JP2011209845A - Autonomous mobile body, self-position estimation method and map information creation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an autonomous mobile body, a self-position estimation method, a map information creation system and a map information creation method, allowing improvement of accuracy of self-position estimation.SOLUTION: The autonomous mobile body includes: a laser sensor measuring distance data corresponding to a distance to a peripheral obstacle; a map information storage part 22 storing a plurality of pieces of two-dimensional estimating map information in different measurement faces of movement environment; a map information selection part 26 selecting one piece of the estimating map information from the plurality of pieces of the estimating map information stored in the map information storage part 22 based on a measurement angle of the laser sensor or a self-position; and a self-position estimation part 23 estimating the self-position based on the estimating map information selected in the map information selection part 26 and the distance data measured by the laser sensor.

Description

  The present invention relates to an autonomous mobile body, a self-position estimation method, a map information creation system, and a map information creation method.

  In recent years, mobile bodies such as autonomous mobile robots that can move autonomously within a limited area inside a building or outdoors based on the surrounding environment are being developed. Since such a moving body enables autonomous movement by creating a movement route from the current position to a specific target point on a movement map relating to a moving area stored in advance, it normally moves. It has a function of recognizing its own position in the area to be operated. A self-position estimation method for controlling such an autonomous mobile body is disclosed in Patent Document 1.

JP 2009-176031 A

  By the way, in a well cab vehicle (a vehicle that can be boarded by a wheelchair), the user gets into the vehicle while riding in the wheelchair. At this time, in the conventional wheelchair, it is necessary to assist the assistant so that the winch does not operate the winch or the wheelchair falls from the vehicle. In addition, it takes time to board. Therefore, it becomes a big burden for a wheelchair user and an assistant.

  Therefore, the present applicant has developed a function capable of fully automatically boarding a well cab vehicle using an autonomous mobile body as a wheelchair. Obstacles such as seats are arranged in the well cab vehicle. For this reason, the moving environment becomes complicated, and autonomous movement based on self-position estimation is difficult. However, in the self-position estimation method described in Patent Document 1, it is necessary to provide an infrared irradiation unit in the environment. In practice, it is difficult to provide an infrared irradiation unit in a well cab vehicle. Therefore, there arises a problem that it is difficult to accurately perform self-position estimation.

  The present invention has been made in view of the above problems, and provides an autonomous mobile body, a self-position estimation method, a map information creation system, and a map information creation method that can improve the accuracy of self-position estimation. For the purpose.

  The autonomous mobile body according to the first aspect of the present invention is an autonomous mobile body that moves within the mobile environment by estimating its own position based on estimation map information in the mobile environment, and surrounding obstacles A map information storage unit for storing a plurality of two-dimensional estimation map information according to the arrangement of obstacles in the moving environment, and a sensor unit for measuring distance data according to the distance to Based on a map information storage unit that stores map information for estimation on different measurement planes, and a plurality of map information for estimation stored in the map information storage unit based on the self-position or the measurement angle of the sensor unit A self-position that estimates a self-position based on a map information selection section that selects one map information for estimation, map information for estimation selected by the map information selection section, and distance data measured by the sensor section And an estimation unit It is. According to this configuration, since the appropriate map information for estimation can be selected, the accuracy of self-position estimation can be improved.

  An autonomous mobile body according to a second aspect of the present invention is the above-described autonomous mobile body, further comprising an angle sensor for measuring a measurement angle of the sensor unit, wherein the map information storage unit has a measurement surface with a different angle. The map information for estimation is stored, and the position information selection unit selects the map information for estimation according to the measurement result of the angle sensor. Thereby, since the suitable map information for estimation can be selected, the accuracy of self-position estimation can be improved.

  The autonomous mobile body according to the third aspect of the present invention is the autonomous mobile body described above, and the map information selection section selects when the self-position estimated by the self-position estimation section exceeds a predetermined position. The map information for estimation is switched. Thereby, since the map information for estimation can be switched at an appropriate timing, the accuracy of self-position estimation can be improved.

  An autonomous mobile body according to a fourth aspect of the present invention is the above-described autonomous mobile body, and a position for switching the estimation map information is preset according to the position of an obstacle in the mobile environment. Is. Thereby, since the suitable map information for estimation can be selected, the accuracy of self-position estimation can be improved.

  An autonomous mobile body according to a fifth aspect of the present invention is the above-described autonomous mobile body, wherein the movement of the autonomous mobile body stops when the map information for estimation is switched. Thereby, the precision of self-position estimation can be improved also in the timing which switches the map information for estimation.

