TWI598854B - Obstacle detecting device and moving object provided therewith - Google Patents

Description

Obstacle detecting device and moving body having the same

The present invention relates to an obstacle detecting device and a moving body provided therewith.

There is already a golf cart with autonomous driving function that travels along a predetermined travel path within the golf course. Autonomous driving is performed by detecting a magnetic field generated by an electromagnetic induction line. Further, the work vehicle or the person crossing the traveling path exists as an obstacle of the golf cart on the traveling path of the golf course. In the case of autonomous driving without a driver, it is necessary to detect such obstacles. Therefore, there is also a golf cart equipped with an obstacle detecting device provided on the front side, and the obstacle detecting device is provided with an obstacle sensor such as an ultrasonic wave to detect an obstacle in front of the golf cart.

The previous obstacle detecting device sometimes detects an object that does not exist on the traveling path as an obstacle, and as a result, decelerates the golf cart. Therefore, the technique of Patent Document 1 determines whether or not an obstacle exists on a traveling path. Thereby, the stop control is suppressed by the obstacle which does not exist in a travel path.

Patent Document 1

Patent Document 1 discloses that a moving body coordinate system is obtained based on coordinate positions of a moving body obtained as an overall coordinate system and coordinate position data of points along a traveling path obtained as an overall coordinate system. The coordinate position data of the driving path in the middle, the driving road surface in the three-dimensional image in the moving body coordinate system is specified, and the detection is detected. Whether the obstacle is on the road.

[Previous Technical Literature] [Patent Literature]

[Patent Document 1]

Japanese Patent Laid-Open No. Hei 10-141954

However, since the traveling road surface in the three-dimensional image is specified by using the three-dimensional data, the calculation load is imposed on the CPU (Central Processing Unit) of the obstacle detecting device, and the calculation time is required. When the calculation time is extended, since the moving body is traveling, the distance from the obstacle is shortened, and the possibility of contact with the obstacle is increased. Further, in the case of a device for passing between vehicles of various trees like a driving path of a golf course, if it is inferred that the error of the current position is large, it is determined that various trees around the traveling path are determined as the traveling path. A situation in which an obstacle is stopped. Therefore, it is necessary to infer the current position with high precision. In Patent Document 1, a three-dimensional position coordinate is used for the current position of the moving body. However, if the current position is to be estimated three-dimensionally and accurately, an expensive GPS (Global Positioning System) or a sensor is required. Moreover, such GPSs, sensors, etc., which are equivalent to the number of vehicles, are required.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an obstacle detecting device and a moving body including the same for reducing an amount of calculation and detecting an obstacle on a traveling path.

In order to achieve the above object, the present invention adopts the constitution as described below.

In other words, the first aspect of the present invention is an obstacle detecting device that is mounted on a moving body that travels along a predetermined traveling path, and includes a distance information detecting unit that faces an object from the obstacle detecting device to the front of the moving body. Distance information detection; three-dimensional information production a living part that generates three-dimensional information in front of the moving body based on the distance information; a memory unit that stores traveling path position information corresponding to a position of the moving body from a predetermined position; and a position estimating unit that estimates a position of the moving body on the traveling path; a three-dimensional information capturing unit on the traveling path that captures three-dimensional information on the traveling path in the three-dimensional information based on the traveling path position information; and an obstacle detecting unit based on The obstacle is detected by the three-dimensional information on the travel path, and the travel route position information is specified by the distance information and the coordinate information of the horizontal direction of the mobile body.

According to the invention, the distance information detecting unit detects the distance information from the obstacle detecting device to the object in front of the moving body. The three-dimensional information generating unit generates three-dimensional information in front of the moving body based on the detected distance information. The travel path information is stored in the memory unit, and the travel route position information specifies a travel route on the three-dimensional information based on the position of the mobile body calculated from the predetermined position. The position estimating unit estimates the position of the moving body on the traveling path. The three-dimensional information capturing unit on the driving path extracts three-dimensional information on the driving path in the three-dimensional information according to the inferred position and the traveling path position information. The obstacle detecting unit detects an obstacle based on the three-dimensional information on the traveling path.

The travel path position information used to capture the three-dimensional information on the travel path is two-dimensional information of the coordinate information of the distance information and the horizontal direction of the mobile body, and thus can be reduced compared with the previous use of the three-dimensional information amount. The amount of information processed can reduce the computational load. Thereby, the calculation time can be shortened, and it can be determined whether or not the obstacle is in the traveling path at the time when the distance is secured with the obstacle, so that contact with the obstacle can be avoided.

Moreover, it is preferable to include a person determination unit that determines whether or not the obstacle on the travel route is a person. Alternatively, it is preferable to include a person determining unit that determines whether or not the obstacle that has been determined to be on the traveling path is a person. It is detected whether the obstacle on the travel path is a person, thereby shortening the processing time of the human detection. Further, since the person who is away from the moving body is not detected, the moving body can smoothly travel.

Moreover, it is preferable to include an extended area three-dimensional information capturing unit, and the extended area is three-dimensionally funded. The information acquisition unit captures three-dimensional information in a travel path extension area that is adjacent to the travel path, based on the travel route position information, and the obstacle detection unit also detects an obstacle on the travel path extension area, and the person determines The part determines whether the obstacle detected in the travel path extension area is a person. Thereby, the detection rate can be improved for a person in an area in the vicinity of both ends in the width direction of the traveling path which is easily confused with various trees around the traveling path.

Furthermore, the present invention may further include: an extended area three-dimensional information capturing unit that captures three-dimensional information in a predetermined area adjacent to the traveling path, that is, in the driving path expansion area, based on the traveling path position information; and a human detecting unit based on the three-dimensional Information, in the above-mentioned driving path expansion area to detect people. The human detection is performed within a range in which the predetermined travel path is larger than the predetermined travel path, whereby the processing time of the human detection can be shortened. Further, since it is possible to detect a person who is in the vicinity of the traveling route, it is also possible to detect a person who wants to enter the traveling route. Thereby, contact with people can be avoided. Further, since the obstacle detecting unit and the human detecting unit are separately included, the obstacle on the traveling path with high possibility of contact and the traveling path which is difficult to determine the possibility of contact and which is easily confused with various trees around the traveling path can be used. People in the area near the ends of the width are treated differently.

