Obstacle detection device, obstacle detection method, and obstacle detection program

Abstract

This obstacle detection device comprises: a first acquisition unit that acquires, from an air pressure sensor mounted on a vehicle or on a tire of the vehicle, tire internal pressure data indicating a change in the internal pressure of the tire while the vehicle is traveling on a road surface; a second acquisition unit that acquires, from an imaging device and / or a three-dimensional scanner mounted on the vehicle, road surface condition data indicating the condition of the road surface while the vehicle is traveling on the road surface; and a detection unit that detects an obstacle on the road surface on the basis of the tire internal pressure data and / or the road surface condition data.

Description

Obstacle detection device, obstacle detection method, and obstacle detection program

[0001] The present disclosure relates to an obstacle detection device, an obstacle detection method, and an obstacle detection program.

[0002] Tire damage is affected by the condition of the road surface on which a vehicle equipped with the tire is traveling. For this reason, a technique for acquiring road surface conditions while a vehicle is traveling is known (see, for example, Patent Document 1 (Japanese Unexamined Patent Application Publication No. 2022-007616)).

[0003] In the technique described in Patent Document 1, the size of an obstacle is not taken into consideration, so it is conceivable to detect the size of an obstacle based on the internal pressure of the tire. However, in order to detect obstacles with high accuracy, the sampling frequency of the sensor may be constantly increased. In this case, since tire internal pressure data is constantly acquired at a high sampling frequency, a large load is imposed on data analysis, such as an increase in data retention time and an increase in data analysis man-hours.

[0004] An object of the present disclosure is to provide an obstacle detection device, an obstacle detection method, and an obstacle detection program that can accurately detect obstacles on a road surface and reduce the load applied to data analysis when a vehicle is traveling on the road surface.

[0005] In order to achieve the above object, a first aspect is an obstacle detection device, comprising: a first acquisition unit that acquires tire internal pressure data representing a change in internal pressure of a tire of a vehicle from an air pressure sensor mounted on the tire while the vehicle is traveling on a road surface; a second acquisition unit that acquires road surface condition data representing a condition of the road surface while the vehicle is traveling on the road surface from at least one of an imaging device and a three-dimensional scanner mounted on the vehicle; and a detection unit that detects an obstacle on the road surface based on at least one of the tire internal pressure data and the road surface condition data.

[0006] The second embodiment further comprises a modification unit that, when an obstacle on the road surface is detected, changes the sampling frequency of the tire internal pressure data from a first sampling frequency before the obstacle on the road surface is detected to a second sampling frequency that is higher than the first sampling frequency.

[0007] A third embodiment is an obstacle detection device according to the first or second embodiment, further comprising: a determination unit that, when it detects an obstacle on the road surface, estimates the size of the obstacle and determines whether or not there is a risk related to the obstacle based on the estimated size of the obstacle; and a warning unit that issues a warning related to the risk when the determination unit determines that there is a risk.

[0008] The fourth aspect is an obstacle detection device according to the third aspect, wherein the warning unit issues a warning regarding the risk when the determination unit determines that the risk exists and detects a fluctuation of the tire internal pressure data above a certain level.

[0009] The fifth aspect is an obstacle detection device according to the third or fourth aspect, wherein the determination unit determines the risk at multiple levels, and the warning unit provides different warnings regarding the risk according to each of the multiple levels.

[0010] The sixth aspect is an obstacle detection device according to the first aspect, further comprising: a modification unit that, when an obstacle on the road surface is detected, changes the sampling frequency of the tire pressure data from a first sampling frequency before the obstacle on the road surface is detected to a second sampling frequency higher than the first sampling frequency; and a determination unit that, when an obstacle on the road surface is detected, estimates the size of the obstacle and determines whether or not there is a risk related to the obstacle based on the estimated size of the obstacle, wherein if the determination unit determines that there is no risk, the modification unit returns the sampling frequency of the tire pressure data from the second sampling frequency to the first sampling frequency.

[0011] The seventh embodiment is an obstacle detection device according to any one of the first to sixth embodiments, further comprising a notification unit that, when an obstacle on the road surface is detected, notifies other vehicles in the vicinity of the vehicle to encourage them to avoid the obstacle.

[0012] The eighth aspect is an obstacle detection device according to any one of the first to sixth aspects, wherein the second acquisition unit detects the obstacle on the road surface and, when the vehicle passes over the obstacle, acquires road surface condition data including the state of the obstacle, and further comprises a notification unit that notifies other vehicles in the vicinity of the vehicle to avoid the obstacle according to the state of the obstacle obtained from the road surface condition data.

[0013] The ninth aspect is an obstacle detection device according to any one of the first to sixth aspects, further comprising a notification unit that, when an obstacle is detected on the road surface, notifies a specific vehicle capable of removing the obstacle to prompt it to remove the obstacle.

[0014] The tenth aspect is an obstacle detection method, wherein a computer performs a process to detect obstacles on the road surface, which involves acquiring tire pressure data representing changes in the internal pressure of the tires from an air pressure sensor mounted on the tires of the vehicle while the vehicle is traveling on the road surface, acquiring road surface condition data representing the condition of the road surface from at least one of a camera and a three-dimensional scanner mounted on the vehicle while the vehicle is traveling on the road surface, and based on at least one of the tire pressure data and the road surface condition data.

[0015] The eleventh aspect is an obstacle detection program, which acquires tire pressure data representing changes in the internal pressure of the tires from an air pressure sensor mounted on the tires of the vehicle while the vehicle is driving on the road surface, acquires road surface condition data representing the condition of the road surface from at least one of a camera and a three-dimensional scanner mounted on the vehicle while the vehicle is driving on the road surface, and causes a computer to perform a process of detecting obstacles on the road surface based on at least one of the tire pressure data and the road surface condition data.

[0016] According to this disclosure, the system has the effect of accurately detecting obstacles on the road surface while a vehicle is traveling on the road, while reducing the load required for data analysis.

[0017] This figure shows the overall configuration including the obstacle detection device and vehicle according to the embodiment. This block diagram shows the hardware configuration of the obstacle detection device according to the embodiment. This block diagram shows an example of the functional configuration of the obstacle detection device according to the embodiment. This figure schematically shows an example of the relationship between the vehicle and an obstacle according to the embodiment. This figure schematically shows another example of the relationship between the vehicle and an obstacle according to the embodiment. This flowchart shows the flow of the obstacle detection process. This flowchart shows the flow of the inspection instruction process. This flowchart shows the flow of the road surface maintenance instruction process. This flowchart shows the flow of the warning process. This flowchart shows the flow of the report process.

