Vehicle control method and device, vehicle and storage medium
Patent Information
- Application Number
- CN202311084901.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-08-25
- Publication Date
- 2025-02-25
Smart Images
Figure CN119502890A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of autonomous driving technology, and in particular to a vehicle control method, device, vehicle, and storage medium. Background Art
[0002] With the rapid development of intelligent driving technology, automatic parking has gradually been applied in low-speed intelligent driving scenarios. Automatic parking allows the car to automatically find and accurately park at an appropriate parking position. It uses environmental perception sensors to sense the position and status of parking spaces around the vehicle, find a suitable parking space, and then plan the optimal parking path to control the vehicle to automatically complete parking. Summary of the invention
[0003] The present disclosure aims to solve one of the technical problems in the related art at least to some extent.
[0004] The first embodiment of the present disclosure provides a vehicle control method, including:
[0005] Acquire multiple images around the vehicle, wherein at least one image around the vehicle contains the target storage location;
[0006] Input the plurality of vehicle circumference images into a state prediction model to obtain a target heading angle of the vehicle relative to the target storage location, a first target distance between the vehicle and the target storage location in a first direction, and a second target distance between the vehicle and the target storage location in a second direction;
[0007] Determining a target steering torque, a first target speed, and a second target speed of the vehicle according to the target heading angle, the first target distance, and the second target distance;
[0008] The vehicle is controlled based on the target steering torque, the first target speed, and the second target speed.
[0009] A second aspect of the present disclosure provides a vehicle control device, including:
[0010] A first acquisition module is used for a plurality of vehicle surrounding images, wherein at least one of the vehicle surrounding images contains a target storage location;
[0011] A second acquisition module is used to input the multiple vehicle circumference images into a state prediction model to obtain a target heading angle of the vehicle relative to the target storage location, a first target distance between the vehicle and the target storage location in a first direction, and a second target distance in a second direction;
[0012] a determination module, configured to determine a target steering torque, a first target speed, and a second target speed of the vehicle according to the target heading angle, the first target distance, and the second target distance;
[0013] A control module is used to control the vehicle based on the target steering torque, the first target speed and the second target speed.
[0014] The third aspect embodiment of the present disclosure proposes a vehicle, comprising: a processor; a memory for storing processor executable instructions; wherein the processor is configured to: perform the steps of the vehicle control method proposed in the first aspect embodiment of the present disclosure.
[0015] The fourth aspect embodiment of the present disclosure proposes a computer-readable storage medium. When the instructions in the storage medium are executed by the processor of the mobile terminal, the mobile terminal can execute the steps of the vehicle control method proposed in the first aspect embodiment of the present disclosure.
[0016] The vehicle control method, device, vehicle and storage medium provided by the present disclosure have the following beneficial effects:
[0017] In the disclosed embodiment, a plurality of images around the vehicle are first obtained, wherein at least one of the images around the vehicle contains the target storage location, and then the plurality of images around the vehicle are input into the state prediction model to obtain the target heading angle of the vehicle relative to the target storage location, the first target distance between the vehicle and the target storage location in the first direction, and the second target distance in the second direction, and then the target steering torque, the first target speed, and the second target speed of the vehicle are determined according to the target heading angle, the first target distance, and the second target distance, and finally the vehicle is controlled based on the target steering torque, the first target speed, and the second target speed. Thus, the state prediction model can be used to directly predict the images around the vehicle to obtain the target heading angle, the first target distance, and the second target distance, and then the steering torque, the first target speed, and the second target speed are determined to accurately control the vehicle, so that the vehicle can accurately enter the target storage location and realize automatic parking.
[0018] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description or learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] The above and / or additional aspects and advantages of the present disclosure will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:
[0020] Figure 1 A schematic flow chart of a vehicle control method provided by an embodiment of the present disclosure;
[0021] Figure 2 A schematic diagram of a coordinate system provided by an embodiment of the present disclosure;
[0022] Figure 3A schematic diagram of the structure of a state prediction model provided by an embodiment of the present disclosure;
[0023] Figure 4 A schematic flow chart of a vehicle control method provided by another embodiment of the present disclosure;
[0024] Figure 5 A schematic diagram of the structure of a vehicle control device provided by an embodiment of the present disclosure;
[0025] Figure 6 A block diagram of an exemplary electronic device suitable for implementing embodiments of the present disclosure is shown. DETAILED DESCRIPTION
[0026] Embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.
