JP2009109219A - Obstacle position recognition method for mobile unit and mobile unit - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To recognize the position of an obstacle by scanning the measurement light in two dimensional directions different from the moving direction of a mobile unit, using a small-sized, low-cost one-dimensional laser scanner. <P>SOLUTION: Arms 71, 73 of a traveling wheel arm 7 connecting a main unit 3 and a traveling wheel 5 of a mobile robot 1 moving on a mobile surface A are bent and stretched appropriately and independently by the rotation of a rotary actuator 75 to oscillate the main unit 3 in a tilt direction, thereby scanning the measurement light, which is output from an external sensor 13 for recognizing obstacles on the mobile surface A and is scanned in the lateral direction of the mobile robot 1, in the vertical direction as well. From a deviation angle between the direction of a vertical coordinate axis in the local coordinate system of the coordinate value group of the mobile surface A which is calculated from reflected light from the mobile surface A received by the external sensor 13 by oscillating the main unit 3 downward in the tilt direction and a gravity direction calculated from the output of an acceleration sensor 153, calibration data for coordinate value conversion from the local coordinate system to the global coordinate system are obtained. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a moving body that moves on a moving surface, and in particular, in order to move on the moving surface while avoiding an obstacle on the moving surface, the obstacle is detected using measurement light scanned in a two-dimensional direction. The present invention relates to a technology for performing position recognition.

  Conventionally, a self-traveling robot (moving body) that performs monitoring work in an unattended monitoring area or the like is known. This type of robot, for example, performs autonomous running on the moving surface of the monitoring area in a pattern according to the purpose while comparing its position obtained by GPS (Global Positioning System) with map information. At that time, when an unexpected obstacle that does not exist in the map information exists on the moving surface, an external sensor is mounted on the robot so that the robot can run independently while avoiding interference with the obstacle.

  One of the popular external sensors is a two-dimensional scanning laser scanner. This two-dimensional scan type laser scanner scans measurement light (ranging light) by a laser in a two-dimensional direction (main scanning direction, sub-scanning direction), and irradiates the moving direction forward of the robot. When an obstacle is present, the reflected light from the obstacle is received, and the position and shape of the obstacle are recognized by a phase difference method, a time difference method, or the like.

  Therefore, in the two-dimensional scan type laser scanner, it is necessary to further scan the measurement light scanned in the main scanning direction by a polygon mirror or the like in the sub scanning direction by a galvano mirror or the like. Therefore, a two-dimensional scanning laser scanner needs to be provided with a mirror rotation drive source in each of the main and sub scanning directions, and the increase in the size and cost of the laser scanner is inevitable.

Therefore, when topographic aerial survey is performed by laser scanning from an aircraft in the sky, scanning at the measurement light output source is limited to one-dimensional scanning in a direction different from the traveling direction of the aircraft (for example, a direction orthogonal to the traveling direction). In regard to the traveling direction, it has been proposed to realize two-dimensional scanning of measurement light by substantially scanning laser light by moving the aircraft itself (for example, Patent Document 1).
JP-A-8-145668

  However, since the above-described conventional technology is such that the moving direction of the aircraft as the moving body is one scanning direction of the two-dimensional scanning of the measurement light, the external sensor in the self-running robot that is the subject of the present invention. In such a field where the traveling direction does not coincide with the scanning direction of the measurement light, it cannot be used.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to use a small and low-cost one-dimensional laser scanner as the moving body moves on the moving surface, and the moving direction of the moving body. An object of the present invention is to provide an obstacle position recognition method for a moving body and a moving body that can scan the measurement light in different two-dimensional directions to recognize the position of the obstacle.

  In order to achieve the above object, the obstacle position recognition method for a moving object according to the first aspect of the present invention is a method for recognizing the position of an obstacle on the moving surface from a moving object moving on the moving surface. The measurement light scanned in the first direction intersecting the moving direction of the moving body is output from the one-dimensional laser scanner on the moving body, and the posture of the moving body with respect to the moving surface, or the moving body By changing the attitude of the one-dimensional laser scanner with respect to the moving surface and changing at least the elevation angle or the depression angle of the one-dimensional laser scanner with respect to the moving surface, the first direction intersects with the moving direction of the moving body and is orthogonal to the first direction. The measurement light is scanned in the direction of 2, and a change in at least the elevation angle or depression angle of the one-dimensional laser scanner with respect to the moving surface is fixed relative to the one-dimensional laser scanner. By detecting the scanning angle of the measurement light in the second direction by detecting it with an angular velocity sensor on the moving body, the reflected light from the irradiation object of the measurement light is received by the one-dimensional laser scanner. The reflected light received by the one-dimensional laser scanner, the known scanning angle of the measurement light by the one-dimensional laser scanner in the first direction, and the measurement light in the second direction. Based on the detected scanning angle, a coordinate value group of the obstacle in a local coordinate system based on the position of the one-dimensional laser scanner is determined, and a coordinate value group of the calculated obstacle in the local coordinate system; and A coordinate value group in the global coordinate system of the obstacle is determined based on a known coordinate value in the global coordinate system of the one-dimensional laser scanner; Based on the coordinate value group in the global coordinate system of the obstacle indexing, characterized in that to perform the position recognition of the obstacle.

