Method and apparatus for predicting a state of a vehicle based on dynamic data

Abstract

A method and apparatus for predicting a state of a vehicle based on dynamic data are disclosed. According to one aspect of the disclosure, a method for predicting a state of a vehicle based on dynamic data is provided, wherein the method includes collecting driving data of the vehicle, detecting a change in a driving environment of the vehicle based on the driving data, generating dynamic data based on the driving data, predicting a current state of the vehicle based on the dynamic data, and generating a final prediction of the current state of the vehicle based on a prediction of an initial state of the vehicle and the prediction of the current state of the vehicle.

Description

[0001] Cross-references to related applications

[0002] This application claims priority and benefit to Korean Patent Application No. 10-2025-0009546, filed on January 22, 2025, the entire disclosure of which is incorporated herein by reference for all purposes. Technical Field

[0003] This disclosure relates to methods and apparatus for predicting the state of a vehicle based on dynamic data. Background Technology

[0004] The statements in this section are provided only as background information in relation to this disclosure and need not constitute prior art.

[0005] Regarding vehicle control systems, the importance of technologies for predicting vehicle conditions during operation and performing adaptive control in real time based on this is increasing. In particular, dynamic variables such as vehicle speed need to be accurately predicted to maintain vehicle stability and efficiency.

[0006] Existing vehicle state prediction technologies are designed to operate by predicting vehicle states under specific conditions based on linear models. A limitation of these technologies is that nonlinear external factors such as changes in road gradient, weather conditions, and vehicle load are not reflected in the predictions, resulting in poor accuracy. Furthermore, existing technologies also suffer from poor prediction accuracy because the data structure remains unchanged even when the vehicle's driving environment alters.

[0007] Therefore, a technology is needed to generate dynamic data reflecting changes in the driving environment of a vehicle and to predict the vehicle's state based on this data. Summary of the Invention

[0008] The purpose of this disclosure is to provide a method and apparatus capable of more accurately predicting the state of a vehicle. Specifically, the main objective of this invention is to provide a method and apparatus that generates dynamic data reflecting changes in the vehicle's driving environment, and combines the results of predicting the vehicle's state based on basic data generated from the initial driving environment with the results of predicting the vehicle's state based on the dynamic data to generate a final prediction of the vehicle's state.

[0009] The technical objectives of this disclosure are not limited to those described above, and those skilled in the art can clearly understand other technical objectives not mentioned above from the description given below.

[0010] According to one aspect of this disclosure, a method for predicting the state of a vehicle based on dynamic data is provided, wherein the method includes: collecting driving data of the vehicle; detecting changes in the driving environment of the vehicle based on the driving data; generating dynamic data based on the driving data; predicting the current state of the vehicle based on the dynamic data; and generating a final prediction of the current state of the vehicle based on the prediction of the initial state of the vehicle and the prediction of the current state of the vehicle.

[0011] According to one aspect of this disclosure, an apparatus for predicting the state of a vehicle based on dynamic data is provided, wherein the apparatus includes: at least one memory storing instructions; and at least one processor, wherein the at least one processor is configured to execute the instructions to perform the following steps: collecting driving data of the vehicle; detecting changes in the driving environment of the vehicle based on the driving data; generating dynamic data based on the driving data; predicting the current state of the vehicle based on the dynamic data; and generating a final prediction of the current state of the vehicle based on the prediction of the initial state of the vehicle and the prediction of the current state of the vehicle.

[0012] According to embodiments of this disclosure, by using an extended Kalman filter to predict the initial state of a vehicle, nonlinear driving patterns can be considered when performing the prediction, thereby improving the accuracy of the prediction.

[0013] According to embodiments of this disclosure, the accuracy of predictions can be improved by generating dynamic data that includes features that indicate the changing driving environment whenever the vehicle's driving environment changes.

[0014] According to embodiments of this disclosure, dynamic data is generated by adding features indicating changes in the vehicle's driving environment based on basic data, enabling predictions that take into account nonlinear driving patterns to be performed, thereby improving the accuracy of the predictions.

[0015] According to embodiments of this disclosure, by generating dynamic data in real time and reflecting various environmental changes that occur while the vehicle is in motion, it can adapt to changing driving environments in real time.

