Vehicle driving evaluation method and device based on SDK combined with multi-dimensional data

Abstract

The application discloses a vehicle driving evaluation method and device based on SDK and multi-dimensional data. The method comprises the following steps: loading a first SDK for receiving original collection data sent by a vehicle on a terminal, loading a second SDK on a server; inputting data in a specified data format after noise data processing in the first SDK into the second SDK; generating effective data after the second SDK analyzes and processes the data in the specified data format; and performing driving evaluation according to the effective data. The application can maximize the safety and integrity of data, and can also effectively provide accurate data sources for data analysis and driving evaluation modules, so that the driving evaluation result is more accurate.

Description

[0001] This application is a divisional application of the invention patent application filed on November 7, 2019, entitled "Method, Apparatus and System for Data Acquisition and Driving Evaluation Based on SDK" with application number 201911084262.X. Technical Field

[0002] This invention relates to the field of intelligent driving technology, and in particular to a vehicle driving evaluation method and device based on SDK combined with multi-dimensional data. Background Technology

[0003] With social development and technological progress, the development of intelligent connected vehicle technology, especially the maturing application of IoT and vehicle-to-everything (V2X) technologies, and the increasingly rich and diverse data from V2X, there is a serious lack of analysis and application of massive amounts of big data, such as user driving behavior data, vehicle data, and driving environment data. Given that existing in-vehicle terminals (devices that transmit vehicle and driver data back to a server, such as TBOX) use different data formats, communication protocols, data accuracy, and collection frequencies, data standardization is extremely difficult. Data access overhead is therefore high and prone to problems. To address, prevent, and mitigate these problems and risks, we adopt an SDK (Software-Defined Kernel) solution. Using an SDK solution to collect data and conduct driving evaluations allows for the rapid processing and analysis of data from data providers (vehicle manufacturers, third-party data companies, and other data providers) to obtain evaluation results for the vehicle and corresponding driver.

[0004] This functionality can be achieved without an SDK, but the development cycle is lengthy and prone to leaking source code and core algorithms. Furthermore, different data providers may require multiple sets of code, necessitating repeated development and integration. Directly accessing the source code can easily lead to finger-pointing and blame-shifting when problems arise, making responsibility determination extremely difficult. Using an SDK, however, provides a clear and modular solution, making it much easier to define responsibilities for both parties during debugging and usage.

[0005] Traditional driving evaluation methods in the current technology mainly fall into two categories: one is where the data processing company directly embeds software (or code) into the data provider, and the other is where the data provider sends data to the data processing company according to certain rules. Both of these methods may involve issues of core technologies from multiple parties or data leakage. Moreover, these two methods also require a significant amount of research and development and testing time. Summary of the Invention

[0006] In view of this, the main objective of the present invention is to propose a vehicle driving evaluation method and device based on SDK combined with multi-dimensional data.

[0007] As a first aspect of the present invention, the present invention provides a vehicle driving evaluation method based on SDK combined with multi-dimensional data, the method comprising:

[0008] The first SDK for receiving raw collected data sent by the vehicle is loaded on the terminal, and the second SDK is loaded on the server.

[0009] Input the data in the specified data format after noise removal processing in the first SDK into the second SDK;

[0010] After analyzing and processing the data in the specified data format using the second SDK, valid data is generated.

[0011] Driving tests will be conducted based on the valid data.

[0012] The step of inputting data in a specified data format after noise removal processing from the first SDK into the second SDK includes:

[0013] The raw collected data is preprocessed according to the required accuracy and frequency for data analysis.

[0014] According to the specified data format, the preprocessed raw collected data in the first SDK is converted into data in the specified data format;

[0015] After the data is converted to the specified data format, it undergoes noise reduction processing and is then input into the second SDK.

