A method and device for controlling automatic driving of a vehicle

Abstract

The present invention discloses a method and device for controlling an automatic driving of a vehicle, and relates to the field of computer technology. A specific implementation of the method includes: collecting a first environmental image of the environment in which the vehicle is located; using a deep learning model to identify whether there is a traffic sign in the first environmental image; the deep learning model is trained based on multiple sample data of a second environmental image, wherein the sample data indicates whether there is a traffic sign in the second environmental image and the type of the traffic sign; when the traffic sign is identified to exist in the first environmental image, determining a first distance between the vehicle and the traffic sign; and controlling the driving state of the vehicle according to the recognition result of the traffic sign in the first environmental image and the first distance. This implementation reduces the hardware requirements and costs required for automatic driving control, improves the vehicle's ability to perceive the environment, and thus improves the accuracy of automatic driving control.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for controlling automatic driving of a vehicle. Background Art

[0002] With the development of computer technology, the development of autonomous driving technology has also become increasingly mature, and it has gradually been applied to the fields of children's toys, robot teaching and automatic control of motor vehicles. However, the existing autonomous driving control methods require high hardware costs, such as chips with strong computing power and sensors with strong perception capabilities, and the accuracy of the existing control methods in actual application is also low. Summary of the invention

[0003] In view of this, an embodiment of the present invention provides a method and device for controlling an automatic driving of a vehicle, which reduces the amount of calculation in the automatic driving process through a deep learning model, reduces the hardware requirements and costs required for automatic driving control, improves the vehicle's perception of the environment, and thus improves the accuracy of automatic driving control.

[0004] To achieve the above objective, according to one aspect of an embodiment of the present invention, a method for controlling automatic driving of a vehicle is provided.

[0005] A method for controlling an automatic driving of a vehicle according to an embodiment of the present invention includes:

[0006] Acquiring a first environment image of the environment in which the vehicle is located;

[0007] Using a deep learning model to identify whether there is a traffic sign in the first environment image; the deep learning model is trained based on a plurality of sample data of the second environment image, wherein the sample data indicates whether the second environment image contains a traffic sign and the type of the traffic sign;

[0008] When the traffic sign is identified as existing in the first environment image, determining a first distance between the vehicle and the traffic sign;

[0009] The driving state of the vehicle is controlled according to the recognition result of the traffic sign in the first environment image and the first distance.

[0010] Optionally, controlling the driving state of the vehicle according to the recognition result of the traffic sign in the first environment image and the first distance includes:

[0011] The deep learning model is used to identify the type of the traffic sign, and the driving state of the vehicle is controlled according to the type of the traffic sign and the first distance.

[0012] Optionally, the method further comprises:

[0013] Collecting driving data of the vehicle, determining a current position of the vehicle according to the driving data, and constructing a driving map corresponding to the current position;

[0014] A driving path of the vehicle is determined according to the driving map.

[0015] Optionally, controlling the driving state of the vehicle according to the recognition result of the traffic sign in the first environment image and the first distance further includes:

[0016] The speed at which the vehicle travels along the travel path is controlled according to a first distance between the current position of the vehicle and the traffic sign and the type of the traffic sign.

[0017] Optionally,

[0018] The determining a first distance between the vehicle and the traffic sign comprises:

[0019] The first distance is calculated according to the projection position of the first environment image corresponding to the acquisition point at the current position when the first environment image is acquired, and the position of the traffic sign corresponding to the first environment image at the current position.

[0020] Optionally, the method further comprises:

[0021] When it is recognized that the traffic sign does not exist in the first environment image, the vehicle is controlled to travel in the current state.

[0022] Optionally,

[0023] The deep learning model is a Mobilenet-SSD model.

[0024] To achieve the above objective, according to another aspect of an embodiment of the present invention, a vehicle automatic driving control device is provided.

