JPH07281753A - Moving robot - Google Patents

Abstract

PURPOSE:To provide the moving robot improved in the exactness of self-position identifying processing and further able to be processed at high speed even in a complicated environment by providing the self-position identifying processing unnecessitated with processing for complicated teaching or a plant by adopting an external sensor instead of visual information without exerting an adverse effect upon the plant. CONSTITUTION:Concerning the moving robot having a mobile truck, mounted with a check sensor, equipped with a moving mechanism for moving the robot back and forth and turning it while moving inside the building of the plant to inspect and monitor the status of the plant equipment, moving amount detecting means 14 and 15 for detecting the amounts of horizontal, vertical and turning moving of the moving robot 1 and a self-position identifying means 23 for identifying a present position based on the moving amount data obtd. from these moving amount detecting means 14 and 15 in correspondence with CAD data showing the arrangement of instruments inside the building of the plant.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mobile robot for autonomous control.

[0002]

2. Description of the Related Art Generally, in a nuclear power plant or other various plants in a bad environment, when an abnormality such as heating or leakage occurs in equipment or piping installed in the plant, a worker can detect the abnormality early. Is moving and inspecting. In addition, in order to monitor places that are difficult to understand by fixed inspection sensors and to perform detailed inspections as needed, workers move around to perform inspection inspections. This inspection inspection work has problems such as damage to the human body and cost, and more recently, the problem of securing workers, and there is a strong demand for a mobile inspection device that replaces workers.

In particular, for a robot for patrol inspection, a detailed inspection work is carried out when a phenomenon considered to be abnormal occurs, and a detailed inspection work is carried out on parts and locations of related systems for investigating the cause of the abnormality. It is expected that something can be done easily. Therefore, there is a strong demand for a mobile robot that can move to an arbitrary place other than a fixed movement route. Further, it is desired that the operation can be performed by one person and is easy.

For this reason, various mobile robots in which inspection sensors such as a television camera and a microphone are attached to a platform are currently being developed in various fields in order to meet these demands. So far, the operator directly operates the drive mechanism by remote control, or decide in advance the course to which the robot will move, attach a guide wire to that course, or equip it with a mark. A teaching playback system that teaches and moves has been developed. However, the control method makes remote control difficult and imposes a heavy burden on the operator. Further, the teaching playback method is complicated and difficult to teach, and cannot be moved to an arbitrary place. These mobile robots do not satisfy the required functions of the patrol inspection robot. Therefore, recently, as a mobile robot that responds to demands, development of a wireless mobile robot that autonomously controls with the aim of expanding a moving range and improving operability has been advanced.

[0005]

In the above-mentioned conventional mobile robot, it is difficult to satisfy the function under the operating conditions such as a very complicated and wide area of a nuclear power plant or other various plants under bad environment. It was In order to autonomously control the movement of a mobile robot and reliably move it to the target position, it is necessary to have a function to grasp the moving environment and constantly find the position and orientation of the robot itself (hereinafter referred to as self-position identification processing). Become. Moreover, in order to move the mobile robot smoothly, it is necessary to perform self-position identification processing at high speed. In the case of human beings, self-position identification processing is easily performed in a short time from an image captured by the eyes, that is, passive visual information. On the other hand, when performing a self-location identification process on the robot using passive visual information, it is necessary to attach a TV camera equivalent to human eyes to perform image recognition, but the visual recognition function is very inferior to humans. . for that reason,
It is difficult to perform accurate self-localization in a short time. The current recognition technology is capable of accurately recognizing a simple object in a simple background in a short time to some extent, but the recognition rate becomes extremely low in a complicated environment, and the recognition rate is as long as 10 seconds or more. It takes time and is far from a practical level. It will take quite a while before recognition technology in a complex environment reaches a practical level. For this reason, at present, it is not feasible for a robot that moves in a complicated environment to perform a self-position identification process using passive visual information.

The present invention has been made in consideration of the above conventional circumstances, and an object thereof is to adopt an external sensor instead of visual information without adversely affecting the plant, and to perform complicated teaching and processing for the plant. The object of the present invention is to provide a mobile robot capable of constructing unnecessary self-position identification processing of (1), improving the accuracy of self-position identification processing in a complex environment, and capable of high-speed processing.