[0016] The method for processing noise-removing data converted to the specified data format includes a filtering algorithm for ignition information authenticity, wherein the filtering algorithm for ignition information authenticity includes:

[0017] The ignition timing for obtaining ignition information;

[0018] Obtain voltage data U1 and acceleration data A1 corresponding to the first time period before the ignition moment, and voltage data U2 and acceleration data A2 corresponding to the second time period after the ignition moment;

[0019] Based on the expected value and variance of the voltage data U1, voltage data U2, acceleration data A1, and acceleration data A2, the actual moment corresponding to system ignition is determined.

[0020] If the initial determination result is that the system may ignite, then determining the actual moment corresponding to system ignition based on the expected value and variance value corresponding to the voltage data U1, voltage data U2, acceleration data A1, and acceleration data A2 includes operation B1:

[0021] Obtain the start time of the first time segment corresponding to the last sampling;

[0022] Based on the start time, continue to acquire the location data G1 corresponding to the first time period before the start time and the location data G2 corresponding to the second time period after the start time;

[0023] Calculate the variance DG1 of the location data G1 and the variance DG2 of the location data G2;

[0024] Based on the variance DG1 of the position data G1 and the variance DG2 of the position data G2, determine whether the system should ignite.

[0025] Repeat step B1 above until system ignition is confirmed;

[0026] Specifically, the system ignition is determined by whether the variance DG2 is greater than the variance DG1; DG1 is the variance corresponding to the position data G1, and DG2 is the variance corresponding to the position data G2.

[0027] The method for processing data converted to the specified data format by removing noise includes a filtering algorithm for verifying the authenticity of the engine shutdown information. The filtering algorithm for verifying the authenticity of the engine shutdown information includes:

[0028] The time of engine shutdown is obtained from the shutdown message;

[0029] Obtain voltage data U3 and acceleration data A3 corresponding to the first time period before the engine shutdown time, and voltage data U4 and acceleration data A4 corresponding to the second time period after the engine shutdown time;

[0030] Based on the expected value and variance value corresponding to the voltage data U3, the voltage data U4, the acceleration data A3, and the acceleration data A4, determine the actual moment corresponding to the system shutdown.

[0031] If the initial determination result is that the system may shut down, then the step of determining the actual moment corresponding to the system shutdown based on the expected value and variance value corresponding to the voltage data U3, voltage data U4, acceleration data A3, and acceleration data A4 includes operation B2:

[0032] Obtain the start time of the first time segment corresponding to the last sampling;

[0033] Based on the start time, continue to acquire the location data G3 corresponding to the first time period before the initial time and the location data G4 corresponding to the second time period after the start time;

[0034] Calculate the variance DG3 of the location data G3 and the variance DG4 of the location data G4;

[0035] Based on the variance DG3 of the location data G3 and the variance DG4 of the location data G4, determine whether the system is shut down;

[0036] Repeat step S2 above until the system is confirmed to be off.

[0037] Specifically, the system is determined to be shut down by judging whether the variance DG4 is less than the variance DG3; DG3 is the variance corresponding to the position data G3, and DG4 is the variance corresponding to the position data G4.

[0038] Preferably, the valid data includes driving behavior collection data, driving behavior analysis data, data obtained from the vehicle's own sensors, and data collected from the vehicle terminal.

[0039] Preferably, the method for processing data converted to the specified data format by noise removal includes a vehicle location information filtering algorithm, wherein the vehicle location information filtering algorithm includes:

[0040] Obtain position data by performing a first-stage noise data processing on the raw position data collected by the sensor;

[0041] The location data that has undergone the first filtering of noise data is then subjected to Kalman filtering again to obtain the second filtered location data, which is used as the effective data after noise removal.

[0042] The original location data includes longitude, latitude, speed, heading angle, altitude, and accuracy factor.

[0043] As a second aspect of the present invention, embodiments of the present invention provide a vehicle driving evaluation device based on an SDK and multi-dimensional data, the device comprising:

[0044] At least one processor; and,

[0045] A memory communicatively connected to the at least one processor; wherein,

[0046] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method described in any of the preceding statements.