[0025] An automatic driving control device for a vehicle according to an embodiment of the present invention comprises: a collection module, a recognition module and a control module; wherein:

[0026] The acquisition module is used to acquire a first environment image of the environment in which the vehicle is located;

[0027] The recognition module is used to use a deep learning model to identify whether there is a traffic sign in the first environment image, where the deep learning model is trained based on multiple sample data of the second environment image, wherein the sample data indicates whether the second environment image contains a traffic sign and the type of the traffic sign;

[0028] The control module is used to determine a first distance between the vehicle and the traffic sign when the traffic sign is identified in the first environment image, and control the driving state of the vehicle according to the recognition result of the traffic sign in the first environment image and the first distance.

[0029] Optionally, the control module is used to use the deep learning model to identify the type of the traffic sign, and control the driving state of the vehicle according to the type of the traffic sign and the first distance.

[0030] Optionally, the control module is used to control the vehicle to travel in a current state when it is recognized that the traffic sign does not exist in the first environment image.

[0031] To achieve the above objective, according to another aspect of an embodiment of the present invention, an electronic device for controlling automatic driving of a vehicle is provided.

[0032] An electronic device for controlling vehicle automatic driving according to an embodiment of the present invention comprises: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement a vehicle automatic driving control method according to an embodiment of the present invention.

[0033] To achieve the above objective, according to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided.

[0034] A computer-readable storage medium according to an embodiment of the present invention stores a computer program, which, when executed by a processor, implements an automatic driving control method for a vehicle according to an embodiment of the present invention.

[0035] One embodiment of the above invention has the following advantages or beneficial effects: after collecting the first environmental image of the vehicle's environment, a deep learning model is used to identify whether there is a traffic sign in the first environmental image. When the traffic sign is identified in the first environmental image, the first distance between the vehicle and the traffic sign is determined, and then the driving state of the vehicle is controlled based on the recognition result of the traffic sign in the first environmental image and the first distance. Since the deep learning model has the advantages of low parameter quantity and low computational complexity, it can reduce the amount of computation in the process of automatic driving, thereby reducing the hardware requirements for automatic driving control and the cost of automatic driving control, so that the automatic driving control method of the vehicle can be applied to low-cost automatic driving fields such as children's toys and robot teaching. In addition, by identifying traffic signs through a deep learning model and controlling the driving state of the vehicle based on the recognition results, the vehicle's perception of the environment is improved, thereby improving the accuracy of automatic driving control.

[0036] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0038] Figure 1 is a schematic diagram of main steps of an automatic driving control method for a vehicle according to an embodiment of the present invention;

[0039] Figure 2 is a schematic diagram of main steps of another automatic driving control method for a vehicle according to an embodiment of the present invention;

[0040] Figure 3 is a schematic diagram of main steps of another method for controlling an automatic driving of a vehicle according to an embodiment of the present invention;

[0041] Figure 4 is a schematic diagram of main steps of another method for controlling an automatic driving of a vehicle according to an embodiment of the present invention;

[0042] Figure 5 is a schematic diagram of main modules of an automatic driving control device for a vehicle according to an embodiment of the present invention;

[0043] Figure 6 is a schematic diagram of main modules of a vehicle according to an embodiment of the present invention;

[0044] Figure 7 is a schematic diagram of connection between an automatic driving control device of a vehicle and a vehicle according to an embodiment of the present invention;

[0045] Figure 8 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0046] Fig. 9 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0048] It should be pointed out that the embodiments of the present invention and the technical features therein may be combined with each other without conflict.

[0049] Figure 1 Schematic diagram of main steps of an automatic driving control method for a vehicle according to an embodiment of the present invention.

[0050] like Figure 1 As shown, a method for controlling an automatic driving of a vehicle according to an embodiment of the present invention mainly comprises the following steps:

[0051] Step S101: Acquire a first environment image of the environment in which the vehicle is located.

[0052] The vehicle may be equipped with a camera for collecting the first environment image, such as a Raspberry Pi camera, so as to collect the first environment image of the environment in which the vehicle is located in the form of a video or a static image during the driving of the vehicle. It is understandable that when the camera collects the first environment image in the form of a video, each frame in the video is a first environment image.

