JP2016024685A - Work vehicle for mine - Google Patents

Abstract

A work vehicle for mining that detects an area where an obstacle is likely to exist at the tip of dust in an off-road environment. A stereo camera device 11 including cameras 11a and 11b, a two-dimensional camera acquired image acquired by the camera 11a, and a two-dimensional camera acquired image acquired by the camera 11b are compared, and these two are compared. An image information acquisition unit that identifies a region without parallax between two camera acquired images as a non-parallax region, and when the number of pixels in the non-parallax region in the parallax image is greater than a predetermined value, the other vehicle 1A exists as a non-parallax region The vehicle possibility area | region with high possibility of performing and the vehicle presence area | region detection part which identifies are provided. [Selection] Figure 1

Description

  The present invention relates to a mining work vehicle such as an off-road dump truck.

  Generally, in mines, construction machines such as hydraulic excavators and dump trucks are used for mining and transporting earth and sand. In particular, dump trucks used in mines are required to operate efficiently because the amount of sediment transported per unit time is directly related to the progress of mining.

  However, since the mine traveling road where the dump truck runs is off-road and has many bad roads, there is a concern about collisions with obstacles such as dirt walls and other vehicles compared to ordinary paved roads. If an obstacle occurs on the travel path and the dump truck comes into contact with the obstacle and stops, the subsequent dump truck is prevented from traveling and the operation of the mine is stopped for a long time. Therefore, in order to efficiently transport a large amount of earth and sand outside the mining site, it is necessary to detect obstacles on the front vehicle and the traveling path at an early stage, and to follow the front vehicle and avoid obstacles. is there.

  As a system for detecting a preceding vehicle or an obstacle, an obstacle detection device using a millimeter wave radar, a laser sensor, a stereo camera, or the like is used. Among these, since the stereo camera can acquire three-dimensional image information and can measure a three-dimensional shape from the acquired image information, the traveling road surface and the obstacle can be distinguished.

  However, since the mine travel path is off-road, when the dump truck travels on the travel path, it is assumed that the heavy truck dusts up to the travel path so that the front vehicle cannot be visually confirmed. . Then, due to the rolled up dust, some or all of obstacles such as other vehicles to be detected are hidden from the image information acquired by the stereo camera. In this case, there is a possibility that the image information necessary for detecting the obstacle cannot be obtained and the obstacle cannot be detected.

  An image processing apparatus that takes this type of dust into account is disclosed in, for example, Japanese Patent Application Laid-Open No. H10-228707. In this Patent Document 1, a dust area such as dust is detected in image information based on the brightness of each pixel in image information acquired by a camera, and the brightness difference in the detected dust area is reduced. People who have become difficult to see are displayed on the display with image processing.

JP 2013-222254 A

  The dust rolled up by the dump truck traveling on the mine traveling path has a higher concentration than the naturally occurring dust, and when the image processing apparatus according to Patent Document 1 is used, the image information acquired by the camera. Although it is possible to detect the dust area in the interior, it is not possible to identify the area where the brightness difference in the dust area is reduced, and it is not easy to accurately identify the area where there is a high possibility that an obstacle will exist at the tip of the dust. Absent.

  The present invention has been made from the above-described state of the art, and the object thereof is a mining work vehicle capable of detecting an area where an obstacle is likely to exist at the tip of dust in an off-road environment. Is to provide.

  In order to achieve this object, a mining work vehicle according to the present invention includes a stereo camera device including first and second cameras capable of acquiring image information in front of a traveling direction, and the first camera. A difference no-area specifying unit that compares the acquired two-dimensional image information with the two-dimensional image information acquired by the second camera, and specifies an area having no difference between the image information as a no-difference area; An obstacle region specifying unit that specifies the difference-free region as an obstacle region that is highly likely to have an obstacle when the number of pixels of the difference-free region in the image information is larger than a predetermined value. It is characterized by that.

  In general, dust becomes a region where there is no difference between the image information of the first and second cameras of the stereo camera device. Therefore, the present invention is, for example, that the mining work vehicle is hidden by dust that has been rolled up rearward when the mining work vehicle travels, and is visually confirmed from the mining work vehicle that runs behind the mining work vehicle. Even if this is not possible, the two-dimensional image information acquired by the first and second cameras of the stereo camera device is compared with the area where the dust is generated, and there is no difference between the image information. Can be specified. Also, the dust that the mining work vehicle whirls when traveling is higher in concentration than the naturally occurring dust and has a larger generation range than the size of the mining work vehicle. Therefore, when the number of pixels of the difference-free area between the image information acquired by the first and second cameras is larger than the predetermined value, there is a possibility that an obstacle, particularly a mining work vehicle exists in the difference-free area. Can be identified as a high obstacle area. Therefore, it is possible to detect an area where there is a high possibility that an obstacle exists in front of the dust.

  According to the present invention, in the above invention, a road surface area specifying unit that specifies a road surface area in the image information based on the image information acquired by the first and second cameras, and the difference in the image information is omitted. A dust region specifying unit that specifies the non-difference region as a dust region when the region is in contact with the road surface region, wherein the obstacle region specifying unit has a predetermined number of pixels of the dust region in the image information; When the value is larger than the value, the dust area is specified as an obstacle area.

  In general, when a no-difference area is in contact with a road surface area in image information, there is a high possibility of an obstacle traveling ahead, for example, dust that a mining work vehicle lifts up during traveling. Therefore, the present invention can identify a dust region that hides an obstacle located in front by specifying a dust region when the no-difference region is in contact with the road surface region, and there is an obstacle ahead of the dust region. A highly likely area can be detected more accurately.

  Moreover, the present invention provides the obstacle detection unit for detecting an obstacle present on the road surface area in the image information, the obstacle information detected by the obstacle detection unit, and the obstacle area in the above-described invention. An obstacle identifying unit that identifies the type, size, and position of the obstacle based on the position and the number of pixels of the obstacle region in the image information identified by the identifying unit. It is said.

  In the present invention configured as described above, obstacle information on the road surface area is detected by the obstacle detection unit, and the position and the number of pixels of the obstacle area in the image information specified by the obstacle area specification unit are By adding the obstacle information on the road surface area detected by the object detection unit, the type, size, and position of the obstacle can be identified.

  Also, in the present invention according to the invention described above, the obstacle detection unit detects the obstacle present on the road surface area in the image information based on the radar and measurement information measured by the radar. Part.

  The present invention configured as described above can detect the position of an obstacle hidden by dust through the dust by using the radar. Therefore, based on the measurement information measured by the radar, obstacles existing on the road surface area in the image information are detected and used as obstacle information by the obstacle detection unit. The obstacle information can be detected with higher accuracy.