[0047] As a third aspect of the present invention, the embodiments of the present invention also provide a vehicle driving evaluation system based on SDK and multi-dimensional data, characterized in that the system further includes: a server, a terminal communicatively connected to the server, and an SDK, wherein the SDK is loaded onto the server and / or the terminal to implement the method described in any of the preceding items.

[0048] In summary, the method, apparatus, and system for data acquisition and driving evaluation based on SDK provided by this invention, by loading the SDK on the terminal and / or server, and processing the raw data acquired by the SDK to obtain effective data in a format recognizable by the terminal and / or server and removing noise, has the following beneficial effects:

[0049] (1) Data providers do not need complex or cumbersome development work for their data servers or vehicle terminal software. They only need to configure the corresponding interfaces of the SDK and ensure that data collection is normal to use the various functions provided by the SDK. For data providers, using the SDK solution can ensure that there is no risk of data leakage and has the advantages of simple development and strong functional scalability. For companies that provide SDKs, using the SDK can protect their core algorithms from being illegally stolen and their business logic from being illegally used to the greatest extent.

[0050] (2) This invention does not affect the original data collection and uploading, nor does it affect the original functions of the data server.

[0051] (3) The present invention becomes very clear and modular after using the SDK, and it is very easy to define the responsibilities of both parties during debugging and use.

[0052] (4) This invention can ensure the security and integrity of data to the greatest extent, and can also provide accurate data sources for data analysis and driving evaluation modules. This makes the driving evaluation results more accurate. Attached Figure Description

[0053] Figure 1 This is a flowchart illustrating the method for data collection and driving evaluation based on SDK in Embodiment 1 of the present invention.

[0054] Figure 2 This is a diagram illustrating the data flow for loading the SDK on the server.

[0055] Figure 3 This is a diagram illustrating the data flow for loading the SDK in the terminal.

[0056] Figure 4 This diagram illustrates the data flow for loading the SDK simultaneously on both the terminal and the server.

[0057] Figure 5 This is a schematic diagram of the device for data acquisition and driving evaluation based on SDK in Embodiment 2 of the present invention.

[0058] Figure 6 This is a schematic diagram of the system for data acquisition and driving evaluation based on SDK in Embodiment 3 of the present invention.

[0059] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0060] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0061] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "part," or "unit" may be used interchangeably.

[0062] Implementation Method 1

[0063] Please see Figure 1 This invention provides a method for data acquisition and driving evaluation based on an SDK, belonging to a multi-dimensional, multi-space data acquisition and driving evaluation method. This method loads a data preprocessing SDK on the in-vehicle data acquisition terminal, a data processing SDK on the data server, and simultaneously loads the data preprocessing SDK on the in-vehicle terminal and the data processing SDK on the data processing server. This method does not affect the original data acquisition and uploading processes, nor does it affect the original functions of the data server.

[0064] This invention provides a method for data collection and driving evaluation based on an SDK, comprising:

[0065] Load the SDK in the terminal and / or on the server;

[0066] The raw data collected by the input SDK is processed to obtain valid data in a format recognizable by the terminal and / or server and after removing noise. The valid data stored in the SDK includes: driving behavior collection data, driving behavior analysis data, data obtained from the vehicle's own sensors, and data collected from the vehicle terminal.

[0067] Output valid data to the server or terminal for processing to achieve driving evaluation.

[0068] Preferably, when the server loads the SDK, the process of processing the raw data collected from the input SDK to obtain valid data in a format recognizable by the terminal and / or server, and after removing noise, includes:

[0069] The SDK acquires the raw data collected by the server from the terminal;

[0070] The SDK converts the raw collected data into data in a specified data format.

[0071] The SDK determines whether to perform noise removal processing on data of a specified data format;

[0072] If necessary, noise removal processing is performed on the data in the specified data format, and the effective data is obtained after noise removal and analysis.

[0073] If not required, the valid data is obtained by analyzing and processing the data in the specified data format.

[0074] Preferably, when the SDK is loaded on the terminal, the process of processing the raw data collected from the input SDK to obtain valid data in a format recognizable by the terminal and / or server, and after removing noise, includes:

[0075] Receive raw data collected from vehicles;

[0076] The raw collected data is preprocessed according to the required accuracy and frequency for data analysis.