[0053] Step S102: Use a deep learning model to identify whether there is a traffic sign in the first environmental image; the deep learning model is trained based on multiple sample data of the second environmental image, wherein the sample data indicates whether the second environmental image contains a traffic sign and the type of the traffic sign.

[0054] The deep learning model can be trained in advance. Specifically, the second environment image marked with whether there is a traffic sign and the type of traffic sign can be used as sample data to train the deep learning model. For example, sample data A is a second environment image marked with a red light, sample data B is a second environment image marked with whether there is a traffic sign of a red light, and sample data C is a second environment image marked with a green light. Therefore, compared with the machine learning algorithm, the use of labeled sample data to train the deep learning model reduces the steps of manually extracting features and rules, and reduces the loss of image edges in the feature extraction process, thereby not only reducing the amount of calculation in the model training process, but also improving the model prediction accuracy, thereby reducing the hardware requirements for autonomous driving control and reducing the cost of autonomous driving control, making the autonomous driving control method suitable for low-cost autonomous driving fields with limited calculations, such as children's toys and robot teaching.

[0055] It is understandable that in order to improve the prediction accuracy of the deep learning model, when the sample data is a second environment image marked with a traffic sign, the position of the traffic sign in the second image can be further marked. In addition, in addition to traffic lights, traffic signs can also be other traffic signs such as speed limit signs or turn signs.

[0056] Among them, the deep learning model can be a Mobilenet-SSD model. The Mobilenet-SSD model is a lightweight deep network model proposed for mobile terminals, which can be used for fast target detection. It has the advantages of low parameter count and low computational complexity, and is particularly suitable for low-cost deep learning scenarios. Therefore, it is suitable for low-cost autonomous driving fields such as children's toys and robot teaching. In an embodiment of the present invention, the depthwise separable convolution Depthwise SeparableConvolution is mainly used to decompose and calculate the standard convolution kernel to reduce the amount of calculation.

[0057] When using a deep learning model to identify traffic signs in the first environment image, the first environment image captured by the camera can be preprocessed first, for example, the resolution of the first environment image can be reduced to reduce the amount of calculation in the process of traffic sign recognition, thereby improving the feedback speed of traffic signs, thereby meeting the real-time requirements of autonomous driving. In an embodiment of the present invention, the Mobilenet-SSD model can be transplanted to the Jetson Nano embedded motherboard, and the detection of traffic signs can be completed under the framework of Tensorflow.

[0058] Step S103: when it is recognized that the traffic sign exists in the first environment image, a first distance between the vehicle and the traffic sign is determined.

[0059] Step S104: controlling the driving state of the vehicle according to the recognition result of the traffic sign in the first environment image and the first distance.

[0060] In an embodiment of the present invention, when it is recognized that the traffic sign exists in the first environment image, the type of the traffic sign can be identified using the deep learning model, and the driving state of the vehicle can be controlled according to the type of the traffic sign and the first distance between the vehicle and the traffic sign. When it is recognized that the traffic sign does not exist in the first environment image, the vehicle can be controlled to drive in the current state.

[0061] Taking the traffic sign as a red, yellow, and green light as an example, the deep recognition model can not only identify whether there is a red, yellow, and green light in the first environment image, but also identify the type of the red, yellow, and green light, that is, which of the red, yellow, and green lights the traffic sign is specifically. When the deep recognition model identifies that there is no red, yellow, and green light in the first environment image, the vehicle can be controlled to continue driving in the current state (current driving direction and current driving speed, etc.). When the red, yellow, and green lights are identified in the first environment image, the driving state of the vehicle is controlled according to the type of red, green, yellow, etc. and the distance between the vehicle and the red, green, and yellow lights. For example, when the traffic sign is a red light, the vehicle is controlled to stop driving, and when the traffic sign is a green light, the vehicle is controlled to continue driving in the current state. During the control process, the vehicle's driving speed can be controlled in combination with the distance between the vehicle and the traffic sign. In the actual application process, the recognition accuracy of the traffic sign can reach more than 85%, indicating that the vehicle control method provided by the embodiment of the present invention is highly reliable.

