutional neural networks CNN, recurrent neural networks RNN, Transformer, etc.) to model and predict the input and output of the system, thereby achieving intelligent control of the system.

[0101] Step S303: During the vehicle parking process, multi-sensor redundancy and Markov chain method are used to perform safety monitoring and fault prediction.

[0102] It should be noted that the use of multi-sensor redundancy and Markov chain methods for safety monitoring and fault prediction ensures that the vehicle can operate safely in various emergency situations. The Markov chain model can predict the transition probability of the system state, thereby identifying potential safety risks in advance. Through this step, the machine learning model is used to match user parking preferences to improve user experience, and multi-sensor redundancy and Markov chain methods are used for safety monitoring and fault prediction to enhance system safety.

[0103] Specifically, the Markov Chain Method is a mathematical method based on the Markov Chain theory. The Markov Chain is a sequence of random variables X1, X2, X3, ... The range of these variables, that is, the set of all their possible values, is called the "state space", and the value of Xn is the state at time n. If the conditional probability distribution of Xn+1 for the past state is only a function of Xn, then the process is said to have the Markov property, that is, the future state depends only on the current state and has nothing to do with the past state. Use multi-sensor redundancy and Markov chain methods for safety monitoring and fault prediction to ensure that the vehicle can operate safely in various emergency situations. The Markov chain model can predict the transition probability of the system state, thereby identifying potential safety risks in advance. During the vehicle parking process, multi-sensor redundancy and Markov chain methods are used for safety monitoring and fault prediction to ensure that the vehicle can operate safely in various emergency situations.

[0104] In this embodiment, the vehicle is controlled according to the planned parking path, and the PID controller and MPC model predictive control method are used to adjust the direction and speed in real time. Fuzzy logic control and deep learning control models are used to enable the vehicle to accurately locate and complete parking in a narrow space. At the same time, multi-sensor redundancy and Markov chain methods are used for safety monitoring and fault prediction to ensure that the vehicle can operate safely in various emergency situations.

[0105] Reference Figure 4 , Figure 4 This is a schematic diagram of the structure of an embodiment of the intelligent vehicle automatic parking path planning and optimization system proposed by the present invention. Figure 4 An embodiment of the intelligent vehicle automatic parking path planning optimization system of the present invention is proposed.

[0106] In this embodiment, the intelligent vehicle automatic parking path planning and optimization system includes the intelligent vehicle automatic parking path planning and optimization method as described above, and the intelligent vehicle automatic parking path planning and optimization system includes: a path planning module 10, which is used to use a search algorithm to plan a parking path for the vehicle from the current position to the target parking space; a simulation module 20, which constructs a virtual parking environment, performs simulation tests in the virtual parking environment, simulates different parking scenarios and environmental changes, and verifies whether the parking path is safe; a control execution module 30, which is used to control the vehicle to park according to the parking path when the parking path is verified to be safe; a dynamic path adjustment module 40, which is used to monitor environmental changes in real time through sensors when the vehicle is parking, and dynamically adjust the parking path using a path-finding algorithm and reinforcement learning.

[0107] It should be noted that the path planning module 10 is used to generate the initial parking path, and find the optimal path in a fixed environment based on the A* algorithm and the Dijkstra algorithm. The dynamic path adjustment unit is used to adjust the path according to real-time environmental information, and uses the D*Lite algorithm and the reinforcement learning-based method to ensure the dynamic adaptability of the path. The path planning module 10 also includes an energy consumption evaluation unit, which is used to calculate the energy consumption of different paths based on the vehicle energy model, and use the Bayesian optimization algorithm to evaluate energy efficiency. The path energy optimization unit is used to optimize the path in combination with the path planning unit through a nonlinear programming method to minimize energy consumption.

[0108] It can be understood that the simulation module 20 also includes a parking scene simulation unit and a simulation data analysis unit. The parking scene simulation unit is used to use CARLA and Gazebo to simulate different parking scenes, test and optimize the performance of the path planning algorithm, and the simulation data analysis unit is used to use big data analysis tools to analyze simulation results, extract effective information for system optimization, and use clustering algorithms or decision trees to analyze simulation results.

[0109] Specifically, the control execution module 30 also includes a path tracking control unit and a low-speed fine control unit. The path tracking control unit is used to ensure that the vehicle moves smoothly along the planned path through a PID controller and an MPC. The low-speed fine control unit is used to perform fine control of the vehicle speed and steering during parking, and uses fuzzy logic control and a control model based on deep learning to achieve fine control in complex situations.

[0110] Optionally, the dynamic path adjustment module 40 is used to adjust the path according to real-time environmental information, using the D*Lite algorithm and a method based on reinforcement learning to ensure the dynamic adaptability of the path.

