Vehicle collision identification method and device

Abstract

A vehicle collision recognition method and device are disclosed. The vehicle collision recognition method includes: receiving vehicle data from a vehicle, the vehicle data including accelerator pedal position data, vehicle speed data, driver demand torque data, low-voltage battery voltage data, and battery pack relay status data; determining a time point at which a collision may occur based on the battery pack relay status data; obtaining vehicle data within a predetermined time period including the determined time point, and determining whether a vehicle collision occurs based on the obtained vehicle data by using a deep learning model, wherein, in the case of determining that a collision occurs, the time point at which a collision may occur is determined as a collision time point.

Description

Technical Field

[0001] The present disclosure relates to a vehicle collision identification method and device. Background Art

[0002] The current vehicle collision recognition method is to install an on-board central processing unit and a sensor module in the vehicle, collect sensor data through the sensor module and use the on-board central processing unit to perform vehicle collision recognition processing on the collected sensor data. More specifically, a method such as a fixed threshold method is used in the on-board central processing unit to perform collision recognition processing. Summary of the invention

[0003] According to a first aspect of the present disclosure, a vehicle collision identification method is provided, the method comprising: receiving vehicle data from a vehicle, the vehicle data comprising accelerator pedal position data, vehicle speed data, driver demand torque data, low-voltage battery voltage data and battery pack relay status data; determining a time point at which a collision may occur based on the battery pack relay status data; obtaining vehicle data within a predetermined time period including the determined time point, and determining whether the vehicle collides by using a deep learning model based on the obtained vehicle data, wherein, in the case where a collision is determined to occur, the time point at which the collision may occur is determined as the collision time point.

[0004] According to a second aspect of the present disclosure, a vehicle collision recognition device is provided, the device comprising: a component for executing the vehicle collision recognition method according to the first aspect.

[0005] According to a third aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, which stores instructions, and when the instructions are executed by a processor, the vehicle collision recognition method according to the first aspect is executed.

[0006] According to a fourth aspect of the present disclosure, a computer program product is provided, which stores instructions, and when the instructions are executed by a processor, the vehicle collision recognition method according to the first aspect is executed. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure without limitation. In the various figures, like reference numerals are used to represent like items.

[0008] Figure 1 is a block diagram of an exemplary vehicle collision recognition device according to some embodiments of the present disclosure.

[0009] Figure 2 is a flow chart illustrating an exemplary vehicle collision identification method according to some embodiments of the present disclosure.

[0010] Figure 3 is a flowchart illustrating an exemplary process of determining a potential collision time point according to some embodiments of the present disclosure.

[0011] Figure 4 1 is a diagram illustrating a general hardware environment in which the present disclosure may be applied according to some embodiments of the present disclosure. DETAILED DESCRIPTION

[0012] In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the described exemplary embodiments. However, it is clear to those skilled in the art that the described embodiments may be practiced without some or all of these specific details. In the described exemplary embodiments, well-known structures or processing steps are not described in detail in order to avoid unnecessarily obscuring the concepts of the present disclosure.

[0013] The blocks in each block diagram shown below can be implemented by hardware, software, firmware or any combination thereof to implement the principles of the present disclosure. It should be understood by those skilled in the art that the blocks described in each block diagram can be combined or divided into sub-blocks to implement the principles of the present disclosure.

[0014] The steps of the methods presented in this disclosure are intended to be illustrative. In some embodiments, the method can be completed with one or more additional steps not described and / or completed without one or more steps discussed.

[0015] In addition, in the description of the present disclosure, the terms "first", "second", "third", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance and order. Similarly, although operations are depicted in a particular order in the accompanying drawings, this should not be understood as requiring that such operations be performed in the particular order shown or in a sequential order, or requiring that all illustrated operations be performed to achieve the desired results. In some cases, multitasking and parallel processing can be advantageous.

[0016] Figure 1 is a block diagram of an exemplary vehicle collision recognition device according to some embodiments of the present disclosure.

[0017] like Figure 1As shown, the device 100 may include: a receiving component 110, configured to receive vehicle data from the vehicle, the vehicle data including accelerator pedal position data, vehicle speed data, driver demand torque data, low-voltage battery voltage data and battery pack relay status data; a potential collision time point determination component 120, configured to determine a time point at which a collision may occur based on the battery pack relay status data; a deep learning recognition component 130, configured to obtain vehicle data within a predetermined time period including the determined time point, and based on the obtained vehicle data, determine whether the vehicle collides by using a deep learning model, wherein, in the case of determining that a collision occurs, the time point at which the collision may occur is determined as the collision time point; a collision recognition result providing component 140, configured to provide a collision recognition result including information indicating that a collision has occurred and a collision time point to a vehicle after-sales service provider.