[0062] Based on this, the vehicle control method provided by the embodiment of the present invention may include the following steps S201 to S204:

[0063] Step S201: Acquire a first environment image of the environment in which the vehicle is located.

[0064] Step S202: Use the deep learning model to determine whether there is a traffic sign in the first environment image. If so, execute step S203; otherwise, execute step S204.

[0065] Step S203: Use the deep learning model to identify the type of traffic sign, and control the driving state of the vehicle according to the type of traffic sign and the first distance between the vehicle and the traffic sign.

[0066] For example, when the traffic light is red, the vehicle is controlled to stop driving, and when the traffic light is green, the vehicle is controlled to continue driving in the current state.

[0067] Step S204: Control the vehicle to continue driving in the current state.

[0068] In addition, during the driving process of the vehicle, the vehicle's driving data can be collected through equipment such as laser radar and sensors installed on the vehicle, the current position of the vehicle can be determined based on the driving data, and a driving map corresponding to the current position can be constructed; the driving path of the vehicle can be determined based on the driving map.

[0069] For example, the environmental information of the vehicle's environment can be collected by installing an RPLIDAR laser ranging radar on the vehicle, and the mileage information of the vehicle can be collected by using an ultrasonic sensor installed on the vehicle. Then, the Gmapping algorithm can be used to locate the vehicle based on the information collected by the laser ranging radar and the mileage information collected by the ultrasonic sensor, and a driving map corresponding to the current position of the vehicle can be constructed. In other words, the input of the Gmapping algorithm includes the information collected by the laser ranging radar and the mileage information, and the output of the Gmapping algorithm includes the current position of the vehicle and its corresponding driving map. When the driving state of the vehicle is subsequently controlled, the driving path of the vehicle can be determined based on the driving map.

[0070] Among them, the Gmapping algorithm is a commonly used open source SLAM algorithm based on the filtering SLAM framework. It is currently a relatively reliable and mature algorithm based on laser radar and odometer solutions. When the Gmapping algorithm constructs a small scene map, the required calculation amount is small and the accuracy is high. Therefore, the Gmapping algorithm is suitable for realizing SLAM tasks in low-cost autonomous driving fields such as children's toys and robot teaching, and can realize accurate positioning and map construction of autonomous driving vehicles in this field. In the actual application process, the navigation accuracy of the SLAM task is detected to be accurate to ±4mm, indicating that the vehicle control method provided by the embodiment of the present invention is highly reliable.

[0071] It can be understood that the first environment image is collected when the vehicle is traveling along the driving path, and based on the collected first environment image, the vehicle's driving path can be returned and updated. For example, when a traffic sign such as a turn or a red light is recognized, the vehicle's driving state will change, and the information collected by the laser ranging radar and the ultrasonic sensor will change, so the vehicle's driving map and driving path will also change, thereby realizing automatic driving of the vehicle. Based on this, the vehicle control method provided by the embodiment of the present invention may include the following steps S301 to S305:

[0072] Step S301: collecting driving data of the vehicle, determining the current position of the vehicle according to the driving data, and constructing a driving map corresponding to the current position.

[0073] Step S302: Determine the driving path of the vehicle according to the driving map.

[0074] Step S303: collecting a first environment image of the environment in which the vehicle is located when the vehicle travels along the driving path.

[0075] Step S304: Control the driving state of the vehicle according to the recognition result of the deep learning model on the traffic sign in the first environment image.

[0076] Step S305: updating the driving path of the vehicle according to the driving data corresponding to the driving state of the vehicle.

[0077] Furthermore, when controlling the driving state of the vehicle, the first distance between the current position of the vehicle and the traffic sign can be determined first, and then the speed of the vehicle along the driving path can be controlled based on the first distance and the type of traffic sign.