  An autonomous mobile body according to a sixth aspect of the present invention is the above autonomous mobile body, and estimates the self-position by pattern matching the distance image data corresponding to the distance data and the map information for estimation. To do. Thereby, a self-position can be estimated correctly.

  An autonomous moving body according to a seventh aspect of the present invention is an autonomous moving body that moves within the moving environment by estimating its own position based on map information for estimation in the moving environment, and surrounding obstacles A sensor unit for measuring the distance to the map, a map information storage unit for storing two-dimensional estimation map information according to the arrangement of obstacles in the moving environment, and the sensor unit during movement of the previous autonomous mobile body A map information creation unit that cuts out the distance image data based on the measured distance and combines it with the estimation map information, the estimation map information combined with the distance image data, and the sensor unit after the combination A self-position estimating unit that estimates the self-position based on the measurement result. Thereby, since the suitable map information for estimation can be used, the precision of self-position estimation can be improved.

  An autonomous mobile body according to an eighth aspect of the present invention is the above-described autonomous mobile body, wherein the distance image data is cut out corresponding to a partial region of the mobile environment, and the one of the mobile environments is selected. The area of the part is set based on the movable range of the obstacle. Thereby, since it can synthesize | combine appropriately, the precision of self-position estimation can be improved.

  A map information creation system according to a ninth aspect of the present invention is a map information creation system for creating map information for estimation used for self-position estimation in a mobile environment, wherein the mobile environment is A sensor unit that measures the distance to an obstacle, a position measurement unit that measures the position of the sensor unit in the moving environment, an angle change mechanism that changes the angle of the measurement surface of the sensor unit, and a measurement surface with a different angle And a map information creating unit that creates the map information for estimation by synthesizing distance image data corresponding to the distance measured by the sensor unit. Thereby, since the estimation map information suitable for self-position estimation can be created, the accuracy of self-position estimation can be improved.

  A self-position estimation method according to a tenth aspect of the present invention is a self-position estimation method in a moving body that estimates a self-position based on map information for estimation in a moving environment, and is based on the distance to surrounding obstacles. A step of measuring the corresponding distance data, and a map information storage unit in which two-dimensional estimation map information corresponding to the arrangement of the obstacle in the moving environment is stored for each different measurement surface, The method includes a step of selecting one piece of estimation map information based on a posture, and a step of estimating a self-position based on the selected piece of estimation map information and the distance data. Thereby, since the suitable map information for estimation can be selected, the accuracy of self-position estimation can be improved.

  A self-position estimation method according to an eleventh aspect of the present invention is the self-position estimation method described above, further comprising a step of measuring a measurement angle of a sensor unit that measures the distance data, wherein the map information storage unit is The map information for estimation of the measurement surface of a different angle is stored, and the map information for estimation is selected according to the measurement angle. Thereby, since the suitable map information for estimation can be selected, the accuracy of self-position estimation can be improved.

  A self-position estimation method according to a twelfth aspect of the present invention is the self-position estimation method described above, and switches to another estimation map information when the estimated self-position exceeds a predetermined position. is there. Thereby, since the suitable map information for estimation can be selected, the accuracy of self-position estimation can be improved.

  A self-position estimation method according to a thirteenth aspect of the present invention is the self-position estimation method described above, wherein a position for switching the estimation map information is preset according to the position of an obstacle in the moving environment. It is what. Thereby, since the map information for estimation can be switched at an appropriate timing, the accuracy of self-position estimation can be improved.

  A self-position estimation method according to a fourteenth aspect of the present invention is the self-position estimation method described above, wherein the autonomous mobile body stops moving when the estimation map information is switched. is there. Thereby, the precision of self-position estimation can be improved also in the timing which switches the map information for estimation.

  A self-position estimation method according to a fifteenth aspect of the present invention is the self-position estimation method described above, wherein the distance image data corresponding to the distance data and the estimation map information are subjected to pattern matching, and the self-position is estimated. Is estimated. Thereby, the accuracy of self-position estimation can be improved.

  A self-position estimation method according to a sixteenth aspect of the present invention is a self-position estimation method in a mobile body that estimates a self-position based on map information for estimation in a moving environment, and calculates a distance to surrounding obstacles. A step of measuring, a step of cutting out distance image data based on a distance measured by the sensor unit during the movement of the autonomous mobile body, and combining it with map information for estimation, and the estimation in which the distance image data is combined And a step of estimating a self-position based on the map information and the measurement result of the sensor unit after combining. Thereby, since the suitable map information for estimation can be used, the precision of self-position estimation can be improved.

  A self-position estimation method according to a seventeenth aspect of the present invention is the self-position estimation method described above, wherein the distance image data is cut out corresponding to a partial area of the movement environment, and The partial area is set based on a movable range of the obstacle. Thereby, since it can synthesize | combine appropriately, the precision of self-position estimation can be improved.