[0027] The following describes a vehicle control method, an apparatus, a vehicle, and a storage medium according to embodiments of the present disclosure with reference to the accompanying drawings.
[0028] Figure 1 A flow chart of a vehicle control method provided in an embodiment of the present disclosure.
[0029] The embodiment of the present disclosure takes the vehicle control method being configured in a vehicle control device as an example. The vehicle control device can be applied to any vehicle so that the vehicle can perform vehicle control functions.
[0030] like Figure 1 As shown, the vehicle control method may include the following steps:
[0031] Step 101 : acquiring a plurality of vehicle surrounding images, wherein at least one of the vehicle surrounding images contains a target storage location.
[0032] The vehicle surrounding image may be an environmental image of the vehicle surrounding collected by using multiple surrounding cameras with different viewing angles, each surrounding camera is used to collect a vehicle surrounding image of one viewing angle, and multiple surrounding cameras cover the environmental range of the vehicle surrounding.
[0033] In some embodiments, multiple perimeter cameras can collect multiple vehicle perimeter images from different perspectives in real time, such as images 1, 2, ... N, and send the collected vehicle perimeter images to the vehicle in real time. In one embodiment, the vehicle may include four perimeter cameras. The four perimeter cameras are respectively arranged at the left front, right front, left rear, and right rear of the carrier. The present disclosure does not limit this.
[0034] The target storage location may be selected by the driver of the vehicle, or may be determined by analyzing data collected by sensors (such as laser radar, camera, ultrasonic sensor, etc.). This disclosure does not limit this.
[0035] Optionally, after the target storage location is determined, the target storage location can be marked in the vehicle surrounding image captured by the camera. For example, the target storage location contained in the vehicle surrounding image can be marked in the form of a line frame. In some embodiments, one vehicle surrounding image can contain the target storage location, and two vehicle surrounding images can contain the same target storage location. For example, the vehicle surrounding image captured by the left front camera and the vehicle surrounding image captured by the right front camera both contain the target storage location.
[0036] Step 102 , input a plurality of vehicle peripheral images into a state prediction model to obtain a target heading angle of the vehicle relative to a target storage location, a first target distance between the vehicle and the target storage location in a first direction, and a second target distance in a second direction.
[0037] Optionally, the first direction and the second direction may be directions of two parking space lines that are perpendicular to each other in the target storage space.
[0038] Figure 2 A schematic diagram of a coordinate system provided by an embodiment of the present disclosure is shown in FIG. Figure 2 As shown, a coordinate system can be established with any vertex of the target storage location as the coordinate origin and two parking space lines perpendicular to each other in the target storage location as two coordinate axes. The direction of the first coordinate axis (x-axis) can be the first direction, and the direction of the second coordinate axis (y-axis) can be the second direction.
[0039] like Figure 2 As shown, the first direction may also be referred to as a horizontal direction, and the first target distance may also be referred to as a horizontal distance; the second direction may also be referred to as a vertical direction, and the second target distance may also be referred to as a vertical distance.
[0040] The first target distance may be the distance d1 between the center of mass of the vehicle and the y-axis, and the second target distance may be the distance d2 between the center of mass of the vehicle and the x-axis.
[0041] The target heading angle may be the angle between the vehicle's center of mass velocity and the x-axis.
[0042] Optionally, the state prediction model may include structures such as a deep residual network and a weighted bidirectional feature pyramid network.
[0043] Specifically, multiple images around the vehicle can be first input into the deep residual network (ResNet) in the state prediction model to obtain multiple first feature maps corresponding to each image around the vehicle, and then the multiple first feature maps corresponding to each image around the vehicle are input into the weighted bidirectional feature pyramid network (Bidirectional Feature Pyramid Network, BiFPN) in the state prediction model for feature fusion to obtain the second feature map corresponding to each image around the vehicle, and then the second feature map corresponding to each image around the vehicle is spliced to obtain a bird's-eye view feature map, and finally based on the bird's-eye view feature map, the target heading angle, the first target distance, and the second target distance are determined.
[0044] Among them, each first feature map has a different scale and a different number of channels.
[0045] The deep residual network may be a RegNet800 architecture. As an image space encoder, RegNet800M is a network structure obtained based on Neural Architecture Search (NAS), which takes into account both the feature extraction capability and computational complexity of the model.