[0018] Embodiments of the technology described herein will be described in detail below with reference to the drawings. Components and processes that perform the same function or operation will be given the same reference numerals throughout the drawings, and redundant explanations may be omitted as appropriate. Furthermore, this disclosure is not limited in any way to the embodiments described below, and can be implemented with appropriate modifications within the scope of the purpose of this disclosure.

[0019] The obstacle detection device according to this embodiment acquires tire pressure data representing changes in tire pressure while the vehicle is traveling on the road surface from an air pressure sensor mounted on the vehicle's tire, and acquires road surface condition data representing the road surface conditions while the vehicle is traveling on the road surface from at least one of a camera and a 3D scanner mounted on the vehicle, and detects obstacles on the road surface based on at least one of the tire pressure data and road surface condition data. This makes it possible to detect obstacles on the road surface with high accuracy while reducing the load on data analysis of tire pressure data.

[0020] Figure 1 is a diagram showing the overall configuration including the obstacle detection device 10 and vehicles 50 according to this embodiment. As shown in Figure 1, the obstacle detection system 1 of this embodiment is composed of the obstacle detection device 10 and a plurality of vehicles 50. In Figure 1, a work vehicle used for work in a mine or other workplace is shown as an example of a vehicle 50, but it is not limited to this work vehicle. Each vehicle 50 is equipped with a communication device 25, and the tires of each vehicle 50 are equipped with various sensors for acquiring tire pressure data. The various sensors may include one or more TPMS (Tire Pressure Monitoring System) 20 (details will be described later), etc. The TPMS 20 is an example of an air pressure sensor, which is mounted on the tire itself or on the surrounding part of the tire (rim, valve, etc.). In addition, each vehicle 50 may be equipped with a GPS (Global Positioning System) sensor for acquiring vehicle position. In addition, each vehicle 50 is equipped with a camera 30 for acquiring road surface condition data. The tire pressure data represents changes in tire pressure, and the road surface condition data represents the road surface conditions. Alternatively, a 3D scanner 30A may be installed instead of the imaging device 30, or both the imaging device 30 and the 3D scanner 30A may be installed. The 3D scanner 30A may be, for example, a LiDAR (Laser Imaging Detection and Ranging) or laser scanner. In the case of the imaging device 30, the road surface condition data is image data, while in the case of the 3D scanner 30A, the road surface condition data is 3D point cloud data. The following explanation will use the imaging device 30 as a representative example.

[0021] The obstacle detection device 10 according to this embodiment can communicate wirelessly with the communication device 25 of each vehicle 50. The communication device 25 of the vehicle 50 can communicate with the TPMS 20 and the imaging device 30 (or 3D scanner 30A) mounted on the vehicle 50. In this way, the obstacle detection device 10 receives tire pressure data and road surface condition data from the moving vehicle 50.

[0022] Figure 2 is a block diagram showing the hardware configuration of the obstacle detection device 10 according to this embodiment. As shown in Figure 2, the obstacle detection device 10 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, storage 14, an input / output I / F (Interface) 15, and a communication I / F 16. Each component is connected to the others via a bus 17 so as to be able to communicate with each other.

[0023] The CPU 11 is a central processing unit that executes various programs and controls various components. Specifically, the CPU 11 reads programs from the ROM 12 and executes them using the RAM 13 as a working area. The CPU 11 controls the above-mentioned components and performs various calculations according to the programs stored in the ROM 12. The CPU 11 is also responsible for processing each of the functional units shown in Figure 3.

[0024] The storage device, comprised of ROM 12, stores various programs, including the operating system, and various data. ROM 12 also stores processing programs for executing inspection instruction processing, road surface maintenance instruction processing, warning processing, and report processing, which will be described later.

[0025] The memory, comprised of RAM 13, temporarily stores programs and data as a working area.

[0026] The storage device 14 is composed of an HDD (Hard Disk Drive) and an SSD (Solid State Drive), and stores various types of data. The storage device 14 also functions as a data storage unit 108 as shown in Figure 3, and stores various types of data such as tire pressure data and road surface condition data. The storage device 14 does not necessarily have to be built into the obstacle detection device 10; for example, it may be a portable storage device that can be attached to or detached from the obstacle detection device 10, or an external cloud server may be used as the storage device.

[0027] The input / output interface 15 is an interface for communicating with input and output devices located outside the obstacle detection device 10. The input device includes a pointing device such as a mouse and a keyboard, and is used for various types of input. The output device includes, for example, a liquid crystal display or an organic EL (Electroluminescence) display, and is a device for outputting various types of information. The output device may also function as an input device by employing a touch panel system. The output device may also be equipped with a speaker or the like as a means of audio output. For example, if the obstacle detection device 10 is installed in a vehicle 50, the output device is used as a monitor inside the vehicle. Also, if the obstacle detection device 10 is installed in a train control room, the output device is used as a monitor inside the train control room.

[0028] The communication interface 16 is an interface for communicating with other devices outside the obstacle detection device 10. For this communication, a wired communication standard such as Ethernet® or FDDI, or a wireless communication standard such as 4G, 5G, or Wi-Fi® can be used. For example, if the obstacle detection device 10 is installed in the operation control room, the obstacle detection device 10 communicates data with each vehicle 50 through a wireless line, which is an example of a communication line connected to the communication interface 16. Alternatively, if the obstacle detection device 10 is installed in a vehicle 50, the obstacle detection device 10 may communicate data with a server (not shown) installed in the operation control room through a wireless line connected to the communication interface 16, and display the internal pressure and temperature status of each tire on a monitor installed in the operation control room.

[0029] The communication interface 16 also communicates wirelessly with the communication device 25 of the vehicle 50. The communication device 25 is installed in the vehicle 50 and communicates with the TPMS 20 and the imaging device 30 of each tire of the vehicle 50. Note that the communication device 25 installed in each vehicle 50 may include a GPS sensor 26. The GPS sensor 26 acquires the position information of the vehicle 50 and transmits it from the communication device 25 to the communication interface 16 of the obstacle detection device 10. In addition, the TPMS 20 installed in each tire of the vehicle 50 includes at least a transmitter 23, a temperature sensor 21, and a pressure sensor 22. The transmitter 23 communicates with the communication device 25 of the vehicle 50. Specifically, the transmitter 23 transmits tire internal pressure data including information on the temperature and pressure of the tires of the vehicle 50. In addition, the measurements by the temperature sensor 21 and the pressure sensor 22 are performed continuously or at a preset sampling interval (sampling frequency). In addition, the imaging device 30 installed in the vehicle 50 includes at least a transmitter 31. The transmitter 31 communicates with the communication device 25 of the vehicle 50. Specifically, the transmitter 31 transmits image data, which is an example of road surface condition data for the vehicle 50. The vehicle may also be equipped with the three-dimensional scanner 30A described above. The three-dimensional scanner 30A includes at least the transmitter 31A. The transmitter 31 communicates with the communication device 25 of the vehicle 50. Specifically, the transmitter 31A transmits three-dimensional point cloud data, which is an example of road surface condition data for the vehicle 50.