  A map information creation method according to an eighteenth aspect of the present invention is a map information creation method for creating map information for estimation used for self-position estimation in a mobile environment, wherein the mobile environment The step of measuring the distance to the obstacle, the step of measuring the position of the sensor unit in the moving environment, the step of changing the angle of the measurement surface of the sensor unit, and the sensor unit on different measurement surfaces Synthesizing distance image data corresponding to the measured distance to create estimation map information. Thereby, since the estimation map information suitable for self-position estimation can be created, the accuracy of self-position estimation can be improved.

  ADVANTAGE OF THE INVENTION According to this invention, it can provide providing the autonomous mobile body which can improve the precision of self-position estimation, the self-position estimation method, the map information creation system, and the map information creation method.

It is a figure which shows typically the whole structure of the robot concerning this invention. It is a top view which shows a mode that the robot concerning this invention approachs a vehicle. It is a side view which shows a mode that the robot concerning this invention approachs a vehicle. It is a figure which shows the map information for estimation used for the self-position estimation of the robot concerning this invention. It is a block diagram which shows the structure of the control part of the moving body concerning this invention. It is a figure which shows the map information creation system concerning Embodiment 1. FIG. It is a figure which shows typically the map information for estimation produced by the map information production system concerning Embodiment 1. FIG. 3 is a flowchart illustrating a map information creation method according to the first exemplary embodiment. It is a figure explaining the self-position estimation by the robot which moves a vehicle. It is a figure explaining switching of the map information for estimation in the robot concerning Embodiment 2. FIG. 6 is a flowchart illustrating a self-position estimation method according to a second exemplary embodiment. It is a figure explaining the self-position estimation by the robot which moves a vehicle. It is a figure which shows a mode that the robot concerning Embodiment 3 is acquiring the map information for estimation. It is a figure explaining a mode that the robot concerning Embodiment 3 reads map information for presumption. 10 is a flowchart illustrating a self-position estimation method according to a third embodiment. 10 is a flowchart showing a flow of switching estimation map information in the self-position estimation method according to the third exemplary embodiment. It is a figure which shows the map information for estimation used for the self-position estimation of the robot concerning Embodiment 4. It is a figure which shows the map information for estimation produced by the self-position estimation of the robot concerning Embodiment 4. 10 is a flowchart illustrating a self-position estimation method according to a fourth embodiment. 14 is a flowchart illustrating a flow of creating map information in the self-position estimation method according to the fourth exemplary embodiment.

  Hereinafter, embodiments of a moving body according to the present invention will be described in detail with reference to the drawings. However, the present invention is not limited to the following embodiments. In addition, for clarity of explanation, the following description and drawings are simplified as appropriate.

    First, a configuration of a robot which is an example of a moving body according to the present invention will be described with reference to FIG. FIG. 1 is a side view schematically showing the configuration of the robot 1. In the present embodiment, the robot 1 will be described as an autonomous mobile body that autonomously moves while estimating its own position in the moving environment. Here, description will be made assuming that the robot 1 moves in the well cab vehicle. Of course, the moving environment is not limited to the well cab vehicle.

  The robot 1 includes a main body 10, a front wheel 11, a rear wheel 12, a laser sensor 13, a laser sensor 14, an angle changing mechanism 15, a control unit 16, and a gyro sensor 17. The front wheel 11 and the rear wheel 12 are rotatably attached to the main body 10. The front wheel 11 and the rear wheel 12 are rotated by, for example, a motor. For example, the front wheel 11 and the rear wheel 12 move forward by rotating forward and move backward by rotating backward. Further, when the front wheels 11 are provided on both the left and right sides, the left and right are turned by changing the left and right rotational speeds. The number of wheels is not particularly limited, and may have two, three, four, or more wheels.

  The laser sensors 13 and 14 measure distance data according to the distance to surrounding obstacles on a predetermined measurement surface. The laser sensors 13 and 14 are, for example, laser range finders and move together with the robot 1. The laser sensor 13 emits laser light at a certain angle on a horizontal plane. Then, the distance to the obstacle is detected according to the time until the laser beam is reflected and returned. The distance to the obstacle is the distance to the outer shape of the obstacle. For example, the laser beam is emitted every 1 degree with respect to all directions. Therefore, the distance to the obstacle is measured in all directions on a certain measurement surface. Distance data is constructed by collecting the distances in the respective directions. Outputs from the laser sensors 13 and 14 are input to the control unit 16.