[0046] Figure 3 A schematic diagram of a state prediction model provided by an embodiment of the present disclosure is shown in FIG. Figure 3 As shown in the figure, img(4,3,576,768) means 4 images around the vehicle, each with 3 channels, 576 pixels high and 768 pixels wide. That is, 4 images around the vehicle with 3 channels and a size of 576*768 can be input into RegNet800, and RegNet800 extracts features from each image around the vehicle to obtain the first feature maps of 5 scales corresponding to each image around the vehicle, which are a feature map with 64 channels and a size of 72*96, a feature map with 128 channels and a size of 72*96, a feature map with 288 channels and a size of 36*48, a feature map with 672 channels and a size of 18*24, and a feature map with 672 channels and a size of 9*12.
[0047] Then, the first feature maps of the five scales corresponding to each vehicle surrounding image are input into BiFPN, and BiFPN unifies the number of channels corresponding to the first feature map of each scale, such as Figure 3 The channels of each first feature map are unified into 128 channels. Then, the five first feature maps after the channels of each vehicle surrounding image are fused to obtain a second feature map corresponding to each vehicle surrounding image. Figure 3 As shown, the second feature map has 128 channels and is 144*192.
[0048] After obtaining the second feature map corresponding to each vehicle surrounding image, multiple second feature maps can be input into a deformable cross-attention module to obtain a spliced bird's-eye view feature map. Figure 3 As shown, an initial bird's-eye view feature map with 128 channels and a size of 96*96 can also be input into the deformable cross attention module to limit the channels of the second feature map to 128 channels and a size of 96*96.
[0049] Finally, the target heading angle (theta), the first target distance (d1) and the second target distance (d2) are predicted based on the bird's-eye view.
[0050] In some embodiments, the TORCS software simulator can be used to obtain training data for the state prediction model. The TORCS software simulator is a vehicle driving simulation software under the Linux platform, which provides a manual control interface for controlling the driving of the game vehicle, and can directly obtain the driving scenes and vehicle parameters built into the game. The training data acquisition process can be achieved by directly obtaining the driving scenes built into the game, but these driving scenes are relatively simple, so some extreme situations can be collected through human control. Rich and diverse training data can make the neural network more robust.
[0051] In some embodiments, the label data corresponding to the training data may include a heading angle label, a first distance label in a first direction, and a second distance label in a second direction. The parameters of the model may then be modified based on the difference between the predicted result and the label data.
[0052] In some embodiments, a Euclidean loss function (MSE) may be used to calculate the difference between the prediction result and the label data.
[0053] For example, the training data contains 732,925 images, the initial learning rate is 0.001, and the mini-batch is 64 random images. The model is iterated 800,000 times.
[0054] Step 103 , determining a target steering torque, a first target speed, and a second target speed of the vehicle according to the target heading angle, the first target distance, and the second target distance.
[0055] The first target speed may be the speed of the vehicle in the first direction, and the second target speed may be the speed of the vehicle in the second direction.
[0056] In some embodiments, a proportional control method may be used to determine the target steering torque. That is, according to the difference in heading angle, the steering torque is set to a value proportional to it. For example, a proportional coefficient K may be set to adjust the magnitude of the steering torque. If the target heading angle deviates greatly from the target storage location direction, the target steering torque will increase accordingly to quickly correct the deviation. When the target heading angle approaches the target storage location direction, the target steering torque will gradually decrease to avoid overcorrection.
[0057] In some embodiments, the first target speed and the second target speed of the vehicle can be determined in combination with a path planning algorithm. The path planning algorithm can calculate an optimal path based on the first target distance between the current position of the vehicle and the target storage location, as well as map information, obstacles and other factors, and then determine the first target speed of the vehicle based on the length of the path and the expected arrival time.
[0058] Similarly, an optimal path is calculated based on the second target distance between the vehicle's current position and the target storage location, as well as map information, obstacles and other factors, and then the vehicle's second target speed is determined based on the length of the path and the expected arrival time.
[0059] Step 104 , controlling the vehicle based on the target steering torque, the first target speed, and the second target speed.
[0060] In the disclosed embodiment, after the target steering torque, the first target speed and the second target speed are determined, the vehicle can be controlled so that the vehicle automatically drives to the target parking position to achieve automatic parking.