  Moreover, the present invention provides the obstacle detection unit for detecting an obstacle present on the road surface area in the image information, the obstacle information detected by the obstacle detection unit, and the obstacle area in the above-described invention. A dust discriminating unit that discriminates whether or not the non-difference area is dust generated by driving of the vehicle based on the position and the number of pixels of the obstacle region in the image information specified by the specifying unit. It is characterized by that.

  Whether the dust is generated by traveling by the vehicle can be determined by the positional relationship between the road surface area and the obstacle area in the image information and the size thereof, that is, the number of pixels. Therefore, the present invention is based on the obstacle information detected by the obstacle detection unit and the position and the number of pixels of the obstacle region in the image information specified by the obstacle region specification unit. The cause of the dust corresponding to the non-difference area can be specified with high accuracy by discriminating whether or not the dust is generated in step S2. Therefore, since the dust covering the vehicle can be identified with higher accuracy, the region where the vehicle is likely to exist can be detected with higher accuracy.

  According to the present invention, in the above invention, a display device for displaying image information, image information acquired by one of the first and second cameras is displayed on the display device, and the obstacle And an image control unit that displays the obstacle region specified by the region specifying unit in an image displayed on the display device.

  The present invention configured as described above causes the display device to display the image information acquired by any one of the first and second cameras, and displays the obstacle region specified by the obstacle region specifying unit. By displaying in the image displayed on the screen, it is possible to confirm a region where there is a high possibility that an obstacle hidden by the dust is present in addition to the position where the dust is generated and the number of pixels.

  According to the present invention, in the above invention, the loading platform, a loading mass detection unit that detects the mass of the load loaded on the loading platform, a speed detection unit that detects a traveling speed, and the first and second cameras. A distance calculating unit that calculates a distance to the non-difference region in the image information based on the acquired image information, and the obstacle region specifying unit is a mass of the load detected by the load mass detecting unit The predetermined value is changed according to the traveling speed detected by the speed detecting unit and the distance to the no-difference area calculated by the distance calculating unit.

  In a mine, a plurality of mining work vehicles are traveling with a predetermined space between them, and the state of other vehicles traveling ahead can often be determined from the state of the host vehicle. The number of pixels of dust in the image information acquired by the first and second cameras in accordance with changes in the mass of the cargo loaded on the loading platform of the mining work vehicle, the traveling speed, and the distance to the obstacle Will change relatively. Therefore, according to the mass of the load detected by the load mass detection unit, the traveling speed detected by the speed detection unit, and the distance to the no-difference region calculated by the distance calculation unit, the obstacle region specifying unit By changing the predetermined value when specifying the area, the obstacle area can be specified with higher accuracy by the obstacle area specifying unit.

  Also, in the present invention according to the above-described invention, the obstacle region specifying unit may obtain images acquired by the first and second cameras when the number of pixels of the no-difference region in the image information is smaller than the predetermined value. The obstacle region is specified from the information.

  According to the present invention configured as described above, when the number of pixels in the non-difference area in the image information is smaller than a predetermined value, the front obstacle is not in the first and second cameras. Can be identified from the acquired image information. Therefore, the present invention specifies an obstacle area from the image information acquired by the first and second cameras when the number of pixels in the difference-free area in the image information is smaller than a predetermined value. As a result, the obstacle area can be specified regardless of the size of the number of pixels in the non-difference area in the image information.

  According to the present invention, two-dimensional image information acquired by the first and second cameras of the stereo camera device is compared, and an area where dust is generated can be specified by a difference-free area between the image information. Moreover, since the dust generated when the mining work vehicle travels has a wider generation range than the size of the mining work vehicle, if the number of pixels in the no-difference area is larger than a predetermined value, an obstacle exists in the no-difference area. It is possible to identify an obstacle region with a high possibility, and it is possible to detect a region where an obstacle is likely to exist ahead of the dust. Problems, configurations, and effects other than those described above will be made clear from the following description of embodiments.

It is the schematic which shows the condition which is detecting the other mine work vehicle which drive | works ahead over dust in the mine work vehicle which concerns on 1st Embodiment of this invention. It is the schematic which shows the mine system where the said working vehicle for mines is used. It is the schematic which shows the drive structure of the said working vehicle for mines. It is the schematic which shows the obstacle detection system mounted in the said working vehicle for mines. It is a figure which shows the camera acquisition image imaged with the 1st and 2nd camera of the said stereo camera apparatus, (a) is a camera acquisition image of a 1st camera, (b) is a camera acquisition image of a 2nd camera. is there. It is a figure which shows the parallax image calculated from the two-dimensional camera acquisition image shown to Fig.5 (a) and FIG.5 (b). It is the schematic which shows the display image of the display apparatus of the said obstacle detection system. It is a flowchart which shows the process by the said obstacle detection system. It is explanatory drawing which shows the specific method in which the predetermined | prescribed object in the camera acquisition image acquired with the 1st camera of the said obstacle detection system is imaged in the camera acquisition image acquired with the 2nd camera. . It is a figure which shows the image for dust detection produced with the said obstacle detection system. It is explanatory drawing which shows the detection method of the dust width by the said obstacle detection system. It is a table | surface which shows the conversion table of the threshold value of the dust width detected in the dust position detection part of the said obstacle detection system. It is the schematic which shows the working vehicle for mines which concerns on 2nd Embodiment of this invention. It is the schematic which shows the obstacle detection system mounted in the said working vehicle for mines. It is a flowchart which shows the process by the said obstacle detection system.

  EMBODIMENT OF THE INVENTION Hereinafter, the form for implementing the obstacle detection system used for the working vehicle for mines which concerns on this invention is demonstrated based on figures.

[First Embodiment]
The first embodiment is an embodiment using a stereo camera device 11. FIG. 1 shows an off-road dump truck that is an example of a mining work vehicle according to the first embodiment of the present invention, and travels forward through dust D in a mining dump truck that is a vehicle 1 for construction machinery. It is the schematic which shows the condition which is detecting 1 A of other vehicles. FIG. 2 is a schematic diagram showing a mine system in which the vehicle 1 is used. FIG. 3 is a schematic diagram showing a drive configuration of the vehicle 1. FIG. 4 is a schematic diagram showing the obstacle detection system 10 mounted on the vehicle 1.

