Automatic parking control method, device and equipment
By using wavelet transform algorithm to process sensor data, the shortcomings of autonomous driving systems in processing multi-scale features and non-stationary signals are addressed, thereby improving the accuracy and response speed of automatic parking.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- MOMENTA (SUZHOU) TECHNOLOGY CO LTD
- Filing Date
- 2024-10-24
- Publication Date
- 2026-04-24
AI Technical Summary
Existing autonomous driving systems struggle to effectively extract multi-scale features and non-stationary signals during data processing, especially in obstacle detection, resulting in low accuracy in automatic parking.
The sensor data is processed using wavelet transform algorithm, and the features of the vehicle environment data are extracted through wavelet coefficient analysis to generate parking control commands.
It improves the recognition rate of autonomous driving systems and the flexibility and accuracy of obstacle avoidance strategies, enabling the accurate extraction of useful features in high-noise environments and achieving faster response.
Smart Images
Figure CN121912947A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of autonomous driving, specifically to an automatic parking control method, apparatus, and device. Background Technology
[0002] During automatic parking, sensors acquire real-time environmental data about the vehicle's surroundings. The autonomous driving system uses this data to determine control signals for automatic parking. Existing data processing methods struggle to effectively extract features from sensor data, particularly regarding multi-scale feature variations and non-stationary signals. Furthermore, current methods cannot perform multi-scale feature analysis on captured time-series data, making it difficult to identify obstacles of varying sizes and distances. Summary of the Invention
[0003] In view of this, this application provides an automatic parking control method, apparatus, and device to help solve the problem of low data processing accuracy in existing autonomous driving systems.
[0004] In a first aspect, embodiments of this application provide an automatic parking control method, including:
[0005] Acquire vehicle environmental data collected by sensors;
[0006] The vehicle environment data is processed using a wavelet transform algorithm to obtain wavelet coefficients.
[0007] Parking control commands are determined based on the wavelet coefficients, and these commands are used to instruct the vehicle to operate autonomously.
[0008] In one optional embodiment, the vehicle environment data includes: radar data, ultrasonic sensor data, camera data, vehicle status data, or weather data.
[0009] In an optional embodiment, before processing the vehicle environment data using the wavelet transform algorithm to obtain wavelet coefficients, the method further includes:
[0010] Remove outliers from the vehicle environment data;
[0011] The vehicle environment data is then subjected to smoothing and filtering processing.
[0012] In one optional embodiment, the process of processing the vehicle environment data based on the wavelet transform algorithm to obtain wavelet coefficients includes:
[0013] Determine a wavelet basis that matches the vehicle environment data;
[0014] Determine the number of decomposition layers that match the vehicle environment data;
[0015] The wavelet coefficients are obtained by performing discrete wavelet transform on the vehicle environment data based on the wavelet basis and the number of decomposition levels.
[0016] In one optional embodiment, determining the wavelet basis that matches the vehicle environment data includes:
[0017] Based on the data type of the vehicle environment data, a wavelet basis matching the vehicle environment data is determined.
[0018] In one optional embodiment, determining the number of decomposition layers matching the vehicle environment data includes:
[0019] The number of decomposition layers that match the vehicle environment data is determined based on the characteristics of the signal, the objective of the analysis, the type of wavelet transform, or the length of the signal.
[0020] In one optional embodiment, determining the parking control command based on the wavelet coefficients includes:
[0021] Determine the wavelet eigenvectors based on the wavelet coefficients;
[0022] Parking control commands are determined based on the wavelet feature vectors.
[0023] Secondly, embodiments of this application provide an automatic parking control device, comprising:
[0024] The acquisition module is used to acquire a pre-configured device driver identifier when a power-off command for the display is detected.
[0025] The processing module is configured to unload the external driver device associated with the pre-configured device driver identifier;
[0026] The control module is used to turn off the power supply to the external drive device that has been unloaded.
[0027] Thirdly, embodiments of this application provide an electronic device, including a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the electronic device is triggered to execute the method described in any of the first aspects above.
[0028] Fourthly, embodiments of this application provide a computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the method described in any of the first aspects.