[0078] Specifically, the first distance may be calculated based on the projection position of the first environment image corresponding to the current position of the acquisition point when the first environment image is acquired, and the position of the traffic sign corresponding to the first environment image at the current position.

[0079] In an embodiment of the present invention, when the Mobilenet-SSD model is transplanted to the Jetson Nano embedded motherboard, that is, the vehicle is equipped with a Jetson Nano embedded motherboard transplanted with the Mobilenet-SSD model, since the Jetson Nano embedded motherboard can only carry a single Raspberry Pi camera, a monocular ranging solution can be used when calculating the distance between the current position of the vehicle and the traffic sign. Specifically, the first distance between the current position of the vehicle and the traffic sign can be calculated using the following calculation formula based on the projection position of the acquisition point (i.e., the position of the Raspberry Pi camera) corresponding to the first environmental image at the current position when the first environmental image is acquired, and the position of the traffic sign corresponding to the first environmental image at the current position:

[0080]

[0081] Where D represents the first distance between the current position of the vehicle and the traffic sign, h represents the height from the acquisition point (the position of the Raspberry Pi camera) to the ground when acquiring the first environment image, y represents the ordinate of the traffic sign in the rectangular coordinate system corresponding to the first environment image, and y 0 represents the ordinate of the origin of the rectangular coordinate system corresponding to the first environment image, f represents the camera focal length of the acquisition device (Raspberry Pi camera) for acquiring the first environment image, and α represents the tilt angle of the acquisition device (Raspberry Pi camera) when acquiring the first environment image.

[0082] Therefore, the vehicle control method provided by the embodiment of the present invention may include the following steps S401 to S404:

[0083] Step S401: collecting driving data of the vehicle, determining the current position of the vehicle according to the driving data, constructing a driving map corresponding to the current position, and determining the driving path of the vehicle according to the driving map.

[0084] Step S402: Acquire a first environment image of the environment in which the vehicle is located when traveling along the driving path.

[0085] Step S403: When a traffic sign is identified in the first environmental image using a deep learning model, a first distance between the current position of the vehicle and the traffic sign is calculated based on the projection position of the first environmental image corresponding to the acquisition point at the current position when the first environmental image was acquired, and the position of the traffic sign corresponding to the first environmental image at the current position.

[0086] Step S404: Determine the speed of the vehicle traveling along the driving path according to the first distance and the type of traffic sign identified by the deep learning model, and control the vehicle to travel according to the speed.

[0087] For example, when a traffic sign is identified as a red light, the vehicle needs to be controlled to stop. It can be understood that the process of controlling the vehicle to stop is the process of controlling the vehicle to decelerate. At this time, the speed of the vehicle during deceleration can be further controlled based on the first distance between the current position of the vehicle and the traffic sign when the traffic sign is identified as a red light.

[0088] In addition, the vehicle's driving path can be determined based on the driving map output by the Gmapping algorithm. For example, if the vehicle's driving path is determined to be a straight path (i.e., the vehicle travels in a straight line) or a curved path (the vehicle needs to turn), then when controlling the vehicle's driving speed, the vehicle's driving speed can be further determined in combination with the driving path. For example, when the driving path is a curved path, the vehicle's driving speed can be determined based on the curvature of the curve.

[0089] After determining the vehicle's current position and driving map, determining the vehicle's driving path and speed, and identifying information such as traffic signs in the vehicle's environment, the vehicle's laser SLAM task is completed and the vehicle's environmental perception is realized. Then, the Movebase navigation package in ROS can be used to complete global and local path planning to achieve vehicle control.

[0090] When controlling the vehicle, specific control operations can be implemented based on the vehicle's control board Arduino. Among them, the control board Arduino pin sends a low signal to indicate that the controller chip pin is grounded, and the control board Arduino sends a high signal to indicate that the resistance between the chip pin and the ground remains unchanged. The control board Arduino can be connected to the Jetson Nano embedded motherboard transplanted with the Mobilenet-SSD model via USB, and then the Jetson Nano embedded motherboard can use the serial interface to output the control command to the control board Arduino, so that the control board Arduino reads the control command and outputs a low signal or a high signal accordingly to drive the vehicle. As a result, the Jetson Nano embedded motherboard directly communicates with the control board Arduino for the recognition results of traffic signs such as traffic lights, thereby directly affecting the behavior of the vehicle at the intersection, improving the vehicle's perception of the environment, and thus improving the accuracy of autonomous driving control.

