An automatic parking control method, device, equipment and storage medium

Abstract

The present invention discloses an automatic parking control method, device, equipment and storage medium. When the driving speed is less than a preset speed, the radar data collected by a millimeter-wave radar is converted into point cloud data, and target detection is performed based on the point cloud data to detect whether there is a parking stopper in the current environment. If there is a parking stopper in the current environment, vehicle state data is obtained, and the driver's intention to move the vehicle is determined based on the vehicle state data. The state of the automatic parking system is controlled based on the driver's intention to move the vehicle, avoiding the problems of reduced driving smoothness and easy occurrence of scratching accidents caused by the frequent activation of the automatic parking system during the vehicle moving process, and improving driving smoothness and safety.

Description

Technical Field

[0001] The present invention relates to parking control technology, and in particular, to an automatic parking control method, device, equipment, and storage medium. Background Art

[0002] The automatic parking function is realized by the coordinated work of the vehicle's braking system, electronic control system, and sensors. The sensors detect the driving state and stationary state of the vehicle. When the vehicle stops and meets certain conditions, the electronic control system controls the braking system to apply appropriate braking force to the wheels to keep the vehicle stationary. When it detects that the driver has the intention to start, the braking force will be quickly released.

[0003] The existing vehicle automatic parking function mainly turns on and off the automatic parking system through a switch. After turning on the automatic parking system, during driving, the driver activates the automatic parking system by stepping on the brake pedal and releases the automatic parking system by stepping on the accelerator pedal.

[0004] However, in scenarios such as parking in or out of a garage, the driver needs to actively turn off the automatic parking system. Otherwise, during this process, the automatic parking system will be frequently activated, reducing driving fluency. In addition, releasing the automatic parking system requires stepping on the accelerator pedal. In a narrow parking lot during a parking maneuver, if the driver is not proficient in controlling the accelerator pedal, it is easy to cause scratching accidents, resulting in unnecessary losses and risks. Summary of the Invention

[0005] The present invention provides an automatic parking control method, device, equipment, and storage medium to improve driving fluency and safety in parking maneuver scenarios.

[0006] In a first aspect, the present invention provides an automatic parking control method, including:

[0007] When the driving speed is less than a preset speed, convert the radar data collected by the millimeter-wave radar into point cloud data;

[0008] Based on the point cloud data, perform target detection to detect whether there is a parking stopper in the current environment;

[0009] If there is a parking stopper in the current environment, obtain the vehicle status data;

[0010] Based on the vehicle status data, determine the driver's parking maneuver intention;

[0011] Based on the driver's parking maneuver intention, control the state of the automatic parking system.

[0012] Optionally, converting the radar data collected by the millimeter-wave radar into point cloud data includes:

[0013] Perform a distance - dimension Fourier transform on the radar data to obtain the distance between the target point and the millimeter - wave radar;

[0014] Perform an angle - dimension Fourier transform on the radar data to obtain the horizontal angle and the pitch angle between the target point and the millimeter - wave radar;

[0015] Calculate the three - dimensional coordinates of the target point in the Cartesian coordinate system based on the distance, horizontal angle, and pitch angle between the target point and the millimeter - wave radar, and form three - dimensional point cloud data.

[0016] Optionally, calculate the three - dimensional coordinates of the target point in the Cartesian coordinate system based on the distance, horizontal angle, and pitch angle between the target point and the millimeter - wave radar. The calculation formula is as follows:

[0017] X = R·cos(φ)·sin(θ);

[0018] Y = R·cos(φ)·cos(θ);

[0019] Z = R·sin(φ);

[0020] Wherein, R is the distance between the target point and the millimeter - wave radar, φ is the horizontal angle between the target point and the millimeter - wave radar, and θ is the pitch angle between the target point and the millimeter - wave radar.

[0021] Optionally, perform target detection based on the point cloud data to detect whether there is a parking stopper in the current environment, including:

[0022] Perform a spatial transformation on the point cloud data to align the point cloud data in three - dimensional space to obtain the original features;

[0023] Perform point - by - point feature extraction on each point in the original features to obtain the first features;

[0024] Perform a spatial transformation on the first features to align the point cloud data in three - dimensional space to obtain the local features;

[0025] Perform point - by - point feature extraction on each point in the local features to obtain the second features;

[0026] Perform global pooling processing on the second features to obtain the global features of the point cloud data;

[0027] Fuse the local features and the global features to obtain the fused features;

[0028] Based on the fused features, perform semantic segmentation on the point cloud data to detect whether there is a parking stopper in the current environment.