[0007]

In order to achieve the above object, the mobile robot of the present invention is, in the invention described in claim 1, a mobile trolley equipped with a moving mechanism for moving in a plant building to perform forward and backward and turning. In a mobile robot that is equipped with an inspection sensor to inspect and monitor the state of plant equipment, a movement amount detection means for detecting the movement amount of the mobile robot in horizontal, vertical, and turning, and the movement amount obtained from this movement amount detection means. Provided is a mobile robot having self-position identifying means for identifying a current position by associating data with CAD data indicating a device layout in a plant building.

According to the second aspect of the present invention, an inspection sensor is mounted on a movable carriage having a moving mechanism for moving in the plant building to perform forward and backward movements and turns. In the robot, movement amount detection means for detecting the movement amount of the mobile robot in horizontal, vertical, and turning directions, and distance measurement means for measuring the distance to the surrounding structure,
Correlation calculation means for performing a correlation calculation between the distance measurement data obtained from the distance measurement means and the movement amount data obtained from the movement amount detection means, and a self position for identifying the current position of the mobile robot using the correlation calculation result. A mobile robot having an identification means.

Further, in the invention according to the third aspect, the distance measuring means has a scanning means for collecting the distance measurement data a plurality of times at every predetermined moving time, and the plurality of storage means are switched at every scanning. 3. The mobile robot according to claim 2, further comprising: averaging processing means for averaging the distance measurement data stored in these storage means to an arbitrary number of accumulations to measure the distance.

[0010]

According to the first aspect of the invention, there is provided a mobile robot having the above structure.
In the invention described above, the moving amount detecting means detects the moving amount of horizontal, vertical and turning, and identifies the self-position of the mobile robot in correspondence with the plant CAD data used for displaying the equipment arrangement of the plant. In the invention according to claim 2,
The movement amount detecting means detects horizontal, vertical and turning movement amounts, and the distance measuring means measures the distance to the surrounding structure to obtain distance measurement data. On the other hand, the moving amount detecting means detects the moving amount of the mobile robot. Correlation calculation is performed on the movement amount and the distance measurement data by the correlation calculation means to identify the position of the mobile robot.

Further, in the invention of claim 3, the distance measuring data is sampled a plurality of times at each moving time by the scanning means, stored in the memory means, and averaged by the averaging processing means at any cumulative number of times. Convert to obtain distance measurement data. On the other hand, the moving amount detecting means detects the moving amount of the mobile robot. Correlation calculation is performed on the movement amount and the distance measurement data by the correlation calculation means to identify the position of the mobile robot.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of a mobile robot according to the present invention will be described below with reference to FIGS. FIG. 1 shows an outline view of a mobile robot in a first embodiment of the mobile robot according to the present invention. The mobile robot 1 is mounted on a mobile carriage 7 driven by wheels 50 having a rotation angle sensor 15 for detecting a movement amount. , A camera platform 8 provided with an inspection sensor 9, and an inspection sensor signal processing device 21. Further, the wireless transmission device 10, the communication control device 20, the scan type laser rangefinder 1 which is an external sensor for the robot in place of the above visual information.
2, ultrasonic sensors 13a and 13b, a gyro 14, and a robot controller 19 are mounted. The mobile robot 1 moves to an inspection object installed in the plant based on a command from the operation panel 3 in the operation room 2 and performs inspection using the inspection sensor 9.

The platform 8 used in this embodiment has a scan type laser range finder 12 mounted on the lower part thereof, and performs a turning drive and an elevation drive based on a posture control command from a robot controller 19. The wireless transmission device 10 is the communication control device 20.
In response to the request and control from the control station, the control signals and status signals necessary for controlling the ground station and the robot and the inspection sensor 9 are wirelessly transmitted. The gyro 14 detects the turning angle and the tilt angle of the robot.

The scan type laser range finder 12 obtains the distance to the object to be measured by measuring the time from the emission of the laser beam to the reflection from the object to be measured. Scanning is 270 degrees horizontally and 1 per second
It is performed about 0 times.