[0018] The following will further describe in detail Figure 1 The operation of each component is shown.

[0019] The method 200 starts from step S210, where the receiving component 110 receives vehicle data from the vehicle, the vehicle data including accelerator pedal position data, vehicle speed data, driver demand torque data, low voltage battery voltage data and battery pack relay status data. In the present disclosure, the vehicle refers to an electric vehicle.

[0020] At least a portion of the vehicle data, such as accelerator pedal position data, vehicle speed data, low-voltage battery voltage data, and battery pack relay status data, may be acquired using onboard sensors, and the driver demand torque data may be acquired by calculation.

[0021] More specifically, accelerator pedal position data can be obtained through sensors such as an accelerator pedal position sensor, and the accelerator pedal position data indicates the position of the accelerator pedal; vehicle speed data can be obtained through sensors such as a speed sensor, and the vehicle speed data indicates the speed of the vehicle; low-voltage battery voltage data can be obtained through sensors such as a voltage sensor, wherein low-voltage batteries are generally used in low-power systems in vehicles, such as door control systems, vehicle starting systems, lighting systems, instrument display systems, etc.; battery pack relay status data can be obtained through a relay status sensor, and the battery pack relay status data indicates the disconnected or connected relay status of the main positive relay or the main negative relay in the battery pack; driver demand torque data can be calculated through, for example, accelerator pedal position data, vehicle speed data, etc., and the driver demand torque data indicates the torque demanded by the driver, and the torque demanded by different drivers under, for example, the same accelerator pedal position may be different.

[0022] In some embodiments, the time interval for collecting vehicle data is 2 seconds. In other embodiments, the time interval for collecting vehicle data is 5 seconds. The time interval for collecting vehicle data is not limited thereto, and the time interval for collecting vehicle data may be changed according to factors such as vehicle sensor performance and vehicle data transmission speed.

[0023] In some embodiments, the receiving component 110 may receive vehicle data from the vehicle via, for example, cellular vehicle-to-everything (LTE-V2X). Figure 1 The device shown may be implemented by a server in the cloud. In other embodiments, the receiving component 110 may receive vehicle data from the vehicle via a data transmission line in the vehicle. In this case, Figure 1 The device shown can be implemented by a vehicle-mounted processor.

[0024] Next, the method 200 proceeds to step S220 , where the potential collision time point determination component 120 determines a time point at which a collision may occur based on the battery pack relay state data.

[0025] The following references Figure 3 The processing at step S220 will be described. Figure 3 The figure shows an exemplary flowchart of the process of determining a potential collision time point. At step S310, the potential collision time point determination component 120 determines a first dynamic representation in a first time period before a target time point and a second dynamic representation in a second time period after the target time point based on the battery pack relay state data, wherein the first time period indicates the time required to collect a predetermined number of data samples before the target time point, and the second time period indicates the time required to collect a predetermined number of data samples after the target time point; the first dynamic representation and the second dynamic representation indicate the dynamic change of the battery pack relay state, wherein the battery pack relay state data includes multiple data samples.

[0026] At step S320, a time point at which a collision may occur is determined based on the first dynamic representation and the second dynamic representation, the determination comprising determining whether the following preset conditions are satisfied, and determining the first time point at which the battery pack relay state is disconnected within the first time period as the time point at which a collision may occur if the following preset conditions are satisfied:

[0027] The data sample at the target time point indicates that the battery pack relay state is disconnected.

[0028] The first dynamic representation indicates that the data samples representing the disconnected battery pack relay state in the first time period account for more than 80% of the total data samples in the first time period, and

[0029] The second dynamic representation indicates that data samples representing the disconnected battery pack relay state in the second time period account for more than 60% of the total data samples in the second time period, and the vehicle speed at the target time point is less than a preset vehicle speed.

[0030] In a specific example, consider the following data sample of battery pack relay status data "1, -1, -1, -1, -1, -1, -1, -1, -1, ...", where "1" indicates that the battery pack relay state is on, and "-1" indicates that the battery pack relay state is off.

[0031] The first dynamic representation and the second dynamic representation in this specific example can be determined by the following expressions (1) and (2):

[0032]

[0033]

[0034] Among them, m i Represents the data sample value of the battery pack relay status data, i takes an integer value greater than zero; k represents a predetermined number and k = 5; OFF i Indicates the first dynamic display, ON i Represents a second dynamic representation.