[0016] According to embodiments of this disclosure, by using a machine learning model to predict the initial state of a vehicle, predictions can be performed considering nonlinear driving patterns, thereby improving the accuracy of the predictions.

[0017] According to embodiments of this disclosure, vehicle status can be predicted based on dynamic data reflecting the latest driving conditions of the vehicle, thereby improving the accuracy of the prediction.

[0018] The technical effects of this disclosure are not limited to those described above, and those skilled in the art to which this disclosure pertains may understand other technical effects not mentioned herein from the following description. Attached Figure Description

[0019] Figure 1 This is a block diagram schematically illustrating the components of a vehicle state prediction device according to an embodiment of the present disclosure.

[0020] Figure 2 This is a diagram illustrating the process of predicting the initial state of a vehicle using a vehicle state prediction device according to an embodiment of the present disclosure.

[0021] Figure 3 This is a diagram illustrating the process of ultimately predicting the current state of a vehicle using a vehicle state prediction device according to an embodiment of the present disclosure.

[0022] Figure 4 This is a diagram illustrating the accuracy of vehicle state prediction based on dynamic data relative to vehicle state prediction based on basic data according to embodiments of the present disclosure.

[0023] Figure 5 This is a flowchart schematically illustrating a vehicle state prediction method based on dynamic data according to an embodiment of the present disclosure.

[0024] Figure 6 This is a block diagram schematically illustrating an exemplary computing device that can be used to implement the devices and methods described in this disclosure. Detailed Implementation

[0025] In the following description, some exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In the following description, although elements are shown in different drawings, the same reference numerals preferably denote the same elements. Furthermore, in the following description of some embodiments, detailed descriptions of known functions and configurations incorporated herein will be omitted for clarity and brevity.

[0026] Furthermore, terms such as first, second, A, B, (a), (b), etc., are used only to distinguish one component from another and do not imply or suggest the nature, order, or sequence of the components. Throughout this specification, when a part “includes” or “comprises” a component, unless specifically stated otherwise, the part means that those components are not excluded and that other components are included. Terms such as “unit”, “module,” etc., refer to one or more units for performing at least one function or operation, which can be implemented by hardware, software, or a combination of hardware and software.

[0027] The following detailed description, together with the accompanying drawings, is intended to describe exemplary embodiments of the invention and is not intended to merely represent embodiments that can be practiced.

[0028] Figure 1 This is a block diagram schematically illustrating the components of a vehicle state prediction device according to an embodiment of the present disclosure.

[0029] refer to Figure 1 The vehicle state prediction device may include a driving data collection unit 101, a situation change detection unit 103, a data feature extraction unit 105, a first vehicle state prediction unit 107, a second vehicle state prediction unit 109, and a final vehicle state prediction unit 111.

[0030] The driving data collection unit 101 can collect driving data of the vehicle. Driving data may include vehicle speed, vehicle acceleration, and vehicle angular velocity. The driving data collection unit 101 can collect driving data based on one or more sensors attached to the vehicle. The one or more sensors attached to the vehicle may include one or more of an accelerometer, a gyroscope, and an inertial measurement unit (IMU) sensor.

[0031] The change-of-condition detection unit 103 can detect changes in the vehicle's driving environment based on driving data. The change-of-condition detection unit 103 can detect changes in the vehicle's driving environment by determining whether at least one or more of a first condition, a second condition, and a third condition are met based on the driving data. The first condition, the second condition, and the third condition can be different from each other. The change-of-condition detection unit 103 can be implemented using some or all of the computing devices in one or more computing devices 60.

[0032] The data feature extraction unit 105 can generate basic data or dynamic data based on driving data. Each of the basic data and dynamic data can include multiple features. The data feature extraction unit 105 can extract one or more features based on driving data, and can generate basic data and dynamic data based on the extracted features. The data feature extraction unit 105 can be implemented using some or all of the computing devices in one or more computing devices 60.

[0033] The data feature extraction unit 105 can extract one or more features based on the initial driving data, and generate basic data based on the extracted features. The basic data can be in the form of a dataset.