  And distance image data is acquired according to the distance to the surrounding obstacle. The distance image data is two-dimensional data having a distance value to an obstacle existing in the moving environment as a pixel value of each pixel. The self-position is estimated using the distance image data and the estimation map information. In addition, the obstacles here are those that may cause the robot 1 to collide or are obstructed to move, and include walls and the like.

  The laser sensor 13 is attached to the front of the robot 1 and the laser sensor 14 is attached to the rear of the robot 1. The laser sensor 13 measures 180 degrees forward, and the laser sensor 14 measures 180 degrees backward. Thereby, distance information can be acquired for all directions around the robot 1. Of course, the measurable range of the laser sensors 13 and 14 is not limited to 180 °. Furthermore, two-dimensional map information for estimation can be created by performing laser measurement with the laser sensors 13 and 14. Hereinafter, the measurement data obtained by the laser sensors 13 and 14 may be referred to as laser data.

  In addition, the distance data is measured using a stereo camera, a distance image camera (for example, SR-4000), a three-dimensional laser irradiation sensor (for example, Legle), etc. other than the laser range finder. Also good. Although a laser sensor that performs optical measurement is used as a sensor that measures distance data, a sensor that measures by other methods may be used. Alternatively, measurement may be performed using only one laser sensor. Furthermore, the number of laser sensors is not limited to two.

  Further, an angle changing mechanism 15 that changes the tilt angle of the laser sensor 13 is provided. The angle changing mechanism is, for example, a joint mechanism, and changes the angle of the laser sensor 13 with respect to the main body 10. Thereby, the laser sensor 13 faces up or down, and the angle around the pitch axis of the robot changes. Therefore, the tilt angle of the laser sensor 13, that is, the angle of the measurement surface changes. The angle changing mechanism 15 is provided with an encoder for measuring the rotation angle. Then, the rotation angle of the angle changing mechanism 15 is input to the control unit 16.

  Further, a control unit 16 and a gyro sensor 17 are built in the main body 10. The gyro sensor 17 is a sensor that detects the attitude angle of the robot 1 and measures the angle of the main body 10 with respect to a horizontal plane (ground), for example. The output of the gyro sensor 17 is input to the control unit 16. The elevation angle of the robot 1 can be obtained from the detection result of the gyro sensor 17. Therefore, the angle of the measurement surface of the laser sensor 13 can be obtained from the detection result of the gyro sensor 17 and the joint angle of the angle changing mechanism 15. The gyro sensor 17 can also detect the angle of the measurement surface of the laser sensor 14. Here, the angle of the measurement surface is an angle with respect to the horizontal plane (ground), and the case where the measurement plane is parallel to the horizontal plane is 0 degree.

  The control unit 16 is an arithmetic processing unit having a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), a communication interface, and the like. The control unit 16 includes a removable HDD, an optical disk, a magneto-optical disk, and the like, stores various programs and control parameters, and supplies the programs and data to a memory (not shown) or the like as necessary. . Of course, the controller 16 is not limited to one physical configuration.

  The control unit 16 performs arithmetic processing for movement control. For example, the control unit 16 performs self-position estimation. And the motor etc. for driving the front wheel 11 and the rear wheel 12 are controlled according to the self-position estimation result. Thereby, it can move autonomously, without colliding with an obstacle. Only one of the front wheel 11 and the rear wheel 12 may be a drive wheel. The control unit 16 stores map information for estimating the moving environment. Then, the self position is estimated based on the estimation map information and the distance information measured by the laser sensors 13 and 14. Specifically, pattern matching between the estimation map information and the distance image data corresponding to the distance information is performed. Then, the position where the pattern is matched is estimated as the self position of the robot 1. And it moves to the movement end point from the present self position so that it may not collide with an obstacle. The robot 1 moves with a predetermined seat position as a movement end point.

  FIGS. 2 and 3 show how the robot 1 moves in the well cab vehicle. As shown in FIGS. 2 and 3, the robot 1 enters the vehicle 50 by going up a slope 54 provided outside the vehicle. The slope 54 is attached to the rear of the vehicle 50. The robot 1 moves from the rear of the vehicle 50 to a predetermined connection position (movement end point). When the robot 1 moves to the connection position, the robot 1 is connected to the vehicle 50. At this time, when moving the vehicle 50 from the slope 54, the robot 1 estimates its own position with reference to the estimation map information.

  For example, the estimation map information is as shown in FIG. 4. In the vehicle 50, the inner wall, gear box, seat, ceiling, etc. of the vehicle 50 are obstacles. The estimation map information is image data corresponding to these obstacles. In FIG. 4, the bold line indicates the outline of the obstacle, and the dotted line indicates the distance image data measured by laser measurement. Then, self-position estimation is performed by matching these patterns.