The ultrasonic sensors 13a and 13b are respectively provided with a transducer 13b for transmitting ultrasonic waves and a receiver 13a for receiving ultrasonic waves reflected from the object to be measured. The distance to the object to be measured is obtained by measuring.

FIG. 2A is a schematic diagram of the inside of the operation room 2 of the mobile robot 1, and FIG. 2B is a schematic diagram of the mobile robot 1. In FIG. 2A, the operation room 2 is provided with a diagnosis device 6, a robot integrated control device 5, and a communication control device 4 in addition to the operation panel 3.

A wireless transmission device 11 with a tracking device for wirelessly transmitting or receiving a communication control signal from the communication control device 4 to the mobile robot 1 is provided. The operation panel 3 is composed of an operating device 17 for remotely operating the moving carriage 7 and the inspection sensor 9, a status display monitor 16 for displaying the inspection result and the status of the mobile robot 1, and a man-machine interface controller 18 for controlling these. Has been done.

The operator inputs a movement request regarding the target position and direction using the operation device 17. The robot integrated control device 5 has a function of receiving a movement request from an operator and generating a route from a current position to a destination using plant CAD (computer-aided design) data,
Further, it manages communication between the operation panel 3 and the mobile robot 1.
This plant CAD data is for designing the layout of equipment, and the content of the data is such that the layout of equipment such as pumps, valves, heat exchangers, and the coordinates indicating the installation position of piping are Is given three-dimensionally in the entire volume occupied by.

The communication control device 4 is a state signal from the traveling controller 24 of the mobile robot 1, the platform controller 25, the inspection sensor signal processing device 21, and the inspection sensor 9.
Input the sensor signal from the robot integrated control device 5,
It has a function of transmitting to the diagnostic device 6. On the contrary, a function for inputting request signals from the robot integrated control device 5 and the diagnostic device 6 and transmitting them to the traveling controller 24, the platform controller 25, and the inspection sensor signal processing device 21 of the mobile robot 1 and a wireless device with a tracking device. It also has a function of controlling the transmission device 11.

The diagnostic device 6 inputs the sensor signal detected at the inspection point, adds it to the built-in database, compares it with the sensor signal stored in advance at the normal time, and detects the temporal fluctuation state. If it is larger than the reference value, it is judged that an abnormality may have occurred, and a detailed inspection method (new inspection location and inspection means) is planned, and the robot is An inspection request is output to the integrated control device 5. Further, when the detailed inspection result is obtained, the abnormality is judged and the abnormal place and the cause of the abnormality are estimated. These processing results are transmitted to the operator via the sequential status display monitor 16.

The wireless transmission device 11 with a tracking device is provided with a tracking device for transmitting a wireless signal having directivity for accurately transmitting a sensor signal such as a video signal of a television camera which has a large capacity and requires high speed. Wireless transmission device. In addition, the wireless transmission device 1 provided in the mobile robot 1
0 has a large size when equipped with a tracking device, and is therefore composed of a plurality of transmitters and receivers.

In FIG. 2B, the mobile robot 1 is
The inspection sensor 9 includes a television camera 27, a microphone 28, a laser vibrometer 29, an infrared camera 30, and an odor sensor 31. As described above, these inspection sensors 9 can be oriented in any inspection direction by remotely operating the platform 8 using the operation device 17.

Control signals for the operating room 2 and the mobile robot 1,
In order to make the signal transmission of the inspection sensor 9 wireless,
Since the mobile robot 1 does not have cables and therefore does not have a power cable, the mobile robot 1 is equipped with a battery 26.

The robot controller 19 is shown in FIG.
As shown in FIG. 3, the robot sensor signal processing circuit 22, the position identification controller 23, the traveling controller 24, and the platform controller 25 are included.

The robot sensor signal processing circuit 22 inputs the detection signals from the scanning laser range finder 12, the ultrasonic sensor 13, the gyro 14, and the rotation angle sensor 15 at the same time, and converts them into digital signals (not shown). No.), a buffer memory (not shown) for storing input data, and an output circuit (not shown) for outputting based on a command from the position identification controller 23.

