JP2022165563A - Autonomous travel inspection robot - Google Patents

Abstract

To provide an autonomous travel inspection robot capable of autonomous travel without being affected by sunlight or shadows and suitable for monitoring outdoor facilities.SOLUTION: An autonomous travel inspection robot that patrols a facility to be monitored along a preset prescribed route to monitor the state of the facility, and comprises travel means, a video camera that can image a road surface, a scanning type distance sensor, measurement means that acquires the state of the facility, and a control device, also includes: a first travel mode in which the control device detects the distance and direction of an object by sensing with the distance sensor and controls the travel means so as to move along the route; and a second travel mode in which the direction is specified by detecting a strip body set on the road surface and the travel means is controlled to move along the strip body. Either the first travel mode or the second travel mode is set as the basic travel mode and when a prescribed condition is met, switching to the second travel mode or the first travel mode is repeated by automatic operation according to the state of the travel location.SELECTED DRAWING: Figure 9

Description

The present invention relates to an autonomous patrol robot that monitors the status of facilities while moving within a predetermined area, and more particularly to an autonomous patrol robot that monitors the status of facilities while moving within the premises of outdoor facilities such as substations. It relates to a technology useful for applying to.

At railway substations, facilities such as circuit breakers, transformers, and rectifiers are installed on a relatively large site. Although work to ensure soundness is being carried out by confirming the presence or absence of such work, it is expected that it will become difficult to secure personnel to undertake such work as the domestic population declines in recent years. Therefore, it is conceivable to patrol using a moving body equipped with detection sensors such as a camera, a microphone, and an odor sensor.
Conventionally, there is an invention relating to a monitoring self-propelled machine described in Patent Document 1, for example, as an invention relating to a device for monitoring the state of facilities while moving in a building.

The self-propelled monitoring machine described in Patent Document 1 includes a main body, a drive mechanism for running the main body by remote control, a camera attached to the main body and operated by remote control, and an abnormal state of a monitoring point. a detection sensor (acoustic sensor, odor sensor, etc.) that detects the It is designed to be equipped with
In addition, in Patent Document 1, as a specific example of a self-propelled machine, one that runs in a suspended state along a rail arranged on the ceiling of a facility and one that runs on a rail or belt-shaped body arranged on the ground It is described that it runs along.

Japanese Patent Application Laid-Open No. 2004-164303 JP 2019-159620 A

When it is desired to monitor equipment in a facility without human intervention, the rail-type self-propelled monitoring machine described in Patent Document 1 can be installed indoors at a relatively low cost. If it is installed outdoors, it costs a lot of money, and since it is exposed to wind and rain, it easily deteriorates, so regular maintenance is also required. In addition, in facilities such as substations, the placement of equipment varies depending on the shape of the land and the environment. There is a problem that work is necessary.

On the other hand, the line trace type, in which the monitoring self-propelled machine runs along the strip, is effective for monitoring the equipment of outdoor facilities because it does not require large-scale rail facilities. It is necessary to lay a strip on the However, in a facility such as a substation, there is a gravel pavement on the route along which the self-propelled machine travels. Therefore, it is difficult to lay a strip (line) that serves as a mark, and maintenance is required to keep the laid strip from deteriorating. In addition, since substations often have relatively large equipment installed beside the driving route, self-position estimation cannot be performed because the belt-shaped object cannot be recognized due to the effects of outdoor sunlight and shadows of large equipment. There is a problem that there are many situations where it is not possible to

Therefore, in facilities such as substations where there are facilities to be monitored indoors and outdoors, the patrol equipment is divided into indoor and outdoor, and the outdoor patrol equipment runs not on a track system like a rail, but on a trackless system. It is desirable to adopt a system in which the vehicle autonomously travels the route. However, in autonomous driving, it is necessary for the mobile object to always know its own position and to accurately sense the equipment to be monitored. With current technology, the accuracy of position detection is low and the error in self-positioning is large, so there is a problem that appropriate autonomous driving cannot be expected when moving through narrow routes or places where high accuracy is required. .