[0029] Optionally, the vehicle state data includes driving speed, acceleration, steering wheel angle, and brake signal duration. Determining the driver's intention to move the vehicle based on the vehicle state data includes:

[0030] Processing the vehicle state data of multiple time steps closest to the current time step including the current time step by using a long short-term memory neural network to obtain time series features;

[0031] Determining the driver's intention to move the vehicle based on the time series features.

[0032] Optionally, the long short-term memory neural network includes N transfer cells. Processing the vehicle state data of multiple time steps closest to the current time step including the current time step by using a long short-term memory neural network to obtain time series features includes:

[0033] The first transfer cell receives the vehicle state data of the first time step for calculation to obtain the hidden state and cell state output by the first transfer cell;

[0034] The i-th transfer cell receives the vehicle state data of the i-th time step, and combines the hidden state and cell state output by the (i - 1)-th transfer cell for calculation to obtain the hidden state and cell state output by the i-th transfer cell, where i is a positive integer greater than 1 and less than N;

[0035] The N-th transfer cell receives the vehicle state data of the N-th time step, and combines the hidden state and cell state output by the (N - 1)-th transfer cell for calculation to obtain the cell state output by the N-th transfer cell as the time series feature.

[0036] Optionally, controlling the state of the automatic parking system based on the driver's intention to move the vehicle includes:

[0037] When the automatic parking system is in the on state, if the driver has the intention to move the vehicle, then control the automatic parking system to turn off;

[0038] When the automatic parking system is in the on state, if the driver does not have the intention to move the vehicle, then maintain the on state of the automatic parking system;

[0039] When the automatic parking system is in the off state, if the driver has the intention to move the vehicle, then maintain the off state of the automatic parking system;

[0040] When the automatic parking system is in the off state, if the driver does not have the intention to move the vehicle, then control the automatic parking system to turn on.

[0041] In a second aspect, the present invention further provides an automatic parking control device, including:

[0042] A point cloud data acquisition module, configured to convert radar data collected by a millimeter-wave radar into point cloud data when the driving speed is less than a preset speed;

[0043] A vehicle stopper detection module, configured to perform target detection based on the point cloud data to detect whether there is a vehicle stopper in the current environment;

[0044] A vehicle state data acquisition module, configured to acquire vehicle state data if there is a vehicle stopper in the current environment;

[0045] A vehicle moving intention determination module, configured to determine the driver's vehicle moving intention based on the vehicle state data;

[0046] A state control module, configured to control the state of the automatic parking system based on the driver's vehicle moving intention.

[0047] In a third aspect, the present invention further provides an electronic device, including:

[0048] One or more processors;

[0049] A storage device, configured to store one or more programs;

[0050] When the one or more programs are executed by the one or more processors, the one or more processors implement the automatic parking control method provided in the first aspect of the present invention.

[0051] In a fourth aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the automatic parking control method provided in the first aspect of the present invention.

[0052] The automatic parking control method provided by the present invention converts the radar data collected by a millimeter-wave radar into point cloud data when the driving speed is less than a preset speed, performs target detection based on the point cloud data to detect whether there is a vehicle stopper in the current environment, acquires vehicle state data if there is a vehicle stopper in the current environment, determines the driver's vehicle moving intention based on the vehicle state data, and controls the state of the automatic parking system based on the driver's vehicle moving intention, avoiding the problems of reduced driving fluency and easy occurrence of scraping accidents caused by frequent activation of the automatic parking system during the vehicle moving process, and improving driving fluency and safety.

[0053] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understandable through the following description. Description of the Drawings

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0055] Figure 1 It is a flowchart of an automatic parking control method provided by the present invention;

[0056] Figure 2 It is a schematic structural diagram of a PointNet provided by the present invention;

[0057] Figure 3 It is a schematic structural diagram of an automatic parking control device provided by the present invention;

[0058] Figure 4 It is a schematic structural diagram of an electronic device provided by the present invention.