The structure of the position identification controller 23 is shown in FIG. The data of the gyro 14 and the rotation angle sensor 15 for detecting the amount of movement are the sensor signal processing circuit 2 for robot.
2 is input to the rough position detection circuit 34 for detecting the rough position of the robot. The approximate position obtained by this circuit is input to the sensor simulation circuit 37,
Sensor simulation is performed using the plant CAD data stored in the sensor simulation circuit 37, and the result is stored in the image memory 36a as position data on the plant CAD data.

Here, the sensor simulation is a change of the gyro 14 and the rotation angle sensor 15 based on the movement of the mobile robot 1, using the plant CAD data composed of the position data of the inside of the building where the mobile robot 1 inspects and travels. It is the operation of tracking the quantity and obtaining the current position on the plant CAD data.

On the other hand, the distance data measured by the scanning laser range finder 12 and the ultrasonic sensors 13a and 13b indicate the actually measured distances. This distance data is coordinate-converted from the sensor coordinate system to the robot coordinate system using the coordinate conversion circuit 35. Here, the sensor coordinate system means the scanning laser range finder 12 or the ultrasonic sensor 13a,
13b means a relative polar coordinate system centered on 13b, and the robot coordinate system means a relative orthogonal coordinate system centered on the mobile robot 1. The data converted into the robot coordinate system is stored in the image memories 36b and 36c.

Further, the distribution of the sensor simulation result data and the distribution of the actually measured sensor data are associated with each other using the pixel correlation circuit 38 to obtain the difference. In addition, the position identification circuit 39 calculates the difference using the above-mentioned rough position detection circuit 34.
The self-position of the robot is determined by adding it to the approximate position obtained in.

Further, in the sensor simulation circuit 37, the plant CAD data is provided with the reflection coefficient with respect to the distance of the scanning laser range finder 12 and the ultrasonic sensor 13, the relatable angle data and the detectable level as attribute data. When the distance is actually measured using the scanning laser range finder 12 and the ultrasonic sensors 13a and 13b at the position of the robot obtained by the rough position detection circuit 34, these attribute data have an influence on the measured distance. Is shown.

For example, the reflection coefficient means the ratio of the amount of laser light reflected by projecting laser light onto an orthogonal object. In addition, the reflectable angle means that when the projection angle of the object and the laser beam becomes sharp, the amount of the reflected laser beam decreases, and when the angle exceeds a certain angle, the reflection is completely lost. Say. What is the detectable level?
Processing is performed when the amount of reflection exceeds a certain threshold level (threshold level), and this threshold level is called a detectable level.

These attribute data change depending on the structure, material, etc. of the object to be measured in the plant building, and are therefore stored corresponding to each position on the plant CAD data.

In associating the sensor simulation result in the pixel correlation circuit 38 with the distribution of the sensor data, the sensor data from the image memories 36b and 36c is first judged by this attribute data, and then the difference is obtained.

On the other hand, in order to use it in a place where accurate plant CAD data is not prepared, the shape data and the attribute data (reflection coefficient and reflection) are used by using the scan type laser range finder 12 and the ultrasonic sensor 13 in a previously moved place. Plant CAD data having a possible angle may be created.

Returning to FIG. 2, the traveling controller 24 is
From the trajectory data up to the target position and the self-position identification result, an arithmetic circuit (not shown) for calculating the traveling mode, the steering angle with respect to the wheels 50 of the mobile robot 1, and the traveling speed, and the calculated result to the traveling driver 33. Output and a control state and self-position information to the communication control device 20 (not shown). Here, the traveling modes include straight / backward movement of the mobile robot 1, turning on the spot, and parallel movement.

The motor 32 is a power supply source for the turning drive of the wheels 50 and the turning drive of the steering shaft. Also,
The driving driver 33 drives and controls the motor 32, and inputs speed command data, turning angle command data, start / stop command data transmitted from the driving controller 24, and specifies the rotation speed and the turning angle of the motor 32. The drive current is supplied to the motor 32 so as to be as described above.