  In order to achieve the above object, the moving body of the present invention described in claim 4 avoids an obstacle on the moving surface and moves on the moving surface in a first direction and a first direction orthogonal to each other. A moving body for recognizing the position of the obstacle using measurement light scanned in the direction of 2, the main body and the main body crossing the first direction and the second direction on the moving surface A moving means for moving in a direction; a one-dimensional laser scanner provided in the main body, for outputting the measurement light to scan in the first direction, and for receiving reflected light from an irradiation target of the measurement light; Attitude changing means for changing the attitude of the main body relative to the moving surface or the attitude of the one-dimensional laser scanner relative to the main body to change at least the elevation angle or depression angle of the one-dimensional laser scanner relative to the moving surface; Based on a detection result of the angular velocity sensor, an angular velocity sensor that is provided in the main body with a fixed relative position with respect to the laser scanner and detects at least a change in elevation angle or depression angle with respect to the moving surface of the one-dimensional laser scanner. Scanning angle detecting means for detecting a scanning angle of the measuring light in the direction of 2; the reflected light received by the one-dimensional laser scanner; and the measuring light from the one-dimensional laser scanner in the first direction The obstacle in the local coordinate system based on the position of the one-dimensional laser scanner based on a known scanning angle and a scanning angle of the measurement light in the second direction detected by the scanning angle detection means Local coordinate system coordinate value indexing means for determining a coordinate value group of the object, and the local coordinate system coordinate value indexing means A global coordinate system coordinate value for determining a coordinate value group in the global coordinate system of the obstacle based on a coordinate value group in the local coordinate system of the harmful object and a known coordinate value in the global coordinate system of the one-dimensional laser scanner Indexing means, and recognizing the position of the obstacle based on a coordinate value group in the global coordinate system of the obstacle determined by the global coordinate system coordinate value acquisition means. .

  According to the obstacle position recognition method of the moving object of the present invention described in claim 1 and the moving object of the present invention described in claim 4, the posture of the moving object with respect to the moving surface, or one-dimensional with respect to the moving object. By changing the attitude of the laser scanner and changing at least the elevation angle or depression angle of the one-dimensional laser scanner with respect to the moving surface, the measurement light in the second direction intersecting the moving direction of the moving body and orthogonal to the first direction Since the scanning is realized, it is not necessary for the laser scanner itself to have a configuration for scanning the measurement light in the second direction.

  In addition, since the scanning angle of the measuring light in the second direction is detected by the output of the angular velocity sensor whose relative position with respect to the one-dimensional laser scanner is fixed, only the configuration for scanning the measuring light in the second direction is used. In addition, it is not necessary for the laser scanner itself to have a configuration for detecting the scanning angle of the measurement light in the second direction.

  For this reason, the position recognition of the obstacle on the moving surface performed by scanning the measurement light in the two-dimensional direction of the first direction and the second direction intersecting the moving direction of the moving body is not a two-dimensional laser scanner but a small size. And a low-cost one-dimensional laser scanner.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a schematic configuration diagram showing a mobile robot according to an embodiment of a moving body to which an obstacle position recognition method according to the present invention is applied.

  A mobile robot 1 (corresponding to a mobile body in the claims) of the present embodiment that moves on the moving surface indicated by reference symbol A in FIG. The four traveling wheels 5 connected to each other via the traveling wheel arms 7, the pan heads 9 and 11 installed on the main body 3, the external sensor 13 and the internal world installed on the pan head 9, respectively. A sensor 15, a monitoring camera 17 installed on the camera platform 11, and a controller unit 19 for various controls mounted on the main body 3 are included.

  Each of the traveling wheels 5 for movement incorporates, for example, an outer rotor type in-wheel motor 51 (see FIG. 9). The in-wheel motor 51 outputs a rotary encoder 51b (FIG. 9) that outputs an encoder signal indicating the amount of rotation angle. 9). The in-wheel motor 51 of each traveling wheel 5 is individually driven and controlled by the controller unit 19 described later.

  In the present embodiment, each traveling wheel arm 7 is constituted by connecting two arm portions 71 and 73 to each other by a rotary actuator 75, and one arm portion 71 is pivotally attached to the traveling traveling wheel 5, The other arm portion 73 is connected to the main body 3 via a rotary actuator 77.

  Each of the rotary actuators 75 and 77 is constituted by a high torque type motor such as a DC servo motor, for example, and rotary encoders 75a and 77a (see FIGS. 5 and 9) for outputting an encoder signal indicating a rotation angle amount. Built-in. The rotary actuators 75 and 77 will be described later so that the traveling wheel arm 7 bends and stretches into a shape that matches the posture of the target mobile robot 1 (or the main body 3), including the bent shape at the home position shown in FIG. The drive is individually controlled by the controller unit 19.

  In the present embodiment, the rotary actuators 75 and 77 of each traveling wheel arm 7 correspond to the actuators in the claims, and each traveling wheel 5 and each traveling wheel arm 7 are included in the claims. The moving means is configured.

  The pan head 9 is fixedly arranged on the front upper part of the main body 3 in the traveling direction F (corresponding to the moving direction in the claims) of the mobile robot 1, and an enlarged perspective view is shown in FIG. As shown in the figure, the two external sensors 13 (corresponding to the one-dimensional laser scanner in the claims) are installed at an interval in the left-right direction of the mobile robot 1.

  Each external sensor 13 includes a laser light source, a polygon mirror, a polygon mirror rotation motor, and a photodiode. This external sensor 13 scans the measuring light from the laser light source in the left-right direction of the mobile robot 1 (corresponding to the first direction in the claims) by the polygon mirror and outputs it to the front of the mobile robot 1. In each external sensor 13, the reflected light from the irradiation target irradiated with the measurement light is reflected toward the photodiode by the polygon mirror of the external sensor 13 and is received by the photodiode.

  Then, each external sensor 13 generates a signal having a content in which the distance value obtained from the output signal of the photodiode and the angle value of the polygon mirror are associated with the scanning angle data of the measurement light in the left-right direction of the mobile robot 1 at that timing. Output as measurement signal.

  While each external sensor 13 outputs a measurement signal, the main body 3 of the mobile robot 1 shown in FIG. 1 is bent and stretched by the traveling wheel arms 7 driven by the rotary actuators 75 and 77. The upper surface is swung up and down in the tilt direction around the home position in FIG.