[0059] Through the above accompanying drawings, the clear embodiments of the present application have been shown, and there will be more detailed descriptions hereinafter. These accompanying drawings and text descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. Detailed Embodiments

[0060] In order to enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0061] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above accompanying drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0062] Figure 1The flowchart of an automatic parking control method provided by the present invention. This embodiment is applicable to the automatic parking control during the process of moving the vehicle in a narrow space. This method can be executed by the automatic parking control device provided by the present invention, and this device can be implemented in a software and / or hardware manner and is usually configured in an electronic device, such as Figure 1 As shown, the automatic parking control method includes the following steps:

[0063] S101. When the driving speed is less than a preset speed, convert the radar data collected by the millimeter-wave radar into point cloud data.

[0064] Exemplarily, in the embodiment of the present invention, the vehicle controller obtains the driving speed of the vehicle in real time through a speed sensor. When the driving speed is less than the preset speed (for example, 3 km / h) and the continuous duration exceeds the preset duration (for example, 3 seconds), convert the radar data collected by the millimeter-wave radar into point cloud data. Exemplarily, during the driving process of the vehicle, the radar data collected by the millimeter-wave radar can be continuously obtained, and when the driving speed is less than the preset speed (for example, 3 km / h) and the continuous duration exceeds the preset duration (for example, 3 seconds), convert the subsequently obtained radar data of the millimeter-wave radar into point cloud data.

[0065] The radar signal emitted by the millimeter-wave radar is a frequency-modulated continuous-wave radar signal. The interference radar periodically emits radar wave signals with continuous frequencies, and this period is called the frequency sweep period. Within one frequency sweep period, the frequency of the frequency-modulated continuous-wave radar signal increases linearly with time.

[0066] In the embodiment of the present invention, for the echo signal received by the millimeter-wave, perform a mixing process on the echo signal and the transmitted signal to obtain an intermediate-frequency signal. Mixing is to mix the transmitted signal and the echo signal in terms of frequency and phase to obtain an intermediate-frequency signal that reflects the motion information of the target.

[0067] In order to reduce the amount of data processed and improve the processing efficiency, the intermediate-frequency signal can be sampled, and then the sampled data is mapped into a two-dimensional matrix diagram. Each row of data in this two-dimensional matrix diagram represents the intermediate-frequency signal within each frequency sweep period of the transmitted signal, and multiple rows of data are arranged in the column direction in the order of the frequency sweep periods. Exemplarily, the sampled data of the intermediate-frequency signal in the first frequency sweep period is mapped to the first row, and the sampled data of the intermediate-frequency signal in the second frequency sweep period is mapped to the second row, and so on, to obtain a two-dimensional matrix diagram.

[0068] Perform a distance-dimensional Fourier transform on the two-dimensional matrix diagram (i.e., perform a Fourier transform in the row direction) to obtain the distance between the target point and the millimeter-wave radar. The distances of multiple target points form a distance distribution diagram of the target in terms of distance. Exemplarily, the Fourier transform can be a fast Fourier transform.

[0069] Performing a Fourier transform on the distance distribution map in the column direction can obtain the range-Doppler map of the target's distribution in range and velocity (Doppler frequency).

[0070] The time when the same echo signal arrives at different receiving antennas is different, which is called the path difference, resulting in a phase difference in the echo signals of different receiving antennas. By extracting the phase differences of multiple receiving antennas, the horizontal angle and elevation angle between the target point and the millimeter-wave radar can be calculated. Exemplarily, performing another Fourier transform on the range-Doppler map in the receiving antenna dimension (angle dimension) can obtain the horizontal angle and elevation angle between the target point and the millimeter-wave radar.

[0071] Based on the distance, horizontal angle, and elevation angle between the target point and the millimeter-wave radar, calculate the three-dimensional coordinates of the target point in the Cartesian coordinate system to form three-dimensional point cloud data. Specifically, the millimeter-wave radar is in spherical coordinate data and needs to be converted to the Cartesian coordinate system. Exemplarily, the conversion process is as follows:

[0072] X = R·cos(φ)·sin(θ);

[0073] Y = R·cos(φ)·cos(θ);

[0074] Z = R·sin(φ);

[0075] Where R is the distance between the target point and the millimeter-wave radar, φ is the horizontal angle between the target point and the millimeter-wave radar, and θ is the elevation angle between the target point and the millimeter-wave radar.

