Intelligent network connection practical training trolley and control method thereof

Abstract

The invention relates to an intelligent network connection practical training trolley and a control method thereof, and the method comprises the steps: inputting video data into a traffic sign detection model of the practical training trolley, so as to obtain the traffic sign information of the video data; inputting the radar data into an obstacle detection model of the practical training trolley, and if detecting that an obstacle exists in the initial driving route, obtaining obstacle data; according to the obstacle data and the traffic sign information, modifying the initial driving route to obtain a target driving route; and controlling the driving speed of the training trolley along the modified target driving route according to the traffic sign information, so that the training trolley bypasses the obstacle according to the target driving route. According to the invention, the unmanned driving capability of the training trolley can be improved.

Description

Technical Field

[0001] This application relates to the technical field of driverless vehicles, and particularly to an intelligent connected training vehicle and its control method. Background Art

[0002] Existing vehicles can obtain image data and radar data of the vehicle's environment, but fail to apply the image data and radar data to the vehicle's autonomous driving. As a result, in the prior art, the driverless of the vehicle cannot be combined with the image data and radar data, resulting in the lack of the ability to bypass obstacles when the vehicle is driverless, that is, the driverless ability of the existing vehicles is poor. Summary of the Invention

[0003] Based on this, the purpose of this application is to overcome the deficiencies of the prior art and provide an intelligent connected training vehicle and its control method, which can improve the driverless ability of the training vehicle.

[0004] To achieve the above purpose, the technical solution adopted in this application is as follows:

[0005] The first embodiment of this application provides an intelligent connected training vehicle, including: a drive-by-wire chassis, a control main board, a 360-degree scanning lidar, a front-view intelligent camera, a millimeter-wave radar, and a combined positioning unit integrating an inertial gyroscope and GNSS;

[0006] The drive-by-wire chassis is provided with a steering wheel group, and the control main board, the 360-degree scanning lidar, the front-view intelligent camera, the millimeter-wave radar, and the combined positioning unit integrating an inertial gyroscope and GNSS are arranged above the drive-by-wire chassis;

[0007] The control main board is respectively connected to the drive-by-wire chassis, the 360-degree scanning lidar, the front-view intelligent camera, the millimeter-wave radar, and the combined positioning unit integrating an inertial gyroscope and GNSS. The control main board is used for data processing of image recognition, SLAM positioning, environment perception, obstacle detection, traffic sign recognition, multi-sensor fusion, autonomous driving decision-making and control.

[0008] The second embodiment of this application provides an intelligent connected training vehicle control method, which is applied to the intelligent connected training vehicle as described above. The method includes:

[0009] Obtain the initial driving route of the training vehicle, as well as the video data and radar data during the driving process of the training vehicle along the initial driving route;

[0010] Input the video data into the traffic sign detection model of the training vehicle to obtain the traffic sign information of the video data;

[0011] Inputting the radar data into the obstacle detection model of the training vehicle, and obtaining obstacle data if an obstacle is detected on the initial driving route;

[0012] Modifying the initial driving route according to the obstacle data and the traffic sign information to obtain a target driving route;

[0013] The driving speed of the training vehicle along the modified target driving route is controlled according to the traffic sign information, so that the training vehicle bypasses the obstacle according to the target driving route.

[0014] As an implementation manner, the step of modifying the initial driving route according to the obstacle data and the traffic sign information to obtain the target driving route includes:

[0015] Modifying the driving route according to the obstacle data to obtain a left detour route and a right detour route around the obstacle;

[0016] Determining the validity of the left detour route and the right detour route according to the traffic sign information;

[0017] The target driving route is determined according to the effective left detour route and / or the right detour route.

[0018] As an implementation manner, the step of determining the target driving route according to the valid left detour route and / or the right detour route includes:

[0019] If only the left detour route is valid, determine the left detour route as the target driving route; if only the right detour route is valid, determine the right detour route as the target driving route;

[0020] If both the left detour route and the right detour route are valid, obtaining a first offset of the left detour route relative to the initial driving route, and a second offset of the right detour route relative to the initial driving route;

[0021] If the first offset is less than or equal to the second offset, the left detour route is determined to be the target driving route; if the first offset is greater than the second offset, the right detour route is determined to be the target driving route.

[0022] As an implementation manner, the step of determining the validity of the left detour route and the right detour route according to the traffic sign information includes:

[0023] If the traffic sign information does not contain a sign prohibiting lane switching to the left, it is determined that the left detour route is valid; if the traffic sign information does not contain a sign prohibiting lane switching to the right, it is determined that the right detour route is valid.