The present invention has been made with a focus on the above-described problems, and is an autonomously traveling patrol robot that can autonomously travel without being affected by sunlight or shadows and is suitable for monitoring equipment in outdoor facilities. intended to provide
Another object of the present invention is to provide an autonomous traveling patrol robot suitable for monitoring facilities in outdoor facilities where rough roads and narrow places exist on the traveling route.

In order to achieve the above object, the present invention
A driving means consisting of driving wheels and driving means for rotating the driving wheels, a video camera capable of photographing the road surface, a scanning distance sensor, a measuring means for acquiring the state of equipment, and the video camera. and a control device capable of controlling the driving means based on the captured image or the signal from the distance sensor, and patrols the facility to be monitored along a predetermined route to monitor the condition of the facility. In an autonomous patrol robot,
The control device is
a first traveling mode in which the distance and direction of an object are detected by sensing by the distance sensor and the vehicle travels by controlling the driving means so as to move along the route;
a second running mode in which a direction is specified by detecting a strip provided on the road surface of the route and the vehicle travels by controlling the driving means so as to move along the strip;
While either the first running mode or the second running mode is set as the basic running mode, the operation of automatically switching to the second running mode or the first running mode when a predetermined condition is satisfied is performed according to the situation of the running place. It is configured to repeat

According to the autonomous traveling patrol robot having the configuration as described above, the control device has the first traveling mode in which the robot travels in autonomous operation based on the sensing of the distance sensor and the second traveling mode in which it travels in line tracing. When monitoring the equipment of the facility, in areas where sunlight and shadows occur and where it is difficult to set up a strip (line), the first driving mode is used to avoid the effects of sunlight and shadows and on the road surface. It can run autonomously without depending on the strip (line). In addition, when monitoring equipment of a facility where there is a narrow place on the travel route, by setting a line in the narrow place and running in the second travel mode, the detection accuracy of the self-position is improved. Even if it is low, it can be made to travel accurately along the desired route.

Here, desirably, the control device, in the second running mode,
An image of a predetermined region is cut out from the image captured by the video camera, normalization processing is performed on the cutout image, and the normalized image is binarized using a predetermined threshold value, From the binarized image data, a pattern having a width of a predetermined width or more and the longest continuous pattern in the running direction is extracted, and the extracted pattern is regarded as the belt-like body and is moved along the belt-like body. controlling the driving means;
In the binarization process, a binarization threshold value is set so that the number of bright spots falls within a predetermined range.

According to the configuration described above, in the binarization process, the binarization threshold value is set so that the number of bright spots falls within a predetermined range. Also, the pattern corresponding to the line can be more reliably extracted from the photographed image, and the line can be prevented from being lost during the second traveling mode in which the vehicle travels in line tracing.

Further, desirably, the control device, in the second running mode,
When the pattern cannot be extracted, the immediately preceding travel control is maintained until the vehicle travels a predetermined time or a predetermined distance, and when the pattern cannot be extracted during that time, the vehicle is switched to the first travel mode. to control the driving means so as to move toward a target set in advance on the route.
According to the configuration as described above, even if the belt-shaped body (line) is lost during the second running mode in which the line trace is run, it is possible to avoid the running stop state, and to quickly follow the prescribed running route. can return to

Further, preferably, the belt-shaped body is provided at a narrow part of the route or at a part where it is necessary to travel a preset traveling route with high accuracy. A marker is provided for instructing switching of the running mode.
According to such a configuration, switching between the first travel mode and the second travel mode can be reliably performed at a desired position, and autonomous travel can be performed accurately along a predetermined route.

Preferably, the facility to be monitored is a substation, and the measuring means includes at least a video camera for measurement, a thermo camera, a sound collecting microphone and an odor sensor,
The control device is configured to operate any one of the measuring means at a preset position on the route to detect the state of the facility and store the detection result in a storage device.