[0061] In the disclosed embodiment, a plurality of images around the vehicle are first obtained, wherein at least one of the images around the vehicle contains the target storage location, and then the plurality of images around the vehicle are input into the state prediction model to obtain the target heading angle of the vehicle relative to the target storage location, the first target distance between the vehicle and the target storage location in the first direction, and the second target distance in the second direction, and then the target steering torque, the first target speed, and the second target speed of the vehicle are determined according to the target heading angle, the first target distance, and the second target distance, and finally the vehicle is controlled based on the target steering torque, the first target speed, and the second target speed. Thus, the state prediction model can be used to directly predict the images around the vehicle to obtain the target heading angle, the first target distance, and the second target distance, and then the steering torque, the first target speed, and the second target speed are determined to accurately control the vehicle, so that the vehicle can accurately enter the target storage location and realize automatic parking.
[0062] Figure 4 A flow chart of a vehicle control method provided by an embodiment of the present disclosure is shown as follows: Figure 4 As shown, the vehicle control method may include the following steps:
[0063] Step 401 : acquiring a plurality of vehicle surrounding images, wherein at least one of the vehicle surrounding images contains a target storage location.
[0064] Step 402 , input a plurality of vehicle peripheral images into a state prediction model to obtain a target heading angle of the vehicle relative to a target storage location, a first target distance between the vehicle and the target storage location in a first direction, and a second target distance between the vehicle and the target storage location in a second direction.
[0065] The specific implementation forms of step 401 and step 402 can refer to the detailed descriptions in other embodiments of the present disclosure, and will not be described in detail here.
[0066] Step 403, determining a first target parameter corresponding to the target heading angle, a second target parameter corresponding to the first target distance, and a third target parameter corresponding to the second target distance.
[0067] In some embodiments, based on the target heading angle, the first target distance, and the second target distance, a first mapping table is queried to obtain the first target parameter, the second target parameter, and the third target parameter, wherein the first mapping table includes each heading angle, the first distance in the first direction, the second distance in the second direction, and the corresponding relationship between the first parameter, the second parameter, and the third parameter.
[0068] In the disclosed embodiment, the target heading angle, the first target distance, and the second target distance can be associated with each other, that is, the first target parameter, the second target parameter, and the third target parameter are associated with each other. Thus, the first target parameter, the second target parameter, and the third target parameter in the current state of the vehicle can be determined more accurately, so that the steering torque, the first target speed, and the second target speed can be determined more accurately.
[0069] For example, the target heading angle is A1, the first target distance is B1, and the second target distance is C1, then the first target parameter is a1, the second target parameter is b1, and the third target parameter is c1; if the target heading angle changes, the changed target heading angle is A2, the first target distance and the second target distance remain unchanged, but the first target parameter, the second target parameter and the third target parameter may all change, for example, the first target parameter is a2, the second target parameter is b2, and the third target parameter is c2.
[0070] Optionally, the first mapping table may be obtained by following the steps below:
[0071] First, a first sample data set is obtained, wherein the first sample data set includes a sample heading angle between a sample vehicle and a sample parking space, a first sample distance between the sample vehicle and the sample parking space in a first direction, a second sample distance in a second direction, a sample steering torque corresponding to the sample heading angle, a first sample speed corresponding to the first sample distance, and a second sample speed corresponding to the second sample distance. Then, the first sample data set is processed to obtain the first parameter, the second parameter, and the third parameter corresponding to each group of sample heading angles, the first sample distance, and the second sample distance. Finally, a first mapping table is generated based on each group of sample heading angles, the first sample distance, the second sample distance, and the corresponding first parameter, the second parameter, and the third parameter.
[0072] Optionally, a statistical analysis method may be used to jointly analyze the sample heading angle, the first sample distance, the second sample distance and the corresponding first steering torque, the first sample speed, and the second sample speed in the first sample data set.
[0073] In some embodiments, a TORCS software simulator may be used to obtain a first sample data set, that is, to collect sample heading angles, first sample distances, and second sample distances between the vehicle and the target storage location at each position.
[0074] In some embodiments, the second mapping table can be queried based on the target heading angle to obtain the first target parameters, the third mapping table can be queried based on the first target distance to obtain the second target parameters, and the fourth mapping table can be queried based on the second target distance to obtain the third target parameters.
[0075] That is, the first target parameter corresponding to the target heading angle, the second target parameter corresponding to the first target distance, and the third target parameter corresponding to the second target distance are independent of each other. If any one of them changes, it will only affect the corresponding target parameter, and will not affect other target parameters. For example, if the target heading angle changes, it will only cause the first target parameter to change, while the second and third target parameters will not change.