[0024] As an implementation manner, the step of controlling the driving speed of the training vehicle along the modified target driving route according to the traffic sign information includes:

[0025] If the traffic sign information signal light is marked, and the signal light is a red signal light, the training vehicle is driven to stop; if the signal light is a green signal light, the training vehicle is driven to travel normally.

[0026] As an implementation manner, after the step of inputting the video data into the traffic sign detection model of the training vehicle to obtain the traffic sign information of the video data, the method further includes:

[0027] Using the video data and the traffic sign information as training samples to train the traffic sign detection model, and recording the number of times the traffic sign detection model is trained;

[0028] When the training times reach a preset times threshold, the model parameter change data of the traffic sign detection model is uploaded to the car model server of the road where the training car is currently located, so that the car model server trains and optimizes the base station traffic sign detection model according to the model parameter change data uploaded by multiple cars;

[0029] According to the trained and optimized base station traffic sign detection model sent by the car model server, a new traffic sign detection model for the training car is determined.

[0030] As an implementation manner, the step of using the video data and the traffic sign information as training samples to train the traffic sign detection model, and recording the number of times the traffic sign detection model is trained, includes:

[0031] According to a current video frame in the video data, and traffic sign information corresponding to the current video frame;

[0032] Comparing the traffic sign information corresponding to the previous video frame for training the traffic sign detection model with the traffic sign information corresponding to the current video frame to obtain a comparison similarity;

[0033] If the comparison similarity is greater than a preset similarity threshold, the traffic sign detection model is trained according to the current video frame and the corresponding traffic sign information, so that the number of training times of the traffic sign detection model is increased by 1.

[0034] As an implementation manner, the step of comparing the traffic sign information corresponding to the previous video frame for training the traffic sign detection model with the traffic sign information corresponding to the current video frame to obtain the comparison similarity includes:

[0035] If the compared similarity is less than or equal to the similarity threshold, the next video frame is determined as a new current video frame.

[0036] Compared with the traditional technology, the beneficial effects described in this application are:

[0037] The control motherboard of the intelligent networked training car of the present application is respectively connected to the wire-controlled chassis, 360-degree scanning laser radar, forward-looking intelligent camera, millimeter-wave radar, integrated inertial gyroscope and GNSS combined positioning unit, which can perform image recognition, SLAM positioning, environmental perception, obstacle detection, traffic sign recognition, multi-sensor fusion, automatic driving decision-making and control data processing, so the control motherboard can be combined with the radar data obtained by the 360-degree scanning laser radar or millimeter-wave radar, and the image data obtained by the forward-looking intelligent camera to control the wire-controlled chassis to control the driving direction of the intelligent networked training car, so that the intelligent networked training car can bypass obstacles when unmanned. Moreover, the present application can also obtain the traffic sign information and obstacle data encountered by the training car when driving along the initial driving route through the traffic sign detection model and the obstacle detection model, so as to modify the initial driving route according to the obstacle data and traffic sign information, and control the driving speed of the training car along the modified target driving route according to the traffic sign information, so that the training car bypasses the obstacles according to the target driving route, and improves the unmanned driving ability of the training car.

[0038] For better understanding and implementation, the present application is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A schematic diagram of an intelligent network-connected training vehicle according to an embodiment of the present application;

[0040] Figure 2 This is a flow chart of a control method for an intelligent network-connected training vehicle according to an embodiment of the present application;

[0041] 10. Wire-controlled chassis; 20. Control main board; 30. 360-degree scanning laser radar. DETAILED DESCRIPTION

[0042] In order to make the objectives, technical solutions and advantages of the present application more clear, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0043] It should be clear that the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the embodiments of the present application.

[0044] When the following description relates to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances. The singular forms of "a", "said" and "the" used in the present application and the appended claims are also intended to include the majority form, unless the context clearly indicates other meanings. The words "if" / "if" used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0045] In addition, in the description of this application, unless otherwise specified, "plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.

[0046] See also Figure 1 , which is a schematic diagram of an intelligent network-connected training vehicle of the first embodiment of the present application, the training vehicle comprises: a wire-controlled chassis 10, a control mainboard 20, a 360-degree scanning laser radar 30, a forward-looking intelligent camera, a millimeter-wave radar, and a combined positioning unit integrating an inertial gyroscope and a GNSS;

[0047] The control-by-wire chassis 10 is provided with a steering wheel set, and the control main board 20, the 360-degree scanning laser radar 30, the front-view intelligent camera, the millimeter-wave radar, and the combined positioning unit integrating the inertial gyroscope and the GNSS are arranged above the control-by-wire chassis;

[0048] The control main board 20 is respectively connected to the wire-controlled chassis 10, the 360-degree scanning laser radar 30, the forward-looking intelligent camera, the millimeter-wave radar, and the combined positioning unit of the integrated inertial gyroscope and the GNSS. The control main board is used for image recognition, SLAM positioning, environmental perception, obstacle detection, traffic sign recognition, multi-sensor fusion, and data processing for autonomous driving decision-making and control.