[0077] The SDK converts the preprocessed raw collected data into data in a specified data format.

[0078] The SDK determines whether to perform noise removal processing on data of a specified data format;

[0079] If necessary, noise removal processing is performed on the data in the specified data format, and the effective data is obtained after noise removal and analysis.

[0080] If not required, the valid data is obtained by analyzing and processing the data in the specified data format.

[0081] Preferably, when the terminal loads the first SDK and the server loads the second SDK, the step of processing the raw data collected from the input SDK to obtain valid data in a format recognizable by the terminal and / or server and after removing noise includes:

[0082] The terminal loads the first SDK;

[0083] The server loads the second SDK, and there are multiple terminals. All the first SDK and the second SDK are combined to form the SDK.

[0084] The first SDK receives the raw data collected by the vehicle;

[0085] The raw collected data is preprocessed according to the required accuracy and frequency for data analysis.

[0086] The first SDK converts the preprocessed raw collected data into data in a specified data format;

[0087] The first SDK determines whether to perform noise removal processing on data of a specified data format;

[0088] If necessary, noise removal processing will be performed on data in the specified data format;

[0089] If not required, input the data in the specified data format into the second SDK;

[0090] The second SDK analyzes and processes data in a specified data format to obtain the valid data;

[0091] Outputting valid data to a server or terminal for processing to achieve driving evaluation includes:

[0092] The second SDK outputs valid data to the server or terminal for processing to achieve driving evaluation.

[0093] Preferably, the method for removing noise data includes a noise data correction step, specifically including:

[0094] The difference between the new raw data and the raw data within a specified historical period is calculated. If the difference is higher than a first preset value, the probability of the new raw data being dirty data is determined.

[0095] Calculate the DX of the raw data collected within a specified historical period. If the DX is higher than the second preset value, it is determined that the raw data collected within the specified historical period contains dirty data.

[0096] Based on the probability of the dirty data or by discarding or correcting the noisy data, the DX satisfies the preset conditions.

[0097] The method to correct noisy data is to assign EX (EX of other data in this time period excluding itself) to the noisy data.

[0098] EX = i p i

[0099] DX= i -EX) 2 * p i

[0100] Preferably, the method for processing noise-removing data includes an acceleration data filtering algorithm, which includes the following steps:

[0101] Acquire acceleration data from the accelerometer, and define it as A1;

[0102] First, filter the data A1 using noise data to obtain the filtered data A2;

[0103] A2 is then processed using a Kalman filter algorithm to obtain A3; the Kalman filter algorithm includes the following steps:

[0104] Prediction phase

[0105] x = (F*x) + (B*u)

[0106] P = (F*P*F T ) + Q

[0107] Correction phase

[0108] y = z – (H*x)

[0109] S = (H*P*H T ) + R

[0110] K = P*H T *S -1

[0111] x = x + (K*y)

[0112] P = (I – (K*H)) * P

[0113] Where y is the measurement margin, S is the measurement margin covariance matrix, x represents the system state; P represents the error covariance matrix, K represents the Kalman gain; Q and R are the optimal solutions selected based on the actual situation.

[0114] Preferably, the method for processing noise-removed data includes a second filtering algorithm, which includes the following steps:

[0115] Acquire data from various sensors and define it as A1;

[0116] First, filter the data A1 using noise data to obtain the filtered data A2;

[0117] A2 is then used to obtain A3 using a Kalman filter algorithm; the second filtering algorithm includes the following steps:

[0118] Optimal value calculation stage

[0119] The prediction value of the previous time T0 for the current time T1 is The bias of Gaussian noise is Gs1, and the bias of A2 is Gs2. Calculate the mean square error H over Gs1 and Gs2. The optimal value at the current time T1 is A3 = +H *(A2 - )

[0120] The optimal value at the previous time step is denoted as The data obtained from A1 in the previous time step using noise data filtering. Then Gs1 = Gs2 = The optimal value at the current moment A1 is the data obtained by filtering noisy data. Then Gs3 = Gs4 = ;

[0121] Predicting the next time value stage

[0122] H The predicted value at the next moment .

