Control method and system for remote takeover of unmanned vehicle and medium

Abstract

The invention relates to a control method and system for remote takeover of an unmanned vehicle and a medium, and the method comprises the steps: M1, obtaining the operation data information of the vehicle and the data information of an automatic driving fault of the vehicle in real time when the vehicle runs on a road, and collecting the data information of traffic participants around the vehicle; and M2, based on the operation data information of the vehicle and the data information of the automatic driving fault of the vehicle, constructing a diffusion model of the fault severity of the vehicle based on deep learning, and representing the fault severity of the vehicle to obtain the data information of the fault severity of the vehicle. According to the method, data of the vehicle and other traffic participants can be analyzed, the self automatic driving technology can be improved conveniently, fault types can be managed in a classified mode, a safety officer can conveniently make a judgment in time, the vehicle takeover rate is reduced, and the vehicle driving smoothness is improved.

Description

Technical Field

[0001] The present invention relates to the field of unmanned driving technology, and in particular to a control method, system and medium for remote takeover of an unmanned vehicle. Background Art

[0002] Current intelligent driving technology cannot fully cover all road scenarios. When exceeding relevant scenarios or encountering risks, the vehicle will be taken over. The takeover is divided into manual takeover and remote takeover. Generally, the owner or safety officer decides whether to take over. There is no classification of faults reported by the vehicle and risk level prompts, resulting in the need to take over the vehicle at any time, and the comfort and experience are poor. Summary of the invention

[0003] In view of the above problems, the present invention provides a control method, system and medium for remote takeover of unmanned vehicles, which can not only analyze the data of the vehicle itself and other traffic participants to facilitate the improvement of its own autonomous driving technology, but also classify and manage fault types to facilitate safety officers to make timely judgments, reduce the vehicle takeover rate, and improve the vehicle's driving smoothness.

[0004] In order to achieve the above-mentioned purpose and other related purposes, the technical solution provided by the present invention is as follows: a control method for remote takeover of an unmanned vehicle, the method comprising:

[0005] M1. When the vehicle is driving on the road, the vehicle's operating data and data on the vehicle's automatic driving failures are obtained in real time, and data on traffic participants around the vehicle are collected;

[0006] M2. Based on the operation data information of the vehicle and the data information of the vehicle's automatic driving fault, a diffusion model of the fault severity of the vehicle based on deep learning is constructed to characterize the fault severity of the vehicle and obtain data information of the fault severity of the vehicle;

[0007] M3. Based on the operation data information of the vehicle and the data information of the traffic participants around the vehicle, a dynamic traffic learning prediction model based on time parameters and space parameters is constructed to predict the traffic accident rate of the vehicle, and obtain the predicted traffic accident rate data information of the vehicle;

[0008] M4. Based on the data information of the fault severity of the vehicle and the predicted traffic accident rate data information of the vehicle, a classification prediction algorithm based on black-winged kite optimization is used to predict and classify the vehicle's emergency degree to obtain classified vehicle emergency degree data information.

[0009] Furthermore, the method further comprises:

[0010] M5. Based on the classified vehicle emergency data information, establish a remote takeover control function P,

[0011]

[0012] Among them, x is the emergency degree data information of the classified vehicle, α1, α2 and α3 are the remote takeover control factors of the vehicle, which control and adjust the remote takeover of the vehicle and output the remote takeover control data information of the vehicle.

[0013] Furthermore, the remote takeover control factors α1, α2 and α3 of the vehicle are:

[0014]

[0015] Among them, x is the emergency data information of the classified vehicle.

[0016] Furthermore, in step M2, the construction of a diffusion model of the fault severity of the vehicle based on deep learning to characterize the fault severity of the vehicle includes:

[0017] M21. Based on the operation data information of the vehicle and the data information of the vehicle's automatic driving failure, establish a fusion function Q of the vehicle failure,

[0018]

[0019] Among them, y1 is the vehicle's operating data information, y2 is the vehicle's automatic driving fault data information, β1, β2 and β3 are feature fusion factors, and the vehicle's operating data and automatic driving fault data are fused to obtain the fused vehicle fault data information;

[0020] M22. Input the fused vehicle fault data information into a diffusion model of vehicle fault severity based on deep learning for training and learning, and determine a characterization function W of the vehicle fault severity.