Further, the second aspect of the present invention is an obstacle detecting device mounted on a moving body that travels along a predetermined traveling path, and includes a distance information detecting unit that faces an object from the obstacle detecting device to the front of the moving body. The distance information is detected; the three-dimensional information generating unit generates the three-dimensional information in front of the moving body based on the distance information; the memory unit stores the traveling path position information, and the traveling path position information is based on the predetermined position a position of the moving body to specify a traveling route on the three-dimensional information; a position estimating unit that estimates a position of the moving body on the traveling path; and a traveling path specifying unit that based on the estimated position and the traveling path position Information indicating a travel route in the three-dimensional information; an obstacle detecting unit that detects an obstacle based on the three-dimensional information; and a determining unit that determines whether the detected obstacle is in a special state The travel route position is determined by the distance information and the coordinate information of the horizontal direction of the mobile body.

According to the invention, the distance information detecting unit detects the distance information from the obstacle detecting device to the object in front of the moving body. The three-dimensional information generating unit generates three-dimensional information in front of the moving body based on the detected distance information. The travel path information is stored in the memory unit, and the travel route position information specifies a travel route on the three-dimensional information based on the position of the mobile body calculated from the predetermined position. The position estimating unit estimates the position of the moving body on the traveling path. The travel route specific portion specifies the travel route in the three-dimensional information based on the estimated position and the travel route position information. The obstacle detecting unit detects an obstacle based on the three-dimensional information. The determination unit determines whether the detected obstacle is on a specific travel route.

The travel route location information of the travel route specifying the three-dimensional information is the two-dimensional information of the coordinate information of the distance information and the horizontal direction of the mobile body, so that the processed information can be reduced compared with the previous use of the three-dimensional information amount. Amount, which reduces the computational load. Thereby, the calculation time can be shortened, and it can be determined whether or not the obstacle is in the traveling path at the time when the distance is secured with the obstacle, so that contact with the obstacle can be avoided.

Further, it is preferable to include an area specifying unit that specifies a travel path extension area that is adjacent to a predetermined area of the specified travel route, and the determination unit determines whether the detected obstacle is specific In the above-described travel path extension area, the person determination unit determines whether or not the obstacle that has been determined to be in the travel path extension area is a person. Thereby, the detection rate can be improved for a person in an area in the vicinity of both ends in the width direction of the traveling path which is easily confused with various trees around the traveling path.

Moreover, the method may further include: an area specifying unit that specifies a travel path extension area that is adjacent to the predetermined area of the specified travel route; and a person detection unit that is specifically configured to travel People are detected in the path extension area. The human detection is performed within a range in which the predetermined travel path is larger than the predetermined travel path, whereby the processing time of the human detection can be shortened. Furthermore, since it is possible for people in the vicinity of the traveling route Detection, it is also possible to detect people who want to enter the driving path. In this way, contact with people can be avoided more. Further, since the obstacle detecting unit and the human detecting unit are separately provided separately, it is possible to make it difficult to determine the obstacle on the traveling path with high possibility of contact and the possibility of contact, and it is easy to be confused with various trees around the traveling path. The person in the area near the both ends of the width direction of the traveling path performs different processing.

Moreover, it is preferable that the position estimating unit estimates the position of the moving body on the traveling path based on the moving distance from the predetermined position. The moving body travels along a predetermined traveling path, and therefore, the position of the moving body on the traveling path can be specified based on the moving distance from the predetermined position. The position estimating unit estimates the position of the moving body on the traveling path based on the moving distance calculated from the predetermined position. Therefore, the position of the moving body can be estimated using only the vehicle coordinate system without using the overall coordinate system. Further, since the moving distance is one-dimensional data amount, the amount of data to be processed can be remarkably reduced.

Further, the distance information detecting unit may include a stereo camera, and the three-dimensional information may be a parallax image. A stereo camera is used as the distance information detecting unit, whereby the parallax image can be used as three-dimensional information. Thereby, the determination of the traveling route and the detection of the obstacle in the parallax image can be appropriately performed. Moreover, the one-dimensional scanning laser radar easily detects the road surface as an obstacle due to the undulation of the road surface, but by using a stereo camera, it is less susceptible to the influence of road surface undulations to detect obstacles.

Further, the distance information detecting unit may include a radar. The radar is used as the distance information detecting unit, whereby three-dimensional information can be appropriately obtained. In comparison with the use of a stereo camera, the distance can be measured with high precision in particular, so that the determination of the travel path and the detection of obstacles in the three-dimensional information can be appropriately performed.

Further, the moving system of the present invention includes the moving body of the above obstacle detecting device. Since the obstacle detecting device described above is included, it is possible to reduce the amount of calculation and determine whether or not there is an obstacle on the traveling path.

Moreover, the moving body preferably includes an alarm, and if the moving path is detected, If the obstacle is mentioned, the alarm will issue a warning. A warning is issued to a driver, an animal, or a driver of another mobile body existing on the traveling path, thereby avoiding contact with the obstacle. Moreover, by issuing a warning, the driver of the mobile body can be evoked.

Further, the moving body preferably includes a speed control unit that decelerates or stops the moving body when the obstacle is detected on the traveling path. Even when the obstacle existing on the traveling path is not a human or an animal, or when the driver does not drive autonomously, the contact with the obstacle can be avoided.

Further, the moving body preferably includes: a speed control unit that decelerates or stops the moving object with respect to the detected obstacle on the traveling path; and an alarm that detects the extended area of the traveling path The person who issued it issued a warning. Thereby, when the obstacle including the person is on the traveling path, the moving body is decelerated or stopped by detecting it. Further, when there is a person in the travel path extension area, the alarm is issued from the alarm by detecting it. By the warning, the person in the extended area of the traveling path can be prevented from entering the traveling path, and as a result, unnecessary deceleration or stopping of the moving body can be suppressed.

According to the present invention, it is possible to provide an obstacle detecting device that reduces an amount of calculation and detects an obstacle on a traveling path, and a moving body including the same.