  5A and 5B are diagrams showing camera acquired images a and b acquired by the cameras 11a and 11b of the stereo camera device 11 of the vehicle 1. FIG. 5A is a camera acquired image a of the camera 11a, and FIG. 5B is a camera 11b. This is a camera acquired image b. FIG. 6 is a diagram showing the parallax image DI calculated by the image information acquisition unit 12 from the two-dimensional camera image information a and b shown in FIGS. 5 (a) and 5 (b). An image that has color information of an object in the parallax image DI and is viewed with the naked eye is called an appearance image. The appearance image includes color information corresponding to each pixel in the parallax image. FIG. 7 is a schematic diagram showing a display image on the display 19 of the obstacle detection system 10.

  As shown in FIG. 2, the vehicle 1 travels on a road surface R that is a travel path (conveyance path) under an off-road environment provided in advance in the mine. The vehicle 1 is a manned traveling type that is travel-controlled by an operator. In the mine, a plurality of vehicles 1 of the same model are traveling. As shown in FIG. 1, another vehicle 1 </ b> A having the same configuration as that of the vehicle 1 has a predetermined traveling interval (inter-vehicle distance) in front of the vehicle 1. Running. In addition, an information center 3 for transmitting and receiving predetermined information to and from the vehicles 1 and 1A is installed in the mine, and a hydraulic excavator 4 for loading a load such as earth and sand on the vehicle 1 is used. Yes. The road surface R is an obstacle detection area by the obstacle detection system 10, and as shown in FIGS. 5 (a), 5 (b) and 6, road shoulders S are provided on both sides in the width direction. Yes. Each road shoulder S is provided by a cut surface that is a cliff formed by embankment or cutting a mine with a hydraulic excavator 4 or the like. These embankments and cutting surfaces are formed at a predetermined height or higher, for example, a height of 1.5 m or higher.

  As shown in FIG. 1, the vehicle 1 includes a vehicle main body 1a, a driver's seat 1b provided on the upper front side of the vehicle main body 1a, a loading platform 1c as a working unit provided on the vehicle main body 1a so as to be raised and lowered, Left and right front wheels 1d and rear wheels 1e that support the vehicle main body 1a so as to travel are provided. On the front deck 1f of the vehicle main body 1a, a stereo camera device 11 for detecting the environment around the vehicle main body 1a, particularly in front of the traveling direction, is attached. The vehicle 1 detects an obstacle on the road surface R recognized by the stereo camera device 11, particularly the other vehicle 1A traveling ahead ahead. A GPS device 1g is attached to the front corner on the deck 1f. The GPS device 1g is a position detection unit for detecting the traveling position of the vehicle main body 1a.

  The stereo camera device 11 includes a pair of cameras 11a and 11b that can acquire image information in front of the traveling direction of the vehicle 1, and uses these two first and second cameras 11a and 11b to create a stereoscopic image of the outside world. Acquire 3D image information. The three-dimensional image information includes up to the imaging target calculated from the difference between the two-dimensional camera acquired images a and b shown in FIGS. 5A and 5B detected by the two cameras 11a and 11b. The disparity image DI having the distance information and the appearance image are included. The appearance image has color information corresponding to each pixel in the parallax image DI.

  The stereo camera device 11 is attached so that the center between the left and right cameras 11a and 11b is located at the center position that is the center portion in the left-right direction on the front side of the vehicle body 1a. Internal parameters such as focal lengths and lens distortions of the cameras 11a and 11b, and external parameters indicating the positional relationship and the installation position on the vehicle body are synchronized with each other. Each camera 11a, 11b is directed to the front of the vehicle 1 so that each optical axis is parallel, and as shown in FIGS. 5 (a) and 5 (b), the imaging regions 11c, It is installed so that a part of 11d may overlap.

  As shown in FIG. 3, the vehicle 1 includes a travel drive device 5. The travel drive device 5 drives an engine 5a as a power source, a generator 5b driven by the engine 5a, a power control device 5c to which power generated by the generator 5b is supplied, and a rear wheel 1e. And a travel motor 5d. The power supplied to the travel motor 5d is controlled from the power control device 5c. The power control device 5 c is controlled by a controller 6 mounted on the vehicle 1.

  The controller 6 controls the drive of a steering motor 5e for steering the vehicle main body 1a and a brake device 5f such as a retarder brake for braking the vehicle main body 1a via the power control device 5c. It is. The vehicle body 1a is equipped with a vehicle-mounted wireless device 5g capable of communicating with the information center 3 for wireless transmission of a predetermined signal. The controller 6 includes a storage unit 6a in which map information related to at least a travel area in which the vehicle 1 travels is stored in advance. The map information stored in the storage unit 6a includes information on the road surface R and the shoulder S on which the vehicle 1 travels.

  Further, the vehicle body 1a includes a vehicle speed sensor 5h for detecting the rotational speed of a predetermined wheel such as the front wheel 1d, and a load sensor 5i for detecting the mass of the load loaded on the loading platform 1c, that is, a load. It is attached. The vehicle speed sensor 5h acquires the rotational speed information of the front wheels 1d and outputs it to the controller 6, and the controller 6 calculates the traveling speed of the vehicle 1. The load sensor 5i is attached to, for example, a suspension (not shown) that supports the front wheel 1d and the rear wheel 1e, and detects a displacement amount of the suspension based on a mass change of the load loaded on the loading platform 1c. The load sensor 5 i outputs the detected displacement amount to the controller 6, and the mass of the load loaded on the loading platform 1 c is detected by the controller 6 based on the displacement amount.

<Obstacle detection system>
The obstacle detection system 10 is an obstacle recognition device for recognizing an obstacle hidden by the dust D generated on the road surface R, for example, another vehicle 1A traveling ahead, which is executed by processing by the controller 6. The controller 6 is also a stereo camera system as an obstacle detection system 10 as an image analysis device for detecting obstacles such as work vehicles, rocks, minerals, earth and sand other than the other vehicle 1A on the road surface R, for example. .

  Specifically, the obstacle detection system 10 includes an image information acquisition unit 12 including a stereo camera device 11, a distance calculation unit 13, a travel path recognition unit 14, a dust detection unit 15, and a dust position detection unit 16. The vehicle presence area detection unit 17, the image control unit 18, and a display 19 as a display device are provided. These processes by the image information acquisition unit 12, the distance calculation unit 13, the travel path recognition unit 14, the dust detection unit 15, the dust position detection unit 16, the vehicle presence region detection unit 17, and the image control unit 18 are performed by the controller 6. Is called.