[0030] Referring to Figure 3, the functional configuration of the obstacle detection device 10 according to this embodiment will be described. As shown in Figure 3, the obstacle detection device 10 of this embodiment includes a first acquisition unit 101, a second acquisition unit 102, a detection unit 103, a modification unit 104, a determination unit 105, a warning unit 106, and a notification unit 107. The first acquisition unit 101 and the second acquisition unit 102 may be configured as a single acquisition unit. A data storage unit 108 is also provided in a predetermined storage area of ​​the storage 14. The CPU 11 executes the obstacle detection program stored in the ROM 12, thereby enabling the first acquisition unit 101, the second acquisition unit 102, the detection unit 103, the modification unit 104, the determination unit 105, the warning unit 106, and the notification unit 107 to function. The data storage unit 108 stores the threshold values ​​for the size of obstacles and the classification of obstacle sizes, which will be described later. Furthermore, the data storage unit 108 stores various data acquired by the first acquisition unit 101 and the second acquisition unit 102. In this embodiment, the collection of various information referred to as a "list" includes the meaning of a table in a database. This database is stored, for example, in the data storage unit 108.

[0031] The first acquisition unit 101 acquires tire information, including tire pressure data, from the moving vehicle 50. The acquisition of tire pressure data is performed via the communication I / F 16 of the obstacle detection device 10. The tire pressure data includes at least the internal pressure value of the vehicle 50's tires. The first acquisition unit 101 acquires tire pressure data from the TPMS 20 installed in the vehicle 50's tires. The reason the first acquisition unit 101 acquires tire pressure data is that, for example, if the tire pressure acquired by the first acquisition unit 101 becomes higher than the normal level, it can be inferred that the tire has run over an obstacle. In other words, the first acquisition unit 101 can acquire information about the tire that has run over an obstacle. The first acquisition unit 101 stores the acquired information in the tire central inspection list, which is a list of candidate tires for inspection.

[0032] The first acquisition unit 101 also acquires the position information of the vehicle 50 along with the tire pressure data mentioned above. The first acquisition unit 101 acquires position information when the position information acquired by the GPS sensor 26 installed in each vehicle 50 is transmitted from the communication device 25 to the communication I / F 16 of the obstacle detection device 10. The first acquisition unit 101 records the position information of the vehicle 50 when the tire that has driven over an obstacle in the road surface maintenance list, which is a list of candidate locations for maintaining the road surface on which the vehicle 50 is traveling.

[0033] The second acquisition unit 102 acquires road surface condition data of the moving vehicle 50. The road surface condition data is acquired via the communication I / F 16 of the obstacle detection device 10. Road surface condition data is image data representing the condition of the road surface on which the vehicle 50 is traveling. The second acquisition unit 102 acquires road surface condition data from the camera 30 installed on the vehicle 50. The reason why the second acquisition unit 102 acquires road surface condition data is, for example, that it can detect obstacles in front of or behind the vehicle 50 from the image data acquired by the second acquisition unit 102.

[0034] The detection unit 103 detects obstacles on the road surface based on at least one of the tire pressure data and road surface condition data. "Obstacles" here include, for example, rocks and fallen objects. For obstacle detection, object recognition using machine learning can be used. Specifically, machine learning can be used to train a model to recognize the correlation between image data (an example of past road surface condition data) and obstacles as training data. Then, using the trained model, image data obtained during the current drive can be input, and obstacles can be presented as output. Even if the road surface condition data is 3D point cloud data, obstacles can be detected using machine learning in the same way as with image data. However, both tire pressure data and road surface condition data may be used for obstacle detection, or only tire pressure data may be used.

[0035] When the detection unit 103 detects an obstacle on the road surface, the modification unit 104 changes the sampling frequency of the tire pressure data from a first sampling frequency (used before detecting an obstacle on the road surface) to a second sampling frequency (used to be higher than the first sampling frequency). Specifically, the change in sampling frequency is performed by changing the sampling frequency of the pressure sensor 22. In other words, it is possible to lower the sampling frequency of the tire pressure data before detecting an obstacle based on road surface condition data while the vehicle 50 is in motion, and to raise the sampling frequency after detecting an obstacle. The sampling frequency of the tire pressure data may be, for example, the frequency measured by the pressure sensor 22, or the frequency used when the pressure sensor 22 records the measured data after decimation. The timing for raising the sampling frequency of the tire pressure data may be immediately after detecting an obstacle, or after the determination unit 105 (described later) determines that there is a risk.

[0036] When the detection unit 103 detects an obstacle on the road surface, the determination unit 105 estimates the size of the obstacle and determines whether or not there is a risk related to the obstacle based on the estimated size of the obstacle. Specifically, the determination unit 105 estimates the size of the obstacle on the road surface from road surface condition data. For example, machine learning can be used to train the system to learn the correlation between obstacles obtained from past road surface condition data and the actual size of obstacles as training data. Using the trained model, image data obtained during the current driving is input, and the size of the obstacle is presented as output. If a 3D scanner 30A is provided, the size of the obstacle on the road surface may be measured from 3D point cloud data. The determination unit 105 then determines that obstacles whose estimated size is greater than or equal to a threshold pose a risk, and obstacles whose size is less than the threshold pose no risk. Here, the "threshold" may be determined by calculating the size of obstacles that could potentially damage the tires based on information such as the tire size, tire pressure, and load of the vehicle 50, and setting this as the threshold. For example, machine learning can be used to train a model that uses data to determine the correlation between information such as the tire size, tire pressure, and load of a vehicle 50 and the size of obstacles that could potentially damage the tires. Then, using the trained model, information such as the tire size, tire pressure, and load of a target vehicle 50 can be input, and the size of obstacles that could potentially damage the tires (i.e., the threshold) can be presented as output.

[0037] The warning unit 106 issues a warning regarding a risk when the determination unit 105 determines that there is a risk. Specifically, the warning unit 106 may issue a warning regarding a risk when the determination unit 105 determines that there is a risk and it detects a fluctuation of a certain amount or more in the tire pressure data. However, the sampling frequency of the tire pressure data may be the first sampling frequency or the second sampling frequency. The warning unit 106 outputs an alert to the output device via the input / output I / F 15 of the obstacle detection device 10. The alert is not particularly limited to voice, text, or warning sound, but may display or output a message such as "There is a dangerous obstacle ahead."