[0029] Fifthly, embodiments of this application provide a computer program product comprising executable instructions that, when executed on a computer, cause the computer to perform the method described in any of the first aspects.
[0030] The scheme provided in this application acquires vehicle environmental data collected by sensors, processes the data using a wavelet transform algorithm to obtain wavelet coefficients, and determines parking control commands based on these coefficients. These commands are used to instruct the vehicle to automate its driving process. Utilizing the multi-scale analysis capabilities of wavelet transform, relevant obstacle features are extracted by decomposing the signal at different scale levels. This allows the autonomous driving system to better adapt to obstacle detection of varying sizes and distances, enhancing the flexibility and accuracy of obstacle avoidance strategies. Applying wavelet transform as the primary tool to process and analyze vehicle environmental data collected during autonomous driving extracts useful time-frequency feature information, improving the autonomous driving system's environmental recognition rate. Even in high-noise environments, useful features can be extracted from noise, resulting in more accurate and faster real-time parking. Attached Figure Description
[0031] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments 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.
[0032] Figure 1 A flowchart illustrating an automatic parking control method provided in an embodiment of this application;
[0033] Figure 2 A flowchart illustrating another automatic parking control method provided in an embodiment of this application;
[0034] Figure 3 A schematic diagram of an automatic parking control device provided in an embodiment of this application;
[0035] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation
[0036] To better understand the technical solution of this application, the embodiments of this application will be described in detail below with reference to the accompanying drawings.
[0037] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.
[0038] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.
[0039] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.
[0040] Autonomous vehicles are equipped with multiple sensors, such as radar, ultrasonic sensors, and cameras. During automatic parking, these sensors acquire real-time environmental data. Based on this data, the autonomous driving system determines the vehicle's status, including its speed, acceleration, position, and distance to surrounding obstacles. The system then outputs appropriate parking control commands to enable the vehicle to perform the parking maneuver.
[0041] This application provides an automatic parking control method that extracts features from vehicle environmental data based on wavelet transform to output parking control commands. This method improves the system's environmental recognition rate and can still extract useful features from noise in high-noise environments, making the parking strategy more accurate and the real-time response faster. In automatic parking scenarios, the advantages of wavelet transform include the following aspects:
[0042] (1) Multiscale analysis: Wavelet transform can provide localized signal analysis in both the time and frequency domains, making it very effective in extracting features with different scale variations. Since many signals contain structures at multiple scales, wavelet transform can accurately capture these structures, thus providing rich information.
[0043] (2) Good time-frequency localization: Compared with Fourier transform (or other algorithms), wavelet transform is more suitable for analyzing signals with non-stationary or instantaneous characteristics. For the above-mentioned vehicle environment data, such as ultrasonic waves and radar, they often capture sudden obstacles or rapidly changing scenes. In this case, wavelet transform can effectively identify and locate these features.
[0044] (3) Effective data compression: Wavelet transform can represent a signal as a small number of important wavelet coefficients, thereby achieving effective data compression. In feature extraction, this means that the dimensionality of the data can be reduced, the computational burden can be decreased, and the key information required for decision-making can be preserved.
[0045] (4) Noise reduction capability: Wavelet transform naturally has noise reduction characteristics because it can separate the high-frequency noise components of the signal. By selecting an appropriate threshold to process the wavelet coefficients, noise can be effectively removed while retaining useful signal information.
[0046] (5) Adapting to different signal characteristics: By changing the wavelet basis, it is possible to optimize for specific types of signals or features. For example, the Daubechies wavelet with more vanishing moments can be selected to process smooth signals, or the Haar wavelet can be used to detect abrupt changes.
[0047] (6) Simplify the model and algorithm: In some cases, the coefficients after wavelet transform are used as input features. Since the signal has been transformed into a set of representative feature coefficients, the complexity of subsequent machine learning or pattern recognition algorithms can be simplified.
[0048] (7) Fast algorithms are available: Discrete Wavelet Transform (DWT) can be implemented by Fast Wavelet Transform (FWT) algorithm. This algorithm is very fast and occupies little memory, making it suitable for real-time or near-real-time application scenarios.