The platform controller 25 uses an arithmetic circuit (not shown) for calculating the orientation of the inspection sensor toward the inspection object from the self-position identification result of the mobile robot and the plant CAD data. Alternatively, it comprises an output circuit (not shown) that outputs a drive command to a camera platform driver provided on the camera platform 8 based on a command from the operation device 17 of the operation panel 3.

Next, the operation of this embodiment will be described.
The mobile robot 1 moves to the inspection location according to the procedure shown in FIG. 4 and performs a predetermined inspection. The procedure of this inspection will be described below.

Move the mobile robot 1 to the inspection location,
In order to perform a predetermined inspection, first, it is necessary to input a target position and a work request to the mobile robot 1, that is, to determine an inspection route (S1). This processing is input by the operator using the operation device 17 of the operation panel 3. The input target position and work request are input to the robot integrated control device 5 via the man-machine interface control device 18. The robot integrated control device 5 obtains the route to the target position and the posture of the platform 8 at the time of inspection using the input target position and the work CAD data for plant equipment arrangement from the work request data. Further, in this processing, the operator may directly instruct the route to the target position and the posture of the platform 8 at the time of inspection from the operation device 17.

When the route to the target position and the posture of the platform 8 at the time of inspection are obtained, the route generation result to the target position is obtained by the communication control device 4, the wireless transmission device 11 with the tracking device,
It is transmitted to the traveling controller 24 on the mobile robot 1 via the wireless transmission device 10 and the communication control device 20. When the traveling controller 24 receives the route generation result to the target position, the traveling controller 24 drives the mobile robot 1 to perform traveling control (moving speed control and steering angle control) along the route to move to the target position. I do.

This processing is performed by the traveling controller 24 mounted on the mobile robot 1, the traveling driver 33, the rotation angle sensor 15, the motor 32, and the scanning laser range finder 12.
To the target position using A (S2 to S5) shown in FIG.
This processing routine is repeated.

Each of the processing routines A will be described below. The self-position identification process (S2) is performed using the position identification controller 23 as described above.

The travel controller 24 calculates the deviation from the target trajectory (S3). The travel controller 24 inputs the target trajectory data calculated by the robot integrated control device 5 at the ground station via the communication control device 4 and the wireless transmission device 11 with a tracking device, and at the same time, the position identification controller 2
The deviation is calculated from the current position and orientation data of the mobile robot 1 calculated in 3.

After the travel controller 24 calculates the deviation, it determines the travel mode, and executes the steering angle and travel speed calculation processing of each wheel 50 (S4). The current traveling mode is maintained unless a new command or a large deviation is generated from the robot integrated control device 5 in the ground station. When there is a new command from the robot integrated control device 5, the mode is switched to the traveling mode based on the command. In addition, when a large deviation occurs and the current traveling mode cannot cope with the deviation or the deviation increases, for example, the mobile robot 1 is once stopped and then the turning or parallel movement is performed. It is also possible to return to the target track, switch to the original traveling mode again, and resume traveling.

When the traveling mode is determined, the steering angle and traveling speed of each wheel 50 are determined. When there is no deviation, a route is determined from the deviation amount and speed at which point the target trajectory is matched, the steering angle and running speed of each wheel 50 up to the matched point are calculated, and the data is sent to the running driver 33. Output.

Steering drive control, speed control processing (S
5) is performed by the traveling driver 33 provided on each wheel 50. The traveling driver 33 inputs the steering angle and traveling speed data of each wheel 50 sent from the traveling controller 24, and the motor 32 of each wheel 50 is input.
The servo is controlled so that becomes the data.

By repeating the above processing routine A, the mobile robot 1 is moved along the target trajectory to reach the target position. The state of this processing is always converted into transmission data by the traveling controller 24, and the communication control device 20,
It is transmitted to the robot integrated control device 5 via the wireless transmission device 10 and the communication control device 4. Robot integrated control device 5
Receives the processing state data of the mobile robot 1, records the control state of the mobile robot 1, converts it into display data that is easy for the operator to display on the state display monitor 16 via the man-machine interface control device 18. .