[0038] Figure 4 is a schematic diagram showing an example of the relationship between the vehicle 50 and obstacles D1 and D2 according to this embodiment. In the example in Figure 4, the determination unit 105 determines that obstacle D1 poses no risk and obstacle D2 poses a risk. The warning unit 106 does not output an alert for obstacle D1, which poses no risk, but outputs an alert for obstacle D2, which poses a risk, as described above. However, depending on the timing of the alert, the driver's inattention, the working environment of the vehicle 50, etc., it is conceivable that the vehicle may not be able to avoid obstacle D2 and the tires may run over obstacle D2. When the tires run over obstacle D2, the tire pressure fluctuates. In this case, that is, when the determination unit 105 determines that there is a risk and detects a fluctuation of a certain amount or more in the tire pressure data acquired at the second sampling frequency, the warning unit 106 may issue a warning regarding the risk. Furthermore, the determination unit 105 may determine the risk at multiple levels, and the warning unit 106 may issue different warnings regarding the risk according to each of the multiple levels.

[0039] For example, if the size of the obstacle estimated by the determination unit 105 is greater than or equal to the first threshold, the warning unit 106 outputs an alert to the output device via the input / output I / F 15 of the obstacle detection device 10. Furthermore, when the warning unit 106 outputs an alert, it may categorize the risk levels according to the size of the obstacle estimated by the determination unit 105 and change the output content accordingly. Specifically, if the size of the obstacle estimated by the determination unit 105 falls under the "low risk" category, which is smaller than the second threshold, the warning unit 106 outputs an alert indicating that it is level 1. Here, the second threshold is a larger value than the first threshold. Similarly, if the size of the obstacle estimated by the determination unit 105 is greater than or equal to the second threshold and falls under the "medium risk" category, which is smaller than the third threshold, the warning unit 106 outputs an alert indicating that it is level 2. Here, the third threshold is a larger value than the second threshold. Similarly, the warning unit 106 outputs an alert indicating that the level is 3 if the size of the obstacle estimated by the determination unit 105 falls under the "high risk" category, which is a risk classification of the third threshold or higher. The output method and the content of the alert may be changed depending on the level.

[0040] The first acquisition unit 101 also stores information about tires that have driven over obstacles in a concentrated tire inspection list, which is a list of tires that are candidates for inspection. The concentrated tire inspection list is stored in the data storage unit 108. The concentrated tire inspection list contains information about tires that have driven over obstacles, which should be inspected intensively in addition to the normal inspection during periodic tire inspections. If the warning unit 106 contains information about tires that have driven over obstacles in the concentrated tire inspection list, it outputs an instruction to inspect the tires in the list in addition to the normal inspection during periodic inspections. After the person in charge has performed the inspection, the inspection results are input to the warning unit 106 by the person in charge via the input / output I / F 15 of the obstacle detection device 10. Based on the input inspection results, the warning unit 106 outputs the next course of action. The warning unit 106 updates the concentrated tire inspection list to add these series of inspection results and corresponding actions.

[0041] Furthermore, if information about the tire on the vehicle is stored in the centralized tire inspection list, the warning unit 106 will contact the tire check staff at the gas station and instruct the vehicle 50 to proceed to the gas station. In this embodiment, the case of contacting the tire check staff at the gas station will be explained as an example. However, it is not limited to this. Not only when checking tires at a gas station, but also in any scenario where tires are checked at the vehicle 50's maintenance shop, tire shop, specialized tire check department, or elsewhere, the system may contact the appropriate person or guide the vehicle to the appropriate location. After inspection at a gas station, etc., if the tire check staff, etc., determines that a detailed inspection is necessary, and the determination result is input to the warning unit 106 via the input / output I / F 15 of the obstacle detection device 10, the warning unit 106 will contact the tire manager at the tire shop and instruct the vehicle 50 to proceed to the tire shop. If the size of the obstacle estimated by the determination unit 105 corresponds to the classification "high risk", the vehicle 50 may be instructed to contact the tire shop directly and proceed to the tire shop without going through a gas station, etc. After a detailed inspection at the tire shop, if the tire manager or other relevant person determines that the tire needs to be removed, and the determination is input to the warning unit 106 via the input / output interface 15 of the obstacle detection device 10, the warning unit 106 outputs an instruction to replace the tire. The warning unit 106 updates the centralized tire inspection list to add these inspection results, response results, and determination results to the list.

[0042] The warning unit 106 also outputs the location indicated by the vehicle 50's position information when the tire hits the obstacle as the location of the obstacle. For example, the warning unit 106 displays the location of the obstacle on a map of a navigation system shared among multiple vehicles 50. The warning unit 106 also stores the vehicle 50's position information when the tire hits the obstacle in a road surface maintenance list, which is a list of potential locations for road surface maintenance where the vehicle 50 is traveling. Once the location information is stored in the road surface maintenance list, the warning unit 106 outputs a message to the vehicle in charge of road surface maintenance, instructing it to perform road surface maintenance.

[0043] The road surface maintenance list stores the size and position information of the obstacle estimated by the determination unit 105 in association with each other, and stores the priority for road surface maintenance based on the size of the obstacle. The warning unit 106 outputs the priority for road surface maintenance based on the size of the obstacle estimated by the determination unit 105. For example, the larger the size of the obstacle, the higher the priority for road surface maintenance is set. The warning unit 106 outputs an instruction for a vehicle in charge of road surface maintenance to remove obstacles such as rocks on the road surface in accordance with the priority. The warning unit 106 also displays the priority on a map of a navigation system shared among a plurality of vehicles 50. Specifically, as an example, if the size of the obstacle is 15 cm or more, the warning unit 106 displays it in red on the map, and if it is less than 15 cm, it displays it in yellow on the map, that is, displays it in different colors according to the priority. Further, the warning unit 106 may display obstacle icons and the like displayed on the map in different sizes according to the priority. The warning unit 106 updates the road surface maintenance list so as to add the series of position information and the result of road surface maintenance to the list.

[0044] Further, when the size of the obstacle estimated by the determination unit 105 is equal to or larger than a first threshold, the warning unit 106 changes and outputs an alert level for each classification of obstacle size, and outputs position information of the obstacle to be added to the road surface maintenance list. In addition, the warning unit 106 monitors whether the distance between the vehicle 50 and the position information of the obstacle is decreasing. When the vehicle 50 approaches the obstacle by a predetermined value or more, the warning unit 106 outputs a warning to the driver of the vehicle 50. In addition, in the warning by the warning unit 106, it is possible to set in advance from which level of alert according to the classification of obstacle size the warning is to be executed. In the monitoring, for example, based on data obtained by learning a mine route on which the vehicle 50 travels, when the simple two-point distance between the vehicle 50 and the obstacle becomes equal to or less than a predetermined value, the path distance may be calculated, and when the path distance becomes equal to or less than the predetermined value, the warning unit 106 may output the warning.

