Driving assistance systems

Abstract

The present invention provides a driver assistance system that allows a vehicle to continue operating after a malfunction has occurred. [Solution] The driving assistance device includes an acquisition unit that acquires data relating to the driving state of a vehicle while it is in motion, a detection unit that detects abnormal conditions affecting the driving of the vehicle from the data, and a presentation unit that uses artificial intelligence to identify and analyze parameters that have a correlation with the abnormal conditions from the data, and presents to the driver of the vehicle a driving method that avoids the abnormal conditions and allows the vehicle to continue driving.

Description

Technical Field

[0007] , ,

[0001] The present invention relates to a driving support device.

Background Art

[0002] For example, Patent Document 1 discloses notifying a user of advice information indicating a coping method according to the degree of abnormality of a vehicle.

Prior Art Document

Patent Document

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Although driving can sometimes be continued depending on the coping method for vehicle abnormalities, Patent Document 1 does not disclose advice information for continuing driving.

[0005] Therefore, the present invention has been made in view of the above problems, and an object thereof is to provide a driving support device that can continue driving of a vehicle after an abnormality occurs in the vehicle.

Means for Solving the Problems

[0006] The driving support device of the present invention includes an acquisition unit that acquires data related to the driving state of a vehicle from the vehicle during driving, a detection unit that detects an abnormal state affecting the driving of the vehicle from the data, and an artificial intelligence that identifies and analyzes parameters having a correlation relationship with the abnormal state from the data, and a presentation unit that presents a driving method for avoiding the abnormal state and continuing the driving of the vehicle to a driver of the vehicle.

Effects of the Invention

[0007] According to the present invention, the vehicle can continue to run even after an abnormality occurs in the vehicle. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a diagram showing an example of a vehicle system. [Figure 2] Figure 2(A) shows the distribution of oil pressure with respect to engine speed. Figure 2(B) shows the vehicle's position P on the circuit map. Figure 2(C) shows the distribution of engine oil pressure with respect to cornering acceleration. Figure 2(D) shows the changes in engine speed, cornering acceleration, oil temperature, and oil pressure over time. [Figure 3] Figure 3 is a flowchart showing an example of how the generation AI function works. [Figure 4] Figure 4 is a flowchart showing an example of processing by an ECU (Electronic Control Unit). [Modes for carrying out the invention]

[0009] (Vehicle system configuration) Figure 1 is a configuration diagram showing an example of a vehicle system S. The vehicle system S is mounted on a vehicle and includes an ECU (Electronic Control Unit) 1, a sensor system 20, a GPS (Global Positioning System) 21, a display 22, a speaker 23, and a wireless communication device 24.

[0010] The sensor system 20 includes multiple sensors (not shown) that detect various parameters related to the vehicle's driving state. These sensors include, for example, a crank angle sensor that detects the rotational speed of the engine (not shown), an acceleration sensor that detects the vehicle's turning acceleration (hereinafter referred to as turning G), an oil pressure sensor that detects the oil pressure of the lubricating oil, and an oil temperature sensor that detects the temperature of the lubricating oil. The sensor system 20 outputs the above-mentioned parameters as driving data to the ECU 1.

[0011] GPS22 measures the vehicle's current position (e.g., latitude and longitude). GPS22 outputs the current position as position information to ECU1.

[0012] The display 22 and speaker 23 are mounted in the driver's seat of the vehicle. The display 22 displays images containing various information to the vehicle's occupants. The speaker 23 outputs audio containing various information to the vehicle's occupants.

[0013] The wireless communication device 24 communicates with the cloud server 3 via a communication network NW such as the Internet. The cloud server 3 is equipped with a trained generative AI (Artificial Intelligence) function 30 that takes vehicle driving data as input and outputs a driving method that can avoid vehicle malfunctions. The generative AI function 30 is formed by software installed on the cloud server 3. The wireless communication device 24 transmits the driving data input from the ECU 1 to the cloud server 3 and receives information about the driving method generated by the generative AI function 30 (hereinafter referred to as driving information) from the cloud server 3 and outputs it to the ECU 1.