[0034] Table 1

[0035] feature Description of features speed vehicle speed value acceleration vehicle acceleration value angular velocity Rotational motion state of the vehicle Wheel speed Rotation speed of each wheel GPS coordinates Vehicle location information Tire condition Tire pressure and wear Fuel status Fuel quantity and consumption

[0036] Table 1 is a table illustrating examples of features that may be included in the basic data. The basic data may include one or more of the following as features: speed, acceleration, angular velocity, wheel speed, GPS coordinates, tire condition, and fuel condition.

[0037] When a change in the vehicle's driving environment is detected, the data feature extraction unit 105 can extract one or more features based on the driving data, and generate dynamic data based on the extracted features. The initial driving data and the actual driving data may differ in the time of measurement and collection. This will be discussed in the following reference. Figure 2 and Figure 3 Detailed Description. The process by which the data feature extraction unit 105 generates dynamic data may include adding one or more features that differ from those included in the basic data based on changes in the vehicle's driving environment. That is, the data feature extraction unit 105 can generate dynamic data reflecting the changing driving environment by adding one or more features extracted from driving data that are not included in the basic data to the basic data, thereby updating the basic data as an existing dataset according to the changing environment. The dynamic data may take the form of a dataset.

[0038] Table 2

[0039]

[0040]

[0041] Table 2 is a table showing examples of features that can be additionally included in dynamic data compared to the basic data. Dynamic data can also include one or more features that differ from those included in the basic data depending on the context. For example, when determining lane change scenarios, "lane position" and "vehicle position change" can be added.

[0042] Basic data assumes the vehicle is traveling on flat ground. Dynamic data assumes the vehicle is traveling under changed conditions, i.e., when it is not traveling on flat ground. As shown in Table 2, situations where the vehicle is not traveling on flat ground can include entering a ramp, driving on a wet road, adding passengers, braking, driving on a highway, and changing lanes.

[0043] The first vehicle state prediction unit 107 can predict the initial state of the vehicle based on basic data. For example, the first vehicle state prediction unit 107 can predict the initial weight of the vehicle based on basic data (such as Table 1 generated based on driving data collected when the vehicle is driving on flat ground).

[0044] In this disclosure, "state" is a term included in the initial state of the vehicle, the current state of the vehicle, and the final prediction of the current state of the vehicle, and can be any one of the vehicle's speed, weight, and torque. Therefore, the process of predicting the state of the vehicle can be the process of predicting any one or more of the vehicle's speed, weight, and torque.

[0045] The initial state, current state, and final prediction of the current state of a vehicle differ only at the point in time of prediction and can essentially represent the value of the same feature. For example, when the goal is to predict speed as the vehicle's state, the first vehicle state prediction unit 107 can predict the initial speed, the second vehicle state prediction unit 109 can predict the current speed, and the final vehicle state prediction unit 111 can generate the final prediction of the current speed. In other words, the initial state, current state, and final prediction of the current state of a vehicle can all be values ​​of the vehicle's speed.

[0046] The first vehicle state prediction unit 107 can predict the initial state of the vehicle based on basic data. For example, the first vehicle state prediction unit 107 can predict the initial weight of the vehicle based on basic data (such as Table 1 generated based on driving data collected when the vehicle is driving on flat ground). The first vehicle state prediction unit 107 can be implemented using some or all of the computing devices in one or more computing devices 60.

[0047] The second vehicle state prediction unit 109 can predict the current state of the vehicle based on dynamic data. For example, the second vehicle state prediction unit 109 can predict the weight of the vehicle based on dynamic data generated from driving data collected before and after the vehicle enters the ramp. When compared with basic data, the dynamic data may also include characteristics such as the slope and the coefficient of friction. The second vehicle state prediction unit 109 can be implemented using some or all of the computing devices in one or more computing devices 60.