[0045] The warning unit 106 also outputs an automatic report that includes inspection results, road surface maintenance results, image data from an on-board camera captured by a vehicle 50 that passed the location of the obstacle, and images of the scene when the falling object that constitutes the obstacle was dropped. For example, the obstacle detection device 10 acquires image data of obstacles and other objects captured during road surface maintenance, and the road surface maintenance results, analyzes the images, and the determination unit 105 further estimates the size of the obstacle. The obstacle detection device 10 saves the estimated size data in addition to the image data, and the warning unit 106 automatically creates and outputs a report that includes this series of information. For example, the obstacle detection device 10 acquires on-board camera footage from a vehicle 50 that subsequently passed the location of the obstacle stored in the road surface maintenance list, saves the image data of the obstacle, and the determination unit 105 analyzes the images to further estimate the size of the obstacle, saving the size data in addition to the images. The warning unit 106 automatically creates and outputs a report that includes this series of information. For example, the obstacle detection device 10 acquires onboard camera footage from a vehicle 50 that passed before the location of an obstacle stored in the road surface maintenance list, saves the image data of the obstacle, determines whether it corresponds to the vehicle 50 that dropped the obstacle, and if so, saves the scene of the fall as an image, along with the vehicle 50's information, time, and location. The warning unit 106 automatically creates and outputs a report containing this series of information.

[0046] Furthermore, if the determination unit 105 determines that there is no risk, the modification unit 104 returns the sampling frequency of the tire internal pressure data from the second sampling frequency to the first sampling frequency.

[0047] When an obstacle on the road surface is detected, the notification unit 107 sends a notification encouraging other vehicles located around the vehicle 50 to avoid the obstacle. Furthermore, when an obstacle on the road surface is detected, the notification unit 107 may also send a notification urging a specific vehicle capable of removing the obstacle (for example, a vehicle in charge of road surface maintenance) to remove the obstacle. Specifically, for example, when the determination unit 105 determines that there is no risk, and the vehicle 50 comes into contact with the obstacle and is damaged by the obstacle, the notification unit 107 may send a notification encouraging other vehicles located around the vehicle 50 to avoid the obstacle. At this time, the notification unit 107 may also send a notification urging a specific vehicle capable of removing the obstacle to remove the obstacle.

[0048] FIG. 5 is a diagram schematically showing another example of the relationship between the vehicle 50 and the obstacle D3 according to the present embodiment. The example of FIG. 4 described above illustrates the case where the state of the obstacle D2 does not change significantly even after the tire runs over the obstacle D2, whereas the example of FIG. 5 illustrates the case where the state of the obstacle D3 changes significantly after the tire runs over the obstacle D3. In the example of FIG. 5, the vehicle 50 is provided with an imaging device 31 at the rear of the vehicle, and the imaging device 31 can capture images of the road surface condition after the vehicle 50 has passed the obstacle D3. Note that a three-dimensional scanner may be provided at the rear of the vehicle instead of the imaging device 31, or together with the imaging device 31.

[0049] In this case, the second acquisition unit 102 acquires road surface condition data including the state of the obstacle D3 when the obstacle D3 on the road surface is detected and the vehicle 50 has passed the obstacle D3. The road surface condition data is acquired from the imaging device 31, for example. The notification unit 107 sends a notification encouraging other vehicles located around the vehicle 50 to avoid the obstacle D3 in accordance with the state of the obstacle D3 obtained from the road surface condition data.

[0050] In other words, as shown in Figure 5, if a tire drives over obstacle D3, it is possible that obstacle D3 may be damaged or embedded in the ground. In this case, obstacle D3 will no longer affect the vehicle's movement, so sending avoidance notices to other vehicles or requests to road maintenance personnel to remove it would be pointless. Therefore, it is necessary to determine whether there is a risk regarding the condition of obstacle D3, and if there is a risk, avoidance notices should be sent to other vehicles; if there is no risk, no avoidance notices should be sent to other vehicles. Alternatively, if there is a risk, a request for removal should be sent to road maintenance personnel; if there is no risk, no request for removal should be sent to road maintenance personnel.

[0051] Next, the operation of the obstacle detection device 10 according to this embodiment will be described with reference to Figures 6 to 10.

[0052] Figure 6 is a flowchart showing the flow of the obstacle detection process. Figure 7 is a flowchart showing the flow of the inspection instruction process. Figure 8 is a flowchart showing the flow of the road surface maintenance instruction process. Figure 9 is a flowchart showing the flow of the warning process. Figure 10 is a flowchart showing the flow of the report process. The CPU 11 of the obstacle detection device 10 executes the obstacle detection process shown in Figure 6, the inspection instruction process shown in Figure 7, the road surface maintenance instruction process shown in Figure 8, the warning process shown in Figure 9, and the report process shown in Figure 10. Each process in the obstacle detection device 10 is executed by the CPU 11 functioning as a first acquisition unit 101, a second acquisition unit 102, a detection unit 103, a modification unit 104, a determination unit 105, a warning unit 106, and a notification unit 107. The obstacle detection device 10 executes the obstacle detection process and, as the process of step S28 of the obstacle detection process, executes at least one of the inspection instruction process, road surface maintenance instruction process, warning process, and report process.

[0053] The obstacle detection process shown in Figure 6 will now be explained. In step S10 of Figure 6, the CPU 11 acquires tire pressure data from the moving vehicle 50 at a first sampling frequency.

[0054] In step S12, the CPU 11 acquires road surface condition data from the moving vehicle 50 and monitors the road surface condition based on the acquired road surface condition data.

[0055] In step S14, the CPU 11 determines whether or not it has detected an obstacle on the road surface based on the road surface condition data acquired in step S12. If the CPU 11 determines that it has detected an obstacle on the road surface (step S14: YES), it proceeds to step S16. If it determines that it has not detected an obstacle on the road surface (step S14: NO), it remains in standby mode in step S14.

[0056] In step S16, the CPU 11 changes the sampling frequency of the tire pressure data to a second sampling frequency that is higher than the first sampling frequency. Note that the change in sampling frequency in step S16 may be performed after determining that there is a risk in step S20, which will be described later.

[0057] In step S18, the CPU 11 estimates the size of the obstacle detected in step S14.

[0058] In step S20, the CPU 11 determines whether there is a risk related to the obstacle based on the size of the obstacle estimated in step S18. If the CPU 11 determines that there is a risk (step S20: YES), the process proceeds to step S22; if it determines that there is no risk (step S20: NO), the process proceeds to step S30.

[0059] In step S22, the CPU 11 outputs an alert prompting the user to avoid the obstacle.

[0060] In step S24, the CPU 11 analyzes the tire pressure data acquired at the second sampling frequency.