[0076] S102. Perform target detection based on the point cloud data to detect whether there is a parking stopper in the current environment.

[0077] In the embodiments of the present invention, perform target detection on the point cloud data to detect whether there is a parking stopper in the current environment. The present invention does not limit the target detection algorithm, as long as it can detect the parking stopper.

[0078] Exemplarily, use PointNet to perform target detection on the point cloud data. PointNet is a neural network specifically designed for processing point cloud data, which can directly process irregular point cloud data without converting the point cloud data into a regular network or voxel. Figure 2 This is a schematic structural diagram of a PointNet provided by the present invention. Refer to Figure 2 , the process of performing target detection on the point cloud data is as follows:

[0079] 1. Perform a spatial transformation on the point cloud data to align the point cloud data in three-dimensional space to obtain the original features.

[0080] Exemplarily, the point cloud data is input into a T-Net network for processing. The point cloud is transformed such as rotation and translation, and an approximate affine transformation matrix is learned to align the input points in the three-dimensional space, obtaining the original features.

[0081] 2. Point-by-point feature extraction is performed on each point in the original features to obtain the first features.

[0082] Exemplarily, the original features are input into a multi-layer perceptron (MLP) for point-by-point feature extraction. The multi-layer perceptron maps the features of each point from the original 3D to a higher dimension, obtaining the first features.

[0083] 3. Spatial transformation is performed on the first features to align the point cloud data in the three-dimensional space, obtaining the local features.

[0084] Exemplarily, the first features are input into a T-Net network for processing. The first features are transformed such as rotation and translation, and an approximate affine transformation matrix is learned to align the input features in the three-dimensional space, obtaining the local features.

[0085] 4. Point-by-point feature extraction is performed on each point in the local features to obtain the second features.

[0086] Exemplarily, the local features are input into a multi-layer perceptron (MLP) for point-by-point feature extraction. The multi-layer perceptron maps the local features to a higher dimension, obtaining the second features.

[0087] 5. Global pooling processing is performed on the second features to obtain the global features of the point cloud data.

[0088] Exemplarily, the second features are input into the global max pooling layer (GMP) for global max pooling processing to obtain the global features of the point cloud data. The max pooling operation can ensure that the network does not depend on the order of points, guaranteeing the disorder of the point cloud.

[0089] 6. The local features and the global features are fused to obtain the fused features.

[0090] Exemplarily, the global features are replicated multiple times until they have the same dimension as the local features and are fused with the local features to obtain the fused features. By fusing the global features and the local features, the network can simultaneously understand the overall structure and local details of the point cloud, improving the accuracy of object detection.

[0091] 7. Semantic segmentation is performed on the point cloud data based on the fused features to detect whether there is a parking stopper in the current environment.

[0092] Exemplarily, the fused features are input into a fully connected layer (FCL) for processing to predict labels for each point, and the probabilities of the predicted class labels for each point are output respectively to achieve semantic segmentation. If there is a point cloud set with the class label "parking stopper", it is considered that there is a parking stopper in the current environment.

[0093] S103. If there is a parking stopper in the current environment, obtain vehicle state data.

[0094] In an embodiment of the present invention, if there is a parking stopper in the current environment, it indicates the possibility of vehicle relocation, and further vehicle state data is obtained. The vehicle state data may include driving speed, acceleration, steering wheel angle, brake signal duration, etc., which are not limited in this embodiment of the present invention.

[0095] S104. Determine the driver's intention of vehicle relocation based on the vehicle state data.

[0096] In an embodiment of the present invention, the driver's intention of vehicle relocation is determined based on the vehicle state data. Specifically, a long short-term memory neural network, a support vector machine, a decision tree, etc. can be used to determine the driver's intention of vehicle relocation, which are not limited in this embodiment of the present invention.

[0097] Exemplarily, in a specific embodiment of the present invention, a long short-term memory neural network is used to determine the driver's intention of vehicle relocation, which is not limited in this embodiment of the present invention.

[0098] Specifically, the long short-term memory neural network processes the vehicle state data of the nearest multiple time steps including the current time step to obtain time series features. Exemplarily, the long short-term memory neural network includes N transfer cells, and the processing process of the long short-term memory neural network is as follows:

[0099] 1. The first transfer cell receives the vehicle state data of the first time step for calculation to obtain the hidden state and cell state output by the first transfer cell.