[0048] The vehicle state final prediction unit 111 can generate a final prediction of the vehicle's current state based on predictions of the vehicle's initial state and current state. For example, the vehicle state final prediction unit 111 can generate a final prediction of the vehicle's weight based on the initial weight of the vehicle predicted by the first vehicle state prediction unit 107 and the weight of the vehicle predicted by the second vehicle state prediction unit 109. As an example, the vehicle state final prediction unit 111 can determine the final prediction of the vehicle's weight as the value obtained by taking the arithmetic mean of the initial weight of the vehicle predicted by the first vehicle state prediction unit 107 and the weight of the vehicle estimated by the second vehicle state prediction unit 109. The vehicle state final prediction unit 111 can be implemented using some or all of the computing devices in one or more computing devices 60.

[0049] Figure 2 This is a diagram illustrating the process of predicting the initial state of a vehicle using a vehicle state prediction device according to an embodiment of the present disclosure.

[0050] Figure 3This is a diagram illustrating the process of ultimately predicting the current state of a vehicle using a vehicle state prediction device according to an embodiment of the present disclosure.

[0051] Figure 2 and Figure 3 This can be a process executed at different times. For example, it can be executed... Figure 2 Execute after the process Figure 3 The process. In other words, Figure 3 The process can be assumed to have already been executed. Figure 2 The process in.

[0052] Figure 2 This can be based on the initial driving conditions. (Reference) Figure 2 The driving data collection unit 101 can send the driving data to the data feature extraction unit (S210). The data feature extraction unit 105 can extract one or more features based on the driving data, and generate basic data based on the extracted features (S220). The driving data can be initial driving data. In this disclosure, the initial driving situation and initial driving data can refer to the situation where basic data is generated as a reference for dynamic data generation and driving data is collected under this condition. The premise for generating basic data can be the situation where the vehicle is driving on flat ground. That is, the vehicle state prediction device can consider the situation where the vehicle's state is approximately constant and can be used as a reference for state changes as the situation where the vehicle is driving on flat ground.

[0053] The data feature extraction unit 105 can send basic data to the first vehicle state prediction unit 107. The first vehicle state prediction unit 107 can predict the initial state of the vehicle based on the basic data (S230). The first vehicle state prediction unit 107 can predict the initial state of the vehicle based on the basic data by using an extended Kalman filter. The extended Kalman filter can be described using Equations 1 to 5. The prediction result of the initial state of the vehicle can be explained based on the estimated value and the error covariance. In Equation 4, the estimated value can be... Furthermore, in Formula 5, the error covariance can be P. k .

[0054] (Formula 1)

[0055]

[0056] (Formula 2)

[0057] P k -=AP k-1 -A T +Q

[0058] (Formula 3)

[0059] Kk =P k -H T HP k -H T +R) -1

[0060] (Formula 4)

[0061]

[0062] (Formula 5)

[0063] P k =P k - -K k HP k -

[0064] Formula 1 is the formula used to calculate the predicted value of the estimated value. It is the predicted value of the k-th estimate. It is the (k-1)th estimate.

[0065] Formula 2 is the formula used to calculate the predicted value of the error covariance. P k - It is the predicted value of the k-th error covariance. A is the system matrix. P k-1 - It is the predicted value of the (k-1)th error covariance. Q is the system noise.

[0066] Formula 3 is the formula used to calculate the Kalman gain. K k It is the Kalman gain. P k - This is the predicted value of the k-th error covariance. H is the system matrix. R is the sensor noise.

[0067] Formula 4 is the formula used to calculate the estimated value. It is the kth estimate. It is the predicted value of the k-th estimate. k It's the Kalman gain. k This is the data from the kth sensor.

[0068] Formula 5 is the formula used to calculate the error covariance. P k It is the k-th error covariance. P k - It is the predicted value of the k-th error covariance, K k H is the Kalman gain. H is the system matrix.

[0069] As described above, the vehicle state prediction device according to the embodiments of the present disclosure predicts the initial state of the vehicle by using an extended Kalman filter, and thus can perform the prediction by taking into account nonlinear driving modes, thereby improving the accuracy of the prediction.

[0070] Figure 3 We can assume any driving conditions after the initial driving conditions. (Reference) Figure 3 The driving data collection unit 101 can send driving data to the situation change detection unit 103 (S310). Figure 3 The driving data in the system can be any driving data other than the initial driving data. Figure 3 Each of the arbitrary driving situations and arbitrary driving data can represent driving situations that are not the initial driving situation, as well as driving situations that are not the initial driving data. Note that... Figure 3 The driving conditions and driving data in the system can be based on... Figure 2 Based on the initial driving data obtained under the initial driving conditions, the basic data can be used to represent the current driving status of the vehicle and the driving data collected under those conditions.