The site of the substation has many gravel areas, and the equipment is relatively high, so there is a lot of sunlight and shadows on the road surface on which the robot runs. According to this, it is possible to accurately autonomously travel along a predetermined route in a facility and detect and store the state of the desired facility. In addition, since it is equipped with a video camera for inspection, a thermo camera, a sound collecting microphone, and an odor sensor as measurement means, it can read the numerical values of instruments, and can detect heat, noise, and odors caused by abnormalities in circuit breakers, transformers, rectifiers, etc. can be detected, and can contribute to early detection of equipment abnormalities.

According to the autonomous traveling patrol robot of the present invention, it can autonomously travel without being affected by sunlight or shadows, and can monitor the equipment of outdoor facilities. In addition, there is an effect that it is possible to appropriately monitor the condition of equipment in an outdoor facility such as a substation where there are rough roads and narrow places on the travel route.

1 is a front view system configuration diagram showing an appearance of an embodiment of an autonomous traveling patrol robot according to the present invention; FIG. 1 is a side view of an autonomously traveling patrol robot according to an embodiment; FIG. FIG. 4 is a side view showing an arm extension state of the autonomous traveling patrol robot of the embodiment; 1 is a block diagram showing a configuration example of an in-vehicle system for an autonomous traveling patrol robot according to an embodiment; FIG. FIG. 2 is a plan view showing a charging base for charging the autonomous traveling patrol robot of the embodiment and the surroundings thereof; (A) is a diagram showing an image of a video camera for line detection and an image of a cutout area, and (B) is a diagram showing an image after conversion of the cutout area. (A), (B), and (C) are diagrams showing image examples when the original image is binarized by setting different threshold values. (A) is a diagram showing an example of a binarized image to be subjected to line determination processing, and (B) is a diagram showing a histogram generated by counting the number of bright spots in the X direction for the image of (A). 4 is a flowchart showing the procedure of line determination processing by the control device of the autonomous traveling patrol robot according to the embodiment;

An embodiment of an autonomously traveling patrol robot according to the present invention will be described below with reference to the drawings. 1 and 2 are a front view and a side view showing the appearance of an autonomously traveling patrol robot according to an embodiment.
As shown in FIGS. 1 and 2, the autonomous traveling patrol robot 10 according to the present embodiment includes a main body 11 containing a battery, a control device, a traveling drive motor, etc. Equipped with four running wheels 12A, 12B, and laser scanners 13A, 13B such as LIDER (Light Detection and Ranging) provided at the front and rear of the main body 11, respectively. It is configured so that it can autonomously run while recognizing the measure.

Two traveling drive motors are provided corresponding to the left and right traveling wheels 12A and 12B, and the direction of movement can be changed by varying the rotation speed of the two motors or by rotating one motor forward and the other in reverse. In addition, a rotation sensor such as a rotary encoder that detects the number of rotations of the left and right wheels 12A and 12B for traveling is provided, and the moving distance is calculated from the number of rotations, and the direction of movement is calculated from the difference in the number of rotations (rotational angle difference) between the left and right wheels. is configured to be able to grasp its own position by recognizing it.
Furthermore, the autonomous mobile patrol robot 10 is provided with a storage device that stores a control program and map information. It is configured to grasp its own position on the map. Instead of the rotation sensor, an acceleration sensor or an angular velocity sensor may be provided to grasp the own posture and orientation.

In addition, the autonomous traveling patrol robot 10 of the present embodiment has a video camera 14 as an imaging means for acquiring image information, a surface temperature of the facility, and a surface temperature sensor of the facility, in order to measure the state of the facility to be monitored such as a substation. It is equipped with a thermo camera 15 for acquiring information, a sound collecting microphone 16 for detecting abnormal sounds emitted by equipment, an odor sensor 17 for detecting abnormal odors emitted by equipment, and the like. A later-described video camera for detecting lines on the road surface may be provided separately from the video camera 14 for measurement, or may be shared.

Furthermore, as shown in FIG. 3, the autonomous traveling patrol robot 10 is provided with an elevating mechanism 20 for elevating the measuring head 18 on which the various measuring means 14 to 17 are collectively mounted. The elevating mechanism 20 has a telescopic arm 21 composed of a first arm 21a and a second arm 21b which are vertically extendable and extendable in the shape of the letter "<".
Rotation drive mechanisms 22A, 22B, and 22C are provided at the base, tip, and intermediate portions of the extendable arm 21, respectively. can be made Further, by changing the angle of the measurement head 18, the vertical orientation of the video camera 14 and the thermo camera 15 can be changed.