  The image information acquisition unit 12 includes camera acquisition images a and b shown in FIGS. 5A and 5B as image information detected by the cameras 11a and 11b, and FIGS. 5A and 5B. A parallax image DI calculated based on the difference in appearance of the camera acquisition images a and b shown in FIG. 5B is obtained. The parallax image DI is created on the basis of one of the camera acquired images a and b. For example, when the parallax image DI is created on the basis of the camera acquired image a in FIG. A search is made for where the position of the imaged object is located in the camera acquired image b in FIG. At this time, the position difference between the camera acquired images a and b is larger as the object is closer and smaller as the object is farther. Based on this difference, the three-dimensional position of the object photographed by each camera 11a, 11b is acquired. This technique is generally called stereo matching.

  The parallax image DI associates each area in the camera acquired image a acquired by the camera 11a with each area in the camera acquired image b acquired by the camera 11b, and in these camera acquired images a and b. The position of the target object is determined from the stereo camera device 11 as viewed. Therefore, the parallax image DI is obtained by obtaining the distance between each pixel on the two-dimensional image acquired by the cameras 11a and 11b.

  Furthermore, the image information acquisition unit 12 compares the two-dimensional camera acquisition image a acquired by the camera 11a with the two-dimensional camera acquisition image b acquired by the camera 11b, and determines between these camera acquisition images a and b. This is a difference-free area specifying unit that specifies a difference-free area, that is, an area without parallax as a difference-free area (no parallax area). The image information acquisition unit 12 acquires three-dimensional image information (three-dimensional point group) that is three-dimensional distance information in each of the cameras 11 a and 11 b of the stereo camera device 11. The three-dimensional point group is point sequence information in which each point (pixel) in the detected image has information of (x, y, z, r, g, b). When the switch (not shown) for turning on / off the power supply of the obstacle detection system 10 is turned on, the image information acquisition unit 12 can acquire appearance images by the cameras 11a and 11b as initialization processing. At the same time, the calibration data of each camera 11a, 11b is read.

  The distance calculation unit 13 determines each point in the appearance image, that is, the imaging target at each pixel in the parallax image DI based on the three-dimensional image information detected by the image information acquisition unit 12, particularly the parallax image DI information. Get distance information. Therefore, the distance calculation unit 13 calculates the distance from the own vehicle 1 to the dust generation position P based on the distance information of each pixel in the parallax image DI.

  The travel path recognition unit 14 calculates a plane area in the parallax image DI based on a process in which the road surface R is configured with a plane with respect to the parallax image DI obtained by the image information acquisition unit 12, and Based on the relative positional relationship between the position and the mounting positions and mounting angles of the cameras 11 a and 11 b of the stereo camera device 11, the plane area is set as a main plane constituting the road surface R. Further, the traveling road recognition unit 14 is a road surface area specifying unit that specifies a main plane in the parallax image DI based on the camera acquired images a and b acquired by the cameras 11a and 11b, and uses the main plane as a road surface. Identify the area.

  The dust detection unit 15 detects the presence or absence of dust D existing in front of the cameras 11 a and 11 b, that is, ahead of the vehicle 1, from the parallax image DI obtained by the image information acquisition unit 12. The dust position detector 16 searches and detects where the dust D is generated. Specifically, the dust position detection unit 16 searches for a pixel in the parallax image DI in which the dust region detected by the dust detection unit 15 is adjacent to the road surface region specified by the travel path recognition unit 14. Of the road surface areas adjacent to each other, the pixel closest to the host vehicle 1 is detected as the dust generation position P. That is, the dust position detection unit 16 is a dust region specifying unit, and specifies the dust region as the dust generation position P when the dust region is adjacent to the road surface region.

  The vehicle presence region detection unit 17 detects the dust generation position P detected by the dust position detection unit 16 as a region where the preceding other vehicle 1A may exist. That is, the vehicle presence area detection unit 17 is an obstacle area specifying unit, and based on the position and the number of pixels of the parallax-free area in the parallax image DI, there is an obstacle in the parallax-free area, particularly the other vehicle 1A. A vehicle possibility region J that is an obstacle region that is highly likely to be detected is identified.

  As shown in FIG. 7, the image control unit 18 outputs and displays the camera acquired image a acquired by any one of the stereo camera devices 11, for example, the camera 11 a, that is, the appearance image, on the display 19. In the image displayed on the display 19, the range of the vehicle possibility region J detected by the vehicle presence region detection unit 17, that is, the position and size are superimposed and displayed as, for example, a rectangular frame. The display 19 is a visual presentation device such as a liquid crystal display device, and is installed in the driver's seat 1 b of the vehicle 1. Specifically, the display 19 is attached to a position where an operator who sits on the driver's seat 1b and operates the vehicle 1 and the like is easy to see.

<Operation>
Next, processing of the obstacle detection system 10 according to the first embodiment will be described with reference to FIG. FIG. 8 is a flowchart showing processing by the obstacle detection system 10 mounted on the vehicle 1. FIG. 9 shows at which position in the camera acquired image b captured by the camera 11b positioned on the right side a specific object in the camera acquired image a captured by the camera 11a positioned on the left side of the stereo camera device 11 is captured. It is explanatory drawing which shows the specific method. FIG. 10 is a diagram showing a dust detection image SI created by the obstacle detection system 10. FIG. 11 is an explanatory diagram showing a method for detecting the dust width E by the obstacle detection system 10. FIG. 12 is a table showing a conversion table of threshold values of the dust width E detected by the dust position detection unit 16 of the obstacle detection system 10.

  When the power source of the obstacle detection system 10 is turned on by a switch, the image information acquisition unit 12 is initialized (step S01, hereinafter simply referred to as “S01”). In the initialization process of S01, the camera acquisition images a and b shown in FIGS. 5A and 5B can be acquired by the cameras 11a and 11b of the stereo camera apparatus 11, and these cameras 11a and 11b can be acquired. 6 is calculated, and the parallax image DI shown in FIG. 6 can be calculated from the camera acquired images a and b detected by the cameras 11a and 11b.

  Here, a method for identifying at which position in the camera acquired image b the specific object captured in the camera acquired image a will be described with reference to FIG. FIG. 9 relates to the coordinate system of the camera acquired images a and b, wherein the horizontal direction is the u axis and the vertical direction is the v axis. In the camera acquired image a, a rectangular area B surrounded by the positions (u1, v1), (u1, v2), (u2, v1) and (u2, v2) is set with the uv coordinates.

  Next, in the camera acquired image b, an area surrounded by the positions (U, v1), (U, v2), (U + (u2-u1), v1) and (U + (u2-u1), v2) is Is increased from u = 0 to u = u3, and scanning is performed to the right of the camera acquired image b to the position of the rectangular area B1. During this scanning, the correlation value between the image in the rectangular area B2 and the image in the rectangular area B1 is compared, and the rectangular area B2 of the camera acquired image b having the highest correlation with the rectangular area B of the camera acquired image a. At the positions (u4, v1), (u4, v2), (u4 + (u2-u1), v1) and (u4 + (u2-u1), v2), the same object as the object imaged in the rectangular area B Is imaged. In addition, it is assumed that each pixel in the rectangular area B corresponds to each pixel in the rectangular area B2.