[0061] In step S26, the CPU 11 determines whether there is a fluctuation in the tire pressure above a certain level, based on the results of analyzing the tire pressure data in step S24. If the CPU 11 determines that there is a fluctuation in the tire pressure above a certain level (step S26: YES), it proceeds to step S28. If it determines that there is no fluctuation in the tire pressure above a certain level (step S26: NO), it returns to step S10 and repeats the process. In step S26, the CPU determines whether there is a fluctuation in the tire pressure above a certain level, such as when the tire drives over an obstacle.

[0062] In step S28, the CPU 11 outputs an alert and executes at least one of the following: inspection instruction processing, road surface maintenance instruction processing, warning processing, and report processing, and then returns to step S10 to repeat the process.

[0063] Meanwhile, in step S30, the CPU 11 changes the sampling frequency of the tire pressure data back from the second sampling frequency to the first sampling frequency.

[0064] In step S32, the CPU 11 determines whether the vehicle 50 has come into contact with an obstacle and whether the vehicle 50 has been damaged by the obstacle. Damage caused by the obstacle may be determined, for example, based on user input or based on sensor data from various sensors. If the CPU 11 determines that damage has occurred (step S32: YES), the process proceeds to step S34. If it determines that no damage has occurred (step S32: NO), the process returns to step S10 and is repeated.

[0065] In step S34, the CPU 11 notifies other vehicles in the vicinity of vehicle 50 to avoid the obstacle, and then returns to step S10 to repeat the process. At this time, the CPU 11 may also notify specific vehicles that are capable of removing the obstacle to remove it.

[0066] The inspection instruction process shown in Figure 7 will now be explained. In step S102 of Figure 7, the CPU 11 acquires obstacle information. Obstacle information includes the results of estimating the size of the obstacle in step S18 of Figure 6.

[0067] In step S104, the CPU 11 determines whether the estimated size of the obstacle is greater than or equal to the first threshold. If the CPU 11 determines that the estimated size of the obstacle is greater than or equal to the first threshold (step S104: YES), the process proceeds to step S106. On the other hand, if the CPU 11 determines that the estimated size of the obstacle is not greater than or equal to the first threshold, i.e., less than the first threshold (step S104: NO), the process returns to step S10 in Figure 6.

[0068] In step S106, the CPU 11 determines whether the size of the obstacle is classified as "low risk". If the CPU 11 determines that the size of the obstacle is classified as "low risk" (step S106: YES), the process proceeds to step S108. On the other hand, if the CPU 11 determines that the size of the obstacle is not classified as "low risk" (step S106: NO), the process proceeds to step S122.

[0069] In step S108, the CPU 11 outputs a level 1 alert indicating that the size of the obstacle is classified as "low risk". For example, it notifies that the vehicle has run over an obstacle that is small in size.

[0070] In step S110, the CPU 11 stores tire information in the concentrated tire inspection list. As mentioned above, the concentrated tire inspection list is a list that stores information on tires that should be inspected intensively in addition to the regular inspection during periodic tire inspections.

[0071] In step S112, the CPU 11 determines whether to instruct the inspection of the tires on the list during the periodic tire inspection. If the CPU 11 determines to instruct the inspection of the tires on the list (step S112: YES), the process proceeds to step S116. On the other hand, if the CPU 11 determines not to instruct the inspection of the tires on the list (step S112: NO), the process proceeds to step S114.

[0072] In step S114, the CPU 11 instructs only normal inspection and returns.

[0073] In step S116, the CPU 11 instructs the inspection of the riding tires in the list during the normal inspection.

[0074] In step S118, the CPU 11 instructs the appropriate action based on the inspection results. For example, if the tires are damaged due to running over an obstacle, it instructs the system to repair them.

[0075] In step S120, the CPU 11 updates the tire central inspection list and returns. Specifically, the CPU 11 updates the list to add the inspection results and corresponding results to the tire central inspection list.

[0076] In step S122, the CPU 11 determines whether the size of the obstacle is classified as "medium risk". If the CPU 11 determines that the size of the obstacle is classified as "medium risk" (step S122: YES), the process proceeds to step S124. On the other hand, if the CPU 11 determines that the size of the obstacle is not classified as "medium risk" (step S122: NO), the process proceeds to step S134.

[0077] In step S124, the CPU 11 outputs a level 2 alert indicating that the size of the obstacle is classified as "medium risk". For example, it notifies that the obstacle is of moderate size and has been driven over.

[0078] In step S126, the CPU 11 stores the tire information in the centralized tire inspection list.

[0079] In step S128, CPU 11 contacts the person in charge of the initial tire check.

[0080] In step S130, as a process running in parallel with step S128, the CPU 11 instructs the vehicle 50 to go to the person in charge of the initial tire check.

[0081] In step S132, the CPU 11 determines whether a detailed inspection is necessary. Specifically, if a tire checker or the like determines that a detailed inspection is necessary, and the determination result is input from the input device via the input / output I / F 15 of the obstacle detection device 10, the CPU 11 determines that a detailed inspection is necessary. If the CPU 11 determines that a detailed inspection is necessary (step S132: YES), the process proceeds to steps S138 and S140. On the other hand, if the CPU 11 determines that a detailed inspection is not necessary (step S132: NO), the process proceeds to step S120.

[0082] In step S134, the CPU 11 outputs a level 3 alert indicating that the size of the obstacle is classified as "high risk." For example, it notifies that the obstacle is large and has been driven over.

[0083] In step S136, the CPU 11 stores the tire information in the centralized tire inspection list.

[0084] In step S138, CPU 11 contacts the tire manager at the tire shop.

[0085] In step S140, as a process running in parallel with step S138, the CPU 11 instructs the vehicle 50 to go to the tire shop.

[0086] In step S142, the CPU 11 determines whether or not tire removal is necessary after the detailed inspection. Specifically, if the tire manager or the like determines that the tire needs to be removed, and the determination result is input from the input device via the input / output I / F 15 of the obstacle detection device 10, the CPU 11 determines that tire removal is necessary. If the CPU 11 determines that tire removal is necessary (step S142: YES), the process proceeds to step S144. On the other hand, if the CPU 11 determines that tire removal is not necessary (step S142: NO), the process proceeds to step S120.

[0087] In step S144, the CPU 11 instructs the system to change the tires, and the system proceeds to step S120.

[0088] The road surface maintenance instruction process shown in Figure 8 will now be explained. In step S202 of Figure 8, the CPU 11 acquires obstacle information.

[0089] In step S204, the CPU 11 determines whether the estimated size of the obstacle is greater than or equal to the first threshold. If the CPU 11 determines that the estimated size of the obstacle is greater than or equal to the first threshold (step S204: YES), the process proceeds to step S206. On the other hand, if the CPU 11 determines that the estimated size of the obstacle is not greater than or equal to the first threshold, i.e., less than the first threshold (step S204: NO), the process returns to step S10 in Figure 6.