The rotation drive mechanism 22A is composed of, for example, a gear 23A integrated with the rotation shaft of each joint of the telescopic arm 21, a pinion 24A screwed to the gear, and a motor 25A for rotating the pinion. In addition, a torsion spring that supports the weight of each part and exerts a restoring force in the direction of rotation is inserted in the rotating shaft of each joint, and is configured to reduce the load on the motor when the measuring head 18 is raised. It is The rotary drive mechanisms 22B and 22C also have the same configuration. An air cylinder or the like may be used instead of the motor as a drive source for expanding and contracting the arm.

Furthermore, although not shown in FIGS. 1 to 3, the autonomous mobile patrol robot 10 has wireless communication means for wirelessly transmitting the measurement data acquired by the various measurement means 14 to 17 to the monitoring terminal. A line tracing function for processing image information captured by a video camera, recognizing a guidance line drawn in advance on the road surface of the monitoring area and running along the line, and a range sensor 13A. , 13B, and a self-driving function of recognizing its own position based on the three-dimensional space information acquired by 13B and traveling while avoiding obstacles.

Self-location recognition based on three-dimensional space information is performed by storing image data of the facility to be monitored in advance, or three-dimensional point cloud data representing the structure of the facility to be monitored, in a storage device. The three-dimensional space information obtained by the area sensors 13A and 13B is compared with the image data or the three-dimensional point cloud data read out from the storage device.
Recognizing the self-position based on such three-dimensional space information requires an enormous amount of time and cost to perform highly accurate position recognition. Therefore, the autonomous traveling patrol robot 10 of the present embodiment runs in the line trace mode when traveling in a narrow space, and runs in the autonomous driving mode by three-dimensional space recognition by laser scanning in other spaces. , an in-vehicle control device is configured.

FIG. 4 shows a system configuration example of devices mounted on the autonomous traveling patrol robot 10 of the present embodiment. In FIG. 4, a video camera 19 for detecting lines on the road surface is provided separately from the video camera 14 for acquiring images of the equipment.
As shown in FIG. 4, the in-vehicle system 30 of the autonomous traveling patrol robot 10 of the present embodiment includes detection signals from the range sensors 13A and 13B, video camera 14, thermo camera 15, sound collecting microphone 16, odor A control device 31 to which measurement data obtained by the sensor 17 and image data from the video camera 19 for line detection are inputted, a storage device 32 for storing the measurement data obtained by the measurement means (14 to 17) and control programs, A wireless communication module 33 is provided for transmission to the monitoring terminal.

Further, the in-vehicle system 30 includes traveling motors 34A and 34B that rotate the traveling wheels 12A and 12B, arm extension and retraction motors 25A to 25C that extend and retract the arms of the lifting mechanism 20, and the measuring means (14 to 17). , drive sources (motors 25A to 25C, 34A, 34B), control device 31, storage device 32, etc., and a battery (secondary battery) 36 that supplies power, and a non A power receiving unit 37 for receiving power supply from the contact charging device 52 is provided. The power receiving unit 37 charges the battery 36 based on the power received from the power receiver (power receiving coil) 38 provided at the bottom of the main body 11 to receive power by an electromagnetic induction method or the like. It is composed of a charger 39 and the like.

The control device 31 reads the measurement data acquired by the measurement means (14 to 17), stores it in the storage device 32, and processes the measurement data stored in the storage device 32. is read out and transmitted by the wireless communication module 33, a travel control unit 43 for controlling the travel motors 34A and 34B, and an elevation control unit 44 for controlling the arm extension/retraction motors 25A to 25C. These functions are realized by cooperation with a microprocessor (CPU) and a control program read out from the storage device 32 and stored in the RAM.