  Therefore, for example, when the main body 3 of the mobile robot 1 swings downward from the home position described above in the tilt direction as shown in the explanatory diagram of FIG. 3, scanning is performed in the left-right direction of the mobile robot 1 output by each external sensor 13. The measured light is scanned in the vertical direction of the mobile robot 1 (corresponding to the second direction in the claims). For this reason, the measurement signal output from each external sensor 13 eventually corresponds to the three-dimensional shape of the measurement light irradiation target.

  The inner world sensor 15 is installed at an intermediate position between the two outer world sensors 13 and 13 on the pan head 9, and includes a gyroscope 151 (corresponding to an angular velocity sensor in the claims, see FIG. 5), an acceleration sensor 153, (See FIGS. 5 and 9). As the gyroscope 151, for example, an ultra-small mechanical (vibration type) angular velocity sensor using MEMS (Micro Electro Mechanical Systems) technology or an optical angular velocity sensor using FOG (Fiber Optic Gyro) can be used. The acceleration sensor 153 can also be a strain gauge or an optical sensor using MEMS technology.

  The pan head 11 shown in FIG. 1 is configured to be extendable and contractable by a not-shown extension / contraction motor, and supports the monitoring camera 17 so as to be able to pan and tilt.

  The surveillance camera 17 is configured by unitizing a normal camera and an infrared camera so that it can be used throughout the day and night.

  The controller unit 19 includes a GPS (global positioning system) receiver 191, a digital map database 193, a mission information database 195, as shown in a block diagram of an electrical schematic configuration of the mobile robot 1 in FIG. 4. The computer includes an external recognition processing computer 197, a travel control processing computer 199, and a power source B for these computers.

  The GPS receiver 191 is used to acquire and make known a coordinate value in the global coordinate system of the mobile robot 1. The GPS receiver 191 is a conventionally known one that receives signals from a plurality of GPS satellites (not shown) and determines its own position data. However, in the mobile robot 1 of this embodiment, in order to set the coordinate value in the vertical coordinate axis direction as a coordinate value based on the moving plane A, not the altitude, latitude data and position data that can be calculated by the GPS receiver 191 are used. Use only longitude data, not elevation data.

  In the digital map database 193, map information on the moving plane A acquired in advance by aerial survey or the like is stored as digital data by linking with three-direction data of latitude, longitude, and altitude. The mission information database 195 stores data on a route to be traveled by the mobile robot 1, data on the contents of execution instructions for various occurrence conditions (When / Where), and the like.

  As shown in the block diagram of FIG. 5, the outside world recognition processing computer 197 includes a CPU 197a, a RAM 197b, a ROM 197c, and a main body 197A incorporating a nonvolatile memory NVM, a keyboard 197B, and a monitor 197C.

  The keyboard 197B is operated when selecting a menu of processing to be executed by the external recognition processing computer 197. The monitor 197C is necessary when selecting a processing menu by operating the keyboard 197B. This is for displaying the input and selection contents on the screen and the keyboard 197B.

  The CPU 197a of the main body 197A includes a RAM 197b and a ROM 197c, a keyboard 197B and a monitor 197C, a non-volatile memory NVM, rotary encoders 75b and 77b built in the rotary actuators 75 and 77 of each traveling wheel arm 7, A gyroscope 151 and an acceleration sensor 153 of the internal sensor 15 are connected. Furthermore, a travel control processing computer 199 is connected to the CPU 197a via an interface I / F.

  For convenience of space, in FIG. 5, the rotary encoders 75 b and 77 b of the rotary actuators 75 and 77 are shown as a representative for one traveling wheel arm 7, and other encoders connected to the CPU 197 a are shown. The rotary encoders 75b and 77b of the rotary actuators 75 and 77 of the traveling wheel arm 7 are not shown in FIG.

  The RAM 197b has a data area for storing various data and a work area used for various processing operations, and the ROM 199c stores a control program for causing the CPU 197a to perform various processing operations. The non-volatile memory NVM stores calibration data determined by processing to be described later.

  Next, processing executed by the CPU 197a according to the control program stored in the ROM 197c will be described with reference to the flowcharts of FIGS.

  When the external recognition processing computer 197 is activated and the program is started, the CPU 197a first checks whether or not an interrupt input for selecting a calibration processing menu is generated by operating the keyboard 197B as shown in FIG. 6 ( Step S1). If an interrupt input has occurred (Y in step S1), a calibration process is executed. If no interrupt input has occurred (N in step S1), an external recognition process is executed (step S5). If the process of step S3 or step S5 is completed, the process returns to step S1.

  In the calibration process in step S3, as shown in the flowchart of FIG. 7, first, the moving surface A is scanned (step S31).

  In this scanning, a drive command signal for the rotary actuators 75 and 77 of each traveling wheel arm 7 is output to the traveling control processing computer 199 via the interface I / F. In response to this, the main body 3 of the mobile robot 1 is swung from the home position in FIG. 1 to the lowest calibration position in the tilt direction by the later-described control of the computer 199 for traveling control processing. Thereby, the measurement signals according to the reflected light from the moving surface A in the range irradiated with the measurement light are periodically output from the two external sensors 13 on the pan head 9.

  Therefore, the measurement signal periodically output from each external sensor 13 during the swing of the mobile robot 1 in the scanning of step S31, and the gyroscope 151 and the acceleration sensor 153 of the internal sensor 15 in the output period of this measurement signal. And the encoder signals from the rotary encoders 75b and 77b built in the rotary actuators 75 and 77 of each traveling wheel arm 7 are fetched (step S33).

  Then, based on the encoder signals from the rotary encoders 75b and 77b incorporated in the rotary actuators 75 and 77 of the traveling wheel arms 7 that have been taken in, the heights of the external sensors 13 and 13 from the moving surface A are determined ( Step S35).

  Further, based on the measurement signal of each external sensor 13 and the output signal of the gyroscope 151 taken in, and the height from the moving surface A of each external sensor 13, 13 determined in step S35, each external sensor 13 is set to the coordinate axis. A coordinate value group of the moving plane A in the range irradiated with the measurement light in the central local coordinate system is determined.