[0076] Optionally, the second mapping table, the third mapping table, and the fourth mapping table may be obtained in the following manner:
[0077] (1) Obtaining a first sample data set, wherein the first sample data set includes a sample heading angle between a sample vehicle and a sample parking space, a first sample distance between the sample vehicle and the sample parking space in a first direction, a second sample distance in a second direction, a sample steering torque corresponding to the sample heading angle, a first sample speed corresponding to the first sample distance, and a second sample speed corresponding to the second sample distance.
[0078] (2) analyzing at least one sample steering torque corresponding to each sample heading angle to obtain a first parameter corresponding to each sample heading angle;
[0079] It should be noted that the difference between the first sample distance and the second sample distance of the vehicle may result in the same sample heading angle corresponding to multiple sample steering torques.
[0080] Optionally, at least one sample steering torque corresponding to the sample heading angle may be clustered or averaged to determine a target steering torque, and a first parameter corresponding to the sample heading angle may be determined based on the sample heading angle and the target steering torque.
[0081] (3) analyzing at least one first sample speed corresponding to each first sample distance to obtain a second parameter corresponding to each first sample distance;
[0082] It should be noted that the differences in the sample heading angles and the second sample distances of the vehicle may result in the same first sample distance corresponding to multiple first sample speeds.
[0083] Optionally, at least one first sample speed corresponding to the first sample distance may be clustered or averaged to determine a target first sample speed, and a second parameter corresponding to the first sample distance may be determined based on the first sample distance and the target first sample speed.
[0084] (4) analyzing at least one second sample speed corresponding to each second sample distance to obtain a third parameter corresponding to each second sample distance;
[0085] It should be noted that the difference in the sample heading angle and the first sample distance of the vehicle may result in the same second sample distance corresponding to multiple second sample speeds.
[0086] Optionally, at least one second sample speed corresponding to the second sample distance may be clustered or averaged to determine a target second sample speed, and a third parameter corresponding to the second sample distance may be determined based on the second sample distance and the target second sample speed.
[0087] (5) generating a second mapping table based on each sample heading angle and the corresponding first parameter;
[0088] (6) generating a third mapping table based on each first sample distance and the corresponding second parameter;
[0089] (7) Generate a fourth mapping table based on each second sample distance and the corresponding third parameter.
[0090] In the above manner, the sample heading angle and the corresponding sample steering torque may be analyzed separately; the first sample distance and the corresponding first sample speed may be analyzed separately; and the second sample speed corresponding to the second sample distance may be analyzed separately.
[0091] Step 404: determine the product of the first target parameter and the target heading angle as the target steering torque.
[0092] Step 405: determine the product of the second target parameter and the first target distance as the first target speed.
[0093] Step 406: determine the product of the third target parameter and the second target distance as the second target speed.
[0094] Step 407 , controlling the vehicle based on the target steering torque, the first target speed, and the second target speed.
[0095] In the disclosed embodiment, after obtaining the target heading angle of the vehicle relative to the target storage location, the first target distance between the vehicle and the target storage location in the first direction, and the second target distance in the second direction through the state prediction model, the first target parameter corresponding to the target heading angle, the second target parameter corresponding to the first target distance, and the third target parameter corresponding to the second target distance can also be determined, and then the product of the first target parameter and the target heading angle is determined as the target steering torque, the product of the second target parameter and the first target distance is determined as the first target speed, and the product of the third target parameter and the second target distance is determined as the second target speed. In this way, the target steering torque, the first target speed, and the second target speed can be determined more accurately, and the vehicle can be controlled to enter the target storage location more accurately.
[0096] In order to implement the above embodiments, the present disclosure also proposes a vehicle control device.
[0097] Figure 5 A schematic diagram of the structure of a vehicle control device provided in an embodiment of the present disclosure.
[0098] like Figure 5 As shown, the vehicle control device 500 may include:
[0099] The first acquisition module 501 is used for a plurality of vehicle surrounding images, wherein at least one vehicle surrounding image contains a target storage location;
[0100] The second acquisition module 502 is used to input multiple vehicle peripheral images into the state prediction model to obtain a target heading angle of the vehicle relative to the target storage location, a first target distance between the vehicle and the target storage location in the first direction, and a second target distance in the second direction;
[0101] The determination module 503 is used to determine the target steering torque, the first target speed and the second target speed of the vehicle according to the target heading angle, the first target distance and the second target distance;
[0102] The control module 504 is used to control the vehicle based on the target steering torque, the first target speed and the second target speed.