1‧‧‧ Vehicles

3‧‧‧ Stereo Camera

3a‧‧‧Left image sensor

3b‧‧‧Right image sensor

4‧‧‧Steering wheel

5‧‧‧Right front wheel

6‧‧‧Left front wheel

7‧‧‧Reading Department

8‧‧‧Rotary angle sensor

9‧‧‧ Wheel speed sensor

12‧‧‧Touch panel display

14‧‧‧ obstacle detection device

14'‧‧‧ obstacle detection device

15‧‧‧Autonomous Driving Control Department

16‧‧‧Speed Control Department

17‧‧‧Alarm

21‧‧‧Distance Information Inspection Department

21'‧‧‧ Distance measuring device

23‧‧‧ Parallax Image Generation Department

23'‧‧‧Distance Image Generation Department

25‧‧‧ Driving Path Location Information Memory

25'‧‧‧ Driving Path Location Information Memory

27‧‧‧Location Estimation Department

29‧‧‧Special part of the route

31‧‧‧ Obstacle Detection Department

31'‧‧‧The Obstacle Detection Department

33‧‧‧Decision Department

33'‧‧‧Decision Department

34‧‧‧Distance calculation department

35‧‧‧ Personage Department

35'‧‧‧ Personage Department

41‧‧‧buffer

43‧‧‧Location calculation department

54‧‧‧ obstacle detection device

56‧‧‧ obstacle detection device

64‧‧‧Extended Area Capture Department

64'‧‧‧Extended Area Capture Department

66‧‧‧Disparity Image Capture on the Driving Path

Da~dg‧‧‧disparity value

Fp‧‧‧ fixed point

Ob1~Ob4‧‧‧Area

Ob1'‧‧‧ area

P1~P4‧‧‧ Parallax image

Sa‧‧ Location

Sn‧‧ Location

S1‧‧‧Location

S2‧‧ Location

So‧‧‧ starting position

S01~S10‧‧‧Steps

Tr‧‧‧ driving route

Tr'‧‧‧ Driving Path Expansion Area

X‧‧‧ direction

Y‧‧‧ direction

X ₁ ~X ₁₀ ‧‧‧X coordinate area

Figure 1 is a front elevational view of the vehicle of the embodiment.

Fig. 2 is a block diagram showing the configuration of the vehicle of the first embodiment.

Fig. 3 is a block diagram showing the configuration of an obstacle detecting device of the first embodiment.

Fig. 4 is an explanatory view showing the relationship between the stereo camera of the first embodiment and the orientation of the distance information.

Fig. 5 is an explanatory view showing an example of a travel route on which the vehicle of the embodiment travels.

FIG. 6 is a view for explaining the specificity of the travel route in the parallax image of the first embodiment. Ming map.

Fig. 7 is an explanatory view showing an obstacle on a traveling path in the parallax image of the first embodiment.

Fig. 8 is a flow chart showing the flow of obstacle detection in the first embodiment.

Fig. 9 is a block diagram showing the configuration of a vehicle of the second embodiment.

Fig. 10 is a block diagram showing the configuration of a vehicle of the third embodiment.

Fig. 11 is an explanatory view showing a travel path extension area in the parallax image of the third embodiment.

Fig. 12 is a block diagram showing the configuration of a vehicle of the fourth embodiment.

Fig. 13 is an explanatory view showing an obstacle on a traveling path in the parallax image of the fourth embodiment.

Fig. 14 is an explanatory view showing a travel path extension area in the parallax image of the fourth embodiment.

Fig. 15 is a flow chart showing the flow of obstacle detection in the fourth embodiment.

[Example 1]

Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. As an embodiment of the moving body in the present invention, a golf cart that drives autonomously can be cited. The mobile body in the present invention is not limited to a golf cart, and includes an unmanned transport vehicle that travels in a factory or an orchard. Further, the moving body in the present invention is not limited to a four-wheeled vehicle, and may be a tricycle or a monorail type. In the following description, the front, rear, and left and right directions are based on the forward direction of the moving body.

1. The schematic composition of the vehicle

Refer to Figure 1. Fig. 1 is a front elevational view of the vehicle 1 of the embodiment. The vehicle 1 is a golf cart that autonomously travels within a golf course. The vehicle 1 can be driven autonomously by electromagnetic wave induction generated by a sensing line buried in the traveling path. Three-dimensionally disposed at the center of the front of the vehicle 1 Camera 3.

Further, the vehicle 1 includes a right front wheel 5 and a left front wheel 6 that are controlled by the rotation of the steering wheel 4. A reading unit 7 is provided below the vehicle 1, and the reading unit 7 reads a fixed point Fp (FIG. 5) that is previously embedded in the traveling path Tr. In addition to the fixed point Fp embedded in the starting position So, a plurality of fixed points Fp are embedded in the traveling path Tr. The fixed point Fp is composed of a combination of magnetic poles of a plurality of magnets. The reading unit 7 is a magnetic sensor that reads such magnetic field information from the fixed point Fp. The magnetic field information includes position information indicating the position of the start position So or a specific position, and the speed control information such as the temporary stop or the increase or decrease of the vehicle speed. Furthermore, the fixed point Fp can be not only a magnet but also an RF tag used in an RFID (radio frequency identifier).

The right front wheel 5 includes a rotation angle sensor 8 that detects the rotation angle of the right front wheel 5, and a wheel speed sensor 9 that detects the rotation speed of the right front wheel 5. Further, a touch panel display 12 for displaying various kinds of information is provided in the vicinity of the steering wheel 4. The rotation angle sensor 8 detects the rotation angle of the wheel, for example, a rotary encoder. Further, the rotation angle sensor 8 and the wheel speed sensor 9 may be provided to the left front wheel 6 or the rear wheel without being disposed on the right front wheel 5.

Next, refer to FIG. 2. FIG. 2 is a functional block diagram showing the configuration of the vehicle 1. The vehicle 1 includes an obstacle detecting device 14 that detects an obstacle on the traveling path, an autonomous driving control unit 15 that controls autonomous driving of the vehicle 1 along the sensing line, and a vehicle speed control unit 16 that drives by autonomously. And detecting the increase and decrease of the vehicle speed by detecting the obstacle; and the alarm device 17, which warns the front of the vehicle and the passenger by the detection of the obstacle. The autonomous travel control unit 15 uses the previous control method of autonomous travel, and therefore, the description herein is omitted. The alarm 17 is disposed on the front side of the vehicle 1, and it is possible to inform the surroundings by the sound that the vehicle 1 is approaching.

2. Composition of obstacle detection device

Next, the configuration of the obstacle detecting device 14 provided in the vehicle 1 will be described with reference to Fig. 3 . FIG. 3 is a block diagram showing the configuration of the obstacle detecting device 14.

The obstacle detecting device 14 includes: a distance information detecting portion 21 that is paired up to the front of the vehicle 1 The distance information of the object is detected; the parallax image generating unit 23 generates a parallax image in front of the vehicle 1; and the traveling path position information storage unit 25 stores the traveling path position information, and the traveling path position information is based on The travel route on the parallax image is specified from the position of the vehicle 1 at the start position So; the position estimation unit 27 estimates the position of the vehicle 1 on the travel route; and the travel route specifying unit 29 based on the estimated position The traveling path in the parallax image is specified with the traveling route position information; the obstacle detecting unit 31 detects the obstacle based on the parallax image; and the determining unit 33 determines whether the detected obstacle is in the specified driving manner. On the path, the distance calculation unit 34 calculates distance information based on the parallax information, and the person determination unit 35 determines whether or not the detected obstacle is a person.