  Then, by applying an ICP (Iteractive Closet Point) algorithm, for example, to the coordinate value group in the local coordinate system of the determined moving surface A, the coordinate value group of the moving surface A having each external sensor 13 as the center of the coordinate axis is obtained. The coordinate values are combined into one coordinate value group so that there is no shift due to the difference in arrangement. As a result of this synthesis, the coordinate center is the position of a single virtual external sensor placed at the midpoint of the two external sensors 13, 13 in the left-right direction of the mobile robot 1, that is, the position of the internal sensor 15. A coordinate value group of the moving plane A in the local coordinate system is determined (step S37).

  Further, based on the measurement signal of each external sensor 13 taken in and the output signal of the acceleration sensor 153 of the internal sensor 15, a vertical coordinate axis orthogonal to the moving surface A and a vertical axis extending in the direction of the earth's gravity And the deviation angle of the determined vertical coordinate axis with respect to the vertical axis is determined (step S39).

  Then, based on the deviation angle of the vertical coordinate axis determined in step S39 with respect to the vertical axis, the coordinate value group in the local coordinate system of the combined moving surface A obtained in step S37 is converted into two external worlds in the left-right direction of the mobile robot 1. A local coordinate system based on the gravitational direction, in which the intermediate point of the sensors 13, that is, the position of the inner world sensor 15 is the center of the coordinate axis, and the extending direction of the vertical coordinate axis coincides with the gravitational direction. Calibration data necessary for conversion to the coordinate value group of the moving plane A in the converted local coordinate system) is calculated, stored in the nonvolatile memory NVM (step S41), and the calibration process is terminated.

  On the other hand, in the external environment recognition process in step S5 of FIG. 6, as shown in the flowchart of FIG. 8, first, whether or not a scanning command signal from the travel control processing computer 199 is input via the interface I / F. Is confirmed (step S51), and if not input (N in step S51), the external recognition process is terminated.

  On the other hand, when the scanning command signal from the travel control processing computer 199 is input via the interface I / F (Y in step S51), the measurement signal output periodically from each external sensor 13 and Then, the output signal of the gyroscope 151 of the internal sensor 15 and the encoder signals from the rotary encoders 75b and 77b built in the rotary actuators 75 and 77 of each traveling wheel arm 7 in the output cycle of the measurement signal are fetched (step) S53).

  Then, based on the encoder signals from the rotary encoders 75b and 77b incorporated in the rotary actuators 75 and 77 of the traveling wheel arms 7 that have been taken in, the heights of the external sensors 13 and 13 from the moving surface A are determined ( Step S55).

  Further, based on the measurement signal of each external sensor 13 taken in step S53 and the output signal of the gyroscope 151, and the height from the moving surface A of each external sensor 13, 13 determined in step S55, each external sensor 13 is displayed. In the local coordinate system with the coordinate axis as the center, the coordinate value group of the moving surface A in the range irradiated with the measurement light is determined and synthesized into one coordinate value group in the same manner as in step S37 (step S57). .

  Further, the coordinate value group in the local coordinate system of the combined moving surface A determined in step S57 and the calibration data stored in the nonvolatile memory NVM are used for the travel control processing computer 199 via the interface I / F. (Step S59), and the external environment recognition process is terminated.

  As is clear from the above description, in the mobile robot 1 of the present embodiment, step S37 in the flowchart of FIG. 7 and step S57 in the flowchart of FIG. The process corresponds to the value indexing means, step S39 in FIG. 7 is the process corresponding to the deviation angle indexing means in the claims, and step S41 in FIG. This process corresponds to the calibration data indexing means.

  As shown in the block diagram of FIG. 9, the travel control processing computer 199 has a main body 199A incorporating a CPU 199a, a RAM 199b, a ROM 199c, and a nonvolatile memory NVM, a keyboard 199B, and a monitor 199C.

  The keyboard 199B is operated when selecting a menu of processing to be executed by the computer 199 for traveling control processing. The monitor 199C is necessary for selecting a processing menu by operating the keyboard 199B. This is for displaying the input and selection contents by the screen and the keyboard 199B.

  The CPU 199a of the main body 199A includes a RAM 199b and ROM 199c, a keyboard 199B and monitor 199C, a non-volatile memory NVM, a driver 51a of the in-wheel motor 51 of each moving traveling wheel 5, a rotary encoder 51b, and each traveling wheel. The drivers 75a and 77a of the rotary actuators 75 and 77 of the arm 7 and the rotary encoders 75b and 77b, the acceleration sensor 153 of the internal sensor 15, the GPS receiver 191, the digital map database 193, and the mission information database 195 are connected. . Furthermore, an external recognition computer 197 is connected to the CPU 199a via an interface I / F.

  For convenience of space, in FIG. 9, the driver 51 a and the rotary encoder 51 b of the in-wheel motor 51 are representatively shown for the one traveling wheel 5 and are connected to the CPU 199 a. The driver 51a and the rotary encoder 51b of the in-wheel motor 51 of the traveling wheel 5 for movement are omitted in FIG.

  For the same reason, in FIG. 9, the drivers 75a and 77a of the rotary actuators 75 and 77 and the rotary encoders 75b and 77b are shown as representative of one traveling wheel arm 7 and connected to the CPU 199a. The description of the drivers 75a and 77a and the rotary encoders 75b and 77b of the rotary actuators 75 and 77 of the other traveling wheel arm 7 is omitted in FIG.

  The RAM 199b has a data area for storing various data and a work area used for various processing operations, and the ROM 199c stores a control program for causing the CPU 199a to perform various processing operations.