[0103] Optionally, the second acquisition module 502 is used to
[0104] Inputting multiple vehicle surrounding images into a deep residual network in a state prediction model to obtain multiple first feature maps corresponding to each vehicle surrounding image, wherein each first feature map has a different scale and a different number of channels;
[0105] Inputting multiple first feature maps corresponding to each vehicle surrounding image into a weighted bidirectional feature pyramid network in a state prediction model for feature fusion to obtain a second feature map corresponding to each vehicle surrounding image;
[0106] The second feature map corresponding to each vehicle surrounding image is spliced to obtain a bird's-eye view feature map;
[0107] Based on the bird's-eye view feature map, the target heading angle, the first target distance, and the second target distance are determined.
[0108] Optionally, the determination module 503 is used to:
[0109] Determine a first target parameter corresponding to the target heading angle, a second target parameter corresponding to the first target distance, and a third target parameter corresponding to the second target distance;
[0110] The product of the first target parameter and the target heading angle is determined as the target steering torque;
[0111] The product of the second target parameter and the first target distance is determined as the first target speed;
[0112] The product of the third target parameter and the second target distance is determined as the second target speed.
[0113] Optionally, the determination module 503 is used to:
[0114] Based on the target heading angle, the first target distance, and the second target distance, the first mapping table is queried to obtain the first target parameter, the second target parameter, and the third target parameter, wherein the first mapping table includes each heading angle, the first distance in the first direction, the second distance in the second direction, and the corresponding relationship between the first parameter, the second parameter, and the third parameter.
[0115] Optionally, the method further includes a first processing module, configured to:
[0116] Acquire a first sample data set, wherein the first sample data set includes a sample heading angle between a sample vehicle and a sample parking space, a first sample distance between the sample vehicle and the sample parking space in a first direction, a second sample distance in a second direction, a sample steering torque corresponding to the sample heading angle, a first sample speed corresponding to the first sample distance, and a second sample speed corresponding to the second sample distance;
[0117] Processing the first sample data set to obtain the first parameter, the second parameter, and the third parameter corresponding to each set of sample heading angles, the first sample distance, and the second sample distance;
[0118] A first mapping table is generated based on each set of sample heading angles, first sample distances, second sample distances, and corresponding first parameters, second parameters, and third parameters.
[0119] Optionally, the determination module 503 is used to:
[0120] Based on the target heading angle, query the second mapping table to obtain the first target parameter;
[0121] Based on the first target distance, querying a third mapping table to obtain a second target parameter;
[0122] Based on the second target distance, a fourth mapping table is queried to obtain a third target parameter.
[0123] Optionally, the method further includes a first processing module, configured to:
[0124] Acquire a first sample data set, wherein the first sample data set includes a sample heading angle between a sample vehicle and a sample parking space, a first sample distance between the sample vehicle and the sample parking space in a first direction, a second sample distance in a second direction, a sample steering torque corresponding to the sample heading angle, a first sample speed corresponding to the first sample distance, and a second sample speed corresponding to the second sample distance;
[0125] Analyze at least one sample steering torque corresponding to each sample heading angle to obtain a first parameter corresponding to each sample heading angle;
[0126] Analyze at least one first sample speed corresponding to each first sample distance to obtain a second parameter corresponding to each first sample distance;
[0127] Analyze at least one second sample speed corresponding to each second sample distance to obtain a third parameter corresponding to each second sample distance;
[0128] Generate a second mapping table based on each sample heading angle and the corresponding first parameter;
[0129] generating a third mapping table based on each first sample distance and the corresponding second parameter;
[0130] A fourth mapping table is generated based on each second sample distance and the corresponding third parameter.
[0131] Optionally, the first direction and the second direction are directions of two parking space lines that are perpendicular to each other in the target storage space.
[0132] The functions and specific implementation principles of the above modules in the embodiments of the present disclosure can be referred to the above method embodiments, and will not be repeated here.
[0133] The vehicle control device of the disclosed embodiment first obtains a plurality of images around the vehicle, wherein at least one of the images around the vehicle contains the target storage location, and then inputs the plurality of images around the vehicle into the state prediction model to obtain the target heading angle of the vehicle relative to the target storage location, the first target distance between the vehicle and the target storage location in the first direction, and the second target distance in the second direction, and then determines the target steering torque, the first target speed, and the second target speed of the vehicle based on the target heading angle, the first target distance, and the second target distance, and finally controls the vehicle based on the target steering torque, the first target speed, and the second target speed. Thus, the state prediction model can be used to directly predict the images around the vehicle to obtain the target heading angle, the first target distance, and the second target distance, and then determine the steering torque, the first target speed, and the second target speed to accurately control the vehicle, so that the vehicle can accurately enter the target storage location and realize automatic parking.
