Driver assistance systems

Abstract

To provide a driver assistance system that can suppress misjudgments regarding dangerous driving. [Solution] The driver assistance system according to this disclosure is a driver assistance system that determines dangerous driving by a vehicle, and determines dangerous driving based on vehicle static information, which is predetermined information about the vehicle, and environmental information, which is information about the surrounding environment of the vehicle.

Description

Technical Field

[0001] The present disclosure relates to a driving support system.

Background Art

[0002] In recent years, the development of technologies for determining dangerous driving by vehicles has been underway. For example, Patent Document 1 discloses a technique for determining a dangerous location where a traffic accident is likely to occur based on vehicle dynamic information indicating an event related to the driving of the vehicle that has occurred in the vehicle. Further, in Patent Document 1, in addition to vehicle dynamic information, for each occurrence location of a dangerous event, a danger level indicating the possibility of occurrence of a traffic accident is determined based on location dynamic information, which is dynamic information regarding the occurrence location of the event, and location static information, which is static information. When the vehicle in motion approaches a dangerous location, the safe driving support device according to Patent Document 1 can provide accident prevention information according to the danger level.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the technique according to Patent Document 1, although vehicle dynamic information and location information are used, in order to suppress errors in danger level determination, it is necessary to make the information input to the safe driving support device more refined. Here, errors in determination include "false negatives", which are errors in which it is impossible to determine that a location is dangerous when it actually is, and "false positives", which are errors in which a location is determined to be dangerous when it actually is not. It is difficult for the safe driving support device according to Patent Document 1 to prevent the occurrence of such misjudgments.

[0005] This disclosure is made to solve these problems and aims to provide a driver assistance system that can suppress misjudgments regarding dangerous driving. [Means for solving the problem]

[0006] The driver assistance system relating to this disclosure is a driver assistance system that determines dangerous driving by a vehicle, and determines the dangerous driving based on vehicle static information, which is predetermined information about the vehicle, and environmental information, which is information about the surrounding environment of the vehicle. Since dangerous driving by a vehicle can be determined based on vehicle static information in addition to environmental information, it is possible to suppress misjudgments of dangerous driving by using the driver assistance system relating to this disclosure. [Effects of the Invention]

[0007] This disclosure makes it possible to provide a driver assistance system that can suppress misjudgments regarding dangerous driving. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a block diagram showing the configuration of the driver assistance system 1 related to this disclosure. [Figure 2] Figure 2 shows an example of the probability of encountering a pedestrian. [Figure 3] Figure 3 is a schematic diagram showing an example of a processing pipeline in the driver assistance system 1. [Figure 4] Figure 4 is a flowchart showing an example of the processing operation by the driver assistance system 1. [Figure 5] Figure 5 is a schematic diagram showing an example of a judgment result by the driver assistance system 1. [Modes for carrying out the invention]

[0009] Embodiments of the present disclosure will be described below with reference to the drawings. Figure 1 is a block diagram showing the configuration of the driver assistance system 1 according to the present disclosure. The driver assistance system 1 is a system that determines whether a vehicle is driving dangerously. Specifically, the driver assistance system 1 is a system that estimates the degree of danger in the vehicle's driving and issues a warning. The driver assistance system 1 comprises a vehicle 2 and a server 4. The vehicle 2 is the target of the driver assistance system 1's determination of dangerous driving. The server 4 can cooperate with the vehicle 2 and provide the vehicle 2 with necessary information.

[0010] First, an overview of the information acquired by the driver assistance system 1 will be described. The driver assistance system 1 according to this embodiment acquires vehicle information and environmental information. Vehicle information is information about vehicle 2. Environmental information is information about the environment surrounding vehicle 2. Both types of information are used to determine whether vehicle 2 is driving dangerously.

[0011] Vehicle information is classified into static vehicle information and dynamic vehicle information. Static vehicle information is predetermined vehicle information for vehicle 2. In other words, static vehicle information is information that does not change moment by moment. To put it another way, static vehicle information is information that does not change while vehicle 2 is in motion.

