Vehicle information storage system

Abstract

This invention provides a vehicle information storage system that improves the convenience of using vehicle information and also reduces the weight of the devices installed in the vehicles. [Solution] The system includes a server 6 that stores vehicle information and performs predetermined information processing before outputting it. The server 6 includes a neural network 7 having an input layer 7a, an intermediate layer 7b, and an output layer 7c. The input layer 7a is configured to acquire vehicle information transmitted from the vehicle's controller. The intermediate layer 7b is configured to perform at least weighting or frequency setting processing on the vehicle information acquired by the input layer 7a. The output layer 7c is configured to output the vehicle information processed by the intermediate layer 7b, divided according to predetermined criteria. Furthermore, an output request unit 9 is provided to change the weighting or frequency of the intermediate layer.

Description

Technical Field

[0001] This invention relates to a system for storing information related to a vehicle, such as the history of the vehicle's operation and the history of diagnosis.

Background Art

[0002] When a vehicle travels, it is subject to mechanical, thermal, and chemical effects, and accordingly, it may become fatigued or its durability may decrease. Also, inevitable abnormalities may occur. In order to ensure the stable running of the vehicle, the vehicle is equipped with a self-diagnosis function, and the results of the diagnosis are regularly or when an abnormality occurs, taken out externally and used for maintenance.

[0003] It is preferable that the vehicle information to be recorded or stored (hereinafter simply referred to as stored) is diverse and large in quantity. On the other hand, the information to be stored includes information that changes temporarily, such as the on / off of the engine and the opening / closing of the door, information that changes continuously, such as the steering angle, the amount of depression of the accelerator pedal and the brake pedal, the engine speed and temperature, and information that is temporary and at the same time highly important (or specific), such as the content and date / time when an event occurs. That is, the information to be stored is diverse and at the same time large in quantity. In contrast, the storage capacity is limited. Therefore, for example, in the device described in Patent Document 1, in order to avoid disappearance, specific processing is performed on predetermined information. That is, with a finite storage capacity, it is normal to overwrite old information with new information in time series, but in doing so, information with high importance or effectiveness may disappear. Therefore, in the device described in Patent Document 1, a process of converting a series of continuously generated information into one operation information is performed.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

[0005] As demonstrated in the device described in Patent Document 1, converting a series of consecutive pieces of information into a single piece of information reduces the likelihood of the series of consecutive pieces of information being overwritten and lost, as the large, continuous pieces of information become one small piece of information. Furthermore, the device described in Patent Document 1 makes it possible to increase the amount and types of information that can be stored. This would increase the effectiveness of the stored information and allow for a greater diversity of information needs. However, the device described in Patent Document 1 does not consider how the stored information can be used, as the stored information is output in the same state as it was stored. Therefore, even though storage has been improved, there is still room for improvement in how the information can be used. In other words, the stored information can include not only information about failures or abnormalities, but also diverse information such as changes in operating status over time, maximum and minimum values, and so on. This information can be used not only for failure analysis, but also for design changes and improvements, and for information on market needs. In that case, if the information is output with the settings set at the design stage of the storage device, the output may be biased towards failure analysis, lacking in information usability. On the other hand, it is conceivable to increase the storage capacity of information storage devices installed in vehicles and to equip the information storage device with functions that enable output according to needs. However, such a configuration would require a computing device to process a massive amount of information, and would also necessitate faster communication speeds in the vehicle, potentially increasing the cost not only of the information storage device but also of the vehicle itself.

[0006] This invention was made in view of the above-mentioned technical problems, and aims to provide a vehicle information storage system that can store a large amount of vehicle information, provides diversification in the output of the stored vehicle information, improves the convenience of using the vehicle information, and also reduces the weight of the device to be installed in the vehicle. [Means for solving the problem]

[0007] To achieve the above objective, this invention provides a vehicle information storage system comprising a controller mounted on a vehicle that transmits vehicle information obtained from a sensor, and a server that stores the vehicle information transmitted from the controller and processes and outputs predetermined information, wherein the server comprises a neural network having an input layer, an intermediate layer, and an output layer, the input layer is configured to acquire the vehicle information transmitted from the controller of the vehicle, the intermediate layer is configured to perform at least weighting or frequency setting processing on the vehicle information acquired by the input layer, the output layer is configured to divide and output the vehicle information processed by the intermediate layer according to predetermined criteria, and further, an output request unit is provided to change the weighting or frequency of the intermediate layer.