[0134] Figure 6 600 is a functional block diagram of a vehicle shown in an exemplary embodiment. For example, vehicle 600 may be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicles. Vehicle 600 may be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle.
[0135] Reference Figure 6 , the vehicle 600 may include various subsystems, for example, an infotainment system 610, a perception system 620, a decision control system 630, a drive system 640, and a computing platform 650. The vehicle 600 may also include more or fewer subsystems, and each subsystem may include multiple components. In addition, each subsystem and each component of the vehicle 600 may be interconnected by wire or wireless means.
[0136] In some embodiments, the infotainment system 610 may include a communication system, an entertainment system, and a navigation system, etc. The perception system 620 may include several sensors for sensing information about the environment around the vehicle 600. For example, the perception system 620 may include a global positioning system (the global positioning system may be a GPS system, a Beidou system, or other positioning systems), an inertial measurement unit (IMU), a laser radar, a millimeter wave radar, an ultrasonic radar, and a camera device.
[0137] The decision control system 630 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system. The drive system 640 may include components that provide power movement for the vehicle 600. In one embodiment, the drive system 640 may include an engine, an energy source, a transmission system, and wheels. The engine may be a combination of one or more of an internal combustion engine, an electric motor, and an air compression engine. The engine can convert the energy provided by the energy source into mechanical energy.
[0138] Some or all functions of the vehicle 600 are controlled by a computing platform 650. The computing platform 650 may include at least one processor 651 and a memory 652, and the processor 651 may execute instructions 653 stored in the memory 652.
[0139] The processor 651 may be any conventional processor, such as a commercially available CPU. The processor may also include a graphics processor (Graphic Process Unit, GPU), a field programmable gate array (Field Programmable Gate Array, FPGA), a system on chip (System on Chip, SOC), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC) or a combination thereof.
[0140] The memory 652 may be implemented by any type of volatile or nonvolatile 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 memory, flash memory, magnetic disk or optical disk.
[0141] In addition to the instructions 653, the memory 652 may also store data, such as road maps, route information, vehicle location, direction, speed, etc. The data stored in the memory 652 may be used by the computing platform 650. In the disclosed embodiment, the processor 651 may execute the instructions 653 to complete all or part of the steps of the above-mentioned vehicle control method.
[0142] The present disclosure also provides a computer-readable storage medium having computer program instructions stored thereon, and the program instructions, when executed by a processor, implement the steps of the vehicle control method provided by the present disclosure.
[0143] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0144] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, "plurality" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0145] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present disclosure belong.
[0146] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.
[0147] It should be understood that the various parts of the present disclosure can be implemented in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0148] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.
[0149] In addition, each functional unit in each embodiment of the present disclosure may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.
[0150] The storage medium mentioned above may be a read-only memory, a disk or an optical disk, etc. Although the embodiments of the present disclosure have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present disclosure. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present disclosure.
Claims
1. A vehicle control method, characterized in that: include: Acquire multiple images around the vehicle, wherein at least one image around the vehicle contains the target storage location; Input the plurality of vehicle circumference images into a state prediction model to obtain a target heading angle of the vehicle relative to the target storage location, a first target distance between the vehicle and the target storage location in a first direction, and a second target distance between the vehicle and the target storage location in a second direction; Determining a target steering torque, a first target speed, and a second target speed of the vehicle according to the target heading angle, the first target distance, and the second target distance; The vehicle is controlled based on the target steering torque, the first target speed, and the second target speed.
2. The method according to claim 1, characterized in that The inputting of the plurality of vehicle peripheral images into the state prediction model to obtain the target heading angle of the vehicle relative to the target storage location, the first target distance between the vehicle and the target storage location in the first direction, and the second target distance between the vehicle and the target storage location in the second direction comprises: Inputting the plurality of vehicle surrounding images into the deep residual network in the state prediction model to obtain a plurality of first feature maps corresponding to each of the vehicle surrounding images, wherein each of the first feature maps has a different scale and a different number of channels; Inputting the plurality of first feature maps corresponding to each of the vehicle surrounding images into the weighted bidirectional feature pyramid network in the state prediction model for feature fusion to obtain a second feature map corresponding to each of the vehicle surrounding images; splicing the second feature map corresponding to each of the vehicle surrounding images to obtain a bird's-eye view feature map; Based on the bird's-eye view feature map, the target heading angle, the first target distance, and the second target distance are determined.