  The nonvolatile memory NVM stores data of coordinate value groups of the moving plane A in the global coordinate system. Further, the nonvolatile memory NVM has a relative positional relationship between the installation location of the GPS receiver 191 and the midpoint of the two external sensors 13 and 13 in the left-right direction of the mobile robot 1, that is, the position of the internal sensor 15. The relative position data indicated by the coordinate value difference in the global coordinate system is also stored. This relative position data is obtained from the coordinate value in the global coordinate system of the mobile robot 1 acquired by the GPS receiver 191, the intermediate point between the two external sensors 13, 13, that is, the position of the internal sensor 15 in the global coordinate system. Used to determine coordinate values and make them known.

  Next, processing relating to the position recognition of the obstacle on the moving surface A, which is performed by the CPU 199a according to the control program stored in the ROM 199c, will be described with reference to the flowchart of FIG.

  When the traveling control processing computer 199 is activated and the program is started, the CPU 199a first sends a drive command signal to the rotary actuators 75 and 77 of each traveling wheel arm 7 from the outside recognition processing computer 197 via the interface I / F. It is confirmed whether or not an input has been made (step S71).

  When a drive command signal is input to the rotary actuators 75 and 77 of each traveling wheel arm 7 (Y in step S71), output of the driving signal to the driver 51a of the in-wheel motor 51 of each traveling wheel 5 is stopped. The mobile robot 1 is stopped and a drive signal is output to the drivers 75a and 77a of the rotary actuators 75 and 77 of the traveling wheel arms 7, so that the main body 3 of the mobile robot 1 is moved in the tilt direction from the home position in FIG. After swinging to the lowest calibration position (step S73), the process returns to step S71.

  On the other hand, if the drive command signal is not input from the external recognition processing computer 197 (N in step S71), the latitude and longitude data output from the GPS receiver 191 and the rotary actuator 75 of each traveling wheel arm 7 are displayed. , 77 from the encoder signals from the rotary encoders 75b, 77b, the coordinate value of the mobile robot 1 in the global coordinate system is determined (step S75).

  Then, it is confirmed whether or not the coordinate value group data of the moving surface A in the global coordinate system related to the periphery of the coordinate value in the calculated global coordinate system is stored in the nonvolatile memory NVM (step S77). If not (N in Step S77), a scanning command signal is output to the outside recognition processing computer 197 via the interface I / F (Step S79).

  Further, the output of the drive signal to the driver 51a of the in-wheel motor 51 of each traveling wheel 5 is stopped to stop the mobile robot 1, and the drivers 75a and 77a of the rotary actuators 75 and 77 of each traveling wheel arm 7 are stopped. A drive signal is output and the main body 3 of the mobile robot 1 is swung by a predetermined angle downward from the home position in FIG. 1 (step S81), and then the synthesized signal from the computer 197 for external recognition processing is displayed. It waits for the coordinate value group and calibration data in the local coordinate system of the moving surface A to be input via the interface I / F (step S83).

  If the coordinate value group in the local coordinate system of the combined moving surface A and the calibration data are input from the outside recognition processing computer 197 via the interface I / F (Y in step S83), they are input. Using the calibration data, the coordinate value group in the local coordinate system of the combined moving plane A that has been input is converted into the coordinate value of the moving plane A in the local coordinate system based on the gravitational direction with the vertical coordinate axis direction coinciding with the gravitational direction. Conversion into a group (step S85).

  Then, using the relative position data stored in the nonvolatile memory NVM, the coordinate value group of the moving surface A in the converted local coordinate system of the gravity direction is used, and the vertical coordinate axis direction with the GPS receiver 191 as the center of the coordinate axis is used. Is converted into a coordinate value group of the local coordinate system based on the gravity direction matched with the gravity direction, and this is further converted into the latitude and longitude data received by the GPS receiver 191 and the rotary actuator 75 of each traveling wheel arm 7. , 77 based on the coordinate values of the mobile robot 1 in the global coordinate system, which are determined from the encoder signals from the rotary encoders 75b, 77b, are converted into a coordinate value group of the moving plane A in the global coordinate system (step S87). .

  Then, after storing the converted coordinate value group of the moving plane A in the global coordinate system in the nonvolatile memory NVM (step S89), the process returns to step S71.

  In step S77, when the coordinate value group data of the moving plane A in the global coordinate system related to the periphery of the coordinate value in the global coordinate system determined in step S75 is stored in the nonvolatile memory NVM (Y). The coordinate value group of the stored data is collated with the digital data of the map information of the digital map database 193 relating to the location corresponding to this coordinate value group, and there is a large difference in the coordinate value in the vertical coordinate axis direction. Whether or not there is an obstacle on the moving surface A is determined (step S91).

  If it is determined that there is an obstacle on the moving surface A (Y in step S91), the movement pattern of the mobile robot 1 for avoiding the obstacle determined to be present is analyzed, and the inversion of each moving traveling wheel 5 is analyzed. Driving signals are output to the driver 51a of the wheel motor 51 and the drivers 75a and 77a of the rotary actuators 75 and 77 of the traveling wheel arms 7, respectively, and the operation corresponding to the analyzed movement pattern is performed for each movement of the mobile robot 1 The in-wheel motor 51 of the traveling wheel 5 and the rotary actuators 75 and 77 of each traveling wheel arm 7 are executed (step S93).

  On the other hand, if it is not determined that there is an obstacle on the moving surface A (N in step S91), digital data of map information near the coordinate value of the mobile robot 1 in the global coordinate system is read from the digital map database 193, Based on this and the execution route for the traveling route indicated by the data stored in the mission information database 195 and various generation conditions, the driver 51a of the in-wheel motor 51 of each traveling wheel 5 and the rotary of each traveling wheel arm 7 are displayed. Driving signals are output to the drivers 75a and 77a of the actuators 75 and 77, respectively, and the movement according to the execution instruction for the traveling route indicated by the data stored in the mission information database 195 and various generation conditions is performed for each movement of the mobile robot 1 ( Of the in-wheel motor 51 of the traveling wheels 5 and the traveling wheel arms 7 Is performed over elementary actuators 75, 77) (step S95).

  In step S93 or step S95, if the mobile robot 1 is caused to execute each operation, the process returns to step S71.