The autonomous patrol robot 10 is usually on standby in a garage called a charging base, and when a predetermined set time comes, it leaves the charging base and moves along a predetermined running route within the facility to be monitored. , when it moves in front of a predetermined facility set in advance, it measures the state of the equipment to be monitored by one of the measuring means (14 to 17). Then, the acquired data is temporarily stored in the storage device 32 . In addition to periodic remote monitoring, in order to enable on-site confirmation in the event of trouble, the status measurement data of the monitoring target equipment by the measuring means (14 to 17) is transmitted by the wireless communication module 33 even during measurement. It is transmitted wirelessly and can be monitored in real time.

FIG. 5 shows the state in the vicinity of the charging base 50. As shown in FIG. As shown in FIG. 5, the charging base 50 is provided with a garage 51 with a roof, and a non-contact charging device 52 is installed approximately in the center of the floor of the garage. In addition, in this embodiment, assuming that the vicinity of the charging base 50 is a narrow space, the guide line 53 is drawn by white paint or the like along the travel route from the doorway of the garage 51 to the position where the space becomes wide. depicted above. Although not shown in FIG. 5, if there is a narrow space on the travel route other than the entrance/exit of the garage, a guide line is also drawn on the road surface at that location. In addition, the code|symbol S is a slope provided in the doorway of a garage.

In addition, a marker 54 is provided on the road surface on the far end side (lower end in FIG. 5) of the guide line 53, and the autonomous mobile patrol robot 10 first starts traveling in the line trace mode when leaving the garage. When it moves along the guide line 53 and reads the marker 54 from the image of the video camera 19, it switches to the self-driving mode, grasps its own position based on the detection signals from the range sensors 13A and 13B, and uses the map information. It is programmed to run according to a predetermined route R. Markers (54) for instructing mode switching are also provided at the front and rear ends of the guidance line provided in a narrow space other than the entrance/exit of the garage. Furthermore, if the guide line is long, a marker indicating the position is also provided in the middle of the line.

The marker 54 is a two-dimensional code such as a QR code (registered trademark) drawn on the road surface in this embodiment, but may be a magnetic marker or the like. Although not shown, the route R is set so as to go around the facility and return to the place where the marker 54 in FIG. 5 is installed.
Further, the main points of the route R are provided with markers (QR codes) serving as route indicators. This marker contains information on the installation position of the marker on the map of the facility. On the other hand, when the autonomous traveling patrol robot 10 loses the line while traveling in the line trace mode, it is programmed to switch to the autonomous operation mode and move to a marker closest to its own position immediately before the line is lost. When the index marker is detected in the driving mode, the self-position can be corrected and the vehicle can accurately travel along the predetermined route.

Furthermore, if the driver fails to recognize the line on the road surface while driving in line trace mode and loses the line, instead of stopping immediately, the vehicle moves for a predetermined time or a predetermined distance, during which the line can be recognized again. If the line is not recognized, the line trace mode is continued, and if the line is not recognized, the vehicle is configured to shift to the autonomous driving mode in which the vehicle travels without depending on the line. Further, the mode is also changed when a marker provided in advance on the travel route instructs to switch the mode.

Next, the image processing procedure of the line detection video camera 19 by the controller 31 of the autonomous traveling patrol robot 10 will be described with reference to FIGS. 6 to 9. FIG.
The control device 31 first performs clipping processing and normalization processing on the image of the video camera 19 for line detection. The reason why such processing is performed is that, in order to accurately detect the line on the road surface, it is desirable to install the camera so that it faces straight down from the vehicle body. Because it was difficult to install such a camera due to the structure of the vehicle body, the camera was installed on the front (rear) surface of the vehicle body so as to face obliquely downward.

Specifically, the control device 31 adjusts the width of the upper side from the lower ⅓ area (for example, about 42% of the height) of the image P0 of the line detection video camera 19 shown in FIG. 6A to the lower side. For example, a trapezoidal area A1 of 4/5 (approximately 80%) of the width of . conduct. This makes it possible to obtain an image in which the vicinity of the vehicle body is viewed from directly above, and facilitates accurate line detection. At this point, the image is a color image.