[0014] ECU1 is an example of a driver assistance system. ECU1 includes a CPU (Central Processing Unit), ROM (Read Only Memory), and RAM (Random Access Memory), and operates the CPU according to the program stored in the ROM.

[0015] When the ECU1 executes a program, it generates an acquisition unit 10, a detection unit 11, and a presentation unit 12 as software functions. The ECU1 also includes a communication interface 13 that processes communication between the acquisition unit 10, the detection unit 11, and the presentation unit 12 and the sensor system 20, GPS 21, display 22, speaker 23, and wireless communication device 24. The communication interface 13 is implemented by hardware and software. The acquisition unit 10, the detection unit 11, and the presentation unit 12 may also be implemented by hardware.

[0016] The acquisition unit 10 acquires driving data from a vehicle in motion. For example, the acquisition unit 10 acquires driving data from the sensor system 20.

[0017] Figures 2(A) to (D) are diagrams showing examples of driving data. Figure 2(A) shows the distribution of hydraulic pressure with respect to the engine speed. The reference line L indicates the boundary between a normal driving state and an abnormal driving state. Hydraulic pressure values higher than the reference line L are determined to be normal, and hydraulic pressure values lower than the reference line L are determined to be abnormal. The hydraulic pressure values are acquired from a hydraulic pressure sensor.

[0018] Figure 2(B) shows the position P of the vehicle on the circuit map. In this example, it is assumed that the vehicle travels on a circuit having the first to fourth corners, and the position P on the circuit course is detected by the GPS 22. Note that the data of the circuit map is acquired from another server (not shown) via the communication network NW by, for example, the wireless communication device 24.

[0019] Figure 2(C) shows the distribution of the engine's hydraulic pressure with respect to the turn G. The turn G is acquired from, for example, an acceleration sensor.

[0020] Figure 2(D) shows the changes in the engine speed, turn G, oil temperature, and hydraulic pressure with respect to time. The oil temperature is acquired from an oil temperature sensor.

[0021] The acquisition unit 10 acquires driving data such as those listed in the above example from the sensor system 20. Note that the driving data is an example of data related to the driving state of the vehicle.

[0022] After the acquisition process by the acquisition unit 10, the detection unit 11 detects an abnormal state that affects the running of the vehicle from the running data. For example, when the hydraulic pressure value below the reference line L described in reference to Fig. 2(A) is a predetermined number or more, the detection unit 11 detects an abnormal state of the vehicle. At this time, the detection unit 11 detects the position P where the abnormality of the vehicle has occurred. The generated position P is, for example, near the third corner of the circuit course. As a cause of the abnormality, it can be cited that the oil inside the engine is biased to one direction side due to a high turning G, and air is mixed into the oil pump and the hydraulic pressure decreases. In addition, the detection unit 11 also detects an abnormal state that does not affect the running of the vehicle. The detection unit 11 discriminates, for example, a case where the position of the vehicle deviates from the circuit course due to a driver's steering error as an abnormal state that does not affect the running of the vehicle.

[0023] The presentation unit 12 identifies and analyzes parameters having a correlation relationship regarding the abnormal state of the vehicle from the running data using the generation AI function 30, and presents a driving method for avoiding the abnormal state of the vehicle and continuing the running of the vehicle to the passengers of the vehicle. Here, the generation AI function 30 is an example of artificial intelligence. The presentation unit 12 transmits an inquiry about the above driving method to the cloud server 3 via the wireless communication device 24 together with the running data. The image information of FIGS. 2(A) to 2(D) described above is input to the generation AI function 30 of the cloud server 3. That is, the generation AI function 30 acquires and analyzes the state quantity regarding the parameter that detected the abnormality, the position P on the circuit course where the abnormality occurred, and the time-series data before the abnormality occurred. As will be described later, the generation AI function 30 uses a learned model to identify the correlation relationship between various parameters related to the abnormal state from the running data and determines a driving method.