  As is apparent from the above description, in this embodiment, step S73 and step S81 in the flowchart of FIG. 10 are processing corresponding to the posture changing means in the claims, and step S85 in FIG. And step S87 is a process corresponding to the global coordinate system coordinate value indexing means excluding the processing part corresponding to the local coordinate system coordinate value conversion means in the claims.

  Note that the outside world recognition processing computer 197 and the travel control processing computer 199 may perform all of the above processing by causing the CPU to execute the control program stored in the ROM. It may be executed by a DSP (Digital Signal Processor).

  In the mobile robot 1 of the present embodiment configured as described above, when the operation of the mobile robot 1 is started, if the calibration processing menu is selected by operating the keyboard 197B of the external environment recognition processing computer 197, the travel control processing Under the control of the computer 199, the measurement light scanned in the left-right direction of the mobile robot 1 is output from the two external sensors 13 while the main body 3 of the mobile robot 1 is swung in the tilt direction.

  When each external sensor 13 receives the reflected light from the moving surface A, which is the measurement light irradiation target, and outputs each measurement signal, each external sensor 13 is placed at the center of the coordinate axis in the external recognition processing computer 197. The coordinate value group of the moving plane A in the local coordinate system obtained by combining the two local coordinate systems is determined, and the vertical coordinate axis of the combined local coordinate system and the output signal of the acceleration sensor 153 of the inner world sensor 15 are determined. The angle of deviation from the vertical axis extending in the direction of gravity of the earth to be determined is determined.

  Further, based on the calculated deviation angle, the coordinate value group of the moving surface A in the local coordinate system based on the gravitational direction is obtained by matching the direction of the vertical coordinate axis of the combined local coordinate system with the gravity direction of the earth. Calibration data necessary for conversion is calculated by the external recognition computer 197 and stored in the non-volatile memory NVM of the external recognition computer 197.

  Once the calibration data has been determined, the mobile robot 1 is controlled under the control of the travel control processing computer 199, and the data of the coordinate value group of the moving plane A in the global coordinate system is stored in the nonvolatile memory of the travel control processing computer 199. By moving to a place not stored in the NVM, the main body 3 of the mobile robot 1 is swung in the tilt direction under the control of the travel control processing computer 199, and the measurement light scanned in the left-right direction of the mobile robot 1 It is output from each of the two external sensors 13.

  When each external sensor 13 receives the reflected light from the moving surface A, which is the measurement light irradiation target, and outputs each measurement signal, each external sensor 13 is placed at the center of the coordinate axis in the external recognition processing computer 197. The coordinate value group of the moving plane A in the local coordinate system obtained by combining the two local coordinate systems is calculated, and the coordinate value group of the moving plane A in the combined local coordinate system is determined by the computer 197 for external recognition processing. Together with the calibration data stored in the non-volatile memory NVM, it is output to the travel control processing computer 199.

  In response to this, the travel control processing computer 199 uses the calibration data input from the external recognition processing computer 197 to convert the coordinate value group of the moving surface A in the combined local coordinate system to the global coordinate system. It is converted into a coordinate value group of the moving surface A in the local coordinate system of the same gravitational direction with the same vertical coordinate axis direction as the gravitational direction, and further converted into a coordinate value group of the moving surface A in the global coordinate system. It is stored in the nonvolatile memory NVM of the control processing computer 199.

  In the travel control processing computer 199, the coordinate value group of the moving plane A in the global coordinate system stored in the nonvolatile memory NVM and the surroundings of the coordinate value of the mobile robot 1 in the global coordinate system, and the digital map database 193 By comparing the map information with the digital data, it is determined whether there is an obstacle on the moving surface A around the mobile robot 1.

  When it is determined that there is an obstacle, the mobile robot 1 flexes and extends each traveling wheel arm 7 individually as appropriate under the control of the traveling control processing computer 199, and the in-wheel motor 51 of each traveling traveling wheel 5 rotates. To move to avoid the obstacle.

  Thus, according to the mobile robot 1 of the present embodiment, as the one-dimensional laser scanner that scans the measurement light only in the left-right direction of the mobile robot 1 by swinging the main body 3 of the mobile robot 1 in the tilt direction. Since the external sensor 13 can be used to scan the measurement light in the two-dimensional direction of the mobile robot 1 that is different from the traveling direction F of the mobile robot 1, the small and low-cost external sensor 13 can be used. The position of the obstacle on the moving surface A can be recognized.

  In the present embodiment, two external sensors 13 and 13 are used, and the movements in the local coordinate system centered on the coordinate axes are determined by using the reflected light from the moving surface A received by each of them. A case has been described in which the coordinate value group of the surface A is synthesized into one coordinate value group such that, for example, by applying an ICP (Iteractive Closet Point) algorithm, a shift due to a difference in mutual arrangement is eliminated.

  However, a single external sensor 13 is used, and the movement in a single local coordinate system is determined by using the reflected light from the moving surface A received by the external sensor 13 and having the external sensor 13 as the center of the coordinate axis. The coordinate value group of the plane A is converted into the coordinate value group of the moving plane A in the local coordinate system based on the gravitational direction using the calibration data, and this is further converted into the coordinate value group of the moving plane A in the global coordinate system. It is good also as a structure to convert.

  However, as in the mobile robot 1 of the present embodiment, the coordinate value group of the moving plane A in the local coordinate system, which is calculated using the two external sensors 13 and 13 and has the external sensors 13 as the centers of the coordinate axes, is 1 If it is configured to synthesize one coordinate value group, there is an advantage that the coordinate value group of the moving surface A can be acquired over a wide range or can be acquired with high accuracy.

