Driver takeover behavior progress prediction model training method, system and terminal, prediction method and system and vehicle
By constructing a multivariate linear regression progress prediction model and using vehicle operation parameters to predict the completion progress of takeover, the problem of insufficient precision prediction of takeover behavior in the existing technology is solved, and the safe and timely transfer of driving rights is achieved, and driving safety and intelligent driving assistance capabilities are improved.
Patent Information
- Application Number
- CN202411593948.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-08
- Publication Date
- 2025-02-28
AI Technical Summary
The prior art has problems such as difficult data collection and poor real-time performance in predicting driver takeover behavior, which leads to inaccurate prediction of takeover behavior progress, which makes it difficult to meet the needs of smooth transfer of driving rights and safe driving.
By collecting vehicle historical takeover data, a multivariate linear regression progress prediction model is constructed, and the docking completion progress is predicted using vehicle operating parameters to achieve accurate prediction of vehicle takeover progress.
It realizes accurate prediction of vehicle takeover progress based on vehicle operating parameters, ensures safe, timely and smooth transfer of driving rights, and improves driving safety and intelligent driving assistance capabilities.
Smart Images
Figure 00000000_0000_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of vehicle takeover control, and in particular to a driver takeover behavior progress prediction model training method, system and terminal, and a prediction method, system and vehicle. Background Art
[0002] In recent years, medium and heavy-duty trucks have made significant progress in intelligent driving technology and have realized many automated functions, such as adaptive cruise control, lane keeping assist, automatic emergency braking, etc. However, at this stage, they cannot meet the full sense of autonomous driving, that is, the L5 level of autonomous driving standard. This means that the vehicle cannot operate independently in all road environments and weather conditions without any intervention from the human driver. Although vehicles currently at the L2 to L3 level of autonomous driving can perform some driving tasks under certain conditions, such as keeping lanes and controlling speed on highways, drivers are still required to remain highly alert and ready to take over control in complex or unexpected situations. It is necessary to achieve a safe transfer of driving rights through driver takeover behavior for the safety of vehicle driving.
[0003] Traditional methods for predicting driver takeover behavior mainly rely on the driver's physiological reaction data (such as eye movements, heart rate, etc.) or behavioral characteristics (such as braking, steering, etc.), but these methods often have problems such as difficulty in data collection and poor real-time performance. The progress prediction of the driver's takeover behavior is not accurate enough, and it is difficult to meet the needs of smooth transfer of driving rights and safe driving. Summary of the invention
[0004] To solve the above problems, the present invention provides a driver takeover behavior progress prediction model training method, system and terminal, prediction method, system and vehicle, which can realize the prediction of vehicle takeover progress according to vehicle operating parameters. The prediction method is simple, reliable and accurate, thereby ensuring safe, timely and smooth transfer of driving rights, improving driving safety and intelligent driving assistance capabilities.
[0005] In a first aspect, the technical solution of the present invention provides a method for training a driver takeover behavior progress prediction model, comprising the following steps: Collect historical vehicle takeover data, which includes the takeover completion time and vehicle operation parameters; The constructed multivariate linear regression progress prediction model is trained using the vehicle historical takeover data to obtain the trained multivariate linear regression progress prediction model.
[0006] In an optional implementation, collecting vehicle historical takeover data specifically includes: Obtain the lateral takeover time and the longitudinal takeover time of the vehicle, and take the longer of the two as the completion takeover time; The vehicle operating parameters from the time the vehicle issues a takeover reminder to the time the takeover is completed are obtained. The vehicle operating parameters include vehicle acceleration value, vehicle yaw angular velocity value, steering wheel hand torque, and takeover reminder level.
[0007] In an optional implementation, the constructed multivariate linear regression progress prediction model is trained using the historical vehicle takeover data, specifically including: Construct vehicle historical takeover data as a training dataset; The constructed multivariate linear regression progress prediction model is trained using the training data set.