  When the rectangular area B1 of the camera acquired image b is scanned and there is no rectangular area whose correlation value exceeds a certain value, the correspondence in the camera acquired image b corresponding to the rectangular area B of the camera acquired image a Let the point be none. If there is no corresponding point, the distance information cannot be calculated, and zero (0) is set to the corresponding pixel in the parallax image DI. For example, when an attempt is made to calculate parallax up to an object with poor visual features, pixels with a parallax value of zero (0) are likely to occur.

  In the mine, when high-concentration sand dust is generated as the vehicle 1 travels, the pixel values in the area where the dust D is captured in the camera acquired images a and b acquired by the cameras 11a and 11b are substantially constant. . For this reason, there are many cases where the corresponding points in the camera acquisition image b corresponding to the rectangular area B of the camera acquisition image a cannot be acquired. And the area | region where the dust D generate | occur | produced is included in the area | region where the pixel value in the parallax image DI is zero (0).

  Thereafter, the camera information a and b shown in FIGS. 5A and 5B are acquired by the image information acquisition unit 12 in each of the cameras 11a and 11b (S02). Based on b, the parallax image DI shown in FIG. 6 is calculated (S03). In the parallax image DI, the two-dimensional camera acquisition images a and b acquired by the cameras 11a and 11b are compared with the pixels in the parallax image DI, and each pixel on the camera acquisition image a is compared. A difference between the pixel value and the pixel value of each pixel on the camera acquired image b, that is, a parallax value is given, and image information based on these parallax values is added to each pixel of the parallax image DI.

  Next, processing for calculating a road surface area in the parallax image DI is executed in the travel path recognition unit 14 from the parallax image DI calculated in S03. The traveling path recognition unit 14 uses, for example, RANSAC (RANdom Sampling Consensus), Hough Transform, or the like based on the process in which the road surface R is a plane with respect to the parallax image D1 calculated in S03. The plane area in the parallax image DI is calculated. And among this plane area, based on the relative positional relationship between the vehicle 1 and each camera 11a, 11b of the stereo camera device 11, the plane area in which the vehicle 1 is traveling is the main plane corresponding to the road surface R. And Further, the main plane and an area having a difference smaller than a predetermined threshold value among the areas adjacent to the main plane are calculated and estimated as a road surface area corresponding to the road surface R ( S04).

  Further, when the difference between the pixel values of each pixel between the main plane and the area adjacent to the main plane is calculated, and the difference between the pixel values at each pixel is equal to or less than a predetermined threshold, it corresponds to the road surface R. The process of estimating the road surface area to be repeated is repeated. This threshold value is set based on the running characteristics of the vehicle 1. For example, even when either the front wheel 1 d or the rear wheel 1 e gets over the unevenness that does not affect the running on the road surface R when the vehicle 1 is running, the vehicle 1. It is set to a value that does not affect the running of the car.

  Next, from the parallax image DI calculated in S03, the dust detection unit 15 detects whether dust D is generated on the road surface R in front of the vehicle 1 (S05). In S05, first, the road surface area estimated in S04 is superimposed on the parallax image DI calculated in S03 to create the dust detection image SI shown in FIG. At this time, the pixel value of each pixel constituting the road surface area on the dust detection image SI is not a detection target of whether or not the dust D is generated from the dust detection image SI, that is, a value that is not used. To do. Each pixel constituting this road surface area is referred to as a road surface area pixel G1.

  Thereafter, a search for the road surface area pixel G1 on the dust detection image SI is executed, and when the road surface area pixel G1 is found, the parallax value of each pixel adjacent to the found road surface area pixel G1 is searched. At this time, when a pixel having a parallax value of zero (0), that is, a pixel included in a non-parallax region is found, the pixel is set as a dust region pixel G2. When the dust area pixel G2 is found, the parallax value of the pixel in the dust detection image SI adjacent to the found dust area pixel G2 is searched.

  However, in the detection of the dust region pixel G2 by the dust detection unit 15, in addition to the region where the dust D is highly likely to occur, for example, a pixel in a region where the parallax value such as sky is zero (0) may be searched. There is. For this reason, no search is made for a height on the road surface area pixel G1 in the parallax image DI that is equal to or higher than a predetermined threshold value (dust area pixel search range). As a result of the search for the dust area pixel G2 in S05, if the number of the dust area pixels G2 is equal to or greater than a predetermined threshold value in S05, it is determined that there is dust (S06 / YES).

  When it is determined that there is dust in S06 (YES), the dust position detection unit 16 searches for which point in the parallax image DI, that is, in which position the dust D is generated (S07). At this time, since each parallax region pixel G2 in the parallax image DI has a parallax value of zero (0), the distance from the vehicle 1 cannot be calculated based on these dust region pixels G2, and the dust D from the vehicle 1 can be calculated. Cannot be calculated directly from the parallax image DI. Therefore, in S07, the position of the road surface area pixel G1 adjacent to the dust area pixel G2 is detected as the dust generation position P. Specifically, among the dust region pixels G2 detected in S05, the dust region pixel G2 to which the road surface region pixel G1 is adjacent in the downward direction is searched. Of the dust area pixels G2 that are adjacent to the road surface area pixel G1 in the downward direction, the dust area pixel G2 that is closest to the vehicle 1, that is, the most downward direction is set as the dust generation position P.

  Thereafter, the dust position detector 16 detects the width of the dust region in the parallax image DI, that is, the dust width E (S08). In S08, the dust area pixel G2 adjacent to each other in the parallax image DI is searched, and the dust area pixel G2 located at the leftmost position in the parallax image DI among these adjacent dust area pixels G2. And the dust region pixel G2 located at the right end, and the difference in the horizontal direction (u-axis direction) between the dust region pixel G2 located at the left end and the dust region pixel G2 located at the right end, ie, these The distance between the left and right dust area pixels G2 is detected as the dust width E.

  The detected dust width E is compared with a predetermined threshold value, and if the dust width E is larger than the threshold value (S09 / YES), the area hidden by the dust D, that is, the dust concealment area which is the dust concealment range. Is determined to be a region where there is a high possibility that the preceding other vehicle 1A exists (S10).