[0071] The change-of-condition detection unit 103 can detect changes in the vehicle's driving environment based on driving data. The process of detecting changes in the vehicle's driving environment may include determining whether at least one or more of a first condition, a second condition, and a third condition are met based on the vehicle's driving data. The first condition, the second condition, and the third condition may be different from each other. Because the change-of-condition detection unit 103 can detect a change in the vehicle's driving environment by determining whether at least one or more of the first condition, the second condition, and the third condition are met, the process by which the change-of-condition detection unit 103 determines whether each condition is met can be understood as the process by which the change-of-condition detection unit 103 executes a detection algorithm corresponding to each condition.

[0072] The first condition can be a condition set based on the rate of change of driving data. For example, the situation change detection unit 103 can collect multiple driving data points during a predetermined time interval, calculate the rate of change of the driving data based on the driving data, and determine that the first condition is met when the magnitude (e.g., absolute value) of the calculated rate of change is equal to or greater than a threshold. Furthermore, when the first condition is determined to be met, the situation change detection unit 103 can determine that the vehicle's driving environment has changed.

[0073] (Formula 6)

[0074] ΔA(t)=A(t)-A(t-1)

[0075] (Formula 7)

[0076] |△A(t)|≥threshold

[0077] Formula 6 is used to calculate the rate of change. ΔA(t) is the rate of change. A(t) is the driving data for a specific time period. A(t-1) is the driving data for the previous time period. Driving data can be, for example, acceleration.

[0078] Formula 7 is used to determine whether the first condition is met. |△A(t)| is the magnitude of the rate of change. The threshold is the threshold value.

[0079] The second condition can be a condition set based on the magnitude of the driving data. For example, when the magnitude of the driving data is equal to or greater than a threshold, the situation change detection unit 103 can determine that the second condition is met. Furthermore, when the second condition is determined to be met, the situation change detection unit 103 can determine that the vehicle's driving environment has changed. The driving data can be, for example, acceleration or speed.

[0080] The third condition can be a condition set based on the driving data pattern. When the driving data for a specific time period does not fall within the normal variation range of driving data from multiple previous time periods, the situation change detection unit 103 can determine that the third condition is met. For example, the situation change detection unit 103 can calculate the moving average and moving standard deviation of driving data from multiple previous time periods, and when the absolute value of the difference between the specific time period and the moving average is equal to or greater than a certain multiple of the moving standard deviation, it can determine that the third condition is met. Furthermore, when the third condition is determined to be met, the situation change detection unit 103 can determine that the vehicle's driving environment has changed.

[0081] (Formula 8)

[0082]

[0083] (Formula 9)

[0084]

[0085] (Formula 10)

[0086] |x(t)-MA(t)|≥α·σ(t)

[0087] Formula 8 is a formula for calculating the moving average of driving data over multiple previous time periods. MA(t) is the moving average.

[0088] Formula 9 is the formula for calculating the moving standard deviation of driving data over multiple previous time periods. σ(t) is the moving standard deviation.

[0089] Formula 10 is used to determine whether the third condition is satisfied. α is any non-zero constant.

[0090] When a change in driving is detected, the situation change detection unit 103 can send the driving data to the data feature extraction unit 105 (S320).

[0091] The data feature extraction unit 105 can generate dynamic data based on driving data (S330). The data feature extraction unit 105 can obtain basic data. Due to the execution Figure 2 The data feature extraction unit 105 generates basic data through processes included in the process, and this basic data can be stored in any memory (not shown) included in the vehicle state prediction device. The data feature extraction unit 105 can add one or more features different from those included in the basic data based on changes in the vehicle's driving environment. Specifically, the data feature extraction unit 105 can generate dynamic data by extracting one or more features based on driving data and updating the basic data by adding features not included in the basic data from the extracted features. When extracting one or more features based on driving data, the data feature extraction unit 105 can take into account changes in the vehicle's driving environment.