[0035] When the target time point is the time point corresponding to the first data sample, since the predetermined number is 5, there is no predetermined number of data samples in the first time period but there is a predetermined number of data samples in the second time period, so it is impossible to obtain OFF i ON i Similarly, when the target time points are the time points corresponding to the second, third, and fourth data samples, respectively, OFF cannot be obtained in the first time period. i , while ON i Both are -5.

[0036] When the target time point is the time point corresponding to the fifth data sample, there are a predetermined number of data samples in the first time period and a predetermined number of data samples in the second time period, so it can be concluded that OFF i and ON i , at this time OFF i -3,ON i is -5.

[0037] By analogy, when the target time point is the time point corresponding to the 6th, 7th, 8th, and so on data samples, the situation is similar to the above situation and will not be elaborated here.

[0038] Based on the calculated OFFi and ON i , it can be determined whether the following conditions are met, and if the following conditions are met, the first time point when the battery pack relay state is disconnected within the first time period is determined as the time point when the collision may occur:

[0039] a) The data sample at the target time point indicates that the battery pack relay state is disconnected;

[0040] b)OFF i <-2;

[0041] c)ON i <0 and the actual vehicle speed is less than the preset vehicle speed, where the preset vehicle speed may be, for example, 20 m / s.

[0042] When the target time point is the time point corresponding to the 5th data sample, the above conditions a)-c) are met, so the time point corresponding to the second data sample, which is the first time point when the battery pack relay state is disconnected within the first time period, is determined as the time point when a collision may occur.

[0043] It should be understood that the first time period may include the target time point or be adjacent to the target time point. Similarly, the second time period may also include the target time point or be adjacent to the target time point.

[0044] It should be understood that in this example, when the time interval for collecting vehicle data is 2 seconds, the first time period and the second time period may be 10 seconds respectively; and when the time interval for collecting vehicle data is 5 seconds, the first time period and the second time period may be 25 seconds respectively.

[0045] It should be understood that although in this example, the battery pack relay state value is -1 when the battery pack relay state is disconnected and the battery pack relay state value is 1 when the battery pack relay state is connected, the present disclosure is not limited to this, and the battery pack relay state value can also be represented by other numerical values.

[0046] It should be understood that although the first dynamic representation and the second dynamic representation are obtained by summing up in expressions (1) and (2), they are not limited to this. For example, the first dynamic representation and the second dynamic representation can also be obtained by using calculation methods such as weighted summation and integration.

[0047] It should be understood that, although in this example, the first time period and the second time period are the same length, the present disclosure is not limited thereto, and the first time period and the second time period may be different lengths. Although the predetermined number is 5 in this example, it is not limited thereto, and the value of the predetermined number may also be an integer between 4 and 20.

[0048] It should be understood that the above condition c) can be used to exclude the situation when the vehicle is starting.

[0049] It should be understood that the time point at which a collision may occur may be determined in the vehicle's central processor, or the time point at which a collision may occur may be determined at a server in the cloud by receiving vehicle data via, for example, a cellular vehicle network.

[0050] Step S220 (i.e., step S310 and step S320) determines the time point at which a collision may occur based on the battery pack relay status data, and can filter out data related to the potential collision time point from massive input data, thereby reducing the computational load of subsequent machine learning processing.

[0051] Next, method 200 proceeds to step S230, where the deep learning identification component 130 obtains vehicle data within a predetermined time period including the determined time point, the vehicle data including accelerator pedal position data, vehicle speed data, driver demand torque data, low-voltage battery voltage data and battery pack relay status data, and based on the obtained vehicle data, determines whether the vehicle collides by using a deep learning model, wherein, in the event that a collision is determined, the time point at which the collision is likely to occur is determined as the collision time point.

[0052] Specifically, the input data of the deep learning model is vehicle data within a predetermined time period including the determined time point, and the vehicle data includes accelerator pedal position data, vehicle speed data, driver required torque data, low-voltage battery voltage data and battery pack relay status data; the output data of the deep learning model is the vehicle collision recognition result of whether the vehicle collides, and the collision time point determined when a collision is determined to have occurred.

[0053] The deep learning model may include a bidirectional long short-term memory (Bi-LSTM) network based on an attention mechanism, a convolutional neural network (CNN), a generative adversarial network (GAN), and the like.

[0054] In some embodiments, the predetermined time period may be a time period from 1 minute before the time point at which the collision may occur to 1 minute after the time point at which the collision may occur. However, the present disclosure is not limited thereto, and the predetermined time period may also be changed according to actual needs.

[0055] It should be understood that the above-mentioned deep learning model is pre-trained. The training of the deep learning model is known and will not be described in detail here.

[0056] It should be understood that vehicle data can be received via, for example, a cellular vehicle network to use a deep learning model at a server in the cloud to determine whether the vehicle has collided.