[0008] In this invention, the output request unit may be connected to the server via OTA.

[0009] In this invention, the output request unit may be provided in the controller. [Effects of the Invention]

[0010] In this invention, vehicle information obtained by sensors installed in the vehicle is sent to an external server for storage. Therefore, the storage capacity of the controller equipped with a storage function in each vehicle can be reduced. The server assigns weights to the input vehicle information according to its content, or sets the frequency of these assignments. The vehicle information processed with parameters such as weighting and frequency is classified according to predetermined criteria and output. Since the processing (calculations) of this information is performed by the server, there is no particular need to increase the computing power of the controller installed in the vehicle, and in this respect as well, the controller can be made smaller in capacity or less expensive. Furthermore, since an output request unit is provided to change parameters such as weighting and frequency, vehicle information can be output in a format that suits the needs, thereby improving the convenience of using the vehicle information. [Brief explanation of the drawing]

[0011] [Figure 1] This is a schematic diagram conceptually illustrating an example of the configuration of a vehicle information storage system according to this invention. [Figure 2] This is a schematic block diagram showing the control system for vehicle information in an embodiment of this invention. [Figure 3] This is a block diagram conceptually illustrating a neural network in an embodiment of this invention. [Figure 4] This figure shows an example of a list of vehicle information output from the controller. [Modes for carrying out the invention]

[0012] Next, embodiments of this invention will be described with reference to the attached drawings. It should be noted that the embodiments described below are merely examples of how to implement this invention and do not limit it.

[0013] Figure 1 conceptually shows the overall configuration of System 1 according to this invention. The vehicle 2 incorporated into System 1 is equipped with a controller (electronic control unit: ECU) 3 that detects the operating status of each part using sensors and transmits this as vehicle information. Figure 2 schematically shows the information control system in vehicle 2, and the sensors 4 are sensors mounted on engine vehicles and electric vehicles, such as a vehicle speed sensor, engine speed sensor, motor speed sensor, and current sensor. The controller 3 is mainly composed of a microcomputer and is configured to organize the signals input from the sensors 4 as vehicle information, temporarily store it, and transmit it to the outside by a predetermined communication means 5.

[0014] A server 6 is provided that receives vehicle information transmitted from vehicle 2, processes that vehicle information according to predetermined procedures, and stores it. Server 6 is mainly composed of a computer with high processing speed and storage capacity, and is installed in a predetermined facility such as a data center. For processing information, server 6 is equipped with a neural network 7 as an example. Figure 3 conceptually shows the neural network 7, which has an input layer 7a, a hidden layer 7b, and an output layer 7c, each consisting of multiple artificial neurons.

[0015] Vehicle information for each vehicle 2 is input to the input layer 7a. The vehicle information is propagated from the input layer 7a to the intermediate layer 7b, where processing such as weighting and frequency is performed by calculations according to a pre-prepared program. Here, for example, if the vehicle information is used for the purpose of diagnosing vehicle 2, the weighting may be the amount of data, and the frequency may be the recording period.

[0016] For data that is constantly generated and continuously changing, such as current values, CAN (Controller Area Network) received values, or circuit board temperature, the data volume is increased. The frequency is set low for data that changes slowly, such as temperature, and high for data that changes frequently and significantly, such as current. On the other hand, for data that maintains its state after a change, such as ignition switch (IG-SW) on / off, door opening / closing, and selected control mode, the data volume is reduced and the frequency is set to "medium" or "normal". Figure 4 shows examples of data names, data volume, and frequency for each diagnostic item. Furthermore, vehicle information related to each diagnostic item is predetermined, and in the intermediate layer 7b, the vehicle information is divided according to the related diagnostic item.

[0017] The output layer 7c outputs vehicle information propagated from the intermediate layer 7b. In this case, the vehicle information is divided and output according to predetermined criteria for each diagnostic item. Note that the classification of vehicle information according to the predetermined criteria may be performed in advance by the intermediate layer 7b. Furthermore, the output of vehicle information is performed to a predetermined terminal device 8, as shown in Figure 1. The terminal device 8 may be a monitor installed in the vehicle 2, or a computer or printer installed in a repair shop, design and development department, or dealership.