[0024] Examples of the generation AI function 30 include, for example, "Chat GPT" (registered trademark). In this case, the presentation unit 12 transmits, for example, the following inquiry. Note that an image of the running data shown in FIGS. 2(A) to 2(D) is attached to the inquiry.

[0025] <Example of inquiry> The following (see Figures 2(A) to 2(D)) shows the driving data when a malfunction occurred in the vehicle during circuit driving. Based on this data What kind of abnormality is occurring? Please suggest countermeasures that the driver should take to avoid malfunctions. Prerequisites When implementing countermeasures, parts replacement or repairs are not possible. Please propose methods to avoid the malfunction solely through the driver's operation.

[0026] In response to this inquiry, the cloud server 3 sends a response message generated by the generation AI function 30 back to the presentation unit 12.

[0027] <Example Answer> From this data, we can see the following: 1. What kind of abnormality is occurring? - The abnormality is related to a drop in hydraulic pressure. -A tendency for oil pressure to decrease is observed when the engine speed is high (see Figure 2(A)). -A tendency for hydraulic pressure to decrease is also observed when high turning G-forces are applied (see Figure 2(C)). -Looking at the circuit map, we can see that a drop in hydraulic pressure is occurring at the third corner. 2. Measures drivers can take to avoid malfunctions: - Engine speed management: Oil pressure tends to drop especially at high engine speeds, so try to drive in a way that keeps engine speed down. Specifically, shift up earlier or keep the engine speed from getting too high. - Controlling cornering G-force: When high cornering G-forces are applied, hydraulic pressure decreases, so reduce cornering speed. Also, strive for smooth steering and avoid sudden turns. -Measures for Turn 3: Particular caution is needed at Turn 3. Slow down before entering this corner and keep the engine speed low while turning to prevent a drop in hydraulic pressure. By implementing these measures, drivers can avoid malfunctions without having to replace parts or perform repairs.

[0028] The display unit 12 converts the response received from the cloud server 3 into audio data. The display unit 12 outputs the audio data to the speaker 23. The vehicle driver can avoid a drop in hydraulic pressure at the third corner and continue driving the vehicle by following the audio from the speaker 23 and reducing the cornering speed at the third corner or lowering the engine speed. Alternatively, the display unit 12 may display the response on the display 22, either in addition to or along with the audio output.

[0029] <Generation AI function> Figure 3 is a flowchart showing an example of the operation of the generation AI function 30. First, the generation AI function 30 receives an inquiry from the presentation unit 12 (St20). Next, the generation AI function 30 recognizes the images Figures 2(A) to 2(D) attached to the inquiry (St21).

[0030] Next, the generation AI function 30 extracts driving data parameters from each image (St22). For example, the generation AI function 30 extracts engine speed and oil pressure from Figure 2(A).

[0031] Next, the generation AI function 30 recognizes the correlation between parameters (St23). For example, based on Figure 2(A), the generation AI function 30 recognizes that hydraulic pressure increases as rotational speed increases, and based on Figure 2(C), it recognizes that hydraulic pressure decreases as turning G increases. Furthermore, based on Figure 2(B), the generation AI function 30 recognizes that the location where the hydraulic pressure drop occurs is position P on the circuit course.

[0032] Next, the generation AI function 30 identifies abnormal parameters from the correlations described above, for example, based on an anomaly detection model (St24). The anomaly detection model takes the correlations between parameters as input and outputs abnormal parameters. The generation AI function 30 identifies hydraulic pressure as an abnormal parameter. In this way, the generation AI function 30 identifies parameters that have a correlation (rotational speed and hydraulic pressure, turning G and hydraulic pressure) with respect to the abnormal state detected by the detection unit 11.

[0033] Next, the generation AI function 30 determines a driving operation for suppressing the occurrence of the abnormal parameter based on, for example, a driving operation determination model (St25). The correspondence determination model takes the above correlation, abnormal parameter, and the position P of the occurrence of the abnormality as inputs and outputs a driving operation for suppressing the abnormality. When the turning G and the rotational speed of the engine are high at the third corner, the generation AI function 30 determines a shift operation, a steering operation, and deceleration for reducing the turning G and the rotational speed so as to eliminate the decrease in hydraulic pressure.