[0111] like Figure 5 As shown, the intelligent vehicle automatic parking path planning and optimization system also includes: an environmental perception module 50, which is used to collect environmental information of vehicle parking in real time through sensors, and use a Kalman filter algorithm to create a three-dimensional environmental model around the vehicle; the three-dimensional environmental model is used to construct a virtual parking environment; a map generation and update module 60, which is used to construct an identification map based on the environmental information, and continuously update the identification map through a particle filter algorithm; a user preference module 70, which is used to analyze and match the target parking space that best meets the user's preference in the identification map based on the user's parking preference using a machine learning model; a multi-vehicle collaborative communication module 80, when there are multiple vehicles parked at the same time in a parking lot, controls multiple vehicles to share information through a V2V communication protocol, and uses game theory and a multi-agent system algorithm to optimize the parking path of each vehicle.

[0112] It should be noted that the environmental perception module 50 also includes a sensor data fusion unit and an obstacle detection and classification unit. The sensor data fusion unit is used to generate a real-time three-dimensional model of the environment by fusing data from multiple sensors such as lidar, camera, and ultrasonic radar. This uses a Kalman filter algorithm to achieve sensor data fusion and noise suppression. The obstacle detection and classification unit is used to identify and classify obstacles in the surrounding environment, including vehicles, pedestrians, and pillars, through an object detection algorithm based on a convolutional neural network.

[0113] It can be understood that the map generation and update module 60 also includes a high-precision map creation unit and a dynamic map update unit. The high-precision map creation unit is used to generate a high-precision map of the vehicle's surroundings, including detailed information in the parking area, and is constructed and updated using a SLAM algorithm. The dynamic map update unit is used to update the map in real time to reflect changes in the surrounding environment, using a dynamic update algorithm based on particle filtering and combining sensor data to update the environmental model in real time.

[0114] Specifically, the user preference module 70 is used to learn the user's parking preference through a collaborative filtering method, including giving priority to a certain type of parking space or a certain parking method.

[0115] Optionally, the multi-vehicle collaborative communication module 80 also includes an inter-vehicle communication unit and a collaborative path planning unit. The inter-vehicle communication unit is used to enable multiple vehicles to collaboratively park in the same parking lot through the V2V communication protocol to avoid conflicts and improve efficiency. The collaborative path planning unit is used to implement a path planning algorithm based on game theory and a multi-agent system MAS to achieve synchronous path planning for multiple vehicles, avoid conflicts and optimize overall parking efficiency.

[0116] In this embodiment, the system performs sensor data fusion and obstacle detection and classification through the environmental perception module, the map generation and update module creates high-precision maps and updates identification maps, the path planning module performs static path planning, including an energy efficiency optimization module that performs energy consumption assessment and path energy optimization, the control execution module performs path tracking control, low-speed fine control and safety redundant fault detection, the dynamic path adjustment adjusts the parking path in real time, the user preference safety consideration module performs user preference learning to select parking spaces, the multi-vehicle collaborative communication module provides vehicle-to-vehicle communication and collaborative path planning, and the simulation module performs parking scene simulation and simulation data analysis.

[0117] The present application provides an intelligent vehicle automatic parking path planning optimization device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the intelligent vehicle automatic parking path planning optimization method in the above-mentioned embodiment one.

[0118] Reference below Figure 6 , which shows a schematic diagram of the structure of the intelligent vehicle automatic parking path planning and optimization device suitable for implementing the embodiment of the present application. The intelligent vehicle automatic parking path planning and optimization device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The intelligent vehicle automatic parking path planning and optimization device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0119] like Figure 6As shown, the intelligent vehicle automatic parking path planning optimization device may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 to the random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the intelligent vehicle automatic parking path planning optimization device are also stored. The processing device 1001, ROM1002 and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow the intelligent vehicle automatic parking path planning optimization device to communicate wirelessly or wired with other devices to exchange data. Although the figure shows an intelligent vehicle automatic parking path planning optimization device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0120] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0121] The intelligent vehicle automatic parking path planning and optimization device provided by the present application adopts the intelligent vehicle automatic parking path planning and optimization method in the above embodiment, which can enhance the safety of automatic parking and solve the technical problem of identifying obstacles that suddenly appear when parking. Compared with the prior art, the beneficial effects of the intelligent vehicle automatic parking path planning and optimization device provided by the present application are the same as the beneficial effects of the intelligent vehicle automatic parking path planning and optimization method provided by the above embodiment, and the other technical features in the intelligent vehicle automatic parking path planning and optimization device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0122] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0123] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0124] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the intelligent vehicle automatic parking path planning optimization method in the above-mentioned embodiment.

[0125] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0126] The computer-readable storage medium may be included in the intelligent vehicle automatic parking path planning and optimization device; or may exist independently without being assembled into the intelligent vehicle automatic parking path planning and optimization device.

[0127] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0128] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0129] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0130] The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned intelligent vehicle automatic parking path planning optimization method, which can enhance the safety of automatic parking and identify obstacles that suddenly appear when parking. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the intelligent vehicle automatic parking path planning optimization method provided in the above-mentioned embodiment, and will not be repeated here.