  This threshold value is set to be approximately the same as the width dimension of the other vehicle 1A to be detected. The dust width E detected by the dust position detection unit 16 changes according to the distance from the vehicle 1 to the dust D. Therefore, as shown in FIG. 11, this threshold is the inter-vehicle distance L from the position of the vehicle 1 (assumed vehicle position F) in the parallax image DI to be detected to the dust generation position P detected in S07. Change accordingly. Here, when the assumed vehicle position F is within a blind spot area H that is not included in the parallax image DI, the vehicle assumed position F is set to a predetermined position below the parallax image DI based on the mounting positions and mounting angles of the cameras 11a and 11b. .

  Further, on the road surface R of the mine, as shown in FIG. 1, when a plurality of vehicles 1 and 1A of the same vehicle type are separated from each other by a predetermined distance and are traveling at the same traveling speed and loaded state, that is, the traveling state. There are many. For this reason, the threshold is determined based on the rising position of the dust D from the road surface R, and based on a predetermined conversion table, as shown in FIG. 12, the load is detected on the loading platform 1c of the host vehicle 1 by the load sensor 5i. When the loaded state, that is, the loaded state is detected, the dust D generated at the time of traveling becomes darker than the case where no load is loaded on the loading platform 1c of the host vehicle 1, that is, the empty state, and the dust generation position Since the distance from P to the vehicle 1 is increased, the size of the dust area in the parallax image DI is also reduced, so the threshold value is set small.

  Further, as the vehicle speed of the host vehicle 1 increases due to detection by the vehicle speed sensor 5h, the vehicle speed of the other vehicle 1A traveling ahead increases in the same manner, so that the dust D that the other vehicle 1A whirls when traveling is dark. Since the range becomes large, the threshold value is set large. Further, as the inter-vehicle distance L from the host vehicle 1 to the dust generation position P calculated by the distance calculation unit 13 becomes longer, the dust generation position P becomes further away from the vehicle 1, and the dust area of the parallax image DI is displayed. Since the size is small, the threshold is set small.

  Thereafter, based on the dust concealment area determined in S10, the dust generation position P detected in S07 is classified into an area where the preceding other vehicle 1A is likely to exist (vehicle possibility area J), and a vehicle existence area. The detection unit 17 recognizes and outputs the image, and the image control unit 18 superimposes and displays the vehicle possibility region J in the camera acquired image a displayed on the display 19 (S11).

  On the other hand, if it is determined in S05 that the number of the dust area pixels G2 is smaller than a predetermined threshold value and there is no dust D (S06 / NO), no dust D is generated in the first place and the other vehicle is ahead of the dust D. There is no possibility that 1A exists. In addition, when the dust width E detected by the dust position detection unit 16 is equal to or smaller than the threshold (S09 / NO), it is very unlikely that the cause of the generation of the dust D is the other vehicle 1A traveling ahead. . For this reason, when it is determined that there is no dust D in S06 (NO), and when the dust width E is equal to or smaller than the threshold value (NO) in S09, the parallax calculated in S03 by the vehicle presence area detection unit 17 The other vehicle 1A traveling ahead is detected from the image DI (S12).

  After the detection of the other vehicle 1A in S12, the process proceeds to S11, where the vehicle information related to the other vehicle 1A detected in S12 is output as the vehicle possibility region J by the vehicle presence region detection unit 17, and the image control unit 18 The vehicle possibility region J is superimposed on the image displayed on the display 19 and displayed. Note that the processes of S01 to S12 are repeatedly performed at a predetermined frame, for example, about 30 frames per second.

<Effect>
In the mine, when the other vehicle 1A traveling in the front lifts up the dust D rearward during traveling, the other vehicle 1A is hidden by the dust D so lifted up, and the vehicle 1A following the other vehicle 1A travels. In some cases, the vehicle 1 cannot visually confirm the other vehicle 1A. Furthermore, the area where the dust D is generated is a substantially white area, and the parallax image DI is acquired by comparing the two-dimensional camera acquired images a and b acquired by the cameras 11a and 11b of the stereo camera device 11. In this case, a parallax-free area with no parallax value is obtained.

  Therefore, in the obstacle detection system 10 according to the first embodiment, a two-dimensional camera acquisition in which the area where the dust D is generated on the road surface R is acquired by the cameras 11a and 11b of the stereo camera device 11 is acquired. The images a and b are compared and specified in the non-parallax region between the camera acquired images a and b.

  Also, the dust D that the vehicle 1A whirls when traveling is higher in concentration than the naturally occurring dust and has a wider range of generation than the size of the vehicle 1A. Therefore, when the parallax-free area between the camera acquired images a and b acquired by the cameras 11a and 11b is larger than a predetermined threshold, the possibility that the other vehicle 1A is likely to exist in the parallax-free area. The region J is specified.

  As described above, even when the dust D is generated and the other vehicle 1A traveling ahead is hidden by the dust D, an area where the other vehicle 1A may exist can be detected. Therefore, since the other vehicle 1A traveling ahead and the vehicle 1A on the road surface R can be detected at an early stage, it becomes possible to cause the other vehicle 1A traveling ahead to perform follow-up traveling and avoidance traveling, and a highly reliable vehicle. Can be 1.

  Further, dust D generated when the vehicle 1A travels is generated on the road surface R. For this reason, when the non-parallax region is in contact with the road surface region in the parallax image DI, there is a high possibility of the dust D that the other vehicle 1 </ b> A traveling ahead travels up. Therefore, when the non-parallax region is in contact with the road surface region, by specifying the non-parallax region as the dust region, it is possible to appropriately identify the dust region that covers the other vehicle 1A that travels forward. The hidden other vehicle 1A can be detected appropriately.

  Further, as shown in FIG. 7, the camera acquired image a acquired by the camera 11 a, that is, the appearance image is displayed on the display 19, and the other vehicle 1 </ b> A traveling ahead in the image displayed on the display 19 is displayed. The vehicle possibility region J that is likely to exist is superimposed and displayed. As a result, when the operator who drives the vehicle 1 confirms the display 19, in addition to the position and range where the dust D is generated in the front in the traveling direction, there is another vehicle 1A hidden by the dust D. Since a highly likely area can be easily confirmed, even in a manned traveling vehicle 1, it becomes possible to more appropriately maintain an inter-vehicle distance from the other vehicle 1 </ b> A traveling ahead, and contact with the other vehicle 1 </ b> A. Can be prevented more appropriately.