[0049] Compared with the prior art, the control motherboard 20 of the intelligent networked training vehicle of the present application is respectively connected to the wire-controlled chassis 10, the 360-degree scanning laser radar 30, the forward-looking intelligent camera, the millimeter-wave radar, and the combined positioning unit of the integrated inertial gyroscope and the GNSS, and can perform image recognition, SLAM positioning, environmental perception, obstacle detection, traffic sign recognition, multi-sensor fusion, and data processing for automatic driving decision-making and control. Therefore, the control motherboard 20 can combine the radar data obtained by the 360-degree scanning laser radar 30 or the millimeter-wave radar, and the image data obtained by the forward-looking intelligent camera to control the wire-controlled chassis 10 to control the driving direction of the intelligent networked training vehicle, so that the intelligent networked training vehicle can bypass obstacles when unmanned, thereby improving the unmanned driving capability of the training vehicle.

[0050] See also Figure 2 , which is a flow chart of a control method of an intelligent network-connected training vehicle according to a first embodiment of the present application. The method is applied to the intelligent network-connected training vehicle as described above, and the method includes:

[0051] S1: Acquire an initial driving route of the training vehicle, as well as video data and radar data of the training vehicle during its driving along the initial driving route.

[0052] S2: Input the video data into the traffic sign detection model of the training vehicle to obtain traffic sign information of the video data.

[0053] S3: Inputting the radar data into the obstacle detection model of the training vehicle, and obtaining obstacle data if an obstacle is detected on the initial driving route.

[0054] S4: modifying the initial driving route according to the obstacle data and the traffic sign information to obtain a target driving route.

[0055] S5: Controlling the driving speed of the training vehicle along the modified target driving route according to the traffic sign information, so that the training vehicle bypasses the obstacle according to the target driving route.

[0056] Compared with the prior art, the present application can obtain traffic sign information and obstacle data encountered by the training vehicle when traveling along the initial driving route through a traffic sign detection model and an obstacle detection model, so as to modify the initial driving route according to the obstacle data and traffic sign information, and control the driving speed of the training vehicle along the modified target driving route according to the traffic sign information, so that the training vehicle can bypass the obstacles according to the target driving route, thereby improving the unmanned driving capability of the training vehicle.

[0057] In a feasible embodiment, the step S4: modifying the initial driving route according to the obstacle data and the traffic sign information to obtain the target driving route includes:

[0058] S41: modifying the driving route according to the obstacle data to obtain a left detour route and a right detour route for bypassing the obstacle.

[0059] S42: Determine the validity of the left detour route and the right detour route according to the traffic sign information.

[0060] S43: Determine the target driving route according to the effective left detour route and / or the right detour route.

[0061] In a feasible embodiment, the step of S42: determining the validity of the left detour route and the right detour route according to the traffic sign information includes:

[0062] If the traffic sign information does not contain a sign prohibiting left lane switching, it is determined that the left detour route is valid; if the traffic sign information does not contain a sign prohibiting right lane switching, it is determined that the right detour route is valid.

[0063] In a feasible embodiment, the step of S43: determining the target driving route according to the valid left detour route and / or the right detour route comprises:

[0064] S431: If only the left detour route is valid, determine the left detour route as the target driving route; if only the right detour route is valid, determine the right detour route as the target driving route.

[0065] S432: If both the left detour route and the right detour route are valid, obtain a first offset of the left detour route relative to the initial driving route, and a second offset of the right detour route relative to the initial driving route.

[0066] S433: If the first offset is less than or equal to the second offset, the left detour route is determined to be the target driving route; if the first offset is greater than the second offset, the right detour route is determined to be the target driving route.

[0067] In a feasible embodiment, the step S5: controlling the driving speed of the training vehicle along the modified target driving route according to the traffic sign information includes:

[0068] If the traffic sign information signal light is marked, and the signal light is a red signal light, the training vehicle is driven to stop; if the signal light is a green signal light, the training vehicle is driven to travel normally.

[0069] In a feasible embodiment, after the step of S2: inputting the video data into the traffic sign detection model of the training vehicle to obtain the traffic sign information of the video data, the method further includes:

[0070] S21: Using the video data and the traffic sign information as training samples to train the traffic sign detection model, and recording the number of times the traffic sign detection model is trained.