Next, the control device 31 converts the image converted as described above into a grayscale image (black-and-white grayscale image), and then performs Gaussian filter processing to remove noise. After that, a process of binarizing using a predetermined threshold value and counting the number of bright spots is executed so that the lines in the image become white bright spots. At this time, depending on the threshold used, the results are greatly different as shown in FIG. In FIG. 7, (A) is the original image when thresholds are set so that white is 60% and black is 40%, and (B) is the original image when white is 25% and black is 75%. (C) is the case where the thresholds are set so that white is 14% and black is 86%.

Therefore, in this embodiment, the binarization threshold value was changed. By changing the threshold value in this way, it is possible to obtain an accurate number of bright spots, that is, to detect a line, even when the photographing conditions of the camera are changed. It should be noted that the numerical values of the lower limit value and the upper limit value are examples, and the threshold value will be changed if the conditions change.

After the above processing, the control device 31 executes line determination processing. FIG. 9 shows the procedure of this line determination process.
In the line determination process, first, an image is obtained from the video camera 19, and the image is equally divided (for example, divided into 20) into elongated areas in the horizontal direction (X direction) as shown in FIG. 8A (step S1). ). Next, for the lower quarter of the image, the number of bright spots is counted to create a histogram as shown in FIG. P5) is picked up (extracted) (step S3).

Next, the control device 31 examines the continuity in the Y direction for the picked up candidates (P1 to P5) and searches for a continuous pattern from bottom to top (step S4). Specifically, the number of bright spots in the range of ±30 pixels with the center in the X direction of each candidate (P1 to P5) as the midpoint is obtained from the bottom. The process of determining the score is repeated to find consecutive candidates of a specified number or more in the Y direction. In this process, the candidates P2, P4, and P5 are eliminated from the image of FIG. 8(B).

Subsequently, the control device 31 determines whether or not a plurality of candidates remain (step S5), and if there are a plurality of candidates remaining and the line has already been recognized and the vehicle is running, the line closest to the previously recognized line is selected. , and when a line is to be detected first, a line near the center of the image is selected as the detection line (step S6). In the image of FIG. 8B, candidate P3 is selected as the detection line. If only one candidate remains in step S5, step S6 is skipped.

Next, the control device 31 calculates a quadratic approximation over the entire Y direction for the remaining candidates or the selected candidates (step S7). Subsequently, the intersection point C between the curve of the quadratic approximation formula obtained in step S7 and the horizontally divided area at the position of 7/12 from the bottom of the image is obtained, and set as a turning point, and from the quadratic approximation formula The steering direction (left or right) is determined (step S8).
Also, an index, which is a reliability value, is set, and the reliability value counter is incremented or decremented according to the processing results of steps S4 and S5 (step S9). Specifically, when multiple lines are detected or the continuity in the Y direction is interrupted, the reliability value is decreased by one, and when the continuity continues without interruption, the reliability value is increased by one. I do.

Next, it is determined whether or not the reliability value has become equal to or less than a predetermined value (step S10). If the reliability value is not equal to or less than the predetermined value (step S10: No), the process proceeds to step S11, and if it is determined that the vehicle body has moved and the next image has been acquired (Yes), the process returns to step S1 and the above image is obtained. Execute the process repeatedly. Further, when the vehicle body reaches the point set in step S8, the steering is turned by a predetermined amount in the direction determined in step S8.
On the other hand, if it is determined in step S10 that the reliability value has become equal to or less than the predetermined value (Yes), it is determined that the line has been lost. The previous operation is maintained (step S12), and if the line is not detected during that time, the mode is switched from line tracing to autonomous operation (step S13).

Note that when the mode is switched to autonomous operation in step S13, the control device 31 stores the position information of the marker closest to the current self-position among the markers provided on the line in advance on the map stored in the storage device. It acquires from the information and performs control to run toward the position.
By executing the processing as described above, even if it becomes difficult to determine the line in the image of the line on the road surface due to the reflection of sunlight or the shadow of facilities, etc., the autonomous traveling patrol robot 10 can move. can do. Therefore, by moving, the possibility of detecting the line again increases, and even if the line cannot be detected, it switches to the autonomous driving mode and continues running to avoid falling into a complete stop state. can be done.