[0091] It can be understood that the vehicle described in the embodiment of the present invention can be a smart car used in the fields of children's toys and robot teaching. It can be seen from the automatic driving control method of the vehicle according to the embodiment of the present invention that after collecting the first environmental image of the environment in which the vehicle is located, the deep learning model is used to identify whether there is a traffic sign in the first environmental image. When the traffic sign is identified in the first environmental image, the first distance between the vehicle and the traffic sign is determined, and then the driving state of the vehicle is controlled based on the recognition result of the traffic sign in the first environmental image and the first distance. Since the deep learning model has the advantages of low parameter amount and low computation amount, it can reduce the amount of calculation in the automatic driving process, thereby reducing the hardware requirements required for automatic driving control and reducing the cost of automatic driving control, so that the automatic driving control method of the vehicle can be applied to low-cost automatic driving fields such as children's toys and robot teaching. In addition, by identifying traffic signs through deep learning models and controlling the driving state of the vehicle based on the recognition results, the vehicle's perception of the environment is improved, thereby improving the accuracy of automatic driving control, and thus improving the fun of controlling the automatic driving of the vehicle in the fields of children's toys and robot teaching.

[0092] Figure 5 is a schematic diagram of main modules of an automatic driving control device for a vehicle according to an embodiment of the present invention.

[0093] like Figure 5 As shown, the vehicle control device 500 of the embodiment of the present invention includes: a collection module 501, an identification module 502 and a control module 503; wherein,

[0094] The acquisition module 501 is used to acquire a first environment image of the environment in which the vehicle is located;

[0095] The recognition module 502 is used to recognize whether there is a traffic sign in the first environment image by using a deep learning model, wherein the deep learning model is trained based on a plurality of sample data of the second environment image, wherein the sample data indicates whether there is a traffic sign in the second environment image and the type of the traffic sign;

[0096] The control module 503 is used to determine a first distance between the vehicle and the traffic sign when the traffic sign is identified in the first environment image, and control the driving state of the vehicle according to the recognition result of the traffic sign in the first environment image and the first distance.

[0097] Among them, the acquisition module 501 can be implemented based on a Raspberry Pi camera, the recognition module 502 can be implemented based on a Jetson Nano embedded motherboard transplanted with a Mobilenet-SSD model, and the control module 503 can be implemented based on an Arduino control board, which is connected to the Jetson Nano embedded motherboard. In addition, the automatic driving control device 500 for a vehicle provided in an embodiment of the present invention can be set on the vehicle, such as Figure 6 As shown, the embodiment of the present invention further provides a vehicle 600, which includes: a vehicle control device 500 provided in any embodiment of the present invention and a vehicle body 601. Figure 7 As shown, the vehicle control device 500 may also be independent of the vehicle 700. For example, the vehicle's automatic driving control device 500 may serve as a companion device of the vehicle 700 and follow the movement of the vehicle.

[0098] In one embodiment of the present invention, the control module 503 is used to identify the type of the traffic sign using the deep learning model, and control the driving state of the vehicle according to the type of the traffic sign and the first distance.

[0099] In one embodiment of the present invention, the control module 503 is used to control the vehicle to travel in the current state when it is recognized that the traffic sign does not exist in the first environment image.

[0100] In one embodiment of the present invention, the identification module 502 is also used to collect driving data of the vehicle, determine the current position of the vehicle based on the driving data, and construct a driving map corresponding to the current position; determine the driving path of the vehicle based on the driving map.

[0101] In one embodiment of the present invention, the control module 503 is used to control the speed of the vehicle traveling along the driving path according to the first distance between the current position of the vehicle and the traffic sign and the type of the traffic sign.