[0071] S22: When the number of training times reaches a preset threshold, the model parameter change data of the traffic sign detection model is uploaded to the car model server of the road where the training car is currently located, so that the car model server trains and optimizes the base station traffic sign detection model according to the model parameter change data uploaded by multiple cars.

[0072] Among them, the model parameter change data of the traffic sign detection model refers to the model parameters of the current traffic sign detection model after training, relative to the parameter change data of the base station traffic sign detection model sent by the car model server when the training car last received it. It should be noted that when the training car is connected to the car model server, the training car receives the base station traffic sign detection model sent by the car model server as the traffic sign detection model for the detection of traffic sign information and model training.

[0073] S23: Determine a new traffic sign detection model for the training car based on the base station traffic sign detection model after training and optimization sent by the car model server.

[0074] When the training car receives the trained and optimized base station traffic sign detection model sent by the car model server, the training car uses the newly received trained and optimized base station traffic sign detection model as a new traffic sign detection model.

[0075] In this embodiment, the training car can continue to perform model training based on the traffic sign information obtained by the traffic sign detection model to optimize the traffic sign detection model of the training car. The model parameter change data of the traffic sign detection model can also be uploaded to the car model server of the road where the training car is currently located according to the number of training times, so that the car model server can train and optimize the base station traffic sign detection model according to the model parameter change data uploaded by multiple cars, and then send it to the training car as a new traffic sign detection model for the training car, so that the training car can obtain a traffic sign detection model trained according to the model parameter change data of the traffic sign detection models of multiple cars, which can improve the accuracy of traffic sign detection by the training car.

[0076] In a feasible embodiment, the step S21: using the video data and the traffic sign information as training samples to train the traffic sign detection model, and recording the number of times the traffic sign detection model is trained, includes:

[0077] S211: Based on the current video frame in the video data, and the traffic sign information corresponding to the current video frame.

[0078] S212: Compare the traffic sign information corresponding to the previous video frame for training the traffic sign detection model with the traffic sign information corresponding to the current video frame to obtain a comparison similarity.

[0079] S213: If the comparison similarity is greater than a preset similarity threshold, the traffic sign detection model is trained according to the current video frame and the corresponding traffic sign information, so that the number of training times of the traffic sign detection model is increased by 1.

[0080] In a feasible embodiment, after the step of comparing the traffic sign information corresponding to the previous video frame for training the traffic sign detection model with the traffic sign information corresponding to the current video frame to obtain the comparison similarity, the method includes:

[0081] S214: If the compared similarity is less than or equal to the similarity threshold, determine the next video frame as a new current video frame.

[0082] Among them, the similarity threshold can be set by the manufacturer or the user. When the comparative similarity of the traffic sign information is less than or equal to the similarity threshold, it means that the current video frame used to train the traffic sign detection model of the training car is slightly different from the previous video frame used to train the traffic sign detection model. At this time, the parameter change of the traffic sign detection model after training is small, so the effect of training the traffic sign detection model is low, and the training of the traffic sign detection model at this time can save the time cost of data processing and the value of electric energy is low; and when the comparative similarity of the traffic sign information is greater than the similarity threshold, it means that the current video frame used to train the traffic sign detection model of the training car is significantly different from the previous video frame used to train the traffic sign detection model. At this time, the parameter change of the traffic sign detection model after training is large, so the effect of training the traffic sign detection model is high, and the training of the traffic sign detection model at this time can save the time cost of data processing and the value of electric energy is high.

[0083] In this embodiment, the training times of the traffic sign detection model are calculated according to the comparison similarity, so as to improve the time cost and the value of electric energy consumed in training the traffic sign detection model.

[0084] The device embodiments described above are merely illustrative, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application. Ordinary technicians in this field can understand and implement it without creative work.

[0085] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0086] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the function selected in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 function selected in a box or multiple boxes.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 steps for the function selected in a box or multiple boxes.

[0088] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0089] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0090] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0091] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0092] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. An intelligent network-connected training vehicle, characterized in that: include: Drive-by-wire chassis, control mainboard, 360-degree scanning laser radar, front-view smart camera, millimeter-wave radar, and a combined positioning unit with integrated inertial gyroscope and GNSS; The control-by-wire chassis is provided with a steering wheel set, and the control mainboard, 360-degree scanning laser radar, forward-looking intelligent camera, millimeter-wave radar, integrated inertial gyroscope and GNSS combined positioning unit are arranged above the control-by-wire chassis; The control main board is respectively connected to the wire-controlled chassis, 360-degree scanning laser radar, forward-looking intelligent camera, millimeter-wave radar, and a combined positioning unit integrating inertial gyroscope and GNSS. The control main board is used for image recognition, SLAM positioning, environmental perception, obstacle detection, traffic sign recognition, multi-sensor fusion, and data processing for autonomous driving decision-making and control.