[0100] Exemplarily, the first transfer cell receives the vehicle state data of the first time step and calculates through the forget gate, input gate, and output gate inside the transfer cell in sequence to output the hidden state and cell state output by the first transfer cell.

[0101] 2. The i-th transfer cell receives the vehicle state data of the i-th time step and combines the hidden state and cell state output by the (i - 1)-th transfer cell for calculation to obtain the hidden state and cell state output by the i-th transfer cell, where i is a positive integer greater than 1 and less than N.

[0102] Exemplarily, the i-th transfer cell receives the vehicle state data at the i-th time step, the hidden state and cell state output by the (i - 1)-th transfer cell, and performs calculations sequentially through the forget gate, memory gate, and output gate inside the transfer cell, and outputs the hidden state and cell state output by the i-th transfer cell.

[0103] 3. The N-th transfer cell receives the vehicle state data at the N-th time step, and performs calculations in combination with the hidden state and cell state output by the (N - 1)-th transfer cell, and obtains the cell state output by the N-th transfer cell as the time series feature.

[0104] Exemplarily, the N-th transfer cell receives the vehicle state data at the N-th time step, the hidden state and cell state output by the (N - 1)-th transfer cell, performs calculations sequentially through the forget gate, memory gate, and output gate inside the transfer cell, outputs the hidden state and cell state output by the N-th transfer cell, and takes the cell state output by the N-th transfer cell as the time series feature.

[0105] After obtaining the time series feature, the parking intention of the driver is determined based on the time series feature. Exemplarily, the time series feature is input into a fully connected layer for processing, and the processed result is input into a classifier to obtain the parking intention of the driver (there is a parking intention or there is no parking intention).

[0106] S105. Control the state of the automatic parking system based on the parking intention of the driver.

[0107] Exemplarily, when the automatic parking system is in the on state, if the driver has a parking intention, then control the automatic parking system to turn off, avoiding the problems of reduced driving smoothness and easy occurrence of scratching accidents caused by the frequent activation of the automatic parking system during parking, and improving driving smoothness and safety.

[0108] When the automatic parking system is in the on state, if the driver has no parking intention, it means that the vehicle is currently driving normally, then maintain the on state of the automatic parking system.

[0109] When the automatic parking system is in the off state, if the driver has a parking intention, then maintain the off state of the automatic parking system, avoiding the problems of reduced driving smoothness and easy occurrence of scratching accidents caused by the frequent activation of the automatic parking system during parking, and improving driving smoothness and safety.

[0110] When the automatic parking system is in the off state, if the driver has no parking intention, it means that the vehicle is currently driving normally, then control the automatic parking system to turn on so that the automatic parking system can be normally activated.

[0111] The automatic parking control method provided by the present invention converts the radar data collected by the millimeter-wave radar into point cloud data when the driving speed is less than the preset speed, performs target detection based on the point cloud data, detects whether there is a parking stopper in the current environment, and if there is a parking stopper in the current environment, obtains the vehicle state data, determines the driver's intention to move the vehicle based on the vehicle state data, and controls the state of the automatic parking system based on the driver's intention to move the vehicle, avoiding the problems of reduced driving smoothness and easy occurrence of scraping accidents caused by the frequent activation of the automatic parking system during the vehicle moving process, and improving the driving smoothness and safety.

[0112] Figure 3 As shown in the structural schematic diagram of an automatic parking control device provided by the present invention, Figure 3 the automatic parking control device includes:

[0113] A point cloud data acquisition module 201, configured to convert the radar data collected by the millimeter-wave radar into point cloud data when the driving speed is less than the preset speed;

[0114] A stopper detection module 202, configured to perform target detection based on the point cloud data to detect whether there is a parking stopper in the current environment;

[0115] A vehicle state data acquisition module 203, configured to obtain the vehicle state data if there is a parking stopper in the current environment;

[0116] A vehicle moving intention determination module 204, configured to determine the driver's vehicle moving intention based on the vehicle state data;

[0117] A state control module 205, configured to control the state of the automatic parking system based on the driver's vehicle moving intention.