When the mobile robot 1 reaches the target position,
The robot integrated control device 5 calculates the attitude of the platform 8 (S
6). The camera platform controller 25 receives the calculation result and controls the camera platform 8 (S7).

When the robot integrated control device 5 receives the fact that the control of the camera platform 8 is completed from the state data, it performs inspection control. Inspection control is performed by the inspection sensor signal processing device 2
Through 1. The data detected by the inspection sensor 9 is transmitted to the operation room 2 side via the wireless transmission device 10 and input to the diagnostic device 6. The diagnostic device 6 diagnoses an abnormality of the device from the sensor data and informs the robot integrated control device 5 that there is no abnormality. When it is determined that there is a suspicion of an abnormality, a detailed inspection plan is generated, and the result is set to the robot integrated control device 5
To transmit. The robot integrated control device 5 creates a movement plan to the next inspection location based on the diagnosis processing result from the diagnosis device 6 (S8).

In order to enable the mobile robot 1 to move smoothly in a complicated environment, the self-position identification process (S
The processes from 2) to steering drive control and speed control (S5) must be able to be executed at high speed and reliably. The present embodiment can be executed at high speed and reliably without stopping during the self-position identification processing as in the method using a positive external sensor such as a conventional TV camera.

Next, the self-position identification processing method that is executed at high speed and reliably will be described in detail. In the present embodiment, as described above, the self-position is obtained by obtaining the deviation between the sensor simulation result (distance measurement data) obtained by tracking the variation of the gyro 14 and the rotation angle sensor 15 with the plant CAD data and the actually measured distance. Identify.

Under a complicated background, there are many places that cannot be detected by the laser rangefinder. For example, the surface to be measured is a black surface or the surface condition is close to a mirror surface, the scanning laser rangefinder 1
2. In some cases, the angles at which the ultrasonic sensors 13a and 13b and the surface to be measured intersect are nearly parallel. Therefore, in this embodiment, the scanning laser range finder and the ultrasonic sensor are used together.

Further, the scanning type laser range finder 12 having a wide scanning range is adopted. A plurality of ultrasonic sensors 13 are attached around the robot. 5A and 5B, the scanning laser range finder 12 and the ultrasonic sensor 13 are shown.
The arrangement example of a and 13b and the measurement range are shown.

The measuring range of the scanning laser range finder 12 is ± 135 °, 0.5 m to 5 as shown by B in the figure.
It is 0m. The wider the measurement range, the more distance data can be obtained, and the more correspondence can be made, the higher the reliability can be. At a certain height, the object does not change, such as the presence of a uniform wall, which makes it difficult to associate the objects. Therefore, it has a mechanism that allows arbitrary vertical movement and tilt movement.

On the other hand, the measurement range of the ultrasonic sensors 13a and 13b is ± 15 °, 0.1 m-, as shown by C in the figure.
It is 10 m. The larger the number of attachments, the more correspondence can be made, but there are restrictions such as mutual interference (ultrasonic waves are input from another sensor and erroneous detection), control complexity, and cost increase. Considering this, the layout is appropriate.

Depending on the arrangement, there may be a location where the distance cannot be measured even by the scanning laser rangefinder 12 and the ultrasonic sensors 13a and 13b which are external sensors. In this case, the processing procedure shown in FIG. 6 is performed. Accordingly, the approximate position obtained from the gyro 14 and the rotation angle sensor 15, which are inner sensors, is directly used as the position information.

As described above, the attribute data for the scanning laser range finder and the ultrasonic sensor are stored in the plant CAD data for displaying the equipment layout on the plant. Whether or not it can be actually measured by an actual sensor is determined by using this attribute data. In the determination based on this attribute data, it is possible to determine the beam width and measurement range of the scanning laser rangefinder 12 and extract the surface to be measured in advance, so that the number of data can be reduced and high-speed processing can be realized. .

When performing a sensor simulation during robot movement, the position of the robot on the plant CAD is determined by the data of the gyro 14 for detecting the orientation and elevation angle of the robot and the rotation angle sensor 15 for detecting the movement amount of the robot. The rough position calculated from the data is used.