  Specifically, the map information for estimation is created by expressing the two-dimensional space with point sequence data storing the position coordinate values of obstacles, as in the case of laser data. As shown in FIG. 4, the estimation map information stores the position coordinates of an obstacle with respect to a predefined coordinate system. A thick line shown in FIG. 4 is a boundary line between a space where no obstacle exists and a space where an obstacle exists. With reference to the estimated self-position, the vehicle moves forward, turns right, turns left, etc. so as not to collide with an obstacle included in the estimation map information. And it moves, repeating self-position estimation. Thereby, it can move autonomously, avoiding an obstacle. Note that the map information for estimation and the map information for movement may be different map information.

  Next, the configuration of the control unit 16 will be described with reference to FIG. FIG. 5 is a block diagram schematically showing the configuration of the control unit 16. The control unit 16 includes a map information creation unit 21, a map information storage unit 22, a self-position estimation unit 23, a drive control unit 24, a tilt angle control unit 25, and a map information selection unit 26. The map information creating unit 21 creates map information for estimating the moving environment. For example, the distance from a certain measurement point to an obstacle is obtained by measuring the laser sensors 13 and 14. Then, the estimation map information can be created based on the distance data corresponding to the distance from a certain measurement point to the obstacle. For example, the estimation map information is created based on the distance image data acquired by the laser sensors 13 and 14. For example, the map information for estimation is created by synthesizing information of an obstacle that does not exist in the map information with respect to certain map information.

  The map information storage unit 22 stores the estimation map information created by the map information creation unit 21. Furthermore, a plurality of estimation map information may be stored in the amount described later. The estimation map information is two-dimensional data having a moving environment, that is, position coordinates of obstacles in and around the vehicle 50. Note that the map information for estimating the moving environment may be created by another device different from the robot 1. That is, the laser measurement may be performed by a robot or apparatus different from the robot 1. The map information selection unit 26 selects one estimation map information from among a plurality of estimation map information stored in the map information storage unit 22. The map information selection unit 26 selects the map information for estimation according to the position and angle of the robot 1.

  The self-position estimation unit 23 estimates the self-position based on the selected estimation map information. That is, the distance image data obtained by the laser sensors 13 and 14 is pattern-matched with the estimation map information. Then, from the pattern matching result, the robot coordinates in the estimation map information, that is, the self-position in the moving environment is estimated. In this way, the self-position is estimated by comparing the estimation map information with the distance image data. The calculation time can be shortened by matching the two-dimensional map information for estimation with the two-dimensional distance image data.

  The drive control unit 24 controls the driving of the front wheels 11 and the rear wheels 12. That is, the moving direction is determined based on the estimated self-location and the estimation map information. Specifically, the drive control unit 24 refers to the self position and determines which direction in the estimation map information should be advanced. Thereby, for example, a command value corresponding to the rotational speed and rotational torque of the front wheels 11 and the rear wheels 12 is output to the drive motor.

  The tilt angle control unit 25 controls the angle changing mechanism 15. That is, the angle changing mechanism 15 is driven according to the detection result of the gyro sensor 17. Thereby, the tilt angle of the laser sensor 13 is controlled. For example, the tilt angle may be changed according to the position of the robot on the slope 54.

Embodiment 1 FIG.
In the present embodiment, estimation map information is created based on a plurality of laser measurement results. Then, the robot 1 performs self-position estimation based on the estimation map information. The creation of the estimation map information will be described below. First, the system configuration for creating the estimation map information will be described with reference to FIG. FIG. 6 is a diagram illustrating a configuration of a creation system that creates map information for estimation.

  Also in the present embodiment, as shown in FIG. 6, the robot 1 climbs up the slope 54 before entering the vehicle 50. In this system, a three-dimensional position measuring device 61 is provided. The three-dimensional position measuring device 61 measures the robot 1 with respect to the vehicle 50. In other words, the three-dimensional position measuring device 61 measures the positions of the laser sensors 13 and 14 in the moving environment. The three-dimensional position measuring device 61 is, for example, a robot calibration ROCAL system manufactured by Toyo Technica. Further, the vehicle 50 has obstacles 52 and 53. The obstacle 53 is a slope gear box arranged on the floor 51. The obstacle 52 is, for example, a flip-up third sheet provided on the ceiling 55.