[0021]

[0022] Among them, z is the fused vehicle fault data information, δ1, δ2 and δ3 are the weight factors of the model, and the trained diffusion model of vehicle fault severity based on deep learning is obtained;

[0023] M23. Based on the trained deep learning-based diffusion model of vehicle fault severity, the fused vehicle fault data information is input to characterize the vehicle fault severity, and obtain the vehicle fault severity data information.

[0024] Furthermore, the constraints of the weight factors δ1, δ2 and δ3 of the model are:

[0025]

[0026] Furthermore, in step M3, the construction of a dynamic traffic learning prediction model based on time parameters and space parameters to predict the traffic accident rate of vehicles includes:

[0027] M31. Based on the operation data information of the vehicle and the data information of the traffic participants around the vehicle, establish a vehicle traffic mapping function R based on time parameters and space parameters,

[0028] Wherein, a is the operation data information of the vehicle, b is the data information of the traffic participants around the vehicle, η is the time parameter, and γ is the space parameter. The correlation between the operation data of the vehicle and the data of the traffic participants around the vehicle is characterized to obtain the correlation data information between the operation data of the vehicle and the data of the traffic participants around the vehicle;

[0029] M32. Inputting the correlation data information of the vehicle's operating data and the data of traffic participants around the vehicle into the dynamic traffic learning prediction model for training and learning, and determining the vehicle's traffic accident rate prediction function P,

[0030]

[0031] Among them, c is the correlation data information between the vehicle's operating data and the data of traffic participants around the vehicle, λ1, λ2 and λ3 are the vehicle's traffic accident rate prediction constant parameters, and the trained dynamic traffic learning prediction model is obtained;

[0032] M33. Based on the trained dynamic traffic learning prediction model, the correlation data information between the vehicle's operating data and the data of traffic participants around the vehicle is input to predict the vehicle's traffic accident rate, and obtain the predicted vehicle's traffic accident rate data information.

[0033] Furthermore, the time parameter η is,

[0034]

[0035] The spatial parameter γ is,

[0036]

[0037] Among them, a is the operating data information of the vehicle, and b is the data information of traffic participants around the vehicle.

[0038] Furthermore, in step M4, the use of the classification prediction algorithm based on black kite optimization to predict and classify the emergency degree of the vehicle includes:

[0039] M41. Inputting the data information of the fault severity of the vehicle and the predicted traffic accident rate data information of the vehicle into the vehicle emergency classification prediction model for training and learning, initializing the model weight and bias, and obtaining the initialized model weight and bias data information;

[0040] M42. Based on the initialized model weights and bias data information, the black-winged kite population is initialized, population parameters and a maximum number of iterations are determined, and data information of the initialized black-winged kite population is obtained;

[0041] M43. Based on the data information of the initialized black-winged kite population, a fitness function S of the black-winged kite population individuals is established.

[0042]

[0043] Among them, r is the data information of the initialized black-winged kite population, μ1 and μ2 are the fitness determining factors of the population individuals, and the fitness values ​​of the population individuals are calculated to obtain the fitness value data information of the black-winged kite population individuals;

[0044] M44. Based on the fitness value data information of the black-winged kite population individuals, establish a target optimization function F,

[0045]

[0046] Among them, g is the fitness value data information of the black-winged kite population, θ1, θ2 and θ3 are the target optimization factors, and the initialized model weights and biases are optimized to obtain the optimized vehicle emergency classification prediction model;

[0047] M45. Based on the optimized vehicle emergency classification prediction model, the data information of the vehicle's fault severity and the predicted vehicle's traffic accident rate data information are input to predict and classify the vehicle's emergency level to obtain classified vehicle emergency level data information.

[0048] In order to achieve the above-mentioned purpose and other related purposes, the present invention also provides a control system for remote takeover of an unmanned vehicle, including a computer device, which is programmed or configured to execute the steps of any one of the control methods for remote takeover of an unmanned vehicle.

[0049] In order to achieve the above-mentioned purpose and other related purposes, the present invention also provides a computer-readable storage medium, which stores a computer program programmed or configured to execute any one of the control methods for remote takeover of an unmanned vehicle.