The elevation angle detected by the gyro sensor is used for correction when the scanning is normally horizontal when the mobile robot is moving in a passage having steps.
In the correction method of the present embodiment, it is assumed that the object to be measured is standing vertically, and the measured distance data is cosine-calculated at the elevation angle and converted into a horizontal distance. Although it takes some time, there is also a method of performing a sensor simulation considering the elevation angle. The distance measurement data obtained by the simulation is converted into Cartesian coordinates with the center of the mobile robot 1 as the origin, and image data 36a
Record to.

The distance measurement data of the scan type laser range finder 12 and the ultrasonic sensor 13 obtained by actual measurement are coordinate conversion and the image memory 3 combined with the coordinate system of the sensor simulation.
It is stored in 6b and 36c.

The sensor simulation and the actually measured distribution data stored in the memory have the same scale (the size of the pattern is the same) and their positions are deviated from each other. The correspondence between the two is based on image memory image memories 36a, 36b,
It can be easily obtained by performing pixel correlation calculation between 36c. This processing is a correlation calculation by software without using special hardware, and can be executed in a few milliseconds.

According to the mobile robot of the present embodiment, a plurality of sensors measure a wide range and the movement amount and the distance measurement data are associated with each other, so that the self-position identification processing can be reliably performed even in a complicated background. . In the unlikely event that a scanning laser rangefinder 1
2. Even if the ultrasonic sensors 13a and 13b exist at locations that cannot be measured by the external sensors, the gyro 14, which is an internal sensor, and the rotation angle sensor 15 perform self-position identification, and the reliability of the mobile robot is improved. Is high.

Since the plant CAD data for displaying the equipment arrangement of the plant is used, the operator
The operation is very easy because the mobile robot automatically generates and moves a trajectory that can move to the target position without interfering with the structure only by the operation of plotting the target position on the map.

In this embodiment, the distance measurement data obtained by the distance measuring means such as the scanning laser range finder 12, the ultrasonic sensors 13a and 13b, and the gyro 1 are used.
4 and the movement amount of the mobile robot obtained by the rotation angle sensor 15 are calculated, the position may be identified only by the movement amount. In that case, in FIG. 3, the scanning laser range finder 12, the ultrasonic sensors 13a, 13b,
The configuration does not include the coordinate conversion circuit 35, the image memories 36b and 36c, and the pixel correlation circuit 38. Therefore, the image memory 36
The output from a is directly input to the position identification circuit 39.

Next, a second embodiment of the mobile robot according to the present invention will be described with reference to FIG. In the first embodiment, since the scanning laser range finder 12 is used, if the self-position identification is performed while moving, an error due to time lag occurs in the measured distance data. Further, the laser range finder has variations in measured data, and a variation error of about 20 to 100 mm occurs unless averaged. However, if averaged, this error can be suppressed to 10 mm or less. If these errors are large, it becomes difficult to match them, or a large identification error occurs, which hinders travel control.

Therefore, in this embodiment, the variation error is reduced by performing the averaging process shown in FIG. 7 during movement. As a means for averaging, for example, eight image memories 36 for storing coordinate-converted detection distance data are provided. When the detected distance data from the scanning laser range finder 12 is stored in the image memory 36, coordinate conversion including the movement amount of the mobile robot 1 is performed via the coordinate conversion circuit 35, and scanning is performed using the multiplexer 40. Switch every time,
It is stored in the image memories 36 of 1 to 8. Therefore, the image memory 36 always stores the detected distance data at the present time and the detected distance data for seven times before that. The averaging is performed by using the sum total averaging circuit 45, obtaining the sum of these eight detection distance data, and obtaining the peak of the distribution by spatial filter processing.

As a result, in one scanning,
The data of eight times can be averaged without time delay, and the average number of times is not limited to eight times. With a laser rangefinder having a high scanning speed, if the number of image memories is increased, the average number is increased in the same period, so that it is possible to further reduce the variation.

As described above, the position data of the mobile robot 1 obtained by the position identification controller 23 is taken into the traveling controller 24 shown in FIG. 2A as traveling control data.