  The moving robot 1 may shake in the pitch direction due to the deflection of the slope 54 or the like. In a state where the laser sensor 13 is horizontal without being shaken, the laser sensor 13 performs measurement on the measurement surface A. Therefore, the laser beam from the laser sensor 13 hits the obstacle 53 but does not hit the obstacle 52. On the other hand, in the shaken state, the laser sensor 13 measures the measurement surface B. In this case, the laser light emitted from the laser sensor 13 hits the obstacle 52 but does not hit the obstacle 53. Thus, in the situation where the robot 1 is shaken by the deflection of the slope 54, it is not known which of the obstacle 52 and the obstacle 53 can be measured. Therefore, in this embodiment, the map information for estimation is created by combining two pieces of map information. Specifically, the map information of both the measurement surface A and the measurement surface B is synthesized and created as estimation map information. In this case, the estimation map information includes information on the obstacles 52 and 53.

  For this reason, the robot 1 creates map information of the measurement surface A and the measurement surface B using the measurement result of the laser sensor 13. Specifically, the robot 1 is manually moved to a predetermined measurement point on the slope 54. In addition, it is preferable to make a measurement point into the position where shaking becomes large actually. For example, since the shaking is likely to occur at the beginning of the slope 54, the position at the beginning of the slope 54 is set as the measurement point. Furthermore, it is good also considering the robot 1 as a measurement point in another position. Of course, two or more measurement points may be used. Thus, the robot 1 is manually moved to one or a plurality of measurement points.

  At this time, the coordinates of the detailed measurement point of the robot 1 are measured by the three-dimensional position measuring device 61. That is, the three-dimensional coordinates of the robot 1 with the vehicle 50 as a reference are accurately measured by the three-dimensional position measuring device 61. Then, the laser sensor 13 detects the distance to the surrounding obstacle on the two measurement surfaces A and B with different tilt angles. Thereby, the distance data according to the distance to the obstacle in the measurement surfaces A and B is acquired. Distance image data on the measurement surfaces A and B are acquired. Here, since the laser range finder is used as the laser sensor 13, the distance to the obstacle in all directions is detected. Note that it is preferable to measure the distance data after detecting that the shaking has been cured by the measurement of the gyro sensor 17. Further, the tilt angle between the measurement surface A and the measurement surface B may be determined according to the actual shaking. That is, the laser sensor 13 is tilted at a swinging angle at a position where it will actually swing. Of course, measurement may be performed at different tilt points at different measurement points.

Then, in consideration of the coordinates of the measurement point measured by the three-dimensional position measuring device 61 and the tilt angle of the laser sensor 13, map information on each measurement surface is created. Thereby, map information including the distance data of the obstacle measured on the measurement surface A and map information including the distance data of the obstacle measured on the measurement surface B are created. For example, as shown in FIG. 7, the map information of the measurement surface A and the map information M _1, the map information of the measurement surface B with the map information M _2. Here, in the map information M ₁ and M ₂ , the outline of the obstacle is shown as a line. That is, the lines shown in the map information M ₁ and M ₂ indicate obstacles, and when the robot 1 moves, it moves avoiding these lines. The combined map information M _{(1 + 2)} is used as the estimation map information. In estimation map information M _{(1 + 2),} with respect to map information M _1, combining region 91 including an obstacle is not detected is added only in the map information M _2.

  In this way, the synthesized new map information is created as estimation map information in a form that complements the distance data of the missing obstacle. Then, when the robot 1 actually enters the vehicle 50, the self-position is estimated using the estimation map information. That is, the map information for estimation and the distance image data are subjected to pattern matching processing. Thereby, even if there is a disturbance such as the slope 54 shaking, the self-position can be estimated accurately. Even when there is a disturbance such as shaking, the self-position can be accurately estimated. Therefore, it is not necessary to wait for the shaking to subside during movement, and safe and quick movement is possible.

  Next, a map information creation method will be described with reference to FIG. FIG. 8 is a flowchart showing a map information creation method. First, the coordinates of the measurement point are calculated based on the connection position of the robot 1 (step S101). One or a plurality of positions necessary for creating the estimation map information are set as measurement points. Measurement points include the position at which the slope 54 starts to rise, the inside of the vehicle 50, and the like. Here, the measurement point is determined so that the obstacle of the entire vehicle 50 is included. That is, information on all obstacles is included in the estimation map information. Then, a laser sensor is placed at the measurement point (step S102). For example, the robot 1 is manually moved to the measurement point. Here, the position of the robot 1 is measured by the three-dimensional position measuring device 61. For this reason, the robot 1 can be accurately moved to the measurement point. When the robot 1 is moved to the measurement point, laser data is acquired by the laser sensor 13 (step S103). This laser data includes information on the distance of the obstacle.