[0050] The present invention has the following positive effects:

[0051] 1. The present invention characterizes the severity of vehicle faults by constructing a diffusion model of the severity of vehicle faults based on deep learning, and predicts the traffic accident rate of the vehicle by combining the construction of a dynamic traffic learning prediction model based on time parameters and space parameters. It can not only analyze the data of the vehicle itself and other traffic participants to facilitate the improvement of its own autonomous driving technology, but also analyze and study the habits of traffic participants to better adapt technical solutions.

[0052] 2. The present invention predicts and classifies the emergency response of vehicles by adopting a classification prediction algorithm based on black kite optimization, and combines the establishment of a remote takeover control function P to control and adjust the remote takeover of the vehicle, and outputs the remote takeover control data information of the vehicle. It can not only classify and manage the fault types, facilitate safety officers to make timely judgments, reduce the vehicle takeover rate, and improve the vehicle's driving smoothness, but also issue emergency work orders to avoid untimely takeover due to safety officers' misjudgment and improve safety redundancy. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0054] Figure 2 A schematic diagram of a process of constructing a diffusion model of fault severity of a vehicle based on deep learning according to the present invention;

[0055] Figure 3 A schematic diagram of a process of constructing a dynamic traffic learning prediction model based on time parameters and space parameters according to the present invention;

[0056] Figure 4 The present invention is a flow chart of a classification prediction algorithm based on black-winged kite optimization. DETAILED DESCRIPTION

[0057] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure 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 disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0058] Example 1: Figure 1 As shown, a control method for remote takeover of an unmanned vehicle, the method comprising:

[0059] M1. When the vehicle is driving on the road, the vehicle's operating data and data on the vehicle's automatic driving failures are obtained in real time, and data on traffic participants around the vehicle are collected;

[0060] M2. Based on the operation data information of the vehicle and the data information of the vehicle's automatic driving fault, a diffusion model of the fault severity of the vehicle based on deep learning is constructed to characterize the fault severity of the vehicle and obtain data information of the fault severity of the vehicle;

[0061] M3. Based on the operation data information of the vehicle and the data information of the traffic participants around the vehicle, a dynamic traffic learning prediction model based on time parameters and space parameters is constructed to predict the traffic accident rate of the vehicle, and obtain the predicted traffic accident rate data information of the vehicle;

[0062] M4. Based on the data information of the fault severity of the vehicle and the predicted traffic accident rate data information of the vehicle, a classification prediction algorithm based on black-winged kite optimization is used to predict and classify the vehicle's emergency degree to obtain classified vehicle emergency degree data information.

[0063] In this embodiment, the method further includes:

[0064] M5. Based on the classified vehicle emergency data information, establish a remote takeover control function P,

[0065]

[0066] Among them, x is the emergency degree data information of the classified vehicle, α1, α2 and α3 are the remote takeover control factors of the vehicle, which control and adjust the remote takeover of the vehicle and output the remote takeover control data information of the vehicle.

[0067] In this embodiment, the remote takeover control factors α1, α2 and α3 of the vehicle are:

[0068]

[0069] Among them, x is the emergency data information of the classified vehicle.

[0070] In this embodiment, if Figure 2 As shown, in step M2, the construction of a diffusion model of the fault severity of the vehicle based on deep learning and characterization of the fault severity of the vehicle includes:

[0071] M21. Based on the operation data information of the vehicle and the data information of the vehicle's automatic driving failure, establish a fusion function Q of the vehicle failure,

[0072]

[0073] Among them, y1 is the vehicle's operating data information, y2 is the vehicle's automatic driving fault data information, β1, β2 and β3 are feature fusion factors, and the vehicle's operating data and automatic driving fault data are fused to obtain the fused vehicle fault data information;

[0074] M22. Input the fused vehicle fault data information into a diffusion model of vehicle fault severity based on deep learning for training and learning, and determine a characterization function W of the vehicle fault severity.