The distance information detecting unit 21 includes a stereo camera 3 that captures an image in front of the vehicle 1 and a buffer 41 that temporarily stores an image captured by the stereo camera 3.

The stereo camera 3 is composed of two image sensors, a left image sensor 3a and a right image sensor 3b. The left image sensor 3a and the right image sensor 3b are general visible light sensors such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor). The left image sensor 3a, the right image sensor 3b is disposed in the vehicle 1 under predetermined geometric conditions. In the first embodiment, the left image sensor 3a and the right image sensor 3b are disposed in the horizontal direction while maintaining a fixed distance. That is, the left image sensor 3a and the right image sensor 3b are respectively arranged in a positional relationship of parallel solid shapes.

The left image sensor 3a and the right image sensor 3b are arranged such that the positions of the respective columns of the respective captured images are identical, that is, the core lines are aligned. The image captured by the left image sensor 3a is set as the left image, and the image captured by the right image sensor 3b is set as the right image. In the first embodiment, the left image sensor 3a is set as the reference camera, and the left image is set as the reference image. Furthermore, the number of image sensors included in the stereo camera 3 is not limited to two, and may be three or more. Also, the right image may be set as the reference image.

The coordinate system of the image sensors 3a, 3b will be described with reference to Fig. 4 . The X axis is set to stand In the horizontal direction (left-right direction) of the body camera 3, the Y-axis is set to the longitudinal direction (up-and-down direction) of the stereo camera 3, and the Z-axis is set to the front direction (depth direction) of the stereo camera 3. Furthermore, the X-axis and the Y-axis are coordinate axes of the left image and the right image. The parallax information of the left image and the right image is distance information from the obstacle detecting device 14 to the object in front of the vehicle 1.

Returning to Fig. 3 for explanation. The buffer 41 temporarily holds the images transmitted by the stereo camera 3, that is, the images transmitted by the image sensors 3a, 3b. The buffer 41 uses a memory, a flash memory, a hard disk (HDD (Hard-Disk Drive)), or the like. Further, in FIG. 3, the buffers 41 are respectively provided corresponding to the image sensors 3a, 3b, but may be one buffer. Furthermore, it is preferable that each of the stored images is an image obtained by correcting lens distortion and unevenness of the focal length using the correction parameters. With this correction, the texture of the face perpendicular to the optical axis of the image sensor is also projected onto the image projection surface.

The parallax image generating unit 23 generates a parallax image based on each of the input images. That is, a parallax image is generated based on the left image and the right image stored in the buffer 41. The parallax image can be generated by stereo matching such as SAD (Sum of Absolute Differences). As a method of stereo matching, an area correlation method or a Census transform can also be cited. The parallax here represents the offset of the pixels between a plurality of images. In Embodiment 1, the parallax is the offset of the pixel in the lateral direction with respect to the right image of the left image. Furthermore, the X coordinate and the Y coordinate on the parallax image are the same as the X coordinate and the Y coordinate on the reference image. The generated parallax image is output to the obstacle detecting unit 31 and the traveling path specifying unit 29.

The travel route position information storage unit 25 stores the position information of the travel route based on the parallax value and the X coordinate value on the parallax image in accordance with the distance calculated from the start position So of the travel path. As shown in FIG. 5, from the start point So on the travel path Tr, for example, every equidistant interval, the parallax value and the X coordinate value of the travel paths of the respective points S1 to Sn on the parallax image are previously measured and memorized. . Further, in FIG. 5, the display of the place is omitted on the way.

Description will be made with reference to Fig. 6 . Fig. 6 shows, for example, a parallax image P1 of the point Sa at a distance La from the start position So in Fig. 5 . In the travel route position information storage unit 25, the X coordinate value of the travel route Tr in the parallax image P1 is stored in advance in association with the parallax value. That is, with respect to the X coordinate of the region depends on the difference da X ₁ -X _10, X ₂ -X ₉ with respect to the X-coordinate of the region depends on the difference between db, dc X coordinate of the region depends on the difference with respect to X ₃ -X _8, The X coordinate area X ₄ -X _{7 of the} parallax difference dd and the X coordinate area X ₅ -X _{6 of the} parallax value de are stored in the travel path position information memory unit 25 as the travel route position information of the point Sa. Furthermore, the relationship of the disparity values da~dg is da>db>dc>dd>de>df>dg.

In this manner, the travel route position information storage unit 25 stores, in association with each of the points S1 to Sn, the X coordinate value and the parallax value of the travel route Tr in the parallax image of the place. The travel route position information storage unit 25 uses a memory, a flash memory, a hard disk (HDD), or the like.

Returning to Fig. 3 for explanation. The position estimation unit 27 includes a reading unit 7, a rotation angle sensor 8, and a position calculation unit 43. The detected values of the reading unit 7 and the rotation angle sensor 8 are output to the position calculating unit 43.

The position calculating unit 43 calculates the current position of the vehicle 1 from the start position So based on the input detected values using the simplified distance method. In other words, the position information of the start position So or each fixed point Fp is input from the reading unit 7 to the position calculating unit 43. Further, by counting the detected values of the rotation angle sensor 8, the position calculating unit 43 calculates the moving distance from the start position So or each fixed point Fp. The calculated position information of the vehicle 1 is output to the travel route position information storage unit 25, and the travel route position information corresponding to the position information of the vehicle 1 is read. The read travel route position information is output to the travel path specifying unit 29.

The vehicle 1 travels along the predetermined travel route Tr, and therefore, the position of the vehicle 1 on the travel route Tr can be specified based on the travel distance from the predetermined position, that is, the start position So or each fixed point Fp. In this way, the position estimation unit 27 estimates the position of the vehicle 1 on the travel route Tr based on the travel distance calculated from the start position So or each fixed point Fp. Therefore, even if the overall coordinate system is not used, it is possible to estimate only using the vehicle coordinate system. Exit the location of the vehicle 1. Furthermore, since the moving distance is one-dimensional data amount, the position of the specific vehicle can be significantly reduced. The amount of information. Further, the position estimating unit 27 can correct the error of the calculated position by reading the magnetic field information of the fixed point Fp embedded in the position other than the start position So. Thereby, the current position of the vehicle 1 can be estimated with high precision.