  Then, map information associated with the coordinates is created (step S104). That is, the map information considering the coordinates of the measurement point and the tilt angle is created. The map information has a pattern corresponding to the distance image data. Then, it is determined whether or not the measurement point exists (step S105). That is, the above process is repeated until the measurement of all measurement points is completed. When the measurement at all measurement points is completed, all the map information is synthesized and the creation of the map is completed. That is, the map information for estimation is created by merging a plurality of pieces of map information having different tilt angles. In this way, a map for self-position estimation that is resistant to disturbance can be created. Since the estimation map information suitable for self-position estimation can be created, the accuracy of self-position estimation can be improved.

  As described above, the position information (distance image data) on the measurement surfaces at different angles is synthesized to create the estimation map information on the horizontal plane. Of course, not only the estimation map information in the horizontal plane but also the estimation map information for planes other than the horizontal plane may be created. It is also possible to acquire laser data at a tilt angle of 3 or more at one measurement point and synthesize 3 or more distance image data. Note that at some measurement points, laser data may be acquired with only one tilt angle. The combined map information may be used only at a place where the shaking becomes large. For example, the combined estimation map information is used only in the vicinity of the slope 54 where the slope 54 starts to rise. In other places where the shaking is small, the map information measured at one tilt angle is used as the estimation map information. That is, map information in which map information of two tilt angles is not synthesized is used as estimation map information. Thereby, self-position estimation can be performed accurately. Furthermore, distance image data of measurement surfaces having the same angle and different heights may be combined.

Embodiment 2. FIG.
In the present embodiment, the map is switched according to the robot 1's own position. In the present embodiment, map information that has not been combined is used as estimation map information. That is, map information based on laser data measured at a certain tilt angle is used as the estimation map information, instead of the combined map information shown in the first embodiment.

  Self-position estimation in the present embodiment will be described with reference to FIG. FIG. 9 is a diagram illustrating self-position estimation by the robot 1 that moves the vehicle 50. As shown in FIG. 9, while the slope 54 is moving, the height of the laser sensor 13 varies depending on the position of the robot 1. For example, the laser sensor 13 is at the height of the measurement surface C at the beginning of the rise of the slope 54, but the laser sensor 13 is at the height of the measurement surface D at the end of the slope 54. That is, even if the tilt angle of the laser sensor 13 is kept constant, the height of the measurement surface gradually changes. Therefore, the obstacle 53 cannot be detected on the measurement surface D. That is, since the measurement surface of the laser sensor 13 is higher than the obstacle 53, the laser light from the laser sensor 13 does not hit the obstacle 53. Thus, when the laser sensor 13 reaches the height of the measurement surface, the detection result of the laser sensor 13 does not include information on the obstacle 53. Thus, since the obstacle that can be detected by the laser sensor 13 changes according to the height of the measurement surface, the distance information of the obstacle will be different. That is, the distance image pattern varies depending on the height of the measurement surface.

Therefore, in this embodiment, the estimation map information to be used is switched according to the self-position of the robot 1 in the moving environment. That is, different estimation map information is used according to the coordinates of the robot 1. For example, until _{P A} point indicated in FIG. 10 uses the estimation map information _{M a.} Then, the _{P A} point to _{P B} point is used for estimation map information _{M ab.} Furthermore, from point B to point C using the estimated map information _{M bc,} from _{P C} point to _{P D} point is used for estimation map information _{M cd.} Furthermore, when the point D is exceeded, the map information _Md is switched to the estimation map information Md. In addition, since the seat becomes too close after passing point D, self-position estimation is performed using only the measurement result from the laser sensor 14 on the rear side. Thus, the accuracy of self-position estimation can be improved by switching the laser sensor to be used according to the self-position.

  In this way, the map information for estimation is switched according to the estimated coordinates of the self position. That is, when the estimated self-position exceeds the threshold value coordinates, the map is switched to different estimation map information. Thereby, the precision of self-position estimation can be made high. For example, when switching the estimation map information, the movement of the robot 1 may be temporarily stopped. That is, when the coordinate of the self position exceeds the threshold value, the movement of the robot 1 is temporarily stopped. Then, after the switching is completed, the movement of the robot 1 is resumed. Thereby, the precision of self-position estimation can be improved also in the timing which switches the map information for estimation. The coordinates serving as the threshold may be set in advance according to the position of the obstacle. That is, if the laser sensor 13 exceeds an obstacle, the estimation map information may be switched. The coordinates to be switched may be set from a position where the self-position estimation result is good when the actual robot 1 is run.

  In the present embodiment, the map information storage unit 22 stores a plurality of estimation map information according to the height of the measurement surface. And the map information for estimation which the map information selection part 26 selects according to a self-position is switched. The process of switching the estimation map information will be described with reference to FIG. FIG. 11 is a flowchart showing a process of moving while switching the estimation map information.