[0075]

[0076] Among them, z is the fused vehicle fault data information, δ1, δ2 and δ3 are the weight factors of the model, and the trained diffusion model of vehicle fault severity based on deep learning is obtained;

[0077] M23. Based on the trained deep learning-based diffusion model of vehicle fault severity, the fused vehicle fault data information is input to characterize the vehicle fault severity, and obtain the vehicle fault severity data information.

[0078] In this embodiment, the constraints of the weight factors δ1, δ2 and δ3 of the model are:

[0079] In this embodiment, if Figure 3 As shown, in step M3, the construction of a dynamic traffic learning prediction model based on time parameters and space parameters to predict the traffic accident rate of vehicles includes:

[0080] M31. Based on the operation data information of the vehicle and the data information of the traffic participants around the vehicle, establish a vehicle traffic mapping function R based on time parameters and space parameters,

[0081] Wherein, a is the operation data information of the vehicle, b is the data information of the traffic participants around the vehicle, η is the time parameter, and γ is the space parameter. The correlation between the operation data of the vehicle and the data of the traffic participants around the vehicle is characterized to obtain the correlation data information between the operation data of the vehicle and the data of the traffic participants around the vehicle;

[0082] M32. Inputting the correlation data information of the vehicle's operating data and the data of traffic participants around the vehicle into the dynamic traffic learning prediction model for training and learning, and determining the vehicle's traffic accident rate prediction function P,

[0083]

[0084] Among them, c is the correlation data information between the vehicle's operating data and the data of traffic participants around the vehicle, λ1, λ2 and λ3 are the vehicle's traffic accident rate prediction constant parameters, and the trained dynamic traffic learning prediction model is obtained;

[0085] M33. Based on the trained dynamic traffic learning prediction model, the correlation data information between the vehicle's operating data and the data of traffic participants around the vehicle is input to predict the vehicle's traffic accident rate, and obtain the predicted vehicle's traffic accident rate data information.

[0086] In this embodiment, the time parameter n is,

[0087]

[0088] The spatial parameter γ is,

[0089]

[0090] Among them, a is the operating data information of the vehicle, and b is the data information of traffic participants around the vehicle.

[0091] Embodiment 2: Based on the control method for remote takeover of an unmanned vehicle in Embodiment 1, the present invention is further illustrated and described below.

[0092] like Figure 1 As shown, a control method for remote takeover of an unmanned vehicle, the method comprising:

[0093] M1. When the vehicle is driving on the road, the vehicle's operating data and data on the vehicle's automatic driving failures are obtained in real time, and data on traffic participants around the vehicle are collected;

[0094] M2. Based on the operation data information of the vehicle and the data information of the vehicle's automatic driving fault, a diffusion model of the fault severity of the vehicle based on deep learning is constructed to characterize the fault severity of the vehicle and obtain data information of the fault severity of the vehicle;

[0095] M3. Based on the operation data information of the vehicle and the data information of the traffic participants around the vehicle, a dynamic traffic learning prediction model based on time parameters and space parameters is constructed to predict the traffic accident rate of the vehicle, and obtain the predicted traffic accident rate data information of the vehicle;

[0096] M4. Based on the data information of the fault severity of the vehicle and the predicted traffic accident rate data information of the vehicle, a classification prediction algorithm based on black-winged kite optimization is used to predict and classify the vehicle's emergency degree to obtain classified vehicle emergency degree data information.

[0097] In this embodiment, if Figure 4 As shown, in step M4, the use of the classification prediction algorithm based on black kite optimization to predict and classify the emergency degree of the vehicle includes:

[0098] M41. Inputting the data information of the fault severity of the vehicle and the predicted traffic accident rate data information of the vehicle into the vehicle emergency classification prediction model for training and learning, initializing the model weight and bias, and obtaining the initialized model weight and bias data information;

[0099] M42. Based on the initialized model weights and bias data information, the black-winged kite population is initialized, population parameters and a maximum number of iterations are determined, and data information of the initialized black-winged kite population is obtained;

[0100] M43. Based on the data information of the initialized black-winged kite population, a fitness function S of the black-winged kite population individuals is established.