  Further, as shown in FIG. 1, in many cases, a plurality of vehicles 1 and 1A of the same vehicle type are separated from each other by a predetermined traveling interval and are traveling in the same traveling state. Is set smaller when the host vehicle 1 is in a loaded state than when it is in an unloaded state, and is set larger as the vehicle speed of the host vehicle 1 increases. Further, the threshold value is set smaller as the inter-vehicle distance L from the host vehicle 1 to the dust generation position P becomes longer. As a result, since it becomes possible to detect the dust D according to the traveling state of the other vehicle 1A traveling in front and the traveling interval to the other vehicle 1A, there is a possibility that the other vehicle 1 hidden by the dust D exists. A high area can be detected with higher accuracy, and the travel interval with the other vehicle 1A can be more appropriately maintained.

  Further, when it is determined that there is no dust D in S06 (NO), no dust D is generated in the first place, and there is no possibility that the other vehicle 1A exists ahead of the dust D. Further, when the dust width E is equal to or smaller than the threshold value (NO) in S09, the possibility of dust generated when the preceding other vehicle 1A travels is extremely low. Further, in these cases, the other vehicle 1A traveling ahead is captured in the camera acquired images a and b, and can be identified from the parallax image DI. Therefore, when it is determined that there is no dust D in S06, and when the dust width E is equal to or smaller than the threshold value in S09, the other vehicle 1A traveling ahead is detected from the parallax image DI calculated in S03. . As a result, regardless of the size of the parallax-free area in the parallax image DI, it is possible to specify an area where there is a high possibility that the other vehicle 1A traveling ahead is present.

[Second Embodiment]
FIG. 13 is a schematic view showing a vehicle 1 according to the second embodiment of the present invention. FIG. 14 is a schematic diagram showing an obstacle detection system 10 </ b> A mounted on the vehicle 1. The second embodiment differs from the first embodiment described above in that the second embodiment includes a radar measurement system 21, a dust determination unit 22, and a vehicle recognition unit 23. . In the second embodiment, the same or corresponding parts as those in the first embodiment are denoted by the same reference numerals.

<Configuration>
In the second embodiment, the radar measurement system 21 is an obstacle detection unit that is provided in the obstacle detection system 10A and detects the presence or absence of an obstacle through the dust D. The radar measurement system 21 detects an obstacle on the road surface area recognized by the traveling road recognition unit 14, and includes a radar measurement information acquisition unit 21a and a radar obstacle detection unit 21b. The radar measurement information acquisition unit 21a is a laser that passes through dust such as a millimeter wave radar and can detect an obstacle ahead of the dust, and a fault that exists in a sensing region that is a region scanned by this laser. The position information of the object is output to the radar obstacle detection unit 21b as radar measurement information.

  The radar obstacle detection unit 21b inhibits the traveling of the vehicle 1 based on the radar measurement information acquired by the radar measurement information acquisition unit 21a and the road surface area information in the parallax image DI estimated by the travel path recognition unit 14. The position of the obstacle that may be detected, for example, the position of the other vehicle 1 </ b> A is detected, and the position information of the detected obstacle is output to the vehicle recognition unit 23. The dust detection unit 22 detects the position of the obstacle detected by the radar obstacle detection unit 21b and the dust generation position P detected by the dust position detection unit 16 and the number of pixels of the dust D detected by the dust detection unit 15. It is determined whether or not the dust region is dust generated by traveling by the other vehicle 1A traveling ahead.

  The vehicle recognition unit 23 is an obstacle identification unit, and includes a vehicle possibility region J in the parallax image DI detected by the vehicle presence region detection unit 17 and the position of the obstacle detected by the radar obstacle detection unit 21b. And the type, size, and position of the other vehicle 1A traveling ahead are identified. That is, the vehicle recognition unit 23 identifies the other vehicle 1A based on the position and the number of pixels of the vehicle possibility region J and the position of the obstacle in the radar measurement information. Further, the vehicle recognition unit 23 outputs vehicle information regarding the type, size, and position of the identified other vehicle 1 </ b> A to the image control unit 18, for example.

<Operation>
Next, processing of the obstacle detection system 10A according to the second embodiment will be described with reference to FIG. FIG. 15 is a flowchart showing processing by the obstacle detection system 10A. Here, the steps of the obstacle detection system 10A are the same in the steps of the obstacle detection system 10 according to the first embodiment.

  After S04, the radar measurement information acquisition unit 21a acquires radar measurement information (S21). Next, the radar obstacle detection unit 21b travels the vehicle 1 based on the position information of the obstacle in the sensing area in the acquired radar measurement information and the road surface area information in the parallax image DI estimated in S04. The position of the other vehicle 1A is detected, and the detected position information of the obstacle is output to the vehicle recognition unit 23 (S22).

  Further, after S10, the position information of the obstacle detected by the radar obstacle detection unit 21b is referred to (S23), and whether or not there is an obstacle near the dust concealment area determined in S10 is determined. (S24). If it is determined in S24 that there is an obstacle in the vicinity of the dust concealment area (YES), it is determined that there is another vehicle 1A traveling ahead, and the position information of the obstacle detected by the radar obstacle detection unit 21b. And the dust generation position P information detected in S07 is output to the vehicle recognition part 23 as vehicle information (S11).

  On the other hand, if it is determined in S24 that there is no obstacle in the vicinity of the dust concealment area (NO), it is determined that the detected dust concealment area is naturally generated dust (S25), and the processing by the obstacle detection system 10A is performed. finish. As a result, naturally occurring dust can be distinguished from dust D generated when the vehicle 1A travels, and the possibility of erroneous recognition of the vehicle 1A hidden by the dust D can be reduced.

  Furthermore, after S12, the position information of the obstacle in the radar measurement information detected by the radar obstacle detection unit 21b is referred to for the vehicle recognition position of the other vehicle 1A in the parallax image DI detected in S12. The other vehicle 1A is detected (S26), and the detected vehicle information of the other vehicle 1A is output to, for example, the image control unit 18 (S11).

<Effect>
As described above, in the obstacle detection system 10A according to the second embodiment, in addition to the operational effects that can be achieved in the first embodiment, the position and size of the other vehicle 1A are specified by the vehicle presence area recognition unit 17. Even if it is not possible, the position of the obstacle on the road surface area is detected by the radar measurement information acquisition unit 21a, and the position of the vehicle possibility area J in the parallax image DI detected by the vehicle presence area detection unit 17 and By adding the position information of the obstacle detected by the radar obstacle detection unit 21b to the number of pixels, the vehicle recognition unit 23 can identify the type, size, and position of the other vehicle 1A traveling ahead.

  In particular, by using a radar as the radar measurement information acquisition unit 21a, it is possible to accurately detect the position information of the obstacle present in the sensing area scanned by the laser, so that the other vehicle 1A hidden by the dust D is appropriately detected. it can. Therefore, based on the radar measurement information measured by the radar measurement information acquisition unit 21a, the radar obstacle detection unit 21b detects an obstacle present on the road surface area in the parallax image DI, and the detected obstacle is detected. By using the position information, the vehicle recognition unit 23 can more accurately detect the obstacle information on the road surface area.