The travel controller 24 calculates the travel mode, the drive angle of each steering wheel, and the speed from the path to the target position and the position data, and the travel driver 3
Control 3

In addition to the effects of the first embodiment, the mobile robot 1 of the present embodiment realizes smooth autonomous movement by performing the self-position identification processing described in the above embodiment during traveling. Although the image memory is used as the self-position identification controller in the above-described embodiment, the distance data obtained by coordinate conversion may be stored in the memory in a two-dimensional array.
Further, the correspondence can be made not only by the correlation calculation between pixels but also by correspondence of feature points. For example, it can be established by associating the length of continuous data from the laser rangefinder, the distance from the corner of the plant building, the angle value of the corner, and the curvature.

Further, in the above-mentioned embodiment, the robot controller section is mounted on the mobile robot. However, if the robot controller section is provided in the operation room and communication is performed via the communication control device and the wireless transmission device, It goes without saying that it has the functions of the embodiment.

[0072]

As described above, the mobile robot of the present invention is not affected by lighting or the like as in the system using a television camera, does not require complicated teaching, and can output plant CAD data. Since it is used, the operation becomes easy and the self-position identification processing can be performed at high speed.
Further, in the mobile robot having the distance measuring means and the correlation calculating means, the self position can be accurately identified even in a complicated environment and background in the plant. Furthermore, in the mobile robot having the distance measuring means including the scanning means, the storage means, and the averaging processing means, the self-position identification processing can be performed accurately even during movement.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing the configuration of a first embodiment of a mobile robot according to the present invention.

FIG. 2A is a schematic diagram of an operation room of a mobile robot according to the present invention, and FIG. 2B is a schematic diagram showing in detail the configuration of the first embodiment of the present invention.

FIG. 3 is a schematic diagram showing the configuration of a position identification controller of the mobile robot of the first embodiment.

FIG. 4 is a flowchart showing a procedure of inspection processing.

5A and 5B show arrangements of a scanning laser rangefinder and an ultrasonic sensor of a mobile robot in the first embodiment, FIG. 5A is a plan view and FIG. 5B is a side view.

FIG. 6 is a flowchart showing a processing procedure for adopting position information of a mobile robot in the first embodiment.

FIG. 7 is a schematic diagram showing a partial configuration of a position identification controller of a second embodiment of a mobile robot according to the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Mobile robot 7 ... Mobile carriage 8 ... Platform 9 ... Inspection sensor 12 ... Scan type laser rangefinder 13 ... Ultrasonic sensor 14 ... Gyro 15 ... Rotation angle sensor 23 ... Position identification controller 27 ... TV camera 28 ... Microphone 29 ... Laser vibrometer 30 ... Infrared camera 31 ... Odor sensor 36a, 36b, 36c ... Image memory 38 ... Pixel correlation circuit 39 ... Position identification circuit 45 ... Sum average processing circuit 50 ... Wheel

Claims (3)

[Claims] 1. A mobile robot for inspecting and monitoring the state of plant equipment by mounting an inspection sensor on a mobile carriage equipped with a moving mechanism for moving forward and backward and turning in a plant building. Self-position identification for identifying the current position by associating the movement amount detection means for detecting the movement amount of up and down and turning, and the movement amount data obtained from this movement amount detection means with CAD data indicating the equipment arrangement in the plant building. A mobile robot having means. 2. A mobile robot for inspecting and monitoring the state of plant equipment by mounting an inspection sensor on a movable carriage equipped with a moving mechanism for moving forward and backward and turning in a plant building. A movement amount detecting means for detecting a movement amount of vertical and turning, a distance measuring means for measuring a distance to a surrounding structure, distance measurement data obtained from the distance measuring means, and a movement amount obtained from the movement amount detecting means. A mobile robot comprising: a correlation calculation means for performing a correlation calculation of data; and a self-position identification means for identifying the current position of the mobile robot using the result of the correlation calculation. 3. The distance measuring means has a scanning means for collecting distance measurement data a plurality of times for each predetermined movement time, and switches the scanning data for each scanning at any time so as to be stored in a plurality of storage means. The mobile robot according to claim 2, further comprising: averaging processing means for averaging the distance measurement data stored in the means at an arbitrary cumulative number of times.