[0101]

[0102] Among them, r is the data information of the initialized black-winged kite population, μ1 and μ2 are the fitness determining factors of the population individuals, and the fitness values ​​of the population individuals are calculated to obtain the fitness value data information of the black-winged kite population individuals;

[0103] M44. Based on the fitness value data information of the black-winged kite population individuals, establish a target optimization function F,

[0104]

[0105] Among them, g is the fitness value data information of the black-winged kite population, θ1, θ2 and θ3 are the target optimization factors, and the initialized model weights and biases are optimized to obtain the optimized vehicle emergency classification prediction model;

[0106] M45. Based on the optimized vehicle emergency classification prediction model, the data information of the vehicle's fault severity and the predicted vehicle's traffic accident rate data information are input to predict and classify the vehicle's emergency level to obtain classified vehicle emergency level data information.

[0107] The classified vehicle emergency data information is classified according to red, orange, yellow and green levels.

[0108] The platform transmits red, orange, yellow and green level work orders to the remote cockpit test bench through the RCU. When the work order conditions are triggered, the remote test bench interface will pop up a prompt of the work order information, and the safety officer can choose to take over or reject it based on the work order information.

[0109] The platform tracks the takeover status based on the feedback from the safety officer. When the platform administrator determines that the safety officer has made a wrong decision or believes that takeover is necessary, an emergency work order can be manually issued to the rack and the safety officer can be notified to take over.

[0110] In this embodiment, the present invention provides a control system for remote takeover of an unmanned vehicle, including a computer device that is programmed or configured to execute any one of the steps of the control method for remote takeover of an unmanned vehicle.

[0111] In this embodiment, the present invention provides a computer-readable storage medium, on which is stored a computer program programmed or configured to execute any one of the control methods for remote takeover of an unmanned vehicle.

[0112] Any reference to memory, storage, database or other media used in the embodiments provided in the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM) and memory bus dynamic RAM (RDRAM).

[0113] In summary, the present invention can not only analyze the data of the vehicle itself and other traffic participants to facilitate the improvement of its own autonomous driving technology, but also classify and manage the fault types to facilitate safety officers to make timely judgments, reduce the vehicle takeover rate, and improve the vehicle's driving smoothness.

[0114] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A control method for remote takeover of an unmanned vehicle, characterized in that: The method comprises: M1. When the vehicle is driving on the road, the vehicle's operating data and data on the vehicle's automatic driving failures are obtained in real time, and data on traffic participants around the vehicle are collected; M2. Based on the operation data information of the vehicle and the data information of the vehicle's automatic driving fault, a diffusion model of the fault severity of the vehicle based on deep learning is constructed to characterize the fault severity of the vehicle and obtain data information of the fault severity of the vehicle; M3. Based on the operation data information of the vehicle and the data information of the traffic participants around the vehicle, a dynamic traffic learning prediction model based on time parameters and space parameters is constructed to predict the traffic accident rate of the vehicle, and obtain the predicted traffic accident rate data information of the vehicle; M4. Based on the data information of the fault severity of the vehicle and the predicted traffic accident rate data information of the vehicle, a classification prediction algorithm based on black-winged kite optimization is used to predict and classify the vehicle's emergency degree to obtain classified vehicle emergency degree data information.

2. The control method for remote takeover of an unmanned vehicle according to claim 1, characterized in that: The method further comprises: M5. Based on the classified vehicle emergency data information, establish a remote takeover control function P, Among them, x is the emergency degree data information of the classified vehicle, α1, α2 and α3 are the remote takeover control factors of the vehicle, which control and adjust the remote takeover of the vehicle and output the remote takeover control data information of the vehicle.

3. The control method for remote takeover of an unmanned vehicle according to claim 1, characterized in that: The remote takeover control factors α1, α2 and α3 of the vehicle are: Among them, x is the emergency data information of the classified vehicle.