[0049] (8) Boundary processing: When processing signals of finite length, wavelet transform provides a more flexible boundary extension method compared to other transform methods (such as Fourier transform), which helps to reduce the interference of boundary effects on feature extraction.
[0050] (9) Robustness: Wavelet transform is generally robust to small disturbances or outliers in the signal, so it can generate reliable features when the input data contains slight noise.
[0051] Figure 1 This is a flowchart illustrating an automatic parking control method provided in an embodiment of this application. This method can be applied to autonomous driving systems, see below. Figure 1 The method may include:
[0052] Step 101: Acquire vehicle environmental data collected by sensors;
[0053] Step 102: Process the vehicle environment data based on the wavelet transform algorithm to obtain wavelet coefficients;
[0054] Step 103: Determine the parking control command based on the wavelet coefficients. The parking control command is used to instruct the vehicle to perform automatic parking operation.
[0055] When the vehicle is in an automatic parking scenario, sensors installed on the vehicle collect environmental data about the surrounding environment and upload it to the system. Optionally, the environmental data collected by the sensors may include: (1) ultrasonic sensor data, which collects the distance and timestamps of obstacles in various directions; (2) radar data, which collects the speed, angle, distance, and timestamps of targets; and (3) camera data, which collects image data and detects the object type, confidence level, and their position information in the image. In addition, the system can also acquire vehicle status data and environmental condition data. Vehicle status data includes: current vehicle speed, steering angle, gear, braking and accelerator pedal status. Environmental condition data includes: current lighting conditions, weather data, and road conditions. All of the above data can be considered as vehicle environmental data.
[0056] Before performing wavelet transform on the above data, the system needs to perform data preprocessing to organize the original data for subsequent processing.
[0057] In one alternative embodiment, the data preprocessing method may include deleting outliers. For example, for ultrasonic sensors, the system typically sets a reasonable range (e.g., 0.2 meters to 5 meters), and readings outside this range are likely outliers. The system will delete the data determined to be outliers.
[0058] In one optional embodiment, the data preprocessing method may further include: data smoothing filtering. For example, radar data may fluctuate significantly and exhibit instantaneous jumps; the system may use methods such as low-pass filters or moving averages to smooth these instantaneous jumps in the radar data.
[0059] In one alternative embodiment, the data preprocessing method may further include: synchronizing timestamps. The system adjusts the timestamps of the data to ensure that the data from all sensors are based on the same reference time point, in order to enable accurate spatiotemporal analysis.
[0060] In one optional embodiment, the data preprocessing method may further include: encoding conversion. For example, if the image data captured by the camera is Base64 encoded, the system converts the Base64 encoded image data into an actual image matrix for further analysis.
[0061] In one alternative embodiment, the data preprocessing method may further include: data format standardization. The system unifies all vehicle environmental data to the same digital format, for example, uniformly using floating-point numbers to represent distance and speed.
[0062] Through the above preprocessing, errors in the vehicle environmental data have been corrected. For example, abnormal ultrasonic sensor readings (such as negative distance values) have been removed, instantaneous jumps in radar data have been eliminated, and camera data has been converted into an image matrix through time synchronization and Base64 decoding, represented by an RGB pixel matrix. Vehicle status data and environmental condition data are obtained directly from the vehicle itself or the network, ensuring high accuracy.
[0063] The system performs wavelet transform on the preprocessed vehicle environment data to obtain wavelet coefficients. These coefficients are then analyzed and their features extracted to output parking control commands.
[0064] In one optional embodiment, the specific steps of the system processing the vehicle environment data based on the wavelet transform algorithm may include: determining a wavelet basis that matches the vehicle environment data; determining a decomposition level that matches the vehicle environment data; and performing a discrete wavelet transform on the vehicle environment data based on the wavelet basis and the decomposition level to obtain wavelet coefficients. The selection of the wavelet basis and the decomposition level depends on the specific application and needs to balance the efficiency of decomposition and the accuracy of reconstruction.