[0008] In an optional implementation, the vehicle historical takeover data is constructed as a training data set, specifically including: Calculate the average of all completed takeover times to obtain the average takeover time; Divide the average takeover time into M segments, and divide the 100% takeover completion progress into M takeover completion percentage progress. Each segment of the average takeover time corresponds to one takeover completion percentage progress. Obtain the vehicle operation parameters corresponding to each average takeover time; Each takeover completion percentage progress and its corresponding vehicle operation parameter constitute a set of element pairs, and a set of element pairs is a training data sample; A number of training data samples form a training data set.
[0009] In an optional implementation, the constructed multivariate linear regression progress prediction model is trained using a training data set, specifically including: Construct a multivariate linear regression progress prediction model, the function expression is:
[0010] In the formula, , They represent a vehicle operating parameter respectively. , , Respectively represent the linear regression coefficients corresponding to each vehicle operating parameter, represents the estimation error, which is a normally distributed zero-mean random variable, Indicates the percentage progress of the driver taking over; Will test The training data samples are input into the multivariate linear regression progress prediction model, and the regression equation is constructed to estimate the linear regression parameters, and the following regression equation is obtained:
[0011] in, , , , ; The least squares method is used to obtain the regression model parameters. , , Estimated value of , , , so that The variance of is an unbiased estimate:
[0012] The model parameter estimates Substitute the multivariate linear regression progress prediction model function expression into the model to obtain the following optimized regression model of the driver's takeover status, that is, the trained multivariate linear regression progress prediction model, .
[0013] In a second aspect, the technical solution of the present invention provides a driver takeover behavior progress prediction model training system, comprising: Historical data collection module: used to collect historical vehicle takeover data, which includes the takeover completion time and vehicle operation parameters; Prediction model training module: used to train the constructed multivariate linear regression progress prediction model using the vehicle historical takeover data to obtain the trained multivariate linear regression progress prediction model.
[0014] In a third aspect, the technical solution of the present invention provides a terminal, including: A memory, used for storing a driver takeover behavior progress prediction model training program; A processor is used to implement the steps of the driver takeover behavior progress prediction model training method as described in any one of the above items when executing the driver takeover behavior progress prediction model training program.
[0015] In a fourth aspect, the technical solution of the present invention provides a method for predicting the progress of a driver's takeover behavior, comprising the following steps: Respond to vehicle takeover reminders and collect vehicle operating parameters in real time; The trained multivariate linear regression progress prediction model obtained by any of the above methods is used to process the vehicle operation parameters collected in real time to obtain the takeover behavior progress prediction result.
[0016] In a fifth aspect, the technical solution of the present invention provides a driver takeover behavior progress prediction system, which is implemented based on the above-mentioned driver takeover behavior progress prediction method, and includes: Real-time data collection module: used to respond to vehicle takeover reminders and collect vehicle operating parameters in real time; Progress prediction module: used to process the vehicle operation parameters collected in real time using the trained multivariate linear regression progress prediction model obtained by any of the above-mentioned driver takeover behavior progress prediction model training methods to obtain the takeover behavior progress prediction results.
[0017] In a sixth aspect, the technical solution of the present invention provides a vehicle, deploys the trained multivariate linear regression progress prediction model obtained by the above-mentioned driver takeover behavior progress prediction model training method, and executes the above-mentioned driver takeover behavior progress prediction method.
[0018] Compared with the prior art, the present invention provides a driver takeover behavior progress prediction model training method, system and terminal, prediction method, system and vehicle, which have the following beneficial effects: the constructed multivariate linear regression progress prediction model is trained according to the historical completion time of the takeover and the vehicle operating parameters, and the takeover completion progress is predicted according to the vehicle operating parameters during operation, so as to realize the prediction of the vehicle takeover progress according to the vehicle operating parameters, and the data collection is simple and real-time, and the prediction method of the multivariate regression model based on the vehicle operating parameters is simple, reliable and accurate, thereby ensuring the safe, timely and smooth transfer of driving rights, and improving the driving safety and intelligent driving assistance capabilities. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0020] Figure 1 The present invention provides a flowchart of a driver takeover behavior progress prediction model training method.