[0118] In some embodiments of the present invention, the point cloud data acquisition module 201 includes:

[0119] A distance calculation sub-module, configured to perform a distance-dimensional Fourier transform on the radar data to obtain the distance between the target point and the millimeter-wave radar;

[0120] An angle calculation sub-module, configured to perform an angle-dimensional Fourier transform on the radar data to obtain the horizontal angle and the pitch angle between the target point and the millimeter-wave radar;

[0121] A coordinate conversion sub-module, configured to calculate the three-dimensional coordinates of the target point in the Cartesian coordinate system based on the distance, the horizontal angle, and the pitch angle between the target point and the millimeter-wave radar, and form three-dimensional point cloud data.

[0122] In some embodiments of the present invention, the three-dimensional coordinates of the target point in the Cartesian coordinate system are calculated based on the distance, horizontal angle, and pitch angle between the target point and the millimeter-wave radar. The calculation formula is as follows:

[0123] X = R·cos(φ)·sin(θ);

[0124] Y = R·cos(φ)·cos(θ);

[0125] Z = R·sin(φ);

[0126] where R is the distance between the target point and the millimeter-wave radar, φ is the horizontal angle between the target point and the millimeter-wave radar, and θ is the pitch angle between the target point and the millimeter-wave radar.

[0127] In some embodiments of the present invention, the vehicle stopper detection module 202 includes:

[0128] A first transformation sub-module for performing a spatial transformation on the point cloud data to align the point cloud data in three-dimensional space to obtain an original feature;

[0129] A first feature extraction sub-module for performing point-by-point feature extraction on each point in the original feature to obtain a first feature;

[0130] A second transformation sub-module for performing a spatial transformation on the first feature to align the point cloud data in three-dimensional space to obtain a local feature;

[0131] A second feature extraction sub-module for performing point-by-point feature extraction on each point in the local feature to obtain a second feature;

[0132] A pooling sub-module for performing global pooling processing on the second feature to obtain the global feature of the point cloud data;

[0133] A feature fusion sub-module for fusing the local feature and the global feature to obtain a fusion feature;

[0134] A semantic segmentation sub-module for performing semantic segmentation on the point cloud data based on the fusion feature to detect whether there is a vehicle stopper in the current environment.

[0135] In some embodiments of the present invention, the vehicle state data includes driving speed, acceleration, steering wheel angle, and brake signal duration. The vehicle moving intention determination module 204 includes:

[0136] A time series feature calculation sub-module for processing the vehicle state data of the nearest multiple time steps including the current time step by using a long short-term memory neural network to obtain time series features;

[0137] A vehicle moving intention determination sub-module, configured to determine the vehicle moving intention of the driver based on the time series feature.

[0138] In some embodiments of the present invention, the long short-term memory neural network includes N transfer cells, and the time series feature calculation sub-module includes:

[0139] A first calculation unit, configured to calculate by receiving the vehicle state data of the first time step by the first transfer cell, and obtain the hidden state and cell state output by the first transfer cell;

[0140] A second calculation unit, configured to calculate by receiving the vehicle state data of the i-th time step by the i-th transfer cell, and combining the hidden state and cell state output by the (i - 1)-th transfer cell, and obtain the hidden state and cell state output by the i-th transfer cell, where i is a positive integer greater than 1 and less than N;

[0141] A third calculation unit, configured to calculate by receiving the vehicle state data of the N-th time step by the N-th transfer cell, and combining the hidden state and cell state output by the (N - 1)-th transfer cell, and obtain the cell state output by the N-th transfer cell as the time series feature.

[0142] In some embodiments of the present invention, the state control module 205 includes:

[0143] A first control sub-module, configured to control the automatic parking system to be turned off if the driver has a vehicle moving intention when the automatic parking system is in an on state;

[0144] A second control sub-module, configured to maintain the on state of the automatic parking system if the driver does not have a vehicle moving intention when the automatic parking system is in an on state;

[0145] A third control sub-module, configured to maintain the off state of the automatic parking system if the driver has a vehicle moving intention when the automatic parking system is in an off state;

[0146] A fourth control sub-module, configured to control the automatic parking system to be turned on if the driver does not have a vehicle moving intention when the automatic parking system is in an off state.

[0147] The above automatic parking control device can execute the automatic parking control method provided in the foregoing embodiments of the present invention, and has corresponding function modules and beneficial effects for executing the automatic parking control method.