[0065] Different wavelet bases have different time and frequency characteristics, which determine their applicability in various application scenarios. Common ones include: (1) Haar wavelet: the simplest wavelet base, with a piecewise constant shape, good time resolution, suitable for analyzing abrupt signals, but poor frequency resolution. (2) Daubechies wavelet: has a smoother waveform and a longer support interval, can capture more detailed features of the signal, and is suitable for occasions with high requirements for the local features of the signal, such as image processing and speech signal analysis. (3) Morlet wavelet: a continuous wavelet with good time and frequency localization characteristics, often used in time-frequency analysis of signals, such as earthquake data analysis and music signal processing. (4) Coiflets wavelet: has good symmetry and multiple vanishing moments, suitable for processing signals with multi-scale structures or requiring symmetrical analysis, and is often used for signal denoising and data compression. (5) Symlets wavelet: a modified version of the Daubechies wavelet, more symmetrical, Symlets is a better choice when signal processing requires a more symmetrical wavelet shape.
[0066] In the automatic parking scenario of this application embodiment, the system needs to select a suitable wavelet basis based on the characteristics of the vehicle environment data in the parking environment. Considering that the vehicle environment data may include rapid changes in obstacle edges and smooth background information, selecting a wavelet basis with good temporal localization characteristics and the ability to effectively detect abrupt changes may be more suitable. For example, the Daubechies wavelet can provide sufficient smoothness and detail features, making it suitable for processing signals in complex environments and thus suitable as a wavelet basis for the automatic parking scenario.
[0067] In one alternative embodiment, the number of wavelet decomposition layers is typically determined by the following factors: (1) Signal characteristics: If the original signal contains multi-scale structures or details, more decomposition layers may be needed to capture this information. (2) Analysis objectives: The time and frequency resolution requirements during the analysis process; if more detailed high-frequency information is needed, a deeper decomposition is required; conversely, if the focus is mainly on low-frequency trends, fewer layers are needed. (3) Type of wavelet transform: Continuous wavelet transform (CWT) does not limit the number of layers, while the number of layers in discrete wavelet transform is constrained by the signal length. (4) Signal length: For discrete wavelet transform, the maximum number of decomposition layers that can be performed depends on the signal length. Theoretically, for a signal of length N, the maximum number of decomposition layers is log2(N). In practical applications, the number of layers is usually chosen to be much less than this limit.
[0068] After determining the wavelet basis and the number of decomposition levels, the system can perform continuous wavelet transform or discrete wavelet transform on the vehicle environment data. The automatic parking scenario in this application prioritizes computational efficiency, real-time performance, and the ability to process discrete signals. Discrete wavelet transform has significant advantages in these aspects, specifically considering the following factors: (1) Computational efficiency: DWT is generally more computationally efficient because it uses fixed wavelets and scaling functions for multi-level decomposition. This method can utilize fast algorithms, such as Fast Wavelet Transform (FWT), to significantly reduce computation time. (2) Real-time requirements: Vehicle parking systems typically require real-time or near-real-time data processing to react quickly. DWT, due to its lower computational complexity, helps meet this requirement. (3) Discrete data: Vehicle environment data is inherently discrete, with sensors sampling at fixed intervals. DWT is naturally suitable for processing this type of discrete data. (4) Data volume management: DWT facilitates data volume management. By selecting an appropriate number of decomposition levels, the acquisition of detailed information and the growth of data volume can be balanced. (5) Multi-scale feature extraction: DWT provides a natural and effective way to capture the features of a signal at different scales, which is very important in automatic parking scenarios, such as when detecting obstacles of different sizes.
[0069] The system performs discrete wavelet transform on the vehicle's environmental data to obtain wavelet coefficients. These decomposed wavelet coefficients represent information about the signal at different scales and locations, from which features helpful for automatic parking can be extracted. For example, high-frequency wavelet coefficients can reveal the precise location of obstacles, while low-frequency coefficients show the overall distance range.
[0070] The system analyzes wavelet coefficients to determine eigenvectors. Then, combining this with artificial intelligence algorithms (such as machine learning models) or decision logic, it calculates the optimal parking path and corresponding obstacle avoidance strategy. Based on the parking strategy, specific parking control commands are generated, such as steering angle, acceleration, and braking. These parking control commands are directly received by the vehicle's execution system, which then controls the vehicle to complete the automatic parking maneuver.