3. The method according to claim 1, characterized in that Determining a target steering torque, a first target speed, and a second target speed of the vehicle according to the target heading angle, the first target distance, and the second target distance includes: Determine a first target parameter corresponding to the target heading angle, a second target parameter corresponding to the first target distance, and a third target parameter corresponding to the second target distance; Determine the target steering torque by multiplying the first target parameter by the target heading angle; multiplying the second target parameter by the first target distance to determine the first target speed; The product of the third target parameter and the second target distance is determined as the second target speed.
4. The method according to claim 3, characterized in that: The determining of a first target parameter corresponding to the target heading angle, a second target parameter corresponding to the first target distance, and a third target parameter corresponding to the second target distance includes: Based on the target heading angle, the first target distance, and the second target distance, a first mapping table is queried to obtain the first target parameter, the second target parameter, and the third target parameter, wherein the first mapping table includes each heading angle, the first distance in the first direction, the second distance in the second direction, and the corresponding relationship between the first parameter, the second parameter, and the third parameter.
5. The method according to claim 4, characterized in that Also includes: Acquire a first sample data set, wherein the first sample data set includes a sample heading angle between a sample vehicle and a sample parking space, a first sample distance between the sample vehicle and the sample parking space in a first direction, a second sample distance in a second direction, a sample steering torque corresponding to the sample heading angle, a first sample speed corresponding to the first sample distance, and a second sample speed corresponding to the second sample distance; Processing the first sample data set to obtain the first parameter, the second parameter, and the third parameter corresponding to each group of the sample heading angles, the first sample distances, and the second sample distances; The first mapping table is generated based on each group of the sample heading angles, the first sample distances, the second sample distances, and the corresponding first parameters, second parameters, and third parameters.
6. The method according to claim 3, characterized in that: The determining of a first target parameter corresponding to the target heading angle, a second target parameter corresponding to the first target distance, and a third target parameter corresponding to the second target distance includes: Based on the target heading angle, query a second mapping table to obtain the first target parameter; Based on the first target distance, query a third mapping table to obtain the second target parameter; Based on the second target distance, a fourth mapping table is queried to obtain the third target parameter.
7. The method according to claim 6, characterized in that Also includes: Acquire a first sample data set, wherein the first sample data set includes a sample heading angle between a sample vehicle and a sample parking space, a first sample distance between the sample vehicle and the sample parking space in a first direction, a second sample distance in a second direction, a sample steering torque corresponding to the sample heading angle, a first sample speed corresponding to the first sample distance, and a second sample speed corresponding to the second sample distance; Analyze at least one sample steering torque corresponding to each of the sample heading angles to obtain a first parameter corresponding to each of the sample heading angles; Analyze at least one first sample speed corresponding to each of the first sample distances to obtain a second parameter corresponding to each of the first sample distances; Analyze at least one second sample speed corresponding to each of the second sample distances to obtain a third parameter corresponding to each of the second sample distances; Generate the second mapping table based on each of the sample heading angles and the corresponding first parameter; generating the third mapping table based on each of the first sample distances and the corresponding second parameter; The fourth mapping table is generated based on each of the second sample distances and the corresponding third parameter.
8. The method according to claim 1, characterized in that The first direction and the second direction are directions of two parking space lines that are perpendicular to each other in the target parking space.
9. A vehicle control method and device, characterized in that: The device comprises: A first acquisition module is used for a plurality of vehicle surrounding images, wherein at least one of the vehicle surrounding images contains a target storage location; A second acquisition module is used to input the multiple vehicle circumference images into a state prediction model to obtain a target heading angle of the vehicle relative to the target storage location, a first target distance between the vehicle and the target storage location in a first direction, and a second target distance in a second direction; a determination module, configured to determine a target steering torque, a first target speed, and a second target speed of the vehicle according to the target heading angle, the first target distance, and the second target distance; A control module is used to control the vehicle based on the target steering torque, the first target speed and the second target speed.
10. A vehicle, characterized in that: include: processor; A memory for storing processor-executable instructions; wherein the processor is configured to: implement the steps of the method described in any one of claims 1-8.
11. A non-transitory computer-readable storage medium, when instructions in the storage medium are executed by a processor of a mobile terminal, the mobile terminal is enabled to execute the method according to any one of claims 1 to 8.