First, a person gets on the robot 1 and moves it to near the bottom of the slope 54 (step S201). Then, self-position estimation is performed by measurement of the laser sensors 13 and 14 (step S202). That is, the distance image data of the obstacle detected by the laser sensors 13 and 14 is compared with the estimation map information to estimate the self position. Of the plurality of estimation map information, one estimation map information is set to be used first. For example, in the example shown in FIG. 10, it is set as the estimated map information using estimation map information M _a first.

  Then, it is determined whether or not self-position estimation is possible (step S203), and the above processing is repeated until estimation is possible. If the self position can be estimated (YES in step S203), the robot 1 starts moving in the moving environment. Specifically, it moves within the slope 54 or the vehicle 50 while estimating its own position based on the measurement results of the laser sensors 13 and 14 (step S204). As a result, the command values are output to the drive motors for the front wheels 11 and the rear wheels 12. Then, it is determined whether or not the estimated self position has reached the connection position (movement end point) (step S205). If the connection position has been reached (step S206), the movement is terminated.

  If the connection position has not been reached, it is determined whether or not the map switching point has been reached (step S207). Here, whether or not the switching point has been reached is determined based on whether or not the estimated coordinates of the self-position exceeded a predetermined threshold value. If the switching point has not been reached (NO in step S207), the processing from step S204 is repeated using the same estimation map information as it is. When the switching point is reached, the estimation map information for self-position estimation is switched (step S208). Then, the processing from step S204 is repeated with the switched estimation map information. In this way, the robot 1 moves autonomously within the vehicle 50.

  Thus, when the vehicle 50 is a moving environment, the sensing environment changes as the robot 1 moves forward. Therefore, the estimation map information acquired at different heights is switched according to the self-position of the robot 1. Then, by switching the estimation map information, the self position can be estimated with high accuracy. In the above description, the estimation map information is created based on the measurement results on the measurement surfaces having different heights, but may be created based on the measurement results on the measurement surfaces having different tilt angles. For example, the tilt angle of the measurement surface may be adjusted to the angle when the slope 54 is raised. Alternatively, the estimation map information may be created based on measurement results on measurement surfaces having different heights and tilt angles.

  In the present embodiment, optimal map information for estimation can be selected according to the self position. Thereby, the accuracy of self-position estimation can be improved. Therefore, it is possible to move safely and quickly in the well cab vehicle.

Embodiment 3 FIG.
A self-position estimation method according to the third embodiment will be described with reference to FIGS. 12 and 13. FIG. 12 is a diagram illustrating a state in which the robot 1 climbs the slope 54. FIG. 13 is a diagram for explaining estimation map information used in the self-position estimation method according to the present embodiment.

  As shown in FIG. 12, when the robot 1 climbs the slope 54, the robot 1 shakes due to the bending of the slope 54 or the like. When the robot 1 shakes, the elevation angle of the measurement surface of the laser sensor 13 changes. For example, the laser sensor 13 measures on the measurement surface E or measures on the measurement surface F. In particular, the robot 1 shakes due to insufficient strength of the slope 54 or unevenness of the floor 51. When this shaking occurs, the obstacle 53 cannot be detected and the ceiling 55 is detected. That is, the laser beam from the laser sensor 13 is irradiated to the ceiling 55 without being irradiated to the obstacle 53. Therefore, the accuracy of self-position estimation is reduced.

  Therefore, in this embodiment, the estimation map information is created from the measurement results measured on the measurement surfaces with different angles. Here, in order to reduce the influence of the fluctuation in the pitch direction, optimal map information is selected from a plurality of estimation map information.

  For example, as shown in FIG. 13, laser measurement is performed by changing the tilt angle of the laser sensor 13 by the angle changing mechanism 15. Then, map information for estimation is created based on the distance image data in each measurement. Specifically, laser measurement is performed at the same measurement point while changing the tilt angle. Map information for estimation at different angles is created from each measurement result. For example, laser measurement is performed every few degrees. Specifically, laser measurement is performed at −10 °, −5 °, 0 °, + 5 °, and + 10 °. Furthermore, laser measurement is performed by changing the measurement point and changing the tilt angle at each measurement point. In this case, the map information storage unit 22 stores the map information for estimation by associating the position and the angle. That is, the estimation map information is stored in the map information storage unit 22 together with the measurement position tag and the angle tag. The self-position estimation unit 23 stores a map information file for estimation for each measurement position tag and each angle tag. If the measurement position or angle is different, the map information storage unit 22 stores the estimation map information as a different file.

  Then, the estimation map information file is read from the map information storage unit 22 