[0071] In this embodiment, wavelet transform can effectively extract useful features from noisy vehicle environment data, improving the system's environmental recognition rate. Wavelet transform allows signal analysis at different scales, enabling the autonomous driving system to better adapt to obstacle detection of varying sizes or distances. Furthermore, wavelet transform possesses excellent time-frequency localization properties, which can help the autonomous driving system more accurately determine the location of obstacles and their corresponding time points, thereby enabling faster and more accurate responses.
[0072] The wavelet transform process is further illustrated with specific embodiments. (Refer to...) Figure 2 The specific steps of wavelet transform may include:
[0073] Step 201: Determine the wavelet basis and the number of decomposition layers;
[0074] Step 202, initialize the signals;
[0075] Step 203, wavelet basis calculation;
[0076] Step 204, construct the feature vector;
[0077] Step 205: Output parking control command;
[0078] The system can select the Daubechies wavelet as the wavelet basis and determine the appropriate number of decomposition levels. Then, the system initializes the signal, creating a numerical array representation of the signal as input for the wavelet transform. For example, the array `signal = [0.8, 1.5, 0.5]`. Assuming there is data from three sensors, `signal = [0.8, 1.5, 0.5]`, the system can pad or interpolate before signal processing to suit the wavelet transform requirements. For example, the system adds one data point to make it four points, `signal = [0.8, 1.5, 0.5, 1.0]` (the last point is obtained through interpolation). Now, the first-level discrete wavelet transform (DWT) is performed: the average and difference coefficients of the first level are calculated using the following formulas:
[0079]
[0080] Where i is the index of each sample group (in this case, i = 0, 1), and we operate on adjacent points in the signal in pairs. The system calculates the DWT of the above signal using the above formula, and obtains the approximation coefficient Approximation = [1.15, 0.75] and the detail coefficient Detail = [-0.35, -0.25].
[0081] The system can then construct a feature vector based on the above. In the automatic parking scenario, since the detail coefficients (Detail) contain high-frequency information, they are more suitable for detecting edges and abrupt changes. Therefore, the system can construct a feature vector: feature_vector = Detail = [-0.35, -0.25]. This feature vector can provide information about the edges of obstacles detected by the sensors.
[0082] Based on artificial intelligence algorithms, the system can output parking control commands.
[0083] The method provided in this application utilizes the multi-scale analysis capability of wavelet transform to extract relevant features of obstacles by decomposing the signal at different scale levels. This enables the autonomous driving system to better adapt to obstacle detection of different sizes or distances, enhancing the flexibility and accuracy of obstacle avoidance strategies. By applying wavelet transform as the primary tool to process and analyze vehicle environmental data collected during autonomous driving, useful time-frequency feature information is extracted, improving the autonomous driving system's environmental recognition rate. Even in high-noise environments, useful features can still be extracted from noise, making autonomous driving more accurate and reacting faster in real time.
[0084] Figure 3 This is a schematic diagram of an automatic parking control device provided in an embodiment of this application. The device can be deployed in electronic devices, such as… Figure 4As shown, the device may include: an acquisition module 310, a processing module 320, and a control module 330.
[0085] The acquisition module 310 is used to acquire a pre-configured device driver identifier when a power-off command for the display screen is detected.
[0086] Processing module 320 is used to unload the external driver device associated with the pre-configured device driver identifier based on the device driver identifier;
[0087] The control module 330 is used to turn off the power supply to the external drive device that has been unloaded.
[0088] For the specific process, please refer to the description in the flowchart above.
[0089] Corresponding to the above embodiments, this application also provides an electronic device. Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device 400 may include a processor 401, a memory 402, and a communication unit 403. These components communicate through one or more buses. Those skilled in the art will understand that the structure of the electronic device shown in the figure does not constitute a limitation on the embodiment of this application. It may be a bus-shaped structure or a star-shaped structure, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.
[0090] The communication unit 403 is used to establish a communication channel, enabling the electronic device to communicate with other devices. It receives user data from other devices or sends user data to other devices.