[0148] Figure 4Schematic diagram of the structure of an electronic device provided by the present invention. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processors, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0149] As Figure 4 shown, the electronic device includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as read-only memory (ROM) 12, random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device can also be stored. The processor 11, ROM 12, and RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0150] Multiple components in the electronic device are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0151] The processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include but are not limited to a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the automatic parking control method.

[0152] In some embodiments, the automatic parking control method may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed onto the electronic device via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the automatic parking control method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to execute the automatic parking control method by any other suitable means (e.g., by means of firmware).

[0153] The various embodiments of the systems and techniques described above in this document may be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGA), application specific integrated circuits (ASIC), application specific standard products (ASSP), systems on a chip (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: implemented in one or more computer programs that may be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a special-purpose or general-purpose programmable processor that receives data and instructions from a storage system, at least one input device, and at least one output device, and transmits the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0154] The computer program for implementing the method of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on the remote machine or server.

[0155] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0156] In order to provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0157] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0158] A computing system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The relationship between the client and the server is created by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0159] An embodiment of the present invention also provides a computer program product, including a computer program which, when executed by a processor, implements the automatic parking control method provided in any embodiment of the present application.

[0160] In the process of implementing the computer program product, computer program code for performing the operations of the present invention can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent 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 can 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 can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0161] It should be understood that various forms of the flow shown above can be used, reordering, adding or deleting steps. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0162] The above specific embodiments do not constitute a limitation to the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An automatic parking control method, characterized in that, Including: When the driving speed is less than the preset speed, convert the radar data collected by the millimeter-wave radar into point cloud data; Perform target detection based on the point cloud data to detect whether there is a parking stopper in the current environment; If there is a parking stopper in the current environment, obtain the vehicle status data; Determine the driver's intention to move the vehicle based on the vehicle status data; Control the state of the automatic parking system based on the driver's intention to move the vehicle; Converting the radar data collected by the millimeter-wave radar into point cloud data includes: Perform distance-dimensional Fourier transform on the radar data to obtain the distance between the target point and the millimeter-wave radar; Perform angle-dimensional Fourier transform on the radar data to obtain the horizontal angle and pitch angle between the target point and the millimeter-wave radar; Calculate the three-dimensional coordinates of the target point in the Cartesian coordinate system based on the distance, horizontal angle, and pitch angle between the target point and the millimeter-wave radar to form three-dimensional point cloud data; Performing target detection based on the point cloud data to detect whether there is a parking stopper in the current environment includes: Perform spatial transformation on the point cloud data to align the point cloud data in three-dimensional space to obtain the original features; Perform point-by-point feature extraction on each point in the original features to obtain the first features; Perform spatial transformation on the first features to align the point cloud data in three-dimensional space to obtain local features; Perform point-by-point feature extraction on each point in the local features to obtain the second features; Perform global pooling processing on the second features to obtain the global features of the point cloud data; Fuse the local features and the global features to obtain fused features; Perform semantic segmentation on the point cloud data based on the fused features to detect whether there is a parking stopper in the current environment; The vehicle status data includes driving speed, acceleration, steering wheel angle, and brake signal duration. Determining the driver's intention to move the vehicle based on the vehicle status data includes: Use a long short-term memory neural network to process the vehicle status data of the nearest multiple time steps including the current time step to obtain time series features; Determine the driver's intention to move the vehicle based on the time series features; The long short-term memory neural network includes N transfer cells. Using the long short-term memory neural network to process the vehicle status data of the nearest multiple time steps including the current time step to obtain time series features includes: The first transfer cell receives the vehicle status data of the first time step for calculation to obtain the hidden state and cell state output by the first transfer cell; The i-th transfer cell receives the vehicle status data of the i-th time step and combines the hidden state and cell state output by the i-1-th transfer cell for calculation to obtain the hidden state and cell state output by the i-th transfer cell, where i is a positive integer greater than 1 and less than N; The N-th transfer cell receives the vehicle status data of the N-th time step and combines the hidden state and cell state output by the N-1-th transfer cell for calculation to obtain the cell state output by the N-th transfer cell as the time series feature.