[0021] Figure 2 The present invention provides a schematic block diagram of a driver takeover behavior progress prediction model training system.
[0022] Figure 3 It is a schematic diagram of the structure of a terminal provided by an embodiment of the present invention.
[0023] Figure 4 The present invention provides a flowchart of a method for predicting the progress of a driver's takeover behavior.
[0024] Figure 5 The present invention provides a schematic block diagram of a driver takeover behavior progress prediction system. DETAILED DESCRIPTION
[0025] In order to enable those skilled in the art to better understand the scheme of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.
[0027] Figure 1 This is a flow chart of a method for training a driver takeover behavior progress prediction model provided by an embodiment of the present invention. Figure 1 The execution subject may be a driver takeover behavior progress prediction model training system. The driver takeover behavior progress prediction model training method provided in the embodiment of the present invention is executed by a computer device, and accordingly, the driver takeover behavior progress prediction model training system runs in the computer device. According to different requirements, the order of the steps in the flowchart can be changed, and some can be omitted.
[0028] like Figure 1 As shown, the method includes the following steps.
[0029] S1, collecting historical vehicle takeover data, the vehicle historical takeover data set includes the takeover completion time and vehicle operation parameters.
[0030] S2, using the historical vehicle takeover data to train the constructed multivariate linear regression progress prediction model to obtain a trained multivariate linear regression progress prediction model.
[0031] The purpose of this method is to use historical takeover data to train the constructed multivariate linear regression progress prediction model, so that the trained multivariate linear regression progress prediction model can be deployed on the vehicle, and the trained multivariate linear regression progress prediction model can be run on the vehicle to predict the progress of the driver's takeover behavior.
[0032] The vehicle history takeover data set of this embodiment includes the takeover completion time and vehicle operation parameters. The collection of the vehicle history takeover data specifically includes the following steps.
[0033] S1.1, obtain the lateral takeover time and the longitudinal takeover time of the vehicle, and take the longer of the two as the takeover completion time.
[0034] S1.2, obtaining vehicle operating parameters from the time when the vehicle issues a takeover reminder to the time when the takeover is completed, the vehicle operating parameters include vehicle acceleration value, vehicle yaw angular velocity value, steering wheel hand torque, and takeover reminder level.
[0035] In a specific embodiment, the mileage obtained in the data set is not less than 50,000 km, and the driving time is not less than 10,000 h. In different driving scenarios, the driver takeover reminder is triggered to obtain the takeover data set of different drivers at different times, including vehicle data such as vehicle acceleration value, vehicle yaw angular velocity value, steering wheel hand torque, driver takeover reminder level and other data sets. At the same time, the time when the driver completes the lateral takeover and longitudinal takeover of the vehicle is used as the time to complete the takeover, and the time when different drivers complete the safe takeover is obtained. It should be noted that the vehicle completes the lateral and longitudinal takeovers to complete the total takeover, so the lateral takeover time and longitudinal takeover time of the vehicle are obtained, and the longer of the two is used as the time to complete the takeover.
[0036] After obtaining the historical data, the constructed multivariate linear regression progress prediction model is trained using the vehicle historical takeover data, which specifically includes the following steps.
[0037] S2.1, construct the vehicle historical takeover data as a training dataset.
[0038] 1) Calculate the average of all completed takeover times to obtain the average takeover time.
[0039] 2) Divide the average takeover time into M segments, and divide the 100% takeover completion progress into M takeover completion percentage progress. Each segment of the average takeover time corresponds to one takeover completion percentage progress.
[0040] 3) Obtain the vehicle operating parameters corresponding to each average takeover time.
[0041] 4) Each takeover completion percentage and its corresponding vehicle operating parameter constitute a set of element pairs, and a set of element pairs is a training data sample.
[0042] 5) Several training data samples form a training data set.