[0102] In one embodiment of the present invention, the control module 503 is used to calculate the first distance based on the projection position of the first environmental image corresponding to the current position of the acquisition point when acquiring the first environmental image, and the position of the traffic sign corresponding to the first environmental image at the current position.

[0103] In one embodiment of the present invention, the deep learning model is a Mobilenet-SSD model.

[0104] It can be seen from the control device of the vehicle according to the embodiment of the present invention that after collecting the first environmental image of the environment in which the vehicle is located, a deep learning model is used to identify whether there is a traffic sign in the first environmental image. When the traffic sign is identified in the first environmental image, the first distance between the vehicle and the traffic sign is determined, and then the driving state of the vehicle is controlled based on the recognition result of the traffic sign in the first environmental image and the first distance. Since the deep learning model has the advantages of low parameter quantity and low computational complexity, it can reduce the amount of computation in the process of automatic driving, thereby reducing the hardware requirements for automatic driving control and the cost of automatic driving control, so that the automatic driving control method of the vehicle can be applied to low-cost automatic driving fields such as children's toys and robot teaching. In addition, by identifying traffic signs through a deep learning model and controlling the driving state of the vehicle based on the recognition results, the vehicle's perception of the environment is improved, thereby improving the accuracy of automatic driving control.

[0105] Figure 8 An exemplary system architecture 800 is shown to which an autonomous driving control method for a vehicle or an autonomous driving control device for a vehicle according to an embodiment of the present invention may be applied.

[0106] like Figure 8 As shown, system architecture 800 may include terminal devices 801, 802, 803, a network 804, and a server 805. Network 804 is used to provide a medium for communication links between terminal devices 801, 802, 803 and server 805. Network 804 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0107] Users can use terminal devices 801, 802, and 803 to interact with server 805 through network 804 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 801, 802, and 803, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0108] The terminal devices 801 , 802 , and 803 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0109] The server 805 may be a server that provides various services, such as a backend management server that provides support for shopping websites browsed by users using the terminal devices 801, 802, and 803. The backend management server may analyze and process the received data such as product information query requests, and feed back the processing results (such as target push information and product information) to the terminal device.

[0110] It should be noted that the automatic driving control method for the vehicle provided in the embodiment of the present invention is generally executed by the server 805, and accordingly, the automatic driving control device for the vehicle is generally arranged in the server 805.

[0111] It should be understood that Figure 8 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0112] Reference below Fig. 9 , which shows a schematic diagram of the structure of a computer system 900 of a terminal device suitable for implementing an embodiment of the present invention. Fig. 9 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0113] like Fig. 9 As shown, the computer system 900 includes a central processing unit (CPU) 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage part 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the system 900 are also stored. The CPU 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0114] The following components are connected to the I / O interface 905: an input section 906 including a keyboard, a mouse, etc.; an output section 907 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, a modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the I / O interface 905 as needed. A removable medium 911, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 910 as needed, so that a computer program read therefrom is installed into the storage section 908 as needed.

[0115] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 909, and / or installed from the removable medium 911. When the computer program is executed by the central processing unit (CPU) 901, the above-mentioned functions defined in the system of the present invention are executed.

[0116] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0117] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0118] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor, for example, it may be described as: a processor includes an acquisition module, an identification module and a control module. The names of these modules do not constitute a limitation on the modules themselves in some cases, for example, the acquisition module may also be described as "a module for acquiring a first environmental image of the environment in which the vehicle is located".

[0119] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: collecting a first environmental image of the environment in which the vehicle is located; using a deep learning model to identify whether there is a traffic sign in the first environmental image; the deep learning model is trained based on multiple sample data of the second environmental image, wherein the sample data indicates whether there is a traffic sign in the second environmental image and the type of the traffic sign; when the traffic sign is identified in the first environmental image, determining the first distance between the vehicle and the traffic sign; and controlling the driving state of the vehicle according to the recognition result of the traffic sign in the first environmental image and the first distance.