[0012] The static vehicle information may include information regarding the specifications of vehicle 2. This information may include, for example, the main specifications of vehicle 2 or information regarding the vehicle class of vehicle 2. This information may include, but is not limited to, vehicle size, minimum turning radius, and drive system. Here, information regarding vehicle size includes the overall length, overall width, overall height, and weight of vehicle 2.

[0013] Furthermore, the vehicle static information may include information that affects the braking distance of vehicle 2. In other words, the vehicle static information may include information that, when changed, can cause the braking distance of vehicle 2 to change. Information that affects the braking distance of vehicle 2 includes, but is not limited to, vehicle size, tire performance, and brake system performance.

[0014] Furthermore, the static vehicle information may include the history information of vehicle 2. The history information of vehicle 2 may include, for example, the maintenance history, repair history, accident history, and usage history of vehicle 2. The history information of vehicle 2 may also include information regarding the year of manufacture of vehicle 2. The information regarding the year of manufacture of vehicle 2 is information regarding the year of manufacture or year of sale of vehicle 2.

[0015] Furthermore, the vehicle static information may include information regarding the presence or absence of various sensors mounted on vehicle 2. These sensors include, for example, vehicle speed sensors, acceleration sensors, angular velocity sensors, and millimeter-wave sensors. In addition, the vehicle static information may include information regarding the vehicle body, drivetrain, braking system, and electrical system of vehicle 2.

[0016] Vehicle dynamic information is vehicle information that changes as vehicle 2 moves. In other words, vehicle dynamic information is information that changes while vehicle 2 is moving. Vehicle dynamic information is typically information that changes in seconds, but is not limited to this, and may also be information that changes in minutes, hours, or days. Vehicle dynamic information may also be called information that changes moment by moment.

[0017] Vehicle dynamic information may include information about the driving state of vehicle 2. Information about the driving state may include, for example, information about vehicle speed, acceleration, angular velocity, engine speed, and battery level. Vehicle dynamic information may also include information about the driving conditions of vehicle 2. Information about driving conditions may include, for example, information about the steering angle, accelerator pedal position, and brake pedal position. Furthermore, vehicle dynamic information may include information about the undercarriage of vehicle 2. That is, vehicle dynamic information may include information about tire wear and tire pressure. This vehicle dynamic information may also be information obtained immediately before the driver assistance system 1 determines that the driving is dangerous. For example, the vehicle dynamic information may be the vehicle speed, acceleration, and angular velocity of vehicle 2 immediately before the driver assistance system 1 determines that the driving is dangerous.

[0018] Next, environmental information will be described. The environmental information is classified into environmental static information and environmental dynamic information. The environmental static information is environmental information predetermined for the surrounding environment of the vehicle 2. The environmental static information is, for example, information regarding the width of the road on which the vehicle 2 travels, the presence or absence of a sidewalk, the presence or absence of a signal, the presence or absence of a sign, and the presence or absence of traffic measures. In addition, the environmental static information may include information regarding the residential density in the vicinity where the vehicle 2 travels.

[0019] The environmental dynamic information is information regarding the surrounding environment in which the vehicle 2 travels and is information that changes according to the timing of the vehicle 2's travel. That is, the environmental dynamic information is environmental information that cannot be acquired in advance. The environmental dynamic information is, for example, information on pedestrians existing in the vicinity when the vehicle 2 travels, the average vehicle speed on the road on which the vehicle 2 travels, and the ratio of children among pedestrians. The information on pedestrians may be acquired by a camera or sensor mounted on the vehicle 2.

[0020] The information on pedestrians may be represented by the pedestrian encounter probability. The pedestrian encounter probability is the probability that the vehicle 2 encounters a pedestrian. The pedestrian encounter probability may be derived each time when determining the dangerous driving of the vehicle 2, or may be predetermined. The pedestrian encounter probability may be set for a plurality of time zones. That is, the pedestrian encounter probability may be determined by time or by minute, or one probability may be determined for every plurality of hours. Also, the pedestrian encounter probability may be represented by, for example, two values such as "high pedestrian encounter probability" or "low pedestrian encounter probability", or may be represented by a fixed probability for each road.