The travel route specifying unit 29 specifies the travel route Tr on the parallax image generated by the parallax image generation unit 23 based on the input travel route position information. The travel route is specified based on the disparity value corresponding to the pre-memorized X coordinate value. The parallax image P2 shown in FIG. 7 is an image of the same place as the parallax image P1 shown in FIG. 6. In Fig. 7, the X coordinate area of the parallax value de in the travel path position information is X ₅ - X ₆ , but there is no region where the X coordinate is the parallax value de of X ₅ due to the obstacle Ob4. In this case, the range of the disparity value de between the X coordinate regions X ₅ - X ₆ is specified as the travel path. The specified travel route area is output to the determination unit 33.

The obstacle detecting unit 31 detects an area in the parallax image as an obstacle, and the different disparity values in the X direction of the area are close to each other, and the area of the same disparity value has a pixel number in the Y direction of a predetermined value or more. In other words, the disparity values are the same and have a predetermined height. In FIGS. 6 and 7, the regions Ob1, Ob2, Ob3, and Ob4 are detected as obstacles. The regions Ob1 to Ob4 of the obstacles detected are output to the determination unit 33. Furthermore, the region of the obstacle is not limited to the same disparity value, and a disparity value may be used as a reference to process the region of the disparity value of the predetermined range, thereby specifying a plurality of parallaxes. The area of the obstacle of value.

The determination unit 33 determines whether or not the detected obstacle exists on the travel path Tr. The determination unit 33 determines whether or not the range of the X coordinate value at the lower end of the region of the input obstacle is included in the range of the X coordinate of each parallax of the traveling path. Thereby, for example, it is determined that only the obstacle Ob4 in the obstacles Ob1 to Ob4 shown in FIG. 7 is on the traveling path Tr.

Further, the determination unit 33 may perform the two-stage determination of the above-described determination for the obstacle having the rectangular region overlapping the traveling path region after determining whether or not the rectangular region surrounding the obstacle region overlaps with the traveling route region. Faster by this 2-stage decision Quickly determine if an obstacle exists on the travel path. Further, when the value of the Y coordinate (height) at the lower end of the rectangle is larger than the predetermined height, it is determined that a part of the tree such as a bridge or a branch that is placed over the traveling path is not an obstacle, and the obstacle can be determined. Increased accuracy.

When it is determined that the obstacle exists on the traveling path Tr, the distance calculating unit 34 calculates the distance in the actual space up to the obstacle based on the parallax information of the obstacle region. The method of calculating the distance from the parallax image uses the parallel stereo method previously used. The calculated distance is output to the vehicle speed control unit 16. The vehicle speed control unit 16 slows down the vehicle speed of the vehicle 1 or stops the vehicle 1 based on the distance up to the obstacle. The closer the distance between the vehicle 1 and the obstacle is, the more the speed is decelerated. Thereby, the vehicle 1 can be prevented from coming into contact with an obstacle.

The person determination unit 35 determines whether or not the area on the reference image corresponding to the area of the obstacle existing on the traveling path among the parallax images is a person. The human decision method uses a combination of an image local feature quantity (HOG (Histogram of Oriented Gradient) feature quantity) and a statistical learning method. Further, in addition, the person determination can be performed by using the matching of the template prepared in advance. If it is determined that the area of the obstacle in the reference image is a person, the touch panel display 12 evokes the rider's attention to the person in front of the vehicle 1. Further, the alarm 17 sounds a siren to warn a person who is in front of the vehicle 1. When it is determined that the region of the obstacle in the reference image is not a human, the touch panel display 12 evokes the rider to notice that there is an obstacle in front of the travel path.

The parallax image generating unit 23, the traveling path specifying unit 29, the obstacle detecting unit 31, the determining unit 33, the distance calculating unit 34, the person determining unit 35, and the position calculating unit 43 are microprocessors or FPGAs (Field Programmable Gate Array). , field programmable gate array).

Next, the obstacle detecting operation in the first embodiment will be described with reference to Fig. 8 . Fig. 8 is a flow chart showing the processing procedure of distance inference.

After the vehicle 1 starts moving from the start point So of the travel route Tr, the position estimating unit 27 estimates the current position of the vehicle 1 on the travel route Tr (step S01). Also, the stereo camera 3 The image in front of the vehicle 1 is taken at predetermined time intervals. A plurality of images are taken in the above manner, thereby obtaining parallax information (distance information). Further, the parallax image generating unit 23 generates a parallax image based on the plurality of captured images, thereby generating three-dimensional information (step S02). Next, the travel route specifying unit 29 specifies the travel route Tr in the parallax image based on the travel route position information corresponding to the current position of the vehicle 1 (step S03). Further, the obstacle detecting unit 31 detects an obstacle in the generated parallax image (step S04). Next, the determination unit 33 determines an obstacle that determines whether or not the detected obstacle is on the traveling path Tr (step S05).

When it is determined that there is no obstacle on the traveling path Tr in the obstacle determination (NO in step S06), the obstacle detecting operation is ended, and the process starting from step S01 is started. When it is determined that there is an obstacle on the traveling path Tr in the obstacle determination (YES in step S06), the person determining unit 35 determines whether or not the obstacle is a person (step S07). When the obstacle is not in the human state (NO in step S08), the vehicle speed control unit 16 performs deceleration or stop control including the brake brake on the vehicle 1 based on the distance from the obstacle (step S10). When it is determined that the obstacle is a person (YES in step S08), a sound is emitted from the alarm 17 of the vehicle 1 to the front of the vehicle, and the sound warns that the vehicle 1 is approaching (step S09). Further, the vehicle speed control unit 16 performs deceleration or stop control including the brake brake on the vehicle 1 based on the distance from the person (step S10).

As described above, according to the first embodiment, the travel route position information of the travel route Tr on the three-dimensional information is specified as the two-dimensional information of the disparity value of the distance information and the X coordinate value of the coordinate information as the horizontal direction of the vehicle 1. Compared with the previous use of three-dimensional information, the amount of information processed can be reduced, thereby reducing the computational load. Thereby, the calculation time can be shortened, and it can be determined whether or not the obstacle is in the traveling path at the time when the distance is secured with the obstacle, so that contact with the obstacle can be avoided.

Further, by limiting the determination of the obstacle to the traveling path Tr, it is not determined that the three-dimensional object such as a tree beside the traveling path is an obstacle, and therefore, the vehicle 1 can be prevented from being stopped by mistake. The vehicle 1 is driven to travel smoothly. In this way, the vehicle 1 can be surely decelerated and stopped with respect to the obstacle on the traveling path Tr having a high possibility of contact.