  In the mobile robot 1 of the present embodiment, calibration data stored in the non-volatile memory NVM of the external recognition processing computer 197 is transferred from the external sensor 13 while the main body 3 of the mobile robot 1 is swung in the tilt direction. Deviation between the vertical coordinate axis of the local coordinate system calculated from the reflected light and the vertical axis extending in the gravity direction of the earth detected from the output of the acceleration sensor 153 by irradiating the flat moving surface A with the measurement light The angle was determined, and the index was determined based on this deviation angle.

  However, the calibration data is separately determined in advance and stored in advance in the non-volatile memory NVM of the external recognition computer 197, and the coordinate value group of the moving surface A in the local coordinate system is converted to the moving surface A in the global coordinate system. It is good also as a structure utilized when converting into the coordinate value group.

  By the way, if it is guaranteed that the moving surface A on which the mobile robot 1 moves extends vertically with respect to the direction of gravity of the earth, that is, extends in the horizontal direction of the earth, Obviously, it is not necessary to use calibration data when converting the coordinate value group of the moving plane A into the coordinate value group of the moving plane A in the global coordinate system.

  Furthermore, in the mobile robot 1 of the present embodiment, each traveling wheel arm 7 that connects the front, rear, left, and right traveling wheels 5 to the main body 3 includes two arm portions 71 and 73 and two rotary actuators 75 and 77. The configuration in which the main body 3 of the mobile robot 1 is swung in the tilt direction by bending and extending each traveling wheel arm 7 individually as appropriate has been described.

  However, for example, the main body 3 of the mobile robot 1 is tilted by connecting the main body 3 and the traveling wheels 5 for movement using an air cylinder having a rod that can be advanced and retracted, and moving the rod of each air cylinder individually and appropriately. The configuration in the case where the posture of the main body 3 of the mobile robot 1 is changed in the tilt direction by changing the posture of the main body 3 such as a configuration that swings in the tilting direction is described in the present embodiment. It is not limited to 7 and is optional.

  Further, in the mobile robot 1 of the present embodiment, the pan head 9 is fixedly disposed on the front upper part of the main body 3 of the mobile robot 1. For example, the pan head 9 is configured to be swingable in the tilt direction. A movement different from the traveling direction F of the mobile robot 1 using the external sensor 13 as a one-dimensional laser scanner that scans the measurement light only in the left-right direction of the mobile robot 1 by swinging the head 9 in the tilt direction. You may comprise so that measurement light can be scanned in the two-dimensional direction of the robot 1 up and down, right and left.

  In other words, the configuration for changing the posture of the main body 3 of the mobile robot 1 in the tilt direction is not limited to that for swinging the main body 3 itself of the mobile robot 1 described in the present embodiment in the tilt direction.

  In that case, the configuration for swinging the pan head 9 in the tilt direction (for example, a stepping motor driven by the control of the travel control processing computer 199) corresponds to the posture changing means in the claims. become.

  The present invention is not limited to the moving body that is moved by the traveling wheel described in the present embodiment, but can change the posture with respect to the moving surface A, such as a moving body that moves by a crawler or a moving body that moves by walking motion. Needless to say, the present invention is widely applicable as long as it is a movable body.

It is a schematic block diagram which shows the mobile robot which concerns on one Embodiment of the mobile body to which the obstacle position recognition method by this invention is applied. It is an expansion perspective view which shows arrangement | positioning of the external field sensor and internal field sensor on the pan head of FIG. It is explanatory drawing of operation | movement at the time of the main body of the mobile robot of FIG. It is a block diagram which shows the electrical schematic structure of the mobile robot of FIG. It is a block diagram which shows schematic structure of the computer for external environment recognition processing of FIG. It is a flowchart of the process which CPU of the computer for external recognition processing of FIG. 5 performs according to the control program stored in ROM. It is a flowchart of the calibration process of FIG. It is a flowchart of the external field recognition process of FIG. It is a block diagram which shows schematic structure of the computer for driving control processing of FIG. It is a flowchart of the process which CPU of the computer for driving control processing of FIG. 9 performs according to the control program stored in ROM.

Explanation of symbols

1 Mobile robot (moving body)
3 Body 5 Traveling wheel for movement (moving means)
7 Traveling wheel arm (moving means)
75, 77 Rotary actuator (actuator)
13 External sensor (one-dimensional laser scanner)
151 Gyroscope (Angular velocity sensor)
153 Acceleration sensor 197 Computer for external recognition processing (scanning angle detection means, local coordinate system coordinate value indexing means, global coordinate system coordinate value indexing means, deviation angle indexing means, calibration data indexing means)
199 Computer for travel control processing (posture changing means, global coordinate system coordinate value indexing means)
A Moving surface F Traveling direction (moving direction)

Claims (6)