[0057] Compared with using methods such as fixed threshold method for vehicle collision recognition, using deep learning models for collision recognition can significantly improve the accuracy of vehicle collision recognition. By performing vehicle collision recognition processing in the cloud, not only high-accuracy vehicle collision recognition is achieved, but also the need to install a processor with high processing power on the vehicle can be eliminated.

[0058] Next, the method 200 proceeds to step S240, where the collision recognition result providing component 140 provides the collision recognition result including information indicating the occurrence of the collision and the collision time point to the vehicle after-sales service provider. The vehicle after-sales service may include maintenance, rescue, insurance and other services.

[0059] In some embodiments, the collision recognition result providing component 140 opens the collision recognition result including information indicating that a collision has occurred and the collision time point to the vehicle after-sales service provider in the form of a RESTful API.

[0060] Alternatively, the collision recognition result including information indicating that the collision has occurred and the collision time point may be opened to a vehicle after-sales service provider in a form such as a Simple Object Access Protocol Application Programming Interface (SOAP API) form.

[0061] Providing the collision recognition result including information indicating that a collision has occurred and the time point of the collision to the vehicle after-sales service provider enables the vehicle after-sales service provider to provide vehicle repair, rescue, insurance and other services in a timely manner, thereby improving the user experience of the vehicle driver.

[0062] Reference above Figure 1 , Figure 2 and Figure 3 , introduces an exemplary vehicle collision identification method and device according to the present disclosure. It can be understood that the method in the present disclosure can filter out data related to potential collision time points from a large amount of input data, thereby reducing the computational load of machine learning processing; by performing vehicle collision identification processing in the cloud, the need to install a processor with high processing power on the vehicle can be eliminated; and by sharing the vehicle collision identification results with the vehicle after-sales service provider in a timely manner after the vehicle collision occurs, vehicle maintenance, rescue, insurance and other services can be provided in a timely manner, thereby improving the user experience of the vehicle driver.

[0063] Hardware Implementation

[0064] Figure 4 A general hardware environment 400 is shown in which the present disclosure may be applied according to an exemplary embodiment of the present disclosure.

[0065] refer to Figure 4 , a computing device 400 will now be described as an example of a hardware device applicable to various aspects of the present disclosure. The computing device 400 may be any machine configured to perform processing and / or computing, and may be, but is not limited to, a workstation, a server, a desktop computer, a laptop computer, a tablet computer, a personal digital assistant, a smart phone, a portable camera, or any combination thereof. The above-mentioned apparatus 100 may be implemented in whole or at least in part by the computing device 400 or a similar device or system.

[0066] The computing device 400 may include an element that can be connected to or communicate with the bus 402 via one or more interfaces. For example, the computing device 400 may include a bus 402, one or more processors 404, one or more input devices 406, and one or more output devices 408. The one or more processors 404 may be any type of processor, and may include, but are not limited to, one or more general-purpose processors and / or one or more special-purpose processors (such as special-purpose processing chips). The input device 406 may be any type of device that can input information to the computing device, and may include, but are not limited to, a mouse, a keyboard, a touch screen, a microphone, and / or a remote controller. The output device 408 may be any type of device that can present information, and may include, but are not limited to, a display, a speaker, a video / audio output terminal, and / or a printer. The computing device 400 may also include or be connected to a non-transient storage device 410, which may be any storage device that is non-transient and can implement a data repository, and may include but is not limited to a disk drive, an optical storage device, a solid-state storage device, a floppy disk, a flexible disk, a hard disk, a magnetic tape or any other magnetic medium, a compact disk or any other optical medium, a ROM (read-only memory), a RAM (random access memory), a cache memory and / or any other memory chip or cartridge, and / or any other medium from which a computer can read data, instructions and / or code. The non-transient storage device 410 may be detachable from the interface. The non-transient storage device 410 may have data / instructions / code for implementing the above methods and steps. The computing device 400 may also include a communication device 412. The communication device 412 may be any type of device or system capable of communicating with an external device and / or with a network, and may include but is not limited to a modem, a network card, an infrared communication device, a wireless communication equipment and / or a communication device such as Bluetooth. TM Chipsets for devices, 802.11 devices, WiFi devices, WiMax devices, cellular communication facilities, etc.

[0067] The bus 402 may include, but is not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, an Enhanced ISA (EISA) bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0068] Computing device 400 may also include working memory 414, which may be any type of working memory that can store instructions and / or data useful for operation of processor 404, and may include, but is not limited to, random access memory and / or read-only memory devices.