[0043] In a specific embodiment, the average takeover time of different drivers is obtained, and the average takeover time is divided into 5 equal parts. The takeover completion progress corresponding to the 5 time periods is determined as 20%, 40%, 60%, 80%, and 100%. For example, if the average takeover time of the driver is 2s, the takeover completion progress corresponding to the 5 moments of 0.4s, 0.8s, 1.2s, 1.6s, and 2s is 20%, 40%, 60%, 80%, and 100%, respectively. The vehicle operation data sets obtained at these 5 moments are used as historical vehicle operation data.
[0044] S2.2, use the training data set to train the constructed multivariate linear regression progress prediction model.
[0045] 1) Construct a multivariate linear regression progress prediction model, the function expression is:
[0046] In the formula, , They represent a vehicle operating parameter respectively. , , Respectively represent the linear regression coefficients corresponding to each vehicle operating parameter, represents the estimation error, which is a normally distributed zero-mean random variable, Indicates the percentage of completion of the driver taking over.
[0047] 2) Test The training data samples are input into the multivariate linear regression progress prediction model, and the regression equation is constructed to estimate the linear regression parameters, and the following regression equation is obtained:
[0048] in, , , , ; It should be noted that The interval period between each group of data should be greater than 100 , is the number of independent variables in the model, satisfying the condition .
[0049] 3) Use the least squares method to obtain the regression model parameters , , Estimated value of , , , so that The variance of is an unbiased estimate:
[0050] 4) Estimated model parameters Substitute the multivariate linear regression progress prediction model function expression into the model to obtain the following optimized regression model of the driver's takeover status, that is, the trained multivariate linear regression progress prediction model, .
[0051] Through this optimized regression model, after the intelligent driving vehicle triggers the driver takeover reminder, the driver's takeover behavior completion progress value can be calculated by real-time collection of vehicle parameter data such as vehicle acceleration value, vehicle yaw angular velocity value, steering wheel hand torque, driver monitoring system image analysis data and driver takeover reminder level data, so as to predict whether the driver has completed the takeover and realize the safe transfer of vehicle driving rights.
[0052] An embodiment of a method for training a driver takeover behavior progress prediction model is described in detail above. Based on the method for training a driver takeover behavior progress prediction model described in the above embodiment, an embodiment of the present invention also provides a driver takeover behavior progress prediction model training system corresponding to the method.
[0053] Figure 2 is a schematic block diagram of a driver takeover behavior progress prediction model training system provided by an embodiment of the present invention. In this embodiment, the driver takeover behavior progress prediction model training system 200 can be divided into multiple functional modules according to the functions it performs, such as Figure 2 The module referred to in the present invention refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, and is stored in a memory.
[0054] The historical data collection module 210 is used to collect the historical takeover data of the vehicle. The historical takeover data set of the vehicle includes the completion time of the takeover and the vehicle operation parameters.
[0055] Prediction model training module 220: used to train the constructed multiple linear regression progress prediction model using the historical vehicle takeover data to obtain the trained multiple linear regression progress prediction model.
[0056] The driver takeover behavior progress prediction model training system of this embodiment is used to implement the aforementioned driver takeover behavior progress prediction model training method. Therefore, the specific implementation method of the system can be seen in the embodiment section of the driver takeover behavior progress prediction model training method in the previous text. Therefore, its specific implementation method can refer to the description of the corresponding embodiments of each part, which will not be elaborated here.
[0057] In addition, since the driver takeover behavior progress prediction model training system of this embodiment is used to implement the aforementioned driver takeover behavior progress prediction model training method, its function corresponds to that of the aforementioned method and will not be repeated here.
[0058] Figure 3 A schematic diagram of the structure of a terminal 300 provided in an embodiment of the present invention includes: a processor 310, a memory 320 and a communication unit 330. The processor 310 is used to implement the following steps when implementing the driver takeover behavior progress prediction model training program stored in the memory 320: Collect historical vehicle takeover data, which includes the takeover completion time and vehicle operation parameters; The constructed multivariate linear regression progress prediction model is trained using the vehicle historical takeover data to obtain the trained multivariate linear regression progress prediction model.