[0131] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for optimizing automatic parking path planning for intelligent vehicles, characterized in that: The path planning optimization method for automatic parking of an intelligent vehicle comprises: Use a search algorithm to plan a parking path from the vehicle's current position to the target parking space; Constructing a virtual parking environment, performing simulation tests in the virtual parking environment, simulating different parking scenarios and environmental changes, and verifying whether the parking path is safe; When the parking path is verified to be safe, controlling the vehicle to park according to the parking path; When the vehicle is parking, environmental changes are monitored in real time through sensors, and the parking path is dynamically adjusted using a path-finding algorithm and reinforcement learning.

2. The intelligent vehicle automatic parking path planning optimization method according to claim 1, characterized in that: Before the step of using the search algorithm to plan the parking path of the vehicle from the current position to the target parking space, the method includes: The environmental information of the vehicle parking is collected in real time by sensors, and a three-dimensional environmental model around the vehicle is created using a Kalman filter algorithm; the three-dimensional environmental model is used to construct a virtual parking environment.

3. The intelligent vehicle automatic parking path planning optimization method according to claim 1, characterized in that: Before the step of using the search algorithm to plan the parking path of the vehicle from the current position to the target parking space, the method includes: Constructing a marker map according to the environmental information, and continuously updating the marker map through a particle filtering algorithm; In the marked map, a machine learning model is used to analyze and match the target parking space that best meets the user's preferences based on the user's parking preferences.

4. The intelligent vehicle automatic parking path planning optimization method according to claim 1, characterized in that: After the step of using the search algorithm to plan a parking path from the current position of the vehicle to the target parking space, the method further comprises: When there are multiple vehicles parking at the same time in the parking lot, the multiple vehicles are controlled to share information through the V2V communication protocol, and the parking path of each vehicle is optimized by using game theory and multi-agent system algorithm.

5. The intelligent vehicle automatic parking path planning optimization method according to claim 1, characterized in that: When verifying the safety of the parking path, the step of monitoring environmental changes in real time through sensors and dynamically adjusting the parking path using a path-finding algorithm and reinforcement learning includes: A control method for predicting the vehicle is implemented by combining a PID controller on the vehicle and an algorithm model, and controlling the vehicle to park according to the parking path.

6. The intelligent vehicle automatic parking path planning optimization method according to claim 5, characterized in that: The step of predicting the control method of the vehicle by combining the PID controller on the vehicle and the algorithm model, and controlling the vehicle to park according to the parking path, comprises: Using the PID controller and the MPC model on the vehicle to predict the control method of the vehicle, adjusting the direction and speed of the vehicle in real time, and controlling the vehicle to travel according to the parking path; When the vehicle arrives in front of the target parking space, the vehicle is controlled to complete parking using fuzzy logic control and deep learning control models; During the vehicle parking process, multi-sensor redundancy and Markov chain methods are used for safety monitoring and fault prediction.

7. An intelligent vehicle automatic parking path planning optimization system, characterized in that: The intelligent vehicle automatic parking path planning optimization system comprises the intelligent vehicle automatic parking path planning optimization method according to claim 1, and the intelligent vehicle automatic parking path planning optimization system comprises: A path planning module is used to plan a parking path from the current position of the vehicle to the target parking space using a search algorithm; A simulation module is used to construct a virtual parking environment, perform simulation tests in the virtual parking environment, simulate different parking scenarios and environmental changes, and verify whether the parking path is safe; A control execution module, configured to control the vehicle to park according to the parking path when the parking path is verified to be safe; The dynamic path adjustment module is used to monitor environmental changes in real time through sensors when the vehicle is parking, and dynamically adjust the parking path using a path-finding algorithm and reinforcement learning.

8. The intelligent vehicle automatic parking path planning optimization system according to claim 7, characterized in that: The intelligent vehicle automatic parking path planning optimization system also includes: An environment perception module is used to collect the environment information of the vehicle parking in real time through sensors, and use the Kalman filter algorithm to create a three-dimensional environment model around the vehicle; the three-dimensional environment model is used to construct a virtual parking environment; A map generation and updating module, used to construct a marking map according to the environmental information, and continuously update the marking map through a particle filtering algorithm; A user preference module, configured to analyze and match the target parking space that best meets the user's preference in the marked map according to the user's parking preference using a machine learning model; The multi-vehicle cooperative communication module controls multiple vehicles to share information through the V2V communication protocol when there are multiple vehicles parked at the same time in the parking lot, and uses game theory and multi-agent system algorithm to optimize the parking path of each vehicle.

9. An intelligent vehicle automatic parking path planning and optimization device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the intelligent vehicle automatic parking path planning optimization method as described in any one of claims 1 to 6.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the intelligent vehicle automatic parking path planning optimization method as described in any one of claims 1 to 6 are implemented.

Images