[0021] The driving support system 1 may make a determination using both the information on pedestrians acquired by a camera or sensor mounted on the vehicle 2 and the pedestrian encounter probability. Thereby, even when a pedestrian cannot be recognized by a camera or sensor, the potential risk level of the road on which the vehicle 2 travels can be determined by using the pedestrian encounter probability.

[0022] Here, an example of pedestrian encounter probability is shown in the figure. Figure 2 is a diagram showing an example of pedestrian encounter probability. In Figure 2, the horizontal axis represents time, and the vertical axis represents pedestrian encounter probability. In Figure 2, pedestrian encounter probability is defined every hour. The road in Figure 2 has a relatively high pedestrian encounter probability in the morning and evening, and a relatively low pedestrian encounter probability at other times. In other words, the road in Figure 2 is assumed to be a residential road. Thus, pedestrian encounter probability may be defined every hour. Also, as indicated at the top of Figure 2, pedestrian encounter probability may be expressed as a binary value: "high pedestrian encounter probability" or "low pedestrian encounter probability".

[0023] The vehicle information and environmental information acquired by the driver assistance system 1 shown in Figure 1 are typically quantitative information, but may also be qualitative information. In other words, this information may or may not be expressed numerically. By expressing it numerically, it is possible to determine whether or not it is dangerous driving without being influenced by the judgment method.

[0024] Next, the components of the driver assistance system 1 will be explained using Figure 1. The vehicle 2 is equipped with a vehicle speed sensor 21, an acceleration sensor 22, an angular velocity sensor 23, a GPS (Global Positioning System) sensor 24, a vehicle ECU (Electronic Control Unit) 25, a communication unit 26, a road link database 27, a vehicle information database 28, a speaker 29, and a display 30.

[0025] The vehicle speed sensor 21, acceleration sensor 22, and angular velocity sensor 23 are sensors for measuring the vehicle speed, acceleration, and angular velocity of the vehicle 2 while it is in motion, respectively. In other words, these sensors are mainly used to acquire vehicle dynamic information. The presence or absence of these sensors may also be part of the vehicle static information. These sensors each transmit the sensor information they acquire to the vehicle ECU 25.

[0026] The GPS sensor 24 is a sensor for receiving the latitude and longitude of vehicle 2 from GPS satellites. In other words, the GPS sensor 24 is a sensor for acquiring the position information of vehicle 2 while it is in motion. Based on the current position of vehicle 2 measured by the GPS sensor 24, vehicle 2 acquires environmental information. The GPS sensor 24 transmits the acquired position information to the vehicle ECU 25.

[0027] The vehicle ECU 25 is a device that controls various systems in vehicle 2. Specifically, the vehicle ECU 25 determines dangerous driving by vehicle 2 by receiving information from various sensors and databases in vehicle 2. The vehicle ECU 25 may be operated by a microcomputer, MPU (Micro Processing Unit), or CPU (Central Processing Unit).

[0028] The vehicle ECU 25 determines whether vehicle 2 is driving dangerously by calculating the degree of risk associated with its operation. Specifically, the vehicle ECU 25 determines that vehicle 2 is driving dangerously if the calculated degree of risk exceeds a predetermined threshold.

[0029] The vehicle ECU25 uses a machine learning model to derive the risk level of driving vehicle 2. The machine learning model is typically a deep learning model using a Convolutional Neural Network (CNN). The machine learning model may use supervised learning or unsupervised learning. If supervised learning is used, the machine learning model may be trained using vehicle information, environmental information, and the risk level of driving the vehicle. Here, the vehicle information used for training is not limited to information relating to vehicle 2, but may also include information relating to vehicles other than vehicle 2.