[0059] Figure 4 : is a flowchart of a method for predicting the progress of a driver's takeover behavior provided by an embodiment of the present invention, wherein: Figure 1 The execution subject may be a driver takeover behavior progress prediction system. The driver takeover behavior progress prediction method provided in the embodiment of the present invention is executed by a computer device in a vehicle, and accordingly, the driver takeover behavior progress prediction system runs in the computer device of the vehicle. According to different requirements, the order of the steps in the flowchart can be changed, and some can be omitted.
[0060] SS1, responds to the takeover reminder from the vehicle and collects the vehicle operating parameters in real time.
[0061] SS2, use the trained multivariate linear regression progress prediction model to process the real-time collected vehicle operation parameters to obtain the takeover behavior progress prediction results.
[0062] It can be understood that the trained multivariate linear regression progress prediction model obtained by the driver takeover behavior progress prediction model training method of the above-mentioned embodiment is pre-deployed in the vehicle. When the vehicle sends out a takeover system, the vehicle operating parameters are collected in real time. The vehicle operating parameters include vehicle acceleration value, vehicle yaw angular velocity value, steering wheel hand torque, and takeover reminder level. These real-time vehicle operating parameters are then input into the trained multivariate linear regression progress prediction model for processing to obtain the takeover behavior progress prediction result.
[0063] After the vehicle obtains the progress prediction results, it can partially transfer the driving rights at a certain progress, and complete the transfer of driving rights after the takeover progress reaches 100%. Specific driving rights transfer rules can be formulated according to needs to ensure the safe and smooth transfer of vehicle driving rights.
[0064] It should be noted that the multivariate linear regression progress prediction model can be trained separately for different operating conditions or scenarios to further improve the accuracy.
[0065] An embodiment of a method for training a driver takeover behavior progress prediction model is described in detail above. Based on the method for training a driver takeover behavior progress prediction model described in the above embodiment, an embodiment of the present invention also provides a driver takeover behavior progress prediction model training system corresponding to the method.
[0066] Figure 5 is a schematic block diagram of a driver takeover behavior progress prediction system provided by an embodiment of the present invention. In this embodiment, the driver takeover behavior progress prediction system 500 can be divided into multiple functional modules according to the functions it performs, such as Figure 5 The module referred to in the present invention refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, and is stored in a memory.
[0067] Real-time data collection module 510: used to respond to the takeover reminder issued by the vehicle and collect vehicle operating parameters in real time.
[0068] The progress prediction module 520 is used to process the vehicle operation parameters collected in real time using the trained multivariate linear regression progress prediction model to obtain the takeover behavior progress prediction result.
[0069] The driver takeover behavior progress prediction system of this embodiment is used to implement the aforementioned driver takeover behavior progress prediction method. Therefore, the specific implementation method of the system can be seen in the embodiment section of the driver takeover behavior progress prediction method in the previous text. Therefore, its specific implementation method can refer to the description of the corresponding embodiments of each part, which will not be elaborated here.
[0070] In addition, since the driver takeover behavior progress prediction system of the present embodiment is used to implement the aforementioned driver takeover behavior progress prediction method, its function corresponds to that of the aforementioned method and will not be described in detail here.
[0071] This embodiment provides a vehicle, which deploys the trained multivariate linear regression progress prediction model obtained in the above embodiment and executes the driver takeover behavior progress prediction method in the above embodiment.
[0072] The above disclosure is only a preferred embodiment of the present invention, but the present invention is not limited thereto. Any non-creative changes that can be thought of by a person skilled in the art, as well as several improvements and modifications made without departing from the principle of the present invention, should fall within the protection scope of the present invention.
Claims
1. A method for training a driver takeover behavior progress prediction model, characterized in that: The following steps are involved: Collect historical vehicle takeover data, which includes the takeover completion time and vehicle operation parameters; The constructed multivariate linear regression progress prediction model is trained using the vehicle historical takeover data to obtain the trained multivariate linear regression progress prediction model.