2. The automatic parking control method according to claim 1, characterized in that Calculate the three-dimensional coordinates of the target point in the Cartesian coordinate system based on the distance, horizontal angle, and pitch angle between the target point and the millimeter-wave radar. The calculation formula is as follows: X = R·cos(φ)·sin(θ); Y = R·cos(φ)·cos(θ); Z = R·sin(φ); where R is the distance between the target point and the millimeter-wave radar, φ is the horizontal angle between the target point and the millimeter-wave radar, and θ is the pitch angle between the target point and the millimeter-wave radar.

3. The automatic parking control method according to claim 1, wherein Control the state of the automatic parking system based on the driver's intention to move the vehicle, including: When the automatic parking system is in the on state, if the driver has the intention to move the vehicle, then control the automatic parking system to turn off; When the automatic parking system is in the on state, if the driver does not have the intention to move the vehicle, then maintain the on state of the automatic parking system; When the automatic parking system is in the off state, if the driver has the intention to move the vehicle, then maintain the off state of the automatic parking system; When the automatic parking system is in the off state, if the driver does not have the intention to move the vehicle, then control the automatic parking system to turn on.

4. An automatic parking control device, characterized in that, Including: A point cloud data acquisition module, which is used to convert the radar data collected by the millimeter-wave radar into point cloud data when the driving speed is less than the preset speed; A parking block detector module, which is used to perform target detection based on the point cloud data to detect whether there is a parking block in the current environment; A vehicle status data acquisition module, which is used to acquire vehicle status data if there is a parking block in the current environment; A moving vehicle intention determination module, which is used to determine the driver's intention to move the vehicle based on the vehicle status data; A status control module, which is used to control the state of the automatic parking system based on the driver's intention to move the vehicle; The point cloud data acquisition module includes: A distance calculation sub-module, which is used to perform a distance-dimensional Fourier transform on the radar data to obtain the distance between the target point and the millimeter-wave radar; An angle calculation sub-module, which is used to perform an angle-dimensional Fourier transform on the radar data to obtain the horizontal angle and pitch angle between the target point and the millimeter-wave radar; A coordinate conversion sub-module, which is used to calculate the three-dimensional coordinates of the target point in the Cartesian coordinate system based on the distance, horizontal angle, and pitch angle between the target point and the millimeter-wave radar to form three-dimensional point cloud data; The parking block detector module includes: A first transformation sub-module, which is used to perform a spatial transformation on the point cloud data to align the point cloud data in three-dimensional space to obtain the original features; A first feature extraction sub-module, which is used to perform point-by-point feature extraction on each point in the original features to obtain the first features; A second transformation sub-module, which is used to perform a spatial transformation on the first features to align the point cloud data in three-dimensional space to obtain the local features; A second feature extraction sub-module, which is used to perform point-by-point feature extraction on each point in the local features to obtain the second features; A pooling sub-module, which is used to perform global pooling processing on the second features to obtain the global features of the point cloud data; A feature fusion sub-module, which is used to fuse the local features and the global features to obtain the fused features; A semantic segmentation sub-module, configured to perform semantic segmentation on the point cloud data based on the fusion features to detect whether there is a parking stopper in the current environment; The vehicle state data includes driving speed, acceleration, steering wheel angle, and brake signal duration. The parking intention determination module includes: A time series feature calculation sub-module, configured to use a long short-term memory neural network to process the vehicle state data of multiple time steps closest to the current time step including the current time step to obtain time series features; A parking intention determination sub-module, configured to determine the driver's parking intention based on the time series features; The long short-term memory neural network includes N transfer cells. The time series feature calculation sub-module includes: A first calculation unit, configured to calculate the vehicle state data of the first time step received by the first transfer cell to obtain the hidden state and cell state output by the first transfer cell; A second calculation unit, configured to calculate the vehicle state data of the i-th time step received by the i-th transfer cell, and combine the hidden state and cell state output by the (i-1)-th transfer cell for calculation to obtain the hidden state and cell state output by the i-th transfer cell, where i is a positive integer greater than 1 and less than N; A third calculation unit, configured to calculate the vehicle state data of the N-th time step received by the N-th transfer cell, and combine the hidden state and cell state output by the (N-1)-th transfer cell for calculation to obtain the cell state output by the N-th transfer cell as the time series feature.

5. An electronic device, characterized in that, Including: One or more processors; A storage device, configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the automatic parking control method according to any one of claims 1-3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the automatic parking control method according to any one of claims 1-3.

Images