[0030] In unsupervised learning, the machine learning model may be trained using vehicle information and environmental information related to vehicles that are not driving dangerously. In this case, the vehicle ECU 25 directly determines whether driving by vehicle 2 is dangerous by determining whether the vehicle information and environmental information related to the driving vehicle 2 are similar to the data in the trained model. Here, if it is determined that driving by vehicle 2 is dangerous, the machine learning model may output "1" as the risk level, and if it is determined that it is not dangerous, it may output "0" as the risk level. Such unsupervised learning is an effective method when it is difficult to obtain data on dangerous driving.

[0031] The communication unit 26 is an interface that performs bidirectional communication with the server 4. In particular, the communication unit 26 communicates with the arithmetic unit 41 in the server 4. The communication unit 26 can communicate with the server 4 using any V2X (Vehicle to Everything) communication standard. For example, the communication unit 26 receives the current location information of the vehicle 2 from the vehicle ECU 25. The communication unit 26 may transmit the current location information of the vehicle 2 to the server 4, or it may transmit road information on which the vehicle 2 is traveling to the server 4. The communication unit 26 also receives environmental information corresponding to this information from the server 4. Furthermore, the communication unit 26 transmits the received environmental information to the vehicle ECU 25.

[0032] The road link database 27 is a database that stores road information. Road information is also referred to as information related to road links. The road link database 27 obtains the current location information of vehicle 2 from the vehicle ECU 25, refers to the information in the road link database 27, and transmits the road information on which vehicle 2 is traveling to the vehicle ECU 25.

[0033] The vehicle information database 28 is a database that stores vehicle information related to vehicle 2. Typically, the vehicle information database 28 contains static vehicle information, but it may also include dynamic vehicle information. The vehicle information database 28 can transmit vehicle information related to vehicle 2 to the vehicle ECU 25 in response to a request from the vehicle ECU 25.

[0034] Speaker 29 is an audio output device for alerting the occupants of vehicle 2 about dangerous driving. Speaker 29 informs the occupants that vehicle 2 is driving dangerously. In this case, speaker 29 may also inform the occupants of specific measures to reduce the level of danger. Furthermore, if the automated driving system (not shown) controls vehicle 2 to eliminate dangerous driving, speaker 29 may also inform the occupants of this fact.

[0035] The display 30 is a video output device for alerting the occupants of vehicle 2 about dangerous driving. Similar to the speaker 29, the display 30 may inform the occupants that the vehicle 2 is driving dangerously and may also display other information.

[0036] Next, the server 4 will be described. The server 4 comprises a calculation unit 41 and an environmental information database 42. The calculation unit 41 is an interface that performs bidirectional communication with the vehicle 2. In particular, the calculation unit 41 communicates with the communication unit 26 inside the vehicle 2. The calculation unit 41 can perform communication with the vehicle 2 using any V2X communication standard.

[0037] Furthermore, the calculation unit 41 can refer to the environmental information database 42 and extract environmental information about the road on which the vehicle 2 is traveling from the road information related to the road on which the vehicle 2 is traveling. That is, the calculation unit 41 performs the following processing, for example: First, the calculation unit 41 receives the current location information of the vehicle 2 and road information from the communication unit 26. Next, the calculation unit 41 refers to the environmental information database 42 based on this acquired information and extracts environmental information about the road on which the vehicle 2 is traveling. After that, the calculation unit 41 transmits the environmental information to the communication unit 26.

[0038] The environmental information database 42 is a database that stores environmental information related to roads on which vehicle 2 can travel. The environmental information database 42 includes static environmental information and dynamic environmental information. The environmental information database 42 can output the necessary environmental information in accordance with requests from the calculation unit 41.

[0039] Next, the processing pipeline in the driver assistance system 1 will be described. Figure 3 is a schematic diagram showing an example of the processing pipeline in the driver assistance system 1. First, the driver assistance system 1 collects vehicle information and environmental information (S1). For example, the driver assistance system 1 obtains static vehicle information from the vehicle information database 28. The driver assistance system 1 obtains dynamic vehicle information from the vehicle speed sensor 21, acceleration sensor 22, and angular velocity sensor 23. The driver assistance system 1 obtains environmental information from the environmental information database 42.