[0123] Preferably, the method for processing noise-removing data includes a location information filtering algorithm, which includes the following steps:

[0124] The raw location data G1 obtained from the location information sensor is processed through a noise data processing step to obtain the first filtered location data G2; wherein, the raw location data G1 mainly includes: longitude, latitude, speed, heading angle, altitude, and accuracy factor;

[0125] The location data G2 is then subjected to Kalman filtering to obtain the second filtered location data G3.

[0126] Preferably, the method for processing noise-removing data includes: a velocity information filtering algorithm; the velocity information filtering algorithm includes:

[0127] The data S1 obtained from the speed sensor is processed by noise data to obtain S2;

[0128] The data S2 is then subjected to Kalman filtering to obtain data S3.

[0129] Preferably, the method for processing noise-removing data includes: an ignition / shutdown information filtering algorithm, wherein the ignition / shutdown information filtering algorithm includes: an ignition information filtering algorithm and a shut-down information filtering algorithm; wherein, the ignition information filtering algorithm mainly includes:

[0130] After determining the vehicle's ignition message, the algorithm of the SDK is combined with the data collected by the acceleration sensor, battery voltage data, and location information data to comprehensively determine whether the ignition is genuine.

[0131] After the SDK receives the ignition message, it starts to jump forward for a period of time. Let the voltage data at the specified time be U1. Then, it starts to take the voltage data for a period of time forward and waits for a period of time backward, which is U2.

[0132] Calculate EU1, EU2, DU1, and DU2, and calculate R1 = (EU2 > EU1) || (DU2 > DU1);

[0133] After the SDK receives the ignition message, it jumps forward a certain period of time from this point and takes the acceleration data for a certain period of time as A1. It then takes the acceleration value for a certain period of time from this point and waits for the acceleration value data A2 for a certain period of time.

[0134] Calculate EA1, EA2, DA1, and DA2, and give R2 = (EA2 > EA1) || (DA2 > DA1);

[0135] If R1*R2 > 0, then system ignition is confirmed;

[0136] If R1*R2 < 1 and R1 + R2 > 0, then the system is considered to be ignited. At this time, continue to jump forward from this point for a period of time and take the position data for a period of time as G1. Take the position data for a period of time from this point and wait for the position data for a period of time G2.

[0137] Calculate DG1 and DG2, and calculate R3 = DG2 > DG1. If R3 > 0, then confirm system ignition.

[0138] The engine shutdown information filtering algorithm includes:

[0139] After the third-party system detects the engine shutdown message, it combines the SDK algorithm with data collected by the accelerometer, battery voltage data, and location information to determine whether the engine shutdown is genuine.

[0140] After the SDK receives the shutdown message, it jumps backward for a period of time from this point and takes voltage data for a period of time as U1; it then takes voltage data for a period of time from this point and waits for a period of time before taking voltage data U2.

[0141] Calculate EU1, EU2, DU1, and DU2, and determine R1 = (EU2 < EU1) || (DU2 < DU1);

[0142] After the SDK receives the engine shutdown message, it jumps forward a certain period of time from this point and takes the acceleration data for a certain period of time as A1. It then takes the acceleration value for a certain period of time from this point and waits for the acceleration value data A2 for a certain period of time.

[0143] Calculate EA1, EA2, DA1, DA2, and calculate R2 = (EA2) / (DA1). <EA1) || (DA2 < DA1);

[0144] If R1*R2 > 0, then confirm that the system is shut down;

[0145] If R1*R2 < 1 and R1 + R2 > 0, then the system is considered to be potentially shut down. In this case, continue jumping forward from this point for a period of time and take position data for a period of time, which is recorded as G1; take position data forward from this point for a period of time and wait for position data for a period of time, which is recorded as G2.