[0120] According to the technical solution of the embodiment of the present invention, after collecting the first environmental image of the vehicle's environment, a deep learning model is used to identify whether there is a traffic sign in the first environmental image. When the traffic sign is identified in the first environmental image, the first distance between the vehicle and the traffic sign is determined, and then the driving state of the vehicle is controlled based on the recognition result of the traffic sign in the first environmental image and the first distance. Since the deep learning model has the advantages of low parameter quantity and low computational complexity, it can reduce the amount of computation in the process of automatic driving, thereby reducing the hardware requirements for automatic driving control and the cost of automatic driving control, so that the automatic driving control method of the vehicle can be applied to low-cost automatic driving fields such as children's toys and robot teaching. In addition, by identifying traffic signs through a deep learning model and controlling the driving state of the vehicle based on the recognition results, the vehicle's perception of the environment is improved, thereby improving the accuracy of automatic driving control.

[0121] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for controlling an automatic driving of a vehicle, characterized in that: include: Acquiring a first environment image of the environment in which the vehicle is located; Using a deep learning model to identify whether there is a traffic sign in the first environment image; The deep learning model is trained based on a plurality of sample data of the second environment image, wherein the sample data indicates whether the second environment image contains a traffic sign and the type of the traffic sign; When the traffic sign is identified as existing in the first environment image, determining a first distance between the vehicle and the traffic sign; Controlling the driving state of the vehicle according to the recognition result of the traffic sign in the first environment image and the first distance; specifically comprising: using the deep learning model to identify the type of the traffic sign; collecting driving data of the vehicle, determining the current position of the vehicle according to the driving data, and constructing a driving map corresponding to the current position; determining the driving path of the vehicle according to the driving map; determining the driving speed of the vehicle based on the first distance, the type of the traffic sign and the driving path, so as to control the driving speed of the vehicle along the driving path; The driving path is updated according to the first environment image.

2. The method according to claim 1, characterized in that The determining a first distance between the vehicle and the traffic sign comprises: The first distance is calculated according to the projection position of the first environment image corresponding to the acquisition point at the current position when the first environment image is acquired, and the position of the traffic sign corresponding to the first environment image at the current position.

3. The method according to claim 1, characterized in that Also includes: When it is recognized that the traffic sign does not exist in the first environment image, the vehicle is controlled to travel in the current state.

4. The method according to any one of claims 1 to 3, characterized in that: The deep learning model is a Mobilenet-SSD model.

5. An automatic driving control device for a vehicle, characterized in that: include: Acquisition module, identification module and control module; wherein, The acquisition module is used to acquire a first environment image of the environment in which the vehicle is located; The recognition module is used to use a deep learning model to identify whether there is a traffic sign in the first environment image, where the deep learning model is trained based on multiple sample data of the second environment image, wherein the sample data indicates whether the second environment image contains a traffic sign and the type of the traffic sign; The control module is used to determine a first distance between the vehicle and the traffic sign when the traffic sign is identified as existing in the first environmental image, and control the driving state of the vehicle according to the recognition result of the traffic sign in the first environmental image and the first distance; specifically, to identify the type of the traffic sign using the deep learning model; to collect driving data of the vehicle, to determine the current position of the vehicle according to the driving data, and to construct a driving map corresponding to the current position; to determine the driving path of the vehicle according to the driving map; to determine the driving speed of the vehicle based on the first distance, the type of the traffic sign and the driving path, so as to control the driving speed of the vehicle along the driving path; and to update the driving path according to the first environmental image.

6. The device according to claim 5, characterized in that The control module is used to use the deep learning model to identify the type of the traffic sign, and control the driving state of the vehicle according to the type of the traffic sign and the first distance.

7. The device according to claim 5, characterized in that The control module is used to control the vehicle to travel in a current state when it is recognized that the traffic sign does not exist in the first environment image.

8. An electronic device for controlling automatic driving of a vehicle, characterized in that: include: one or more processors; a storage device for storing 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 method according to any one of claims 1 to 4.

9. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images