[0040] Next, the driver assistance system 1 derives the degree of risk regarding the driving of vehicle 2 (S2). The driver assistance system 1 derives the degree of risk by inputting the necessary information into a trained model. After that, the driver assistance system 1 makes a risk determination and outputs the necessary information (S3). That is, the driver assistance system 1 determines whether the derived degree of risk exceeds a preset threshold. If the degree of risk exceeds the threshold, the driver assistance system 1 provides necessary warnings to the in-vehicle user.

[0041] Next, an example of the processing operation by the driver assistance system 1 will be described. Figure 4 is a flowchart of an example of the processing operation by the driver assistance system 1. First, the driver assistance system 1 acquires vehicle information from various sensors and the vehicle information database 28 (S101). Next, the driver assistance system 1 acquires road information from the road link database 27 (S102). After that, the driver assistance system 1 acquires environmental information of the road on which the vehicle 2 is traveling from the environmental information database 42 (S103). Here, the driver assistance system 1 may perform steps S101 to S103 simultaneously, or the order of each step may be reversed.

[0042] Subsequently, the driver assistance system 1 derives the degree of risk of driving the vehicle 2 (S104). Then, the driver assistance system 1 determines whether the derived degree of risk exceeds a preset threshold (S105). If the threshold is not exceeded, the process ends. If the threshold is exceeded, the driver assistance system 1 alerts the in-vehicle user via the speaker 29 and display 30 (S106). After this, the processing by the driver assistance system 1 ends, but the driver assistance system 1 may repeat the above operations. That is, after step S106, the driver assistance system 1 may return to step S101 and execute the process again.

[0043] Thus, the driver assistance system 1 can determine dangerous driving by vehicle 2 based on static vehicle information in addition to environmental information, thereby reducing the likelihood of misjudgments regarding dangerous driving. For example, as the size of the vehicle increases, the braking distance of vehicle 2 tends to increase. This is because the mass of vehicle 2 increases. In addition, information such as tire performance and brake system performance may also affect the braking distance of vehicle 2. Related technologies cannot take into account such vehicle-specific information, so they cannot make precise judgments and may result in misjudgments such as false negatives and false positives. If the judgment result is a false negative, it is not possible to alert the in-vehicle user in a truly dangerous situation. Furthermore, if false positive judgment results occur frequently, the in-vehicle user will not trust the judgment, and related technologies will not be able to improve the in-vehicle user's attention. The driver assistance system 1 can solve these problems.

[0044] Furthermore, the driver assistance system 1 can include information on key specifications and vehicle class, such as vehicle size and minimum turning radius, as static vehicle information. This information affects the handling of the vehicle. In other words, if the vehicle is large, it tends to occupy more road width, requiring careful handling from the driver. By considering information on the specifications of vehicle 2, appropriate warnings can be given to the in-vehicle user according to the specifications of vehicle 2.

[0045] Furthermore, by using information about the driving conditions of vehicle 2, such as vehicle speed, acceleration, and angular velocity, as vehicle dynamic information, the driver's judgment can be taken into consideration when determining the degree of danger. For example, if the vehicle speed of vehicle 2 is high just before the driver assistance system 1 determines that driving is dangerous, the driver may be in a hurry to reach their destination, and in such cases, it is assumed that the situation is prone to accidents. The driver assistance system 1 can make a more reliable judgment based on this vehicle dynamic information.

[0046] In the driver assistance system 1, pedestrian information can be represented by the probability of encountering a pedestrian. Cameras and sensors can be used to acquire pedestrian information, but if a pedestrian is hidden by another vehicle, for example, it is difficult to detect them using cameras and sensors. Even in such cases, by considering the probability of encountering a pedestrian as pedestrian information, the potential level of danger on the road on which vehicle 2 is traveling can be determined.

[0047] The driver assistance system 1 can set the probability of encountering pedestrians for multiple time periods. For example, if the road on which vehicle 2 is traveling is a residential street, the number of pedestrians walking on that road will vary depending on the time of day. By setting the probability of encountering pedestrians for multiple time periods, the driver assistance system 1 can make a more accurate risk assessment.