[0090] In step S206, the CPU 11 determines whether the size of the obstacle is classified as "low risk". If the CPU 11 determines that the size of the obstacle is classified as "low risk" (step S206: YES), the process proceeds to step S208. On the other hand, if the CPU 11 determines that the size of the obstacle is not classified as "low risk" (step S206: NO), the process proceeds to step S210.

[0091] In step S208, the CPU 11 outputs a level 1 alert, and the process proceeds to step S216.

[0092] In step S210, the CPU 11 determines whether the size of the obstacle is classified as "medium risk". If the CPU 11 determines that the size of the obstacle is classified as "medium risk" (step S210: YES), the process proceeds to step S212. On the other hand, if the CPU 11 determines that the size of the obstacle is not classified as "medium risk" (step S210: NO), the process proceeds to step S214.

[0093] In step S212, the CPU 11 outputs a level 2 alert, and the process proceeds to step S216.

[0094] In step S214, the CPU 11 outputs a level 3 alert, and the process proceeds to step S216.

[0095] In step S216, the CPU 11 saves GPS information to the road surface maintenance list. Specifically, the CPU 11 stores the position information of the vehicle 50 when the tire that was driving over the obstacle into the road surface maintenance list, which is a list of candidate locations for road surface maintenance where the vehicle 50 is driving.

[0096] In step S218, the CPU 11 contacts the vehicle responsible for road surface maintenance.

[0097] In step S220, the CPU 11 instructs the assigned vehicle to remove obstacles such as rocks from the road surface according to priority.

[0098] In step S222, the CPU 11 updates the list and returns. Specifically, the CPU 11 updates the list to add the location information of the vehicle 50 at the time of the runaway and the results of the road surface maintenance to the road surface maintenance list.

[0099] The warning process shown in Figure 9 will now be explained. In step S302 of Figure 9, the CPU 11 acquires obstacle information.

[0100] In step S304, the CPU 11 determines whether the estimated size of the obstacle is greater than or equal to the first threshold. If the CPU 11 determines that the estimated size of the obstacle is greater than or equal to the first threshold (step S304: YES), the process proceeds to step S306. On the other hand, if the CPU 11 determines that the estimated size of the obstacle is not greater than or equal to the first threshold, i.e., less than the first threshold (step S304: NO), the process returns to step S10 in Figure 6.

[0101] In step S306, the CPU 11 determines whether the size of the obstacle is classified as "low risk". If the CPU 11 determines that the size of the obstacle is classified as "low risk" (step S306: YES), the process proceeds to step S308. On the other hand, if the CPU 11 determines that the size of the obstacle is not classified as "low risk" (step S306: NO), the process proceeds to step S310.

[0102] In step S308, the CPU 11 outputs a level 1 alert, and the process proceeds to step S316.

[0103] In step S310, the CPU 11 determines whether the size of the obstacle is classified as "medium risk". If the CPU 11 determines that the size of the obstacle is classified as "medium risk" (step S310: YES), the process proceeds to step S312. On the other hand, if the CPU 11 determines that the size of the obstacle is not classified as "medium risk" (step S310: NO), the process proceeds to step S314.

[0104] In step S312, the CPU 11 outputs a level 2 alert, and the process proceeds to step S316.

[0105] In step S314, the CPU 11 outputs a level 3 alert, and the process proceeds to step S316.

[0106] In step S316, the CPU 11 automatically adds the location information of obstacles such as fallen objects to the monitoring list. Specifically, regardless of whether the tire has driven over the obstacle or not, the CPU 11 adds the location information of the obstacle to the monitoring list, either the location information of the vehicle 50 at the time the obstacle was assessed as a risk, or the estimated location of the obstacle obtained through image analysis.

[0107] In step S318, the CPU 11 monitors whether the distance between the vehicle 50 and the positional information of obstacles such as fallen objects is decreasing.

[0108] In step S320, the CPU 11 determines whether the vehicle 50 has come closer to an obstacle such as a fallen object than a predetermined value. If the CPU 11 determines that the vehicle 50 has come closer to the fallen object than a predetermined value (step S320: YES), the process proceeds to step S322. On the other hand, if the CPU 11 determines that the vehicle 50 has not come closer to the fallen object than a predetermined value (step S320: NO), the process returns to step S10 in Figure 6.

[0109] In step S322, the CPU 11 outputs a warning to the driver of the vehicle 50 and returns. Specifically, the CPU 11 notifies the driver that the vehicle 50 is close to an obstacle and urges caution. If the vehicle 50 is an autonomous vehicle, the CPU 11 may notify the vehicle 50 that it is close to an obstacle and also control the direction and speed of the vehicle 50.

[0110] The report processing shown in Figure 10 will now be explained. In step S402 of Figure 10, the CPU 11 acquires obstacle information.

[0111] In step S404, the CPU 11 determines whether the estimated size of the obstacle is greater than or equal to the first threshold. If the CPU 11 determines that the estimated size of the obstacle is greater than or equal to the first threshold (step S404: YES), the process proceeds to step S406. On the other hand, if the CPU 11 determines that the estimated size of the obstacle is not greater than or equal to the first threshold, i.e., less than the first threshold (step S404: NO), the process returns to step S10 in Figure 6.

[0112] In step S406, the CPU 11 outputs an alert, and the process proceeds to steps S408, S412, and S418. As described above, the alert may be divided into risk levels according to the size classification of the obstacles, and the content of the alert may be changed for each level.

[0113] In step S408, the CPU 11 stores the tire information in the centralized tire inspection list.

[0114] In step S410, the CPU 11 obtains the inspection result and proceeds to step S436.

[0115] In step S412, as a process running in parallel with step S408, the CPU 11 contacts the vehicle responsible for road surface maintenance.

[0116] In step S414, the CPU 11 acquires image data of obstacles and other objects captured during road surface maintenance.

[0117] In step S416, the CPU 11 obtains the road surface maintenance results and proceeds to step S424.

[0118] In step S418, as a process in parallel with step S412, the CPU 11 automatically adds location information of obstacles such as falling objects to the list.

[0119] In step S420, the CPU 11 acquires onboard camera footage of a vehicle that subsequently passed the location in question.

[0120] In step S422, the CPU 11 saves the image data of the in-vehicle camera and proceeds to step S424.

[0121] In step S424, the CPU 11 further estimates the size of the obstacle through image analysis.

[0122] In step S426, the CPU 11 saves the image along with the size data, and then proceeds to step S436.

[0123] In step S428, as a process in parallel with step S420, the CPU 11 acquires the onboard camera image of the vehicle 50 that passed the location in front of it.

[0124] In step S430, the CPU 11 saves the image data of the in-vehicle camera.