[0092] As described above, the vehicle state prediction device according to embodiments of the present disclosure can improve the accuracy of prediction by generating dynamic data, which includes features that can indicate changes in the driving environment at any time as the vehicle's driving environment changes.

[0093] As described above, the vehicle state prediction device according to embodiments of the present disclosure generates dynamic data by adding features indicating changes in the vehicle's driving environment based on basic data, and thus can perform predictions taking into account nonlinear driving patterns, thereby improving prediction accuracy.

[0094] As described above, the vehicle state prediction device according to the embodiments of this disclosure can generate dynamic data and reflect various environmental changes that occur in real time when the vehicle is driving, thereby adapting to the changing driving environment in real time.

[0095] The data feature extraction unit 105 can send dynamic data to the second vehicle state prediction unit 109. The second vehicle state prediction unit 109 can predict the current state of the vehicle based on the dynamic data (S340). The second vehicle state prediction unit 109 can predict the current state of the vehicle based on the dynamic data using a machine learning model. The machine learning model can be an XGBoost regression model. The machine learning model can be a pre-trained model. A training module (not shown) separate from the machine learning model can be used for pre-training the machine learning model. The training module can adapt the machine learning model to the training dataset. The training module can train the machine learning model so that the machine learning model corrects the error in the vehicle state prediction. For example, the training module can input the input data into the machine learning model and calculate the loss based on the output data output by the machine learning model and the ground truth (GT) data mapped to the input data. Mean squared error (MSE) can be used as the loss function, but is not limited to it. The training module can update the parameters of the machine learning model to minimize the loss. For example, the training module can update the weights of the layers (or nodes) included in the machine learning model based on the backpropagation algorithm.

[0096] (Formula 11)

[0097]

[0098] (Formula 12)

[0099]

[0100] Equation 11 is a formula used to illustrate an example of the XGBoost regression model. It is the predicted value, f m This is the prediction function for each tree. M is the number of trees. i This is the input data.

[0101] Equation 12 is an example formula used to illustrate the loss function of the XGBoost regression model. L(θ) is the loss function. l is the loss. y i This is GT data. Ω is the normalization term.

[0102] As described above, the vehicle state prediction device according to the embodiments of the present disclosure can improve the accuracy of prediction by using a machine learning model to predict the initial state of the vehicle and by performing the prediction in consideration of nonlinear driving patterns.

[0103] The second vehicle state prediction unit 109 can send the prediction of the vehicle's current state to the vehicle state final prediction unit 111. The vehicle state final prediction unit 111 can generate a final prediction of the vehicle's current state based on the prediction of the vehicle's initial state and the prediction of the vehicle's current state (S350). The vehicle state final prediction unit 111 can obtain the prediction of the vehicle's initial state. The prediction of the vehicle's initial state is generated to execute the following: Figure 2 The result of the process can be stored in any memory (not shown) included in the vehicle state prediction device. As an example, the vehicle state final prediction unit 111 can determine the value obtained by adding different weights to each of the prediction of the initial state of the vehicle and the prediction of the current state of the vehicle as the final prediction of the current state of the vehicle.

[0104] Figure 4 This is a diagram illustrating the accuracy of vehicle state prediction based on dynamic data relative to vehicle state prediction based on basic data according to embodiments of the present disclosure.

[0105] refer to Figure 4 The error 41 in vehicle state prediction based on basic data and the error 42 in vehicle state prediction based on dynamic data are shown.

[0106] When comparing the prediction error 411 based on basic data and the prediction error 421 based on dynamic data relative to the vehicle's speed, the prediction error 411 based on basic data is greater than the prediction error 421 based on dynamic data.

[0107] Comparing the prediction error 413 based on basic data with the prediction error 423 based on dynamic data, relative to the vehicle's weight, the prediction error 413 based on basic data is greater than the prediction error 423 based on dynamic data. Specifically, the prediction error 413 based on basic data is approximately four times or more greater than the prediction error 423 based on dynamic data.

[0108] Comparing the prediction error 415 based on basic data with the prediction error 425 based on dynamic data, relative to the vehicle's torque, the prediction error 415 based on basic data is greater than the prediction error 425 based on dynamic data. Specifically, the prediction error 415 based on basic data is approximately twice as large as the prediction error 425 based on dynamic data.