[0018] As mentioned earlier, vehicle information is both voluminous and diverse. Therefore, the neural network 7 described above uses machine learning to sort the vehicle information according to diagnostic items and requirements. For example, the current value of the motor as a driving force source becomes data for fault diagnosis and analysis, data for determining the vehicle's energy consumption and changes in the energy storage system over time, and may even be used as data for explaining consumption. In other words, it is included in these diverse requirements. However, each piece of vehicle information has varying degrees of importance or priority from the perspective of the requirements. That is, it is desirable that the output vehicle information be classified or arranged according to the importance or priority of the requirements.

[0019] In the system 1 according to the embodiment of this invention, in response to such a request, it is possible to change the weighting and frequency of vehicle information. For example, as schematically shown in FIG. 3, an output request unit 9 for changing parameters such as weighting and frequency is provided. The change of the parameter may be a so-called change of a coefficient, or may be a change of a program. Further, the output request unit 9 may be an operating device connected to the server 6, or may be a terminal device such as an ECU 3 of the vehicle 2 connected to the server by OTA (Over The Air).

[0020] A large amount and various vehicle information are input and stored in the server 6 described above from the controller 3 of the vehicle 2. The vehicle information is processed based on the weighting and frequency set by the output request unit 9 and is also classified for each required item. Since such processing of vehicle information is performed by the server 6, the capacity of a predetermined controller provided separately from the in-vehicle controller 3 or the server 6 can be reduced without impairing the convenience of using the vehicle information, and the cost can be reduced.

[0021] Items required for diagnosing the vehicle 2 and examples of vehicle information corresponding to those items are shown in FIG. 4. FIG. 4 shows examples of data related to the diagnosis of "current sensor failure" and data related to the diagnosis of "CAN communication abnormality", and those vehicle information are arranged in the order of frequency. The frequency is set by the output request unit 9, or the value set in the design is changed by the output request unit 9. Therefore, for example, regarding "current sensor failure", it is output in the order of "current sensor value", "IG-SW information", "substrate temperature sensor value", and others. Regarding "CAN communication abnormality", it is output in the order of "CAN reliability information", "control mode", "CAN communication value", and others. Thus, vehicle information with high importance or priority can be obtained according to the required items. This is the same for various required items such as the needs for evaluating the diagnostic function, the needs for evaluating the abnormalities and durability of vehicle parts, the needs for quality assurance, the needs for understanding how the vehicle is used, and the needs for information collection for MaaS (Mobility as a Service). By appropriately setting parameters such as weighting and frequency, vehicle information with high importance or priority can be obtained.

[0022] Since processing or calculation regarding such information is performed by the server 6, the calculation speed of a predetermined controller provided separately from the in-vehicle controller 3 or the server 6 does not particularly need to be high. As a result, the controller can be cost-reduced without impairing the convenience of using vehicle information.

[0023] Note that the present invention is not limited to the above-described embodiments, and can be appropriately modified and implemented without departing from the gist of the present invention. For example, although the detection or acquisition of vehicle information is performed by the vehicle, the use of vehicle information may be performed exclusively at locations other than the vehicle, such as factories, design and development departments, or dealers.

Explanation of Reference Numerals

[0024] 1 Vehicle information storage system 2 Vehicle 3 Controller 4 sensors 5. Means of communication 6 servers 7 Neural Networks 7a Input Layer 7b Middle layer 7c output layer 8 Terminal devices 9 Output request section

Claims

1. A vehicle information storage system comprising a controller mounted on the vehicle that transmits vehicle information obtained from sensors, and a server that stores the vehicle information transmitted from the controller and processes the information for a predetermined purpose before outputting it, The aforementioned server, It comprises a neural network having an input layer, a hidden layer, and an output layer. The input layer is configured to acquire the vehicle information transmitted from the vehicle's controller. The intermediate layer is configured to perform at least weighting or frequency setting processing on the vehicle information acquired by the input layer. The output layer is configured to output the vehicle information processed in the intermediate layer, divided according to predetermined criteria. Furthermore, an output request unit is provided to change the weighting or frequency of the intermediate layer. A vehicle information storage system characterized by the following features.

2. A vehicle information storage system according to claim 1, The output request unit is connected to the server via OTA. A vehicle information storage system characterized by the following features.

3. A vehicle information storage system according to claim 1 or 2, The output request unit is provided in the controller. A vehicle information storage system characterized by the following features.

Images