Although the embodiments of the present invention have been described above, the present invention is not limited to the above embodiments, and various modifications and changes are possible. For example, in the above embodiment, it was explained that the autonomous traveling patrol robot grasps its own position based on the number of rotations (rotation angle) of the left and right traveling wheels. The self-location may be grasped based on GPS information.
Further, in the above embodiment, the case where the present invention is applied to an autonomously traveling patrol robot that inspects equipment at a substation has been described, but the present invention can also be used for inspecting equipment at facilities other than substations. can do.

10 Autonomous Traveling Patrol Robot 11 Main Body 12A, 12B Traveling Wheels 13A, 13B Range Sensor (Distance Sensor)
14 video camera for measurement (measurement means)
15 Thermo camera (measurement means)
16 Sound collection microphone (measurement means)
17 Odor sensor (measuring means)
18 measurement head 19 video camera for line detection 20 elevating mechanism 21 telescopic arm 22A, 22B, 22C rotary drive mechanism 23A, 23B, 23C gears 24A, 24B, 24C pinion 25A, 25B, 25C motor 30 in-vehicle system 31 control device 32 storage device 33 wireless communication module 34A, 34B traveling motor (driving means)
35 Arm extension/retraction motor 36 Battery 37 Power receiving unit 38 Power receiver 41 Acquisition information processing unit 42 Data transmission unit 43 Driving control unit 44 Lifting control unit 50 Charging base 51 Garage 52 Non-contact charging device 53 Induction line 54 Marker

Claims (5)

A driving means consisting of driving wheels and driving means for rotating the driving wheels, a video camera capable of photographing the road surface, a scanning distance sensor, a measuring means for acquiring the state of equipment, and the video camera. and a control device capable of controlling the driving means based on the captured image or the signal from the distance sensor, and patrols the facility to be monitored along a predetermined route to monitor the condition of the facility. An autonomously traveling patrol robot,
The control device is
a first traveling mode in which the distance and direction of an object are detected by sensing by the distance sensor and the vehicle travels by controlling the driving means so as to move along the route;
a second running mode in which a direction is specified by detecting a strip provided on the road surface of the route, and the driving means is controlled so as to move along the strip,
While either the first running mode or the second running mode is set as the basic running mode, the operation of automatically switching to the second running mode or the first running mode when a predetermined condition is satisfied is performed according to the situation of the running place. An autonomously traveling patrol robot characterized by repeating The control device, in the second travel mode,
An image of a predetermined region is cut out from the image captured by the video camera, normalization processing is performed on the cutout image, and the normalized image is binarized using a predetermined threshold value, From the binarized image data, a pattern having a width of a predetermined width or more and the longest continuous pattern in the running direction is extracted, and the extracted pattern is regarded as the belt-like body and is moved along the belt-like body. controlling the driving means;
2. The autonomous mobile patrol robot according to claim 1, wherein in said binarization process, a binarization threshold value is set so that the number of bright spots falls within a predetermined range. The control device, in the second travel mode,
When the pattern cannot be extracted, the immediately preceding travel control is maintained until the vehicle travels a predetermined time or a predetermined distance, and when the pattern cannot be extracted during that time, the vehicle is switched to the first travel mode. 3. The autonomous traveling patrol robot according to claim 2, wherein said driving means is controlled so as to move towards a target set in advance on said route. The belt-like body is provided at a narrow part of the route or at a part where it is necessary to travel a preset traveling route with high accuracy. 4. The autonomous mobile patrol robot according to claim 1, further comprising a marker for instructing switching. The facility to be monitored is a substation, and the measurement means includes at least a video camera for measurement, a thermo camera, a sound collecting microphone, and an odor sensor,
4. The control device operates any one of the measuring means at a preset position on the route to detect the state of the facility, and stores the detection result in a storage device. Autonomous traveling patrol robot according to any one of the above.

Images