[0109] As described above, the vehicle state prediction device according to the embodiments of this disclosure can predict the vehicle state based on dynamic data by reflecting the latest driving conditions of the vehicle, thereby improving the accuracy of the prediction.

[0110] Figure 5This is a flowchart schematically illustrating a vehicle state prediction method based on dynamic data according to an embodiment of the present disclosure.

[0111] refer to Figure 5 The vehicle status prediction device can collect vehicle driving data (S510).

[0112] The vehicle state prediction device can detect changes in the vehicle's driving environment based on driving data (S520).

[0113] The vehicle state prediction device can generate dynamic data based on driving data (S530).

[0114] The vehicle state prediction device can predict the current state of the vehicle based on dynamic data (S540).

[0115] The vehicle state prediction device can generate a final prediction of the current state of the vehicle based on the prediction of the initial state of the vehicle and the prediction of the current state of the vehicle (S550).

[0116] Figure 6 This is a block diagram schematically illustrating an exemplary computing device that can be used to implement the devices and methods described in this disclosure.

[0117] Computing device 60 may include all or some of a memory 600, a processor 620, a storage device 640, an input / output interface 660, and a communication interface 680. Computing device 60 may be a fixed computing device (such as a desktop computer or server) or a mobile computing device (such as a laptop computer or smartphone). Computing device 60 may include a dedicated hardware accelerator capable of efficiently processing the operations of artificial intelligence models. For example, computing device 60 may include a graphics processing unit (GPU), a tensor processing unit (TPU), or a neural processing unit (NPU).

[0118] Memory 600 may store programs that enable processor 620 to perform methods or operations according to various embodiments of the present disclosure. For example, the program may include multiple instructions executable by processor 620, and the methods or operations described above may be performed by processor 620 executing the multiple instructions. Memory 600 may consist of a single memory or multiple memories. In this case, information required to perform the methods or operations according to various embodiments of the present disclosure may be stored in a single memory or distributed across multiple memories. When memory 600 consists of multiple memories, the multiple memories may be physically separated. Memory 600 may include at least one of volatile memory and non-volatile memory. Volatile memory includes static random access memory (SRAM) or dynamic random access memory (DRAM), and non-volatile memory includes flash memory.

[0119] Processor 620 may include at least one core capable of executing at least one instruction. Processor 620 may execute instructions stored in memory 600. Processor 620 may consist of a single processor or multiple processors.

[0120] Even if the power supply to computing device 60 is cut off, storage device 640 retains the stored data. For example, storage device 640 may include non-volatile memory or may include storage media such as magnetic tape, optical disc, or magnetic disk. Programs stored in storage device 640 can be loaded into memory 600 before being executed by processor 620. Storage device 640 may store files written in a programming language, and programs created from files by a compiler can be loaded into memory 600. Storage device 640 may store data to be processed by processor 620 and / or data that has already been processed by processor 620.

[0121] The input / output interface 660 can provide an interface with an input device such as a keyboard or mouse and / or an output device such as a display device or printer. The user can trigger the processor 620 to execute a program through the input device and / or check the processing results of the processor 620 through the output device.

[0122] The communication interface 680 can provide access to external networks. The computing device 60 can communicate with other devices through the communication interface 680.

[0123] Each element of the device or method according to the invention can be implemented in hardware or software, or a combination of hardware and software. The function of the corresponding element can be implemented in software, and a microprocessor can be implemented to execute the software function corresponding to the corresponding element.

[0124] Various implementations of the systems and techniques described herein can be implemented using digital electronic circuits, integrated circuits, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), computer hardware, firmware, software, and / or combinations thereof. Various implementations may include implementations utilizing one or more computer programs executable on a programmable system. The programmable system includes at least one programmable processor, at least one input device, and at least one output device. The at least one programmable processor may be a dedicated processor or a general-purpose processor, coupled to receive data and instructions from and send data and instructions to a storage system. The computer program (also referred to as a program, software, software application, or code) includes instructions for the programmable processor and is stored in a computer-readable recording medium.