2. The method for training a driver takeover behavior progress prediction model according to claim 1, characterized in that: Collect historical vehicle takeover data, including: Obtain the lateral takeover time and the longitudinal takeover time of the vehicle, and take the longer of the two as the completion takeover time; The vehicle operating parameters from the time the vehicle issues a takeover reminder to the time the takeover is completed are obtained. The vehicle operating parameters include vehicle acceleration value, vehicle yaw angular velocity value, steering wheel hand torque, and takeover reminder level.
3. The method for training a driver takeover behavior progress prediction model according to claim 2, characterized in that: The constructed multivariate linear regression progress prediction model is trained using the historical vehicle takeover data, including: Construct vehicle historical takeover data as a training dataset; The constructed multivariate linear regression progress prediction model is trained using the training data set.
4. The method for training a driver takeover behavior progress prediction model according to claim 3, characterized in that: The vehicle historical takeover data is constructed as a training dataset, including: Calculate the average of all completed takeover times to obtain the average takeover time; Divide the average takeover time into M segments, and divide the 100% takeover completion progress into M takeover completion percentage progress. Each segment of the average takeover time corresponds to one takeover completion percentage progress. Obtain vehicle operation parameters corresponding to each average takeover time; Each completion percentage of the takeover and its corresponding vehicle operation parameter constitute a set of element pairs, and a set of element pairs is a training data sample; A number of training data samples form a training data set.
5. The method for training a driver takeover behavior progress prediction model according to claim 4, characterized in that: The constructed multivariate linear regression progress prediction model is trained using the training data set, including: Construct a multivariate linear regression progress prediction model, the function expression is: In the formula, , They represent a vehicle operating parameter respectively. , , Respectively represent the linear regression coefficients corresponding to each vehicle operating parameter, represents the estimation error, which is a normally distributed zero-mean random variable, Indicates the percentage progress of the driver taking over; Will test The training data samples are input into the multivariate linear regression progress prediction model, and the regression equation is constructed to estimate the linear regression parameters, and the following regression equation is obtained: in, , , , ; The least squares method is used to obtain the regression model parameters. , , Estimated value of , , , so that The variance of is an unbiased estimate: The model parameter estimates Substitute the multivariate linear regression progress prediction model function expression into the model to obtain the following optimized regression model of the driver's takeover status, that is, the trained multivariate linear regression progress prediction model, 。 6. A driver takeover behavior progress prediction model training system, characterized in that: include, Historical data collection module: used to collect historical vehicle takeover data, which includes the takeover completion time and vehicle operation parameters; Prediction model training module: used to train the constructed multivariate linear regression progress prediction model using the vehicle historical takeover data to obtain the trained multivariate linear regression progress prediction model.
7. A terminal, characterized in that: include: A memory, used for storing a driver takeover behavior progress prediction model training program; A processor, used to implement the steps of the driver takeover behavior progress prediction model training method as described in any one of claims 1 to 5 when executing the driver takeover behavior progress prediction model training program.
8. A method for predicting the progress of a driver's takeover behavior, characterized in that: The following steps are involved: Respond to vehicle takeover reminders and collect vehicle operating parameters in real time; The trained multivariate linear regression progress prediction model obtained by the method described in any one of claims 1 to 5 is used to process the vehicle operation parameters collected in real time to obtain the takeover behavior progress prediction result.
9. A driver takeover behavior progress prediction system, characterized in that: The method according to claim 8 is implemented, comprising: Real-time data collection module: used to respond to vehicle takeover reminders and collect vehicle operating parameters in real time; Progress prediction module: used to process the vehicle operation parameters collected in real time using the trained multivariate linear regression progress prediction model obtained by the method described in any one of claims 1 to 5 to obtain the progress prediction result of the takeover behavior.
10. A vehicle, characterized in that: Deploy the trained multivariate linear regression progress prediction model obtained by the method described in any one of claims 1-5, and execute the method described in claim 8.