[0069] The software elements may be located in the working memory 414, including but not limited to an operating system 416, one or more applications 418, a driver and / or other data and code. Instructions for executing the above methods and steps may be included in one or more applications 418, and the components of the above device 100 may be implemented by the processor 404 reading and executing the instructions of one or more applications 418. More specifically, the receiving component 110 may be implemented, for example, by the processor 404 when executing the application 418 having instructions to execute step S210. The potential collision time point determination component 120 may be implemented, for example, by the processor 404 when executing the application 418 having instructions to execute step S220 (or steps S310 and S320). The deep learning recognition component 130 may be implemented, for example, by the processor 404 when executing the application 418 having instructions to execute step S230. The collision recognition result providing component 140 may be implemented, for example, by the processor 404 when executing the application 418 having instructions to execute step S240. The executable code or source code of the instructions of the software elements may be stored in a non-transitory computer-readable storage medium (such as the storage device(s) 410 described above) and may be read into the working memory 414 where it may be compiled and / or installed. The executable code or source code of the instructions of the software elements may also be downloaded from a remote location.

[0070] From the above embodiments, it will be clear to those skilled in the art that the present disclosure can be implemented by software and necessary hardware, or can be implemented by hardware, firmware, etc. Based on this understanding, the embodiments of the present disclosure can be partially implemented in the form of software. Computer software can be stored in a computer-readable storage medium, such as a floppy disk, a hard disk, an optical disk, or a flash memory. Computer software includes a series of instructions that enable a computer (such as a personal computer, a service station, or a network terminal) to run a method according to each embodiment of the present disclosure or a part thereof.

[0071] The disclosure having thus been described, it will be apparent that the same may be varied in many ways. Such variations are not to be regarded as a departure from the spirit and scope of the disclosure, and all such modifications as would be obvious to one skilled in the art are intended to be included within the scope of the following claims.

Claims

1. A vehicle collision identification method, characterized in that: include: receiving vehicle data from the vehicle, the vehicle data including accelerator pedal position data, vehicle speed data, driver demand torque data, low-voltage battery voltage data, and battery pack relay status data; Determine the time point when a collision may occur based on the battery pack relay status data, including: Based on the battery pack relay state data, determining a first dynamic representation within a first time period before a target time point and a second dynamic representation within a second time period after the target time point, the first dynamic representation and the second dynamic representation indicating a dynamic change of the battery pack relay state, wherein the battery pack relay state data includes a plurality of data samples; Determining a time point at which a collision may occur based on the first dynamic representation and the second dynamic representation, the determination includes judging whether the following conditions are satisfied, and determining a first time point at which a battery pack relay state is disconnected within a first time period as a time point at which a collision may occur if the following conditions are satisfied: a data sample at a target time point indicates that the battery pack relay state is disconnected; the first dynamic representation indicates that data samples indicating a disconnected battery pack relay state within the first time period account for more than 80% of total data samples within the first time period; and the second dynamic representation indicates that data samples indicating a disconnected battery pack relay state within the second time period account for more than 60% of total data samples within the second time period, and a vehicle speed at the target time point is less than a preset vehicle speed; Vehicle data within a predetermined time period including the determined time point is obtained, and based on the obtained vehicle data, whether the vehicle collides is determined by using a deep learning model, wherein, if a collision is determined to have occurred, the time point at which the collision is likely to have occurred is determined as the collision time point.

2. The vehicle collision recognition method according to claim 1, wherein: The first time period has the same length as the second time period and corresponds to a predetermined number of data samples, where the predetermined number is an integer value between 4 and 20.

3. The vehicle collision recognition method according to claim 1, further comprising: The collision recognition result including information indicating that a collision has occurred and the time point of the collision is provided to a vehicle after-sales service provider.

4. The vehicle collision recognition method according to claim 1, wherein: The deep learning model includes a bidirectional long short-term memory network based on the attention mechanism.

5. The vehicle collision recognition method according to claim 1, wherein: At least a portion of the vehicle data is collected using onboard sensors.

6. The vehicle collision recognition method according to claim 1, wherein: Vehicle data is received via cellular vehicle networking.

7. A vehicle collision recognition device, characterized in that: include: Means for performing the method according to any one of claims 1-6.

8. A vehicle collision recognition device, characterized in that: include: at least one processor; as well as At least one storage device stores instructions, and when the instructions are executed by the at least one processor, the at least one processor performs the method according to any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium, characterized in that: Instructions are stored which, when executed by a processor, cause the method according to any one of claims 1 to 6 to be performed.

10. A computer program product, characterized in that Instructions are stored which, when executed by a processor, cause the method according to any one of claims 1 to 6 to be performed.

Images