[0125] Computer-readable recording media can include all types of storage devices capable of storing computer-readable data. Computer-readable recording media can be non-volatile media, such as read-only memory (ROM), random access memory (RAM), optical disc ROM (CD-ROM), magnetic tape, floppy disk, or optical data storage devices. Additionally, computer-readable recording media can also include transient media such as data transmission media. Furthermore, computer-readable recording media can be distributed across computer systems connected via a network and can store and execute computer-readable program code in a distributed manner.

[0126] Although the operations are shown as being performed sequentially in the flowcharts / timing diagrams of this specification, this is merely an exemplary description of the technical concept of one embodiment of this disclosure. In other words, those skilled in the art to which one embodiment of this disclosure pertains will understand that various modifications and changes can be made without departing from the essential characteristics of the embodiments of this disclosure; that is, the order shown in the flowcharts / timing diagrams can be changed, and one or more operations can be performed in parallel. Therefore, the flowcharts / timing diagrams are not limited to a temporal order.

[0127] Although exemplary embodiments of this disclosure have been described for illustrative purposes, those skilled in the art will understand that various modifications, additions, and substitutions can be made without departing from the concept and scope of the claimed invention. Therefore, exemplary embodiments of this disclosure have been described for the sake of brevity and clarity. The scope of the technical concept of these embodiments is not limited by the description. Therefore, those skilled in the art will understand that the scope of the claimed invention is not limited to the embodiments explicitly described above, but rather to the claims and their equivalents.

Claims

1. A method for predicting the state of a vehicle based on dynamic data, the method comprising: Collect the vehicle's driving data; The vehicle's driving environment is detected based on the driving data; The dynamic data is generated based on the driving data; The current state of the vehicle is predicted based on the dynamic data; as well as Based on the prediction of the vehicle's initial state and the prediction of the vehicle's current state, a final prediction of the vehicle's current state is generated.

2. The method according to claim 1, in, Detecting changes in the driving environment of the vehicle includes: determining, based on the vehicle's driving data, whether at least one or more of a first condition, a second condition, and a third condition are met, wherein the first condition, the second condition, and the third condition are different from each other.

3. The method according to claim 2, wherein, The first condition is set based on the rate of change of the driving data during a predetermined time interval.

4. The method according to claim 2, wherein, The second condition is a condition set based on the size of the driving data.

5. The method according to claim 2, wherein, The third condition is a condition set based on the driving data.

6. The method according to claim 1, wherein: Generating the dynamic data includes adding one or more features that are different from those included in the basic data based on changes in the vehicle's driving environment, and each feature included in the basic data is a feature extracted based on the vehicle's initial driving environment.

7. The method according to claim 6, wherein, The prediction of the initial state of the vehicle is generated based on the vehicle's basic data.

8. An apparatus for predicting the state of a vehicle based on dynamic data, the apparatus comprising: At least one memory to store instructions; and at least one processor, The at least one processor is configured to execute the instructions to perform the following steps: Collect the vehicle's driving data; The vehicle's driving environment is detected based on the driving data; The dynamic data is generated based on the driving data; Predict the current state of the vehicle based on the dynamic data; and Based on the prediction of the vehicle's initial state and the prediction of the vehicle's current state, a final prediction of the vehicle's current state is generated.

9. The device according to claim 8, wherein, Detecting changes in the driving environment of the vehicle includes: determining, based on the vehicle's driving data, whether at least one or more of a first condition, a second condition, and a third condition are met, wherein the first condition, the second condition, and the third condition are different from each other.

10. The device according to claim 9, wherein, The first condition is set based on the rate of change of the driving data during a predetermined time interval.

11. The device according to claim 9, wherein, The second condition is a condition set based on the size of the driving data.

12. The device according to claim 9, wherein, The third condition is a condition set based on the driving data.

13. The device according to claim 8, wherein: Generating the dynamic data includes adding one or more features that are different from those included in the basic data based on changes in the vehicle's driving environment, and each feature included in the basic data is a feature extracted based on the vehicle's initial driving environment.

14. The device according to claim 13, wherein, The prediction of the initial state of the vehicle is generated based on the vehicle's basic data.

Images