Further, the distance information detecting unit 21 has a stereo camera 3, whereby a parallax image can be used as the three-dimensional information. Thereby, the determination of the traveling path Tr and the detection of the obstacle in the parallax image can be appropriately performed. Further, if a one-dimensional scanning laser radar is used, it is easy to detect the road surface as an obstacle due to the undulation of the road surface. However, by using the stereo camera 3, it is less likely to be affected by the road surface undulation and the obstacle is detected.

Further, in the case where the estimation accuracy of the current position is to be further improved, the interval for correcting the error may be shortened. For example, it is only necessary to increase the number of fixed points Fp buried in the traveling path. Thereby, it is not necessary to mount an expensive GPS or a sensor to each vehicle, and therefore the current position can be estimated at low cost and with high precision.

[Embodiment 2]

Next, an obstacle detecting device of the second embodiment will be described with reference to Fig. 9 . Fig. 9 is a block diagram showing the configuration of an obstacle detecting device in the second embodiment. In the second embodiment, the portions denoted by the same reference numerals as those in the first embodiment are the same as those in the first embodiment, and therefore the description herein will be omitted. Further, the configuration of the vehicle and the obstacle detecting device other than the following description is the same as that of the first embodiment.

The obstacle detecting device 14' of the second embodiment is characterized in that the distance measuring device 21' is used instead of the stereo camera 3 as the distance information detecting portion 21 of the obstacle detecting device 14 in the first embodiment. In the first embodiment, the two pieces of image sensors of the stereo camera 3 are used to obtain the parallax information, thereby detecting the distance information. In the second embodiment, the distance information from the obstacle detecting device 14' to the object in front of the vehicle 1 is directly detected by using the distance measuring device 21'. Therefore, in Embodiment 2, the measured value of the distance information is used instead of the disparity value of Embodiment 1.

The obstacle detecting device 14' in the second embodiment is configured to include a distance measuring device 21' instead of the stereo camera 3. The distance measuring device 21' is, for example, a radar or a laser for distance measurement. Radar (LIDAR). The distance information in the polar coordinates can be obtained by radar scanning. Therefore, in the second embodiment, the scanning angle θ is used instead of the X coordinate value of the first embodiment. Further, in place of the parallax image generating unit 23 of the first embodiment, the distance image generating unit 23' is included in the second embodiment. The distance image generating unit 23' generates a distance image based on the detection result of the distance measuring device 21'. The horizontal axis of the distance image is the scanning angle θ, the vertical axis is the Y axis, and the pixel value is the distance information. Furthermore, it is preferable that the Y coordinate value is at least two points or more. That is, it is preferable to obtain the distance information at different heights by the distance measuring device 21'. Further, in the travel route position information storage unit 25' in the second embodiment, the θ value and the distance value r of the travel route Tr in the distance image of the place are stored in association with each of the points S1 to Sn. For example, the following θ value is stored as a travel route data in advance, and the θ value indicates a range of the travel route Tr in units of 1 m with respect to the distance value r. The travel route specifying unit 29 in the second embodiment specifies the travel route Tr on the distance image generated by the distance image generating unit 23' based on the travel route data stored in the travel route position information storage unit 25'. The travel route is specified based on the distance value corresponding to the pre-memorized scan angle θ. The obstacle detecting unit 31 in the second embodiment first discriminates the measurement point group detected at a position lower than the predetermined Y coordinate on the distance image as the ground, and excludes the measurement point group. The predetermined Y coordinate value changes according to the r value, and the two values are stored in association with the travel path position information memory portion 25'. For the remaining measurement point groups, a plurality of measurement points having a close distance on the θ-r plane are grouped and detected as obstacles. The distance to group is set in advance. Furthermore, when the distance coordinate information is obtained by only one point of the Y coordinate value, the discrimination to the ground can be omitted, and the obstacles can be detected only by grouping on the θ-r plane. Next, the determination unit 33 in the second embodiment determines whether or not the detected obstacle exists on the travel path Tr. The determination unit 33 determines whether or not the range of the θ value and the r value at the lower end of the region of the input obstacle are included in the range of the θ value and the r value of the specific travel path. Further, in the same manner as in the first embodiment, the determination unit 33 may perform the above determination on the obstacle having the rectangular region overlapping the traveling path region after determining whether or not the rectangular region surrounding the obstacle region overlaps with the traveling route region. 2 stages determination. Further, in the second embodiment, the distance calculating unit 34 in the first embodiment is omitted. This is because the distance in the actual space up to the obstacle is calculated by the distance measuring device 21'. Further, the person determining unit 35 in the second embodiment determines whether or not the area on the reference image corresponding to the area of the obstacle existing on the traveling path in the distance image is a person.

As described above, according to the second embodiment, the distance measuring radar is used as the distance information detecting unit, whereby the distance information r with respect to the scanning angle θ can be appropriately obtained. In comparison with the use of the stereo camera 3, the distance can be measured particularly accurately in the distance, so that the determination of the travel path Tr and the detection of obstacles in the three-dimensional information can be appropriately performed.

[Example 3]

Next, an obstacle detecting device of the third embodiment will be described with reference to Fig. 10 . Fig. 10 is a block diagram showing the configuration of an obstacle detecting device in the third embodiment. In the third embodiment, the portions denoted by the same reference numerals as those in the first embodiment are the same as those in the first embodiment, and therefore, the description herein will be omitted. Further, the configuration of the vehicle and the obstacle detecting device other than the following description is the same as that of the first embodiment.

In the first embodiment, it is determined whether or not the obstacle that has been determined to be on the traveling path Tr is a person, thereby detecting the person. On the other hand, the obstacle detecting device 54 of the third embodiment is characterized in that it is determined whether or not the obstacle existing in the range including the traveling path Tr of the traveling path Tr is a person, thereby detecting the person. In the case where the person is detected in the above manner, it is preferable to detect the person in the range of the width of the traveling path, which is larger than the specific traveling path Tr area by, for example, 1 rn in the actual distance. about. Thereby, it is also possible to detect a person whose one part of the body is in the traveling path Tr area, and therefore, the avoidance action can be taken in advance.

The obstacle detecting device 54 of the third embodiment is characterized in that the traveling path extending region specifying portion 64 is provided in the configuration of the obstacle detecting device 14 in the first embodiment. In the first embodiment, it is determined by the determination unit 33 whether or not an obstacle exists in the travel route specified by the travel route specifying unit 29. In the third embodiment, in addition to this, as shown in FIG. 11, the travel route extension area is specified. The portion 64 is formed by expanding the predetermined range from the traveling path Tr outward in the width direction to the traveling path extending region Tr'. In the travel path extension area Tr', the determination of the obstacle and the judgment of the person are performed. The predetermined range of expansion is in the width direction of the traveling path, and the range of, for example, about 1 m is expanded by the actual distance. That is, the traveling path area determined by the obstacle-free object is expanded to the outside by the number of pixels corresponding to the following parallax as compared with the specific traveling path width shown in FIG. 6, and the parallax corresponds to 1 m.