[0146] Calculate DG1 and DG2, and calculate R3 = DG2 < DG1. If R3 > 0, then confirm that the system is shut down.

[0147] Example 1

[0148] Please see Figure 2 The method of loading data preprocessing SDK into the vehicle terminal is as follows: after combining the original data collected by the vehicle terminal, if there is still incomplete data or the data accuracy and collection frequency cannot meet the data processing function, the data preprocessing SDK package will supplement the missing data and combine the algorithm itself to adapt and modify the collection frequency and accuracy of one or more data items.

[0149] The SDK-based solution for terminal devices standardizes terminal data collection because the SDK preprocesses the underlying data according to the required accuracy and frequency for data analysis. The advantage of this approach is that regardless of the type of terminal, operating system, or platform, the same SDK is used, effectively reducing the workload associated with software integration and multi-party data exchange. Data standardization ensures data integrity. Crucially, it does not affect the functionality of existing terminal devices, and data exchange is relatively simple, requiring no significant code modification or debugging; simply connecting the data to the interface provided by the SDK is sufficient.

[0150] Using an SDK solution on terminal devices can maximize the protection of the integrity of the original functions of the terminal devices. In addition, it can effectively avoid the risk of leakage of data processing algorithms, and also ensure that the intellectual property rights of both parties are well protected.

[0151] Example 2

[0152] Please see Figure 3 The data server loads the data processing SDK. After analyzing the original data on the data server, if it finds that the data for the necessary items used in the driving evaluation is missing or that the accuracy or collection frequency does not meet the requirements, it will provide data collection standards and suggestions until the data provided by the data provider meets the algorithm requirements.

[0153] The data processing SDK is primarily used to process and analyze the data collected by the data server, and then load the processed data into the SDK's driving evaluation module. The driving evaluation results are then obtained after the data passes through this module.

[0154] Example 3

[0155] Please see Figure 4 If the data provider cannot load the data analysis SDK on its data server for various reasons, the second technical solution is the most suitable. However, since the data analysis SDK is not loaded on its data server, some of the data collected from the terminal SDK will inevitably be uploaded to the data analysis server, which may pose a risk of terminal data leakage. This concern does not exist if the terminal data is not sensitive. However, if the terminal data is highly sensitive, the data analysis SDK must be loaded on the data server. That is, both the data terminal and the data server must load the SDK. This significantly improves the effective utilization of data and reduces the development work for the data provider. This maximizes data security and integrity and provides a highly accurate data source for data analysis and driving evaluation modules, resulting in more accurate driving evaluation results.

[0156] The method of loading a data preprocessing SDK on the vehicle-mounted terminal and a data processing SDK on the data server effectively addresses issues such as non-standard data collection and discrepancies in accuracy and frequency. Simultaneously, the data server uses the data processing SDK to analyze and process the data. Because the data terminal loads the data preprocessing SDK, the data collected by the data server has already undergone preliminary processing and standardization, resulting in more accurate results from the driving evaluation function. Since both the terminal and the data server load the corresponding SDKs, the development work for the data provider becomes simpler, the effective utilization of data is improved, the transmission of invalid data and the meaningless operation of functional modules are reduced, and the execution efficiency of the driving evaluation module is improved.

[0157] The advantages of using an SDK solution are: Data providers' data servers or vehicle terminal software do not require complex or cumbersome development work; they only need to configure the corresponding SDK interfaces and ensure normal data collection to use the various functions provided by the SDK. For data providers, using an SDK solution ensures that data is not at risk of leakage and offers the advantages of simple development and strong functional scalability. For the company providing the SDK, using an SDK can maximize the protection of its core algorithms from illegal theft and its business logic from unauthorized use.

[0158] The beneficial effects of this invention compared to existing technologies are as follows: Since the software (or code) for data storage, processing, and analysis is implemented by different companies, data processing is essential to realize the value of data providers' data through functions such as mining, processing, analysis, and driving evaluation. In other words, only by organically combining data (materials) and data processing (methods) can their value be unlocked.