[0034] Next, the generation AI function 30 generates a response sentence according to the above determination result (St26). In this way, the generation AI function 30 operates.

[0035] In this way, the ECU 1 analyzes the driving data with the generation AI function 30 and presents a driving method for avoiding abnormalities to the driver of the vehicle. Note that the generation AI function 30 is not limited to the cloud server 3 and may be provided in the ECU 1. Also, instead of the generation AI function 30, other AI may be used. In this case, the AI transmits digital information indicating a driving operation for avoiding an abnormal state to the presentation unit 12 instead of a response sentence.

[0036] <ECU Processing> FIG. 4 is a flowchart showing an example of the processing of the ECU 1. This processing is repeatedly executed, for example, at a fixed cycle.

[0037] First, the acquisition unit 10 determines whether the vehicle is running (St1). At this time, the acquisition unit 10 detects the running state from, for example, the gear position of the vehicle. If the vehicle is not running (No in St1), this processing ends.

[0038] If the vehicle is running (Yes in St1), the acquisition unit 10 acquires driving data from the sensor system 20 (St2). The acquisition process of the driving data may be executed for a predetermined period or only while traveling a predetermined distance.

[0039] Next, the detection unit 11 determines from the driving data whether there are any abnormalities that affect driving (St3). If there are no abnormalities that affect driving (St3 No.), the detection unit 11 determines whether there are any other abnormalities that do not affect driving (St9). If there are no abnormalities that do not affect driving (St9 No.), this process ends.

[0040] Furthermore, if there is an abnormality that does not affect driving (Yes in St9), the detection unit 11 selects a warning message corresponding to the abnormality (St10) and displays it on the display 22 (St11).

[0041] Furthermore, if there is an abnormality affecting driving (Yes in St3), the display unit 12 sends the driving data to the cloud server 3, causing the generating AI function 30 to analyze the driving data (St4).

[0042] Next, the display unit 12 determines whether there is an operating method to avoid the abnormality based on the response message returned from the cloud server 3 (St5). If there is no operating method (No. in St5), the display unit 12 outputs an audio message to the speaker 23 instructing the driver to stop operation (St8).

[0043] Furthermore, if there is a driving method (Yes in St5), the presentation unit 12 generates audio data from the response text of the generation AI function 30 (St6). Next, the presentation unit 12 outputs the audio data to the speaker 23 (St7). In this way, the presentation unit 12 presents the driver of the vehicle with a driving method that avoids vehicle malfunctions and allows the vehicle to continue driving. The ECU 1 performs processing in this manner.

[0044] In this way, the ECU1 analyzes the vehicle's driving data using the generated AI function 30 and presents the driver with a driving method that avoids abnormalities affecting the vehicle's operation and allows it to continue driving. Therefore, the ECU1 can allow the vehicle to continue driving even after an abnormality occurs in the vehicle.

[0045] Although this example describes a vehicle driving on a circuit course, the above driving assistance methods can also be applied when a vehicle is driving on public roads.

[0046] The embodiments described above are preferred examples of the present invention. However, the invention is not limited thereto, and various modifications are possible without departing from the spirit of the invention. [Explanation of Symbols]

[0047] 1 ECU (Driver's Control Unit), 10 Acquisition Unit, 11 Detection Unit, 12 Presentation Unit, 3 Cloud Server, 30 Generation AI Function (Artificial Intelligence)

Claims

[Claim 1] An acquisition unit that acquires data regarding the driving status of a vehicle while it is in motion, A detection unit that detects abnormal conditions affecting the vehicle's operation from the aforementioned data, The system includes a presentation unit that uses artificial intelligence to identify and analyze parameters that have a correlation with the abnormal state from the data, and then presents to the driver of the vehicle a driving method that avoids the abnormal state and allows the vehicle to continue running. Driving assistance system.

Images