[0125] In step S432, the CPU 11 determines whether or not the vehicle 50 that dropped the object is the vehicle 50 that dropped the object. If the CPU 11 determines that the vehicle 50 is the vehicle that dropped the object (step S432: YES), the process proceeds to step S434. On the other hand, if the CPU 11 determines that the vehicle is not the vehicle that dropped the object (step S432: NO), the process returns to step S428.

[0126] In step S434, the CPU 11 saves the fall scene as an image, and also saves vehicle information, time, and location, and then proceeds to step S436.

[0127] In step S436, the CPU 11 outputs an automatic report and returns. Specifically, the CPU 11 outputs an automatic report that includes inspection results, road surface maintenance results, image data from on-board camera footage captured by the vehicle 50 as it passed the location of the obstacle, and images of the scene where the falling object that constitutes the obstacle was dropped.

[0128] In addition, during the report processing shown in Figure 10, the loading of the cargo onto the vehicle 50 using a loader / excavator, etc., may be saved as image data. In this case, an image of the cargo loading immediately after loading may be extracted, and the loading configuration may be analyzed from the image to determine cargo imbalance, overloading, etc. The analysis results of the loading configuration may be saved and an automatic report may be output.

[0129] Thus, according to this embodiment, the sampling frequency of tire pressure data can be lowered before detecting an obstacle using road surface condition data while the vehicle is in motion, and increased after detecting an obstacle. As a result, obstacles on the road surface can be detected with high accuracy while reducing the load required for data analysis of tire pressure data.

[0130] Furthermore, by detecting obstacles based on road surface data while the vehicle is in motion and issuing alerts, it is possible to avoid obstacles before they are driven over. This reduces the damage to the tires caused by driving over obstacles.

[0131] Furthermore, by estimating the size of obstacles from road surface condition data and determining the risk, it becomes unnecessary to constantly avoid obstacles, thus reducing the burden on the driver.

[0132] Furthermore, in the above embodiments, the processor referred to as CPU 11 refers to a broad type of processor, including general-purpose processors (e.g., CPUs) and dedicated processors (e.g., GPUs: Graphics Processing Units, ASICs: Application Specific Integrated Circuits, FPGAs: Field Programmable Gate Arrays, programmable logic devices, etc.).

[0133] Furthermore, the operation of the processor in the above embodiment may not be performed by a single processor, but may be performed by multiple processors located in physically separate locations working together. Also, the order of the processor's operations is not limited to the order described in the above embodiment, but may be changed as appropriate.

[0134] Furthermore, although the obstacle detection system 1 in this embodiment is described as being composed of multiple devices as an example, it may also be composed of a single device that has some of the functions of multiple devices.

[0135] Furthermore, the processing performed by the obstacle detection device 10 according to the above embodiment may be performed by software, by hardware, or by a combination of both. Also, the processing performed by the obstacle detection device 10 may be stored as a program on a storage medium and distributed.

[0136] The obstacle detection program described herein is available as a program product. A program product includes any form of product for providing a program. For example, a program product includes a program provided via a network such as the Internet, and non-temporary computer-readable recording media such as CD-ROMs and DVDs on which the program is stored.

Claims

1. An obstacle detection device comprising: a first acquisition unit that acquires tire pressure data representing changes in the internal pressure of the tire while the vehicle is traveling on the road surface from an air pressure sensor mounted on the tire of the vehicle; a second acquisition unit that acquires road surface condition data representing the condition of the road surface while the vehicle is traveling on the road surface from at least one of a camera and a three-dimensional scanner mounted on the vehicle; and a detection unit that detects obstacles on the road surface based on at least one of the tire pressure data and the road surface condition data.

2. The obstacle detection device according to claim 1, further comprising a modification unit that, when an obstacle on the road surface is detected, changes the sampling frequency of the tire internal pressure data from a first sampling frequency before the detection of the obstacle on the road surface to a second sampling frequency higher than the first sampling frequency.

3. An obstacle detection device according to claim 1, further comprising: a determination unit that, when it detects an obstacle on the road surface, estimates the size of the obstacle and determines whether or not there is a risk related to the obstacle based on the estimated size of the obstacle; and a warning unit that issues a warning related to the risk when the determination unit determines that there is a risk.

4. The obstacle detection device according to claim 3, wherein the warning unit issues a warning regarding the risk when the determination unit determines that the risk exists and detects a fluctuation of the tire internal pressure data above a certain level.

5. The obstacle detection device according to claim 3, wherein the determination unit determines the risk at multiple levels, and the warning unit issues different warnings regarding the risk according to each of the multiple levels.

6. An obstacle detection device according to claim 1, further comprising: a modification unit that, when an obstacle on the road surface is detected, changes the sampling frequency of the tire internal pressure data from a first sampling frequency before the detection of the obstacle on the road surface to a second sampling frequency higher than the first sampling frequency; and a determination unit that, when an obstacle on the road surface is detected, estimates the size of the obstacle and determines whether or not there is a risk related to the obstacle based on the estimated size of the obstacle, wherein if the determination unit determines that there is no risk, the modification unit returns the sampling frequency of the tire internal pressure data from the second sampling frequency to the first sampling frequency.

7. The obstacle detection device according to claim 1, further comprising a notification unit that, when it detects the obstacle on the road surface, notifies other vehicles in the vicinity of the vehicle to encourage them to avoid the obstacle.

8. The obstacle detection device according to claim 1, further comprising a second acquisition unit that detects the obstacle on the road surface, acquires road surface condition data including the state of the obstacle when the vehicle passes over the obstacle, and notifies other vehicles in the vicinity of the vehicle to avoid the obstacle according to the state of the obstacle obtained from the road surface condition data.

9. The obstacle detection device according to claim 1, further comprising a notification unit that, when it detects the obstacle on the road surface, notifies a specific vehicle capable of removing the obstacle to prompt it to remove the obstacle.

10. An obstacle detection method comprising: acquiring tire pressure data representing changes in the internal pressure of a tire from an air pressure sensor mounted on the tire of a vehicle while the vehicle is traveling on a road surface; acquiring road surface condition data representing the condition of the road surface from at least one of a camera and a three-dimensional scanner mounted on the vehicle while the vehicle is traveling on the road surface; and having a computer perform a process to detect an obstacle on the road surface based on at least one of the tire pressure data and the road surface condition data.

11. An obstacle detection program that causes a computer to perform a process of detecting obstacles on the road surface, which involves acquiring tire pressure data representing changes in the internal pressure of the tires from an air pressure sensor mounted on the tires of the vehicle while the vehicle is traveling on the road surface, acquiring road surface condition data representing the condition of the road surface from at least one of a camera and a three-dimensional scanner mounted on the vehicle while the vehicle is traveling on the road surface, and based on at least one of the tire pressure data and the road surface condition data.

Images