In addition to the function of the determination unit 33 of the first embodiment, the determination unit 33' of the third embodiment determines whether or not the obstacle detected by the obstacle detection unit 31 exists in the travel path specified by the travel path extension area specifying unit 64. Extended area Tr'. Further, in addition to the function of the person determination unit 35 of the first embodiment, the person determination unit 35' of the third embodiment determines whether or not the obstacle is a person, and the obstacle is determined to exist in the travel path extension area specifying unit 64. The specific travel path extension area Tr' is specified.

According to the above configuration, the determination unit 33' and the person determination unit 35' can change the respective processing frequencies of the obstacle determination and the person determination on the travel route Tr and the travel route extension region Tr'. That is, the processing frequency of the person detection around the traveling path can be reduced, and the person around the traveling path detects the processing time and the detection result is not used for the speed control. Therefore, the frame rate of the obstacle detection processing on the traveling path Tr can be ensured. As a result, erroneous detection of an obstacle can be avoided, and the vehicle 1 can be stably stopped. Compared with the detection result using only one frame, the detection result using a plurality of frames can suppress the erroneous stop caused by the erroneous detection. Therefore, it is preferable that the obstacle detection processing on the traveling path Tr has a high frame rate.

In this way, the vehicle 1 can be surely decelerated and stopped with respect to the obstacle on the traveling path Tr having a high possibility of contact. Further, a person who is in the vicinity of the end portion in the width direction of the traveling path is easily confused with a stationary object such as various trees around the traveling path. Further, it is difficult to determine the possibility of contact with a part of the body such as a human hand. In such a person, an alarm can be issued by detecting a person to move the person to a position away from the traveling path, whereby the contact rate can be reduced, and an unpleasant erroneous stop can be avoided.

[Example 4]

Next, an obstacle detecting device of the fourth embodiment will be described with reference to Fig. 12 . Fig. 12 is a block diagram showing the configuration of an obstacle detecting device in the fourth embodiment. In the fourth embodiment, the portions denoted by the same reference numerals as those in the first embodiment are the same as those in the first embodiment, and therefore the description herein will be omitted. Further, the configuration of the vehicle and the obstacle detecting device other than the following description is the same as that of the first embodiment.

In the first embodiment, an obstacle is detected from the entire parallax image generated, and it is determined whether or not the detected obstacle is on the traveling path. On the other hand, in the fourth embodiment, the parallax image on the traveling path is extracted from the generated parallax image, and the obstacle is detected from the parallax image on the traveling path that is captured. Thereby, the area where the obstacle is detected is reduced, and therefore, the calculation load can be further reduced.

The parallax image capturing unit 66 on the traveling path of the fourth embodiment captures only the traveling path based on the traveling path position information stored in the traveling path position information storage unit 25 with respect to the parallax image generated by the parallax image generating unit 23. Parallax image. For example, the parallax image shown in FIG. 7 is generated by the parallax image generating unit 23. The parallax image capturing unit 66 on the traveling path captures the parallax image of the traveling path Tr region and the parallax map that is higher in the Y direction than the traveling path Tr having the same parallax value in each of the parallax regions. image. Fig. 13 shows the parallax image P3 on the travel path that is captured.

The area Ob4 has a disparity value dc. The area Ob4 is captured as a parallax image in the Y direction upwards on the traveling path Tr between the X coordinate areas X ₃ -X ₈ having the parallax difference dc on the traveling path Tr. In addition, the region Ob1 'in line as an upper traveling path Tr representing the X-coordinate of the region depends on the difference between de X ₅ -X ₆ of the traveling path Tr closer to the Y-direction parallax image is to be captured. Since only the parallax image on the traveling path Tr is captured, not only the traveling path Tr but also the parallax images of FIG. 7 of the first embodiment are deleted, and Ob2 and Ob3 are deleted, and the obstacle other than the traveling path Tr may be omitted. The object is tested.

The parallax image on the travel path that is captured is transmitted to the obstacle detecting portion 31'. obstacle Similarly to the first embodiment, the object detecting unit 31' detects an obstacle from the parallax image P3 on the traveling path. As a result, the region Ob4 is detected as an obstacle. The region Ob1' is not detected as an obstacle because the minimum value of the Y coordinate value is away from the traveling path Tr. The person determination unit 35 determines whether or not the detected obstacle Ob4 is a person. Further, the obstacle detecting device 56 may further include an extended area capturing portion 64'. As shown in FIG. 14, the extended area capturing unit 64' extracts the parallax information on the traveling path expansion area Tr' in which the predetermined traveling range Tr is expanded outward in the width direction by a predetermined range. FIG. 14 shows, for example, the traveling path Tr and the parallax image P4 on the traveling path extension area Tr' captured when the parallax image generating unit 23 generates the parallax image P1 shown in FIG. 6 . The obstacle detecting unit 31' also detects an obstacle on the above-described captured travel path extension area Tr'. When an obstacle is detected on the travel path extension area Tr', the person determination unit 35 determines whether or not the detected obstacle is a person. Thereby, the same effect as in the case of including the traveling path extension region specifying portion 64 in the third embodiment can be obtained.

Next, the operation of the obstacle detection in the fourth embodiment will be described with reference to Fig. 15 . Fig. 15 is a flow chart showing the processing procedure of obstacle detection. In addition, steps S01 and S02 are the same as those of the first embodiment, and thus the description thereof is omitted.

The parallax image capturing unit 66 on the traveling path extracts the parallax image on the traveling path from the parallax image based on the traveling route position information corresponding to the current position of the vehicle 1 (step S03). Next, the obstacle detecting unit 31' starts detecting an obstacle in the parallax image on the travel route that is captured (step S04). If an obstacle is not detected in the parallax image on the traveling path (NO in step S06), the obstacle detecting operation is ended, and the process starting from step S01 is started. In the obstacle detection, if an obstacle is detected on the travel path Tr (YES in step S06), the person determination unit 35 determines whether or not the obstacle is a person (step S07). In addition, the process after step S07 is the same as that of the first embodiment, and thus the description thereof is omitted.