  Further, whether or not the dust is generated by traveling by the vehicle 1A can be determined by the positional relationship between the road surface area and the dust area in the parallax image DI and the size thereof, that is, the number of pixels. Therefore, based on the position of the vehicle possibility region J and the number of pixels in the parallax image DI detected by the vehicle presence region detection unit 17 and the position information of the obstacle detected by the radar obstacle detection unit 21b, The dust determination unit 22 can determine whether the dust region detected by the detection unit 15 is the dust D generated by the traveling of the other vehicle 1A traveling forward. Therefore, the cause of generation of dust D corresponding to the dust region can be identified with high accuracy. Therefore, since the dust D covering the vehicle 1A can be specified with higher accuracy, the region where the vehicle 1A is likely to exist can be detected with higher accuracy.

[Others]
In addition, this invention is not limited to embodiment mentioned above, Various deformation | transformation aspects are included. For example, the above-described embodiments have been described in order to explain the present invention in an easy-to-understand manner, and the present invention is not necessarily limited to those having all the configurations described.

  Furthermore, in each said embodiment, although the vehicle 1 was demonstrated taking the dump truck for mines as an example, for example, in addition to working vehicles, such as a hydraulic excavator 4, a water truck, a grader, a bulldozer, a load loader, a patrol car, a mine The obstacle detection systems 10 and 10A according to the present invention can also be mounted on the mining work vehicle that travels on the road surface R provided on the road.

  Further, not only the manned traveling vehicle 1 but also the unmanned traveling vehicle 1 capable of autonomous traveling according to an operation signal from the information center 3 may be used. In this case, the vehicle information detected by the obstacle detection systems 10 and 10A is transmitted from the vehicle 1 to the information center 3 and compared with the vehicle information recognized by the information center 3. The operation of a plurality of vehicles 1 and 1A can be managed with higher accuracy.

  In addition, as the obstacle detection systems 10 and 10A according to the above embodiments, for example, other vehicles 1A traveling in front, other vehicles such as watering vehicles, graders, patrol cars traveling on the road surface R of the mine are detected. Further, a vehicle detection system or the like may be used. Further, for example, a system for detecting obstacles other than the vehicles 1 and 1A such as a dropped load, collapsed earth and sand, and falling rocks may be used.

  Furthermore, in each of the above-described embodiments, the controller 6 mounted on the vehicle 1 performs all the processing by the obstacle detection systems 10 and 10A. However, transmission and reception of predetermined information using the in-vehicle wireless device 5g and the like. It is also possible to adopt a configuration in which at least one of the processes of the controller 6 is performed at a place other than the vehicle 1 such as the information center 3.

1,1A vehicle (mine work vehicle)
1c Loading platform 5h Vehicle speed sensor (speed detector)
5i Load sensor (load mass detector)
6 controller 10 obstacle detection system 11 stereo camera device 11a camera (first camera)
11b Camera (second camera)
12 Image information acquisition unit (difference-free region specifying unit)
13 Distance calculation unit 14 Travel path recognition unit (road surface area specifying unit)
DESCRIPTION OF SYMBOLS 15 Dust detection part 16 Dust position detection part 17 Vehicle presence area | region detection part (obstacle area specific | specification part)
18 Image Control Unit 19 Display (Display Device)
21 Radar measurement system (obstacle detection unit)
21a Radar measurement information acquisition unit (radar)
21b Radar obstacle detection unit 22 Dust determination unit 23 Vehicle recognition unit (obstacle identification unit)

Claims (8)

A stereo camera device including first and second cameras capable of acquiring image information in front of the traveling direction;
The two-dimensional image information acquired by the first camera is compared with the two-dimensional image information acquired by the second camera, and an area having no difference between the image information is identified as a no-difference area. A difference-free region specifying unit to
An obstacle area specifying unit for specifying the difference-free area as an obstacle area where an obstacle is likely to exist when the number of pixels of the difference-free area in the image information is larger than a predetermined value;
Mining work vehicle characterized by comprising: The mine work vehicle according to claim 1,
A road surface area specifying unit for specifying a road surface area in the image information based on the image information acquired by the first and second cameras;
A dust region specifying unit for specifying the difference non-region as a dust region when the difference non-region in the image information is in contact with the road surface region;
The obstacle region specifying unit specifies the dust region as an obstacle region when the number of pixels of the dust region in the image information is larger than a predetermined value. The mining work vehicle according to claim 2,
An obstacle detection unit for detecting an obstacle present on the road surface area in the image information;
The type and size of the obstacle based on the obstacle information detected by the obstacle detection unit and the position and the number of pixels of the obstacle region in the image information specified by the obstacle region specifying unit. , And an obstacle identifying unit for identifying the position;
Mining work vehicle characterized by comprising: The mine work vehicle according to claim 3,
The obstacle detection unit includes a radar and a radar obstacle detection unit that detects an obstacle existing on a road surface area in the image information based on measurement information measured by the radar. Mining work vehicle. The mining work vehicle according to claim 2,
An obstacle detection unit for detecting an obstacle present on the road surface area in the image information;
Based on the obstacle information detected by the obstacle detection unit and the position and the number of pixels of the obstacle region in the image information specified by the obstacle region specification unit, the difference-free region is used for driving the vehicle. A dust discriminating unit for discriminating whether or not the dust is generated,
Mining work vehicle characterized by comprising: The mine work vehicle according to claim 1,
A display device for displaying image information;
Image information acquired by one of the first and second cameras is displayed on the display device, and the obstacle region specified by the obstacle region specifying unit is displayed on the display device. An image control unit to be displayed in the image;
Mining work vehicle characterized by comprising: The mine work vehicle according to claim 1,
Loading platform,
A load mass detector for detecting the mass of the load loaded on the cargo bed;
A speed detector for detecting the traveling speed;
A distance calculation unit that calculates a distance to the no-difference area in the image information based on the image information acquired by the first and second cameras;
The obstacle region specifying unit is responsive to the mass of the load detected by the load mass detection unit, the travel speed detected by the speed detection unit, and the distance to the no difference region calculated by the distance calculation unit. The predetermined value is changed. A mining work vehicle. The mine work vehicle according to claim 1,
The obstacle area specifying unit specifies the obstacle area from the image information acquired by the first and second cameras when the number of pixels of the non-difference area in the image information is smaller than the predetermined value. A mining work vehicle characterized by that.