[0159] This invention provides the core algorithms and processing flow to data providers in the form of an SDK. This method effectively protects the algorithms for detailed processing and critical data processing, as well as the core business processes, while also safeguarding the data provider's data from leakage. This is because the data provider only needs to connect to the corresponding data interface, instead of having to perform extensive code integration or transmit data to the data analyst's server as before.

[0160] Implementation Method 2

[0161] Please see Figure 5 Corresponding to the method of Embodiment 1 above, the present invention also provides a device for data acquisition and driving evaluation based on an SDK, which mainly includes:

[0162] At least one processor 401; and,

[0163] Memory 402 communicatively connected to the at least one processor; wherein,

[0164] The memory 402 stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method described in Embodiment 1.

[0165] Please see details. Figure 5 The device for data acquisition and driving evaluation based on SDK according to the present invention includes a processor 401 and a memory 402 storing computer program instructions.

[0166] Specifically, the processor 401 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of the present invention.

[0167] Memory 402 may include mass storage for data or instructions. For example, and not limitingly, memory 402 may include a hard disk drive (HDD), a floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 402 may include removable or non-removable (or fixed) media. Where appropriate, memory 402 may be internal or external to a data processing device. In a particular embodiment, memory 402 is a non-volatile solid-state memory. In a particular embodiment, memory 402 includes read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0168] The processor 401 reads and executes computer program instructions stored in the memory 402 to implement any of the SDK-based data acquisition and driving evaluation methods in the above embodiments.

[0169] In one example, the device for data acquisition and driving evaluation based on the SDK may further include a communication interface 403 and a bus 410. For example, Figure 5 As shown, the processor 401, memory 402, and communication interface 403 are connected through bus 410 and complete communication with each other.

[0170] The communication interface 403 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of the present invention.

[0171] Bus 410 includes hardware, software, or both, that couples together components of a device for data acquisition and driving evaluation based on an SDK. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 410 may include one or more buses. While specific buses are described and illustrated in embodiments of the invention, the invention contemplates any suitable bus or interconnect.

[0172] Furthermore, in conjunction with the SDK-based data acquisition and driving evaluation methods described in the above embodiments, this invention can be implemented using a computer-readable storage medium. This computer-readable storage medium stores computer program instructions; when executed by a processor, these computer program instructions implement any of the SDK-based data acquisition and driving evaluation methods described in the above embodiments.

[0173] For a detailed description of the device, please refer to Embodiment 1, which will not be repeated here.

[0174] Implementation Method 3

[0175] A system for data acquisition and driving evaluation based on an SDK is characterized in that the system further includes: a server, a terminal communicatively connected to the server, and an SDK, wherein the SDK is loaded onto the server and / or the terminal to implement the method described in Embodiment 1. For a detailed description of this system, please refer to Embodiment 1, which will not be repeated here.

[0176] The above is a detailed description of a method, apparatus, and system for data acquisition and driving evaluation based on an SDK provided by the present invention. It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, or article that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, or article. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or printer-based vehicle collision avoidance system that includes that element.

[0177] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0178] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for executing the methods described in the various embodiments of the present invention.