4. The control method for remote takeover of an unmanned vehicle according to claim 1, characterized in that: In step M2, the construction of a diffusion model of the fault severity of the vehicle based on deep learning to characterize the fault severity of the vehicle includes: M21. Based on the operation data information of the vehicle and the data information of the vehicle's automatic driving failure, establish a fusion function Q of the vehicle failure, Among them, y1 is the vehicle's operating data information, y2 is the vehicle's automatic driving fault data information, β1, β2 and β3 are feature fusion factors, and the vehicle's operating data and automatic driving fault data are fused to obtain the fused vehicle fault data information; M22. Input the fused vehicle fault data information into a diffusion model of vehicle fault severity based on deep learning for training and learning, and determine a characterization function W of the vehicle fault severity. Among them, z is the fused vehicle fault data information, δ1, δ2 and δ3 are the weight factors of the model, and the trained diffusion model of vehicle fault severity based on deep learning is obtained; M23. Based on the trained deep learning-based diffusion model of vehicle fault severity, the fused vehicle fault data information is input to characterize the vehicle fault severity, and obtain the vehicle fault severity data information.

5. The control method for remote takeover of an unmanned vehicle according to claim 4, characterized in that: The constraints of the weight factors δ1, δ2 and δ3 of the model are:

6. The control method for remote takeover of an unmanned vehicle according to claim 1, characterized in that: In step M3, the construction of a dynamic traffic learning prediction model based on time parameters and space parameters to predict the traffic accident rate of vehicles includes: M31. Based on the operation data information of the vehicle and the data information of the traffic participants around the vehicle, establish a vehicle traffic mapping function R based on time parameters and space parameters, Wherein, a is the operation data information of the vehicle, b is the data information of the traffic participants around the vehicle, η is the time parameter, and γ is the space parameter. The correlation between the operation data of the vehicle and the data of the traffic participants around the vehicle is characterized to obtain the correlation data information between the operation data of the vehicle and the data of the traffic participants around the vehicle; M32. Inputting the correlation data information of the vehicle's operating data and the data of traffic participants around the vehicle into the dynamic traffic learning prediction model for training and learning, and determining the vehicle's traffic accident rate prediction function P, Among them, c is the correlation data information between the vehicle's operating data and the data of traffic participants around the vehicle, λ1, λ2 and λ3 are the vehicle's traffic accident rate prediction constant parameters, and the trained dynamic traffic learning prediction model is obtained; M33. Based on the trained dynamic traffic learning prediction model, the correlation data information between the vehicle's operating data and the data of traffic participants around the vehicle is input to predict the vehicle's traffic accident rate, and obtain the predicted vehicle's traffic accident rate data information.

7. The control method for remote takeover of an unmanned vehicle according to claim 6, characterized in that: The time parameter η is, The spatial parameter γ is, Among them, a is the operating data information of the vehicle, and b is the data information of traffic participants around the vehicle.

8. The control method for remote takeover of an unmanned vehicle according to claim 1, characterized in that: In step M4, the method of using the classification prediction algorithm based on black kite optimization to predict and classify the emergency degree of the vehicle includes: M41. Inputting the data information of the fault severity of the vehicle and the predicted traffic accident rate data information of the vehicle into the vehicle emergency classification prediction model for training and learning, initializing the model weight and bias, and obtaining the initialized model weight and bias data information; M42. Based on the initialized model weights and bias data information, the black-winged kite population is initialized, population parameters and a maximum number of iterations are determined, and data information of the initialized black-winged kite population is obtained; M43. Based on the data information of the initialized black-winged kite population, a fitness function S of the black-winged kite population individuals is established. Among them, r is the data information of the initialized black-winged kite population, μ1 and μ2 are the fitness determining factors of the population individuals, and the fitness values ​​of the population individuals are calculated to obtain the fitness value data information of the black-winged kite population individuals; M44. Based on the fitness value data information of the black-winged kite population individuals, establish a target optimization function F, Among them, g is the fitness value data information of the black-winged kite population, θ1, θ2 and θ3 are the target optimization factors, and the initialized model weights and biases are optimized to obtain the optimized vehicle emergency classification prediction model; M45. Based on the optimized vehicle emergency classification prediction model, the data information of the vehicle's fault severity and the predicted vehicle's traffic accident rate data information are input to predict and classify the vehicle's emergency level to obtain classified vehicle emergency level data information.

9. A control system for remote takeover of an unmanned vehicle, comprising a computer device, characterized in that: The computer device is programmed or configured to execute the steps of the control method for remote takeover of an unmanned vehicle as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program that is programmed or configured to execute the control method for remote takeover of an unmanned vehicle as described in any one of claims 1 to 8.

Images