[0091] The processor 401 serves as the control center of the electronic device, connecting various parts of the device via interfaces and lines. It executes software programs, instructions, and / or modules stored in the memory 402, and calls data stored in the memory to perform various functions and / or process data. The processor may be composed of integrated circuits (ICs), such as a single packaged IC or multiple packaged ICs with the same or different functions connected together. For example, the processor 401 may consist only of a central processing unit (CPU). In this embodiment, the CPU may have a single processing core or include multiple processing cores.
[0092] The memory 402 is used to store the execution instructions of the processor 401. The memory 402 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.
[0093] When the execution instructions in memory 402 are executed by processor 401, the electronic device 400 is able to perform operations. Figure 1 Some or all of the steps in the illustrated embodiments.
[0094] In a specific implementation, this application also provides a computer storage medium, wherein the computer storage medium may store a program, which, when executed, may include some or all of the steps of the various embodiments of the automatic parking control method provided in this application. The storage medium may be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.
[0095] In a specific implementation, this application also provides a computer program product, wherein the computer program product includes executable instructions, which, when executed on a computer, cause the computer to perform some or all of the steps in various embodiments of the automatic parking control method provided in this application.
[0096] This application also provides a non-transitory computer-readable storage medium that stores computer instructions, which cause the computer to execute the automatic parking control method provided in this application.
[0097] The aforementioned non-transitory computer-readable storage medium may be any combination of one or more computer-readable media. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or flash memory, optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in connection with an instruction execution system, apparatus, or device.
[0098] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including—but not limited to—electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of transmitting, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.
[0099] The program code contained on a computer-readable medium may be transmitted using any suitable medium, including—but not limited to—wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.
[0100] Those skilled in the art will clearly understand that the techniques in the embodiments of this application can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of this application, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application or some parts of the embodiments.
[0101] The same or similar parts between the various embodiments in this specification can be referred to mutually. In particular, the device embodiments and terminal embodiments are basically similar to the method embodiments, so the description is relatively simple, and the relevant parts can be referred to the description in the method embodiments.
Claims
1. An automatic parking control method, characterized in that, include: Acquire vehicle environmental data collected by sensors; The vehicle environment data is processed using a wavelet transform algorithm to obtain wavelet coefficients. The parking control command is determined based on the wavelet coefficients, and the parking control command is used to instruct the vehicle to perform automatic parking operation.
2. The method according to claim 1, characterized in that, The vehicle environmental data includes: radar data, ultrasonic sensor data, camera data, vehicle status data, or weather data.
3. The method according to claim 1, characterized in that, Before processing the vehicle environment data using the wavelet transform algorithm to obtain wavelet coefficients, the method further includes: Remove outliers from the vehicle environment data; The vehicle environment data is then subjected to smoothing and filtering processing.
4. The method according to claim 1, characterized in that, The wavelet transform algorithm is used to process the vehicle environment data to obtain wavelet coefficients, including: Determine a wavelet basis that matches the vehicle environment data; Determine the number of decomposition layers that match the vehicle environment data; The wavelet coefficients are obtained by performing discrete wavelet transform on the vehicle environment data based on the wavelet basis and the number of decomposition levels.
5. The method according to claim 4, characterized in that, The determination of the wavelet basis that matches the vehicle environment data includes: Based on the data type of the vehicle environment data, a wavelet basis matching the vehicle environment data is determined.
6. The method according to claim 4, characterized in that, Determining the number of decomposition layers that match the vehicle environment data includes: The number of decomposition layers that match the vehicle environment data is determined based on the characteristics of the signal, the objective of the analysis, the type of wavelet transform, or the length of the signal.
7. The method according to claim 1, characterized in that, The step of determining the parking control command based on the wavelet coefficients includes: Determine the wavelet eigenvectors based on the wavelet coefficients; Parking control commands are determined based on the wavelet feature vectors.
8. An automatic parking control device, characterized in that, include: The acquisition module is used to acquire vehicle environmental data collected by sensors; The processing module is used to process the vehicle environment data based on the wavelet transform algorithm to obtain wavelet coefficients; The determination module is used to determine parking control commands based on the wavelet coefficients, and the parking control commands are used to instruct the vehicle to drive automatically.
9. An electronic device, characterized in that, The device includes a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the electronic device performs the method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the method according to any one of claims 1 to 7.