[0179] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A vehicle driving evaluation method based on SDK and multi-dimensional data, characterized in that, The method includes: The first SDK for receiving raw collected data sent by the vehicle is loaded on the terminal, and the second SDK is loaded on the server. Input the data in the specified data format after noise removal processing in the first SDK into the second SDK; After analyzing and processing the data in the specified data format using the second SDK, valid data is generated. Driving assessments will be conducted based on the valid data. The step of inputting data in a specified data format after noise removal processing from the first SDK into the second SDK includes: The raw collected data is preprocessed according to the required accuracy and frequency for data analysis. According to the specified data format, the preprocessed raw collected data in the first SDK is converted into data in the specified data format; After the data is converted to the specified data format, it undergoes noise reduction processing and is then input into the second SDK. The method for processing noise-removing data converted to the specified data format includes a filtering algorithm for ignition information authenticity, wherein the filtering algorithm for ignition information authenticity includes: The ignition timing for obtaining ignition information; Obtain voltage data U1 and acceleration data A1 corresponding to the first time period before the ignition moment, and voltage data U2 and acceleration data A2 corresponding to the second time period after the ignition moment; Based on the expected value and variance value corresponding to the voltage data U1, the voltage data U2, the acceleration data A1, and the acceleration data A2, the actual moment corresponding to the system ignition is determined; If the initial determination result is that the system may ignite, then determining the actual moment corresponding to system ignition based on the expected value and variance value corresponding to the voltage data U1, voltage data U2, acceleration data A1, and acceleration data A2 includes operation B1: Obtain the start time of the first time segment corresponding to the last sampling; Based on the start time, continue to acquire the location data G1 corresponding to the first time period before the start time and the location data G2 corresponding to the second time period after the start time; Calculate the variance DG1 of the location data G1 and the variance DG2 of the location data G2; Based on the variance DG1 corresponding to the position data G1 and the variance DG2 corresponding to the position data G2, determine whether the system should ignite. Repeat step B1 above until system ignition is confirmed; Specifically, the system ignition is determined by whether the variance DG2 is greater than the variance DG1; DG1 is the variance corresponding to the position data G1, and DG2 is the variance corresponding to the position data G2. The method for processing data converted to the specified data format by removing noise includes a filtering algorithm for verifying the authenticity of the engine shutdown information. The filtering algorithm for verifying the authenticity of the engine shutdown information includes: The time of engine shutdown is when the shutdown message is received; Obtain voltage data U3 and acceleration data A3 corresponding to the first time period before the engine shutdown time, and voltage data U4 and acceleration data A4 corresponding to the second time period after the engine shutdown time; Based on the expected value and variance value corresponding to the voltage data U3, the voltage data U4, the acceleration data A3, and the acceleration data A4, determine the actual moment corresponding to the system shutdown. If the initial determination result is that the system may shut down, then the step of determining the actual moment corresponding to the system shutdown based on the expected value and variance value corresponding to the voltage data U3, voltage data U4, acceleration data A3, and acceleration data A4 includes operation B2: Obtain the start time of the first time segment corresponding to the last sampling; Based on the start time, continue to acquire the location data G3 corresponding to the first time period before the start time and the location data G4 corresponding to the second time period after the start time; Calculate the variance DG3 of the location data G3 and the variance DG4 of the location data G4; Based on the variance corresponding to the position data G3 and the variance corresponding to the position data G4, determine whether the system is shut down; Repeat step B2 above until you confirm that the system is off; Specifically, the system is determined to be shut down by judging whether the variance DG4 is less than the variance DG3; DG3 is the variance corresponding to the position data G3, and DG4 is the variance corresponding to the position data G4.

2. The vehicle driving evaluation method based on SDK and multi-dimensional data according to claim 1, characterized in that, The valid data includes driving behavior collection data, driving behavior analysis data, data obtained from the vehicle's own sensors, and data collected from the vehicle's onboard terminal.

3. The vehicle driving evaluation method based on SDK and multi-dimensional data according to claim 1, characterized in that, The data processing method for noise removal of data converted to the specified data format includes a vehicle location information filtering algorithm, wherein the vehicle location information filtering algorithm includes: Obtain position data by performing a first-stage noise data processing on the raw position data collected by the sensor; The location data that has undergone the first filtering of noise data is then subjected to Kalman filtering again to obtain the second filtered location data, which is used as the effective data after noise removal. The original location data includes longitude, latitude, speed, heading angle, altitude, and accuracy factor.

4. A vehicle driving evaluation device based on SDK and multi-dimensional data, characterized in that, The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 3.

5. A vehicle driving evaluation system based on SDK and multi-dimensional data, characterized in that, The system further includes: a server, a terminal connected in communication with the server, and an SDK, wherein the SDK is loaded onto the server and / or the terminal to implement the method according to any one of claims 1 to 3.

Images