[0048] The driver assistance system 1 uses a pre-trained model to determine if vehicle 2 is driving dangerously. The driver assistance system 1 derives the degree of danger using multiple parameters such as vehicle information and environmental information. When constructing logic from multiple indicators in this way, it can be difficult to appropriately define the degree of danger. By using a pre-trained model, the driver assistance system 1 can appropriately derive the degree of danger.

[0049] An example of a judgment made by the driver assistance system 1 is described below. Figure 5 is a schematic diagram showing an example of a judgment result by the driver assistance system 1. Figure 5 shows three levels of danger. Specifically, Figure 5 shows three patterns: (A) dangerous driving, (B) moderately dangerous driving, and (C) non-dangerous driving. For example, in pattern (A), vehicle 2 is large in size, the speed of the vehicle directly in front of vehicle 2 is high, the road is narrow, and there are many pedestrians. In this case, the driver assistance system 1 determines that the driving by vehicle 2 is dangerous and issues a warning to the user inside vehicle 2.

[0050] In pattern (B), vehicle 2 is large, the road is narrow, but the speed of the vehicle directly in front of vehicle 2 is low, and there are few pedestrians. In this case, the driver assistance system 1 may or may not judge the driving to be dangerous, based on predetermined thresholds. In pattern (C), vehicle 2 is small, the speed of the vehicle directly in front of vehicle 2 is low, the road is wide, and there are few pedestrians. In this case, the driver assistance system 1 determines that the driving by vehicle 2 is not dangerous.

[0051] This disclosure is not limited to the embodiments described above, and can be modified as appropriate without departing from the spirit of the disclosure. For example, the driver assistance system 1 does not need to use all of the vehicle information and environmental information to determine the degree of risk. That is, the driver assistance system 1 may determine the degree of risk without using any of the vehicle static information, vehicle dynamic information, environmental static information, or environmental dynamic information. For example, the driver assistance system 1 may determine the degree of risk without using vehicle static information, and only using vehicle dynamic information, environmental static information, and environmental dynamic information.

[0052] In this case, the driver assistance system 1 does not need to have a database for storing such information or sensors for acquiring such information. For example, if static vehicle information is not used, the driver assistance system 1 does not need to have a vehicle information database 28. Also, if dynamic vehicle information is not used, the driver assistance system 1 does not need to have a vehicle speed sensor 21, an acceleration sensor 22, and an angular velocity sensor 23.

[0053] Furthermore, the driver assistance system 1 may also use information from other vehicles to determine the degree of danger. That is, the driver assistance system 1 may acquire vehicle information from other vehicles by cooperating with other vehicles using a predetermined communication standard, and perform a determination of the degree of danger based on that information.

[0054] Furthermore, the driver assistance system 1 may store all the information necessary for determining the degree of danger in the vehicle 2. In other words, the information in the server 4 may be in the vehicle 2. In this case, the server 4 does not need to be part of the driver assistance system 1. [Explanation of Symbols]

[0055] 1. Driver assistance system, 2. Vehicle, 4. Server, 21. Vehicle speed sensor, 22. Acceleration sensor, 23. Angular velocity sensor, 24. GPS sensor, 25. Vehicle ECU, 26. Communication unit, 27. Road link database, 28. Vehicle information database, 29. Speaker, 30. Display, 41. Processing unit, 42. Environmental information database

Claims

1. A driver assistance system that determines dangerous driving by a vehicle, The dangerous driving is determined based on vehicle static information, which is predetermined information about the vehicle, and environmental information, which is information about the surrounding environment of the vehicle. Driver assistance system.

2. The vehicle static information includes information regarding the vehicle's specifications. The driver assistance system according to claim 1.

3. The vehicle static information includes information that affects the braking distance of the vehicle. The driver assistance system according to claim 1.

4. The aforementioned environmental information includes the probability of the vehicle encountering a pedestrian. The driver assistance system according to claim 1.

5. The vehicle static information includes information regarding the vehicle size of the vehicle. The driver assistance system according to claim 2 or 3.

Images