As described above, according to the fourth embodiment, since the stationary object outside the traveling path is not detected as an obstacle, the calculation load can be reduced. For example, when detecting a person leaning against a wall such as a driving path, the wall and the person are collectively detected as an obstacle, and therefore, it is difficult The individual is detected as a human. However, according to the fourth embodiment, the fence outside the travel path has been removed from the parallax image, and therefore, only the person inside the travel path can be detected as an obstacle, so that the detection accuracy of the obstacle can be improved. Further, since the processing for the traveling route in steps S03 to S10 can be performed on the traveling path extension area, it is possible to detect a person in the vicinity of the traveling path and issue a warning. Furthermore, in the case of using a radar instead of a stereo camera, it is only necessary to mask the data outside the range of the travel path.

The present invention is not limited to the above embodiments, and can be implemented in the following manner.

(1) In the third embodiment, it is determined whether or not the obstacle that has been determined to be on the traveling path is a person, thereby detecting the person. The present invention is not limited thereto, and may include a person detecting unit that detects a person in a range wider than the traveling path Tr including the traveling path Tr in parallel with the detection of the obstacle. Thereby, it is also possible to detect a person whose one part of the body is in the traveling path Tr area, and therefore, the avoidance action can be taken in advance. Further, since the person detecting unit that is separate from the obstacle detecting unit 31 is included, it is possible to perform different processes in parallel. In other words, it is possible to treat different obstacles in the traveling path with high possibility of contact, and people in the vicinity of both ends in the width direction of the traveling path which is difficult to determine the contact possibility and which is easily confused with various trees around the traveling path. .

(2) In the first to fourth embodiments, the obstacle detecting device obtained by the combination of the respective configurations may be configured.

(3) In the above embodiment, the obstacle detecting device 14 is provided to the vehicle 1, but is not limited thereto. Further, for example, it can be used for a vision system for autonomously driven robots or a support system for visually impaired persons.

3‧‧‧ Stereo Camera

3a‧‧‧Left image sensor

3b‧‧‧Right image sensor

7‧‧‧Reading Department

8‧‧‧Rotary angle sensor

16‧‧‧Speed Control Department

17‧‧‧Alarm

21‧‧‧Distance Information Inspection Department

23‧‧‧ Parallax Image Generation Department

25‧‧‧ Driving Path Location Information Memory

27‧‧‧Location Estimation Department

31'‧‧‧The Obstacle Detection Department

34‧‧‧Distance calculation department

35‧‧‧ Personage Department

41‧‧‧buffer

43‧‧‧Location calculation department

56‧‧‧ obstacle detection device

64'‧‧‧Extended Area Capture Department

66‧‧‧Disparity Image Capture on the Driving Path

Claims (15)

An obstacle detecting device mounted on a moving body traveling along a predetermined traveling path, and comprising: a distance information detecting unit that detects distance information from the obstacle detecting device to an object in front of the moving body; a three-dimensional information generating unit that generates three-dimensional information in front of the moving body based on the distance information; and a memory unit that stores traveling path position information corresponding to a position of the moving body from a predetermined position; and a position estimating unit, And estimating a position of the moving body on the traveling path; and a three-dimensional information capturing unit on the traveling path, which is based on the traveling path position information, and extracts three-dimensional information on the driving path in the three-dimensional information; and an obstacle detecting unit, The obstacle is detected based on the three-dimensional information on the travel path; and the travel route position information is specified by the distance information and the coordinate information of the horizontal direction of the mobile body. The obstacle detecting device of claim 1, comprising a person determining unit that determines whether the obstacle on the traveling path is a person. The obstacle detecting device of claim 2, comprising: an extended area three-dimensional information capturing unit that extracts a predetermined area adjacent to the traveling path, that is, the driving path expansion area, based on the traveling path position information In the three-dimensional information, the obstacle detecting unit detects an obstacle on the travel path extension area, and the person determination unit determines that the obstacle is detected in the travel path extension area. Whether the obstacle is a person. The obstacle detecting device of claim 1, comprising: an extended area three-dimensional information capturing unit that extracts three-dimensional information in a predetermined area adjacent to the traveling path, that is, in the driving path expansion area, based on the traveling path position information; The detecting unit detects the person in the travel path extension area based on the three-dimensional information. An obstacle detecting device mounted on a moving body traveling along a predetermined traveling path, and comprising: a distance information detecting unit that detects distance information from the obstacle detecting device to an object in front of the moving body; The three-dimensional information generating unit generates three-dimensional information in front of the moving body based on the distance information; the memory unit stores traveling path position information, and the traveling path position information is based on the position of the moving body from the predetermined position. Specifying a travel route on the three-dimensional information; the position estimation unit estimates a position of the moving body on the travel path; and the travel route specifying unit specifies the above based on the estimated position and the travel route position information a traveling path in the three-dimensional information; an obstacle detecting unit that detects an obstacle based on the three-dimensional information; and a determining unit that determines whether the detected obstacle is on the specific traveling path; and the driving path position The information is based on the above distance information and the horizontal direction of the above moving body. The specific coordinates information. The obstacle detecting device of claim 5, comprising a person determining unit that determines whether the obstacle that has been determined to be on the traveling path is a person. An obstacle detecting device according to claim 6, comprising: a region specifying portion that specifies a traveling path extending region that is adjacent to a predetermined region of the specified traveling route, wherein the determining portion determines the detected region Whether or not the obstacle is in the specific travel path extension area that has been specified, the person determination unit determines whether or not the obstacle that has been determined to be on the travel path extension area is a person. The obstacle detecting device of claim 5, comprising: a region specifying portion that specifies a travel path extension region that is adjacent to a predetermined region of the specific travel route; and a person detecting portion The person is detected in the above-mentioned travel path extension area that has been specified. The obstacle detecting device according to any one of claims 1 to 8, wherein the position estimating unit estimates the position of the moving body on the traveling path based on a moving distance calculated from a predetermined position. The obstacle detecting device according to any one of claims 1 to 8, wherein the distance information detecting unit includes a stereo camera, and the three-dimensional information is a parallax image. The obstacle detecting device according to any one of claims 1 to 8, wherein the distance information detecting portion includes a radar. A moving body comprising the obstacle detecting device according to any one of claims 1 to 11. The mobile body of claim 12, comprising an alarm that issues a warning when the obstacle is detected on the travel path. A moving body according to claim 12 or 13, comprising a speed control unit that decelerates the moving body when detecting the obstacle on the traveling path stop. A moving body, comprising the obstacle detecting device according to any one of claims 3, 4, 7, and 8, and comprising: a speed control portion for causing the detected obstacle on the traveling path to The moving body is decelerated or stopped; and an alarm that issues a warning to the detected person on the extended path of the travel path.