2. A control method for an intelligent networked training vehicle, characterized in that: Applied to the intelligent networked training vehicle as claimed in claim 1, the method comprises: Acquire an initial driving route of the training vehicle, as well as video data and radar data of the training vehicle during its driving along the initial driving route; Inputting the video data into the traffic sign detection model of the training vehicle to obtain traffic sign information of the video data; Inputting the radar data into the obstacle detection model of the training vehicle, and obtaining obstacle data if an obstacle is detected on the initial driving route; Modifying the initial driving route according to the obstacle data and the traffic sign information to obtain a target driving route; The driving speed of the training vehicle along the modified target driving route is controlled according to the traffic sign information, so that the training vehicle bypasses the obstacle according to the target driving route.

3. The intelligent network-connected training vehicle control method according to claim 2 is characterized in that: The step of modifying the initial driving route according to the obstacle data and the traffic sign information to obtain the target driving route includes: Modifying the driving route according to the obstacle data to obtain a left detour route and a right detour route around the obstacle; Determining the validity of the left detour route and the right detour route according to the traffic sign information; The target driving route is determined according to the effective left detour route and / or the right detour route.

4. The intelligent network-connected training vehicle control method according to claim 3 is characterized in that: The step of determining the target driving route according to the effective left detour route and / or the right detour route comprises: If only the left detour route is valid, determine the left detour route as the target driving route; if only the right detour route is valid, determine the right detour route as the target driving route; If both the left detour route and the right detour route are valid, obtaining a first offset of the left detour route relative to the initial driving route, and a second offset of the right detour route relative to the initial driving route; If the first offset is less than or equal to the second offset, the left detour route is determined to be the target driving route; if the first offset is greater than the second offset, the right detour route is determined to be the target driving route.

5. The intelligent network-connected training vehicle control method according to claim 3 is characterized in that: The step of determining the validity of the left detour route and the right detour route according to the traffic sign information comprises: If the traffic sign information does not contain a sign prohibiting lane switching to the left, it is determined that the left detour route is valid; if the traffic sign information does not contain a sign prohibiting lane switching to the right, it is determined that the right detour route is valid.

6. The intelligent network-connected training vehicle control method according to claim 2 is characterized in that: The step of controlling the driving speed of the training vehicle along the modified target driving route according to the traffic sign information comprises: If the traffic sign information signal light is marked, and the signal light is a red signal light, the training vehicle is driven to stop; if the signal light is a green signal light, the training vehicle is driven to travel normally.

7. The intelligent network-connected training vehicle control method according to claim 2 is characterized in that: After the step of inputting the video data into the traffic sign detection model of the training vehicle to obtain the traffic sign information of the video data, the method further includes: Using the video data and the traffic sign information as training samples to train the traffic sign detection model, and recording the number of times the traffic sign detection model is trained; When the training times reaches a preset times threshold, the model parameter change data of the traffic sign detection model is uploaded to the car model server of the road where the training car is currently located, so that the car model server trains and optimizes the base station traffic sign detection model according to the model parameter change data uploaded by multiple cars; According to the trained and optimized base station traffic sign detection model sent by the car model server, a new traffic sign detection model for the training car is determined.

8. The intelligent network-connected training vehicle control method according to claim 7 is characterized in that: The step of using the video data and the traffic sign information as training samples to train the traffic sign detection model, and recording the number of times the traffic sign detection model is trained, comprises: According to a current video frame in the video data, and traffic sign information corresponding to the current video frame; Comparing the traffic sign information corresponding to the previous video frame for training the traffic sign detection model with the traffic sign information corresponding to the current video frame to obtain a comparison similarity; If the comparison similarity is greater than a preset similarity threshold, the traffic sign detection model is trained according to the current video frame and the corresponding traffic sign information, so that the number of training times of the traffic sign detection model is increased by 1.

9. The intelligent network-connected training vehicle control method according to claim 8 is characterized in that: After the step of comparing the traffic sign information corresponding to the previous video frame for training the traffic sign detection model with the traffic sign information corresponding to the current video frame to obtain the comparison similarity, the method further comprises: If the compared similarity is less than or equal to the similarity threshold, the next video frame is determined as a new current video frame.

Images