A method for recognizing an obstacle on a moving surface from a moving body that moves on a moving surface,
From the one-dimensional laser scanner on the moving body, output measurement light scanned in a first direction intersecting the moving direction of the moving body,
By changing the attitude of the moving body with respect to the moving surface or the attitude of the one-dimensional laser scanner with respect to the moving body, the at least the elevation angle or depression angle of the one-dimensional laser scanner with respect to the moving surface is changed. Scanning the measurement light in a second direction intersecting the body movement direction and perpendicular to the first direction;
A change in at least an elevation angle or depression angle of the one-dimensional laser scanner with respect to the moving surface is detected by an angular velocity sensor on the moving body in which a relative position with respect to the one-dimensional laser scanner is fixed. Detecting a scanning angle of the measuring light;
The reflected light from the measurement light irradiation target is received by the one-dimensional laser scanner,
The reflected light received by the one-dimensional laser scanner, a known scanning angle of the measurement light by the one-dimensional laser scanner in the first direction, and the detection of the measurement light in the second direction Based on the scanned angle, a coordinate value group of the obstacle in a local coordinate system based on the position of the one-dimensional laser scanner is determined,
A coordinate value group in the global coordinate system of the obstacle is determined based on a coordinate value group in the local coordinate system of the obstacle and the known coordinate value in the global coordinate system of the one-dimensional laser scanner,
Based on the coordinate value group in the global coordinate system of the calculated obstacle, the position of the obstacle is recognized.
An obstacle position recognition method for a moving object.   A distance between at least some of the traveling wheels for traveling of the moving body and the main body of the moving body is made closer to each other to change the posture of the moving body relative to the moving surface. 2. The obstacle position recognition method for a moving object according to claim 1, wherein the obstacle position is recognized.   The reflected light from the flat reference surface as the moving surface received by the one-dimensional laser scanner, a known scanning angle of the measurement light by the one-dimensional laser scanner in the first direction, and the first The vertical coordinate axis in the local coordinate system is determined based on the detected scanning angle of the measuring light in the direction of 2, and the acceleration sensor on the moving body whose relative position with respect to the one-dimensional laser scanner is fixed A gravity direction is detected, a deviation angle of the vertical coordinate axis of the calculated local coordinate system with respect to a vertical axis extending in the gravity direction of the detected earth is calculated, and the local coordinate system is determined based on the calculated deviation angle. The coordinate value in is a post-conversion log with the position of the one-dimensional laser scanner as a reference and the direction of gravity of the earth as the direction of the vertical coordinate axis. Calibration data to be converted into coordinate values in the cal coordinate system is determined, and by using the calculated calibration data, the coordinate value group in the local coordinate system of the obstacle is based on the position of the one-dimensional laser scanner, And the coordinate value group in the transformed local coordinate system of the transformed local coordinate system is converted into the coordinate value group in the transformed local coordinate system with the gravity direction of the earth as the direction of the vertical coordinate axis, and the one-dimensional laser 3. The obstacle position recognition method for a moving object according to claim 1, wherein a coordinate value group in the global coordinate system of the obstacle is determined based on a known coordinate value in the global coordinate system of the scanner. A moving body for recognizing the position of the obstacle using measurement light scanned in a first direction and a second direction orthogonal to each other in order to avoid the obstacle on the moving surface and move on the moving surface Because
The body,
Moving means for moving the body in the direction intersecting the first direction and the second direction on the moving surface;
A one-dimensional laser scanner that is provided in the main body, outputs the measurement light, scans in the first direction, and receives reflected light from an irradiation target of the measurement light;
Posture changing means for changing at least the elevation angle or depression angle of the one-dimensional laser scanner relative to the moving surface by changing the posture of the main body relative to the moving surface or the posture of the one-dimensional laser scanner relative to the main body;
An angular velocity sensor that is provided in the main body with a fixed relative position with respect to the one-dimensional laser scanner and detects a change in at least an elevation angle or depression angle with respect to the moving surface of the one-dimensional laser scanner;
A scanning angle detection means for detecting a scanning angle of the measurement light in the second direction based on a detection result of the angular velocity sensor;
The reflected light received by the one-dimensional laser scanner, the known scanning angle of the measurement light by the one-dimensional laser scanner in the first direction, and the second detected by the scanning angle detection means Local coordinate system coordinate value indexing means for determining a coordinate value group of the obstacle in a local coordinate system based on the position of the one-dimensional laser scanner based on the scanning angle of the measurement light in the direction;
Based on the coordinate value group in the local coordinate system of the obstacle determined by the local coordinate system coordinate value indexing means, and the known coordinate value in the global coordinate system of the one-dimensional laser scanner, the obstacle A global coordinate system coordinate value indexing means for determining a coordinate value group in the global coordinate system of
Recognizing the position of the obstacle based on a coordinate value group in the global coordinate system of the obstacle determined by the global coordinate system coordinate value acquisition means;
A moving object characterized by that.   The moving means includes a plurality of traveling wheels for moving the main body, and a plurality of traveling wheel arms respectively connecting the plurality of traveling traveling wheels to the main body. At least some of the traveling wheel arms are provided with an actuator for separating and approaching the distance between the main body and the traveling traveling wheel, and the posture changing means is configured to move the plurality of traveling traveling wheels by the actuator. 5. The at least elevation angle or depression angle of the one-dimensional laser scanner with respect to the moving surface is changed by changing a posture of the main body with respect to the moving surface by making a distance between at least a part of the main body and the main body apart from each other. The moving body described.   An acceleration sensor that is provided in the main body with a fixed relative position with respect to the one-dimensional laser scanner and detects the direction of gravity of the earth, and the local coordinate system with respect to a vertical axis that extends in the direction of gravity of the earth detected by the acceleration sensor A deviation angle indexing means for determining a deviation angle of the vertical coordinate axis, and based on the deviation angle calculated by the deviation angle indexing means, the coordinate value in the local coordinate system is based on the position of the one-dimensional laser scanner, and Calibration data indexing means for calculating calibration data for converting the gravity direction of the earth into a coordinate value in the converted local coordinate system with the direction of the vertical coordinate axis as the direction of the vertical coordinate axis, The calculating means includes the calibration data calculated by the calibration data calculating means. The coordinate value group in the local coordinate system of the obstacle determined by the local coordinate system coordinate value indexing means is based on the position of the one-dimensional laser scanner, and the gravity direction of the earth is a vertical coordinate axis. A local coordinate system coordinate value converting means for converting into a coordinate value group in the converted local coordinate system, and the deviation angle indexing means is used as the moving surface received by the one-dimensional laser scanner. The reflected light from the flat reference surface, the known scanning angle of the measurement light by the one-dimensional laser scanner in the first direction, and the second direction detected by the scanning angle detection means The deviation angle of the vertical coordinate axis of the local coordinate system, calculated by the local coordinate system coordinate value indexing unit, based on the scanning angle of the measurement light with respect to the vertical axis A coordinate value group in the transformed local coordinate system of the obstacle after being transformed by the local coordinate system coordinate value transforming means, and a known value in the global coordinate system of the one-dimensional laser scanner. The moving body according to claim 4, wherein a coordinate value group in a global coordinate system of the obstacle is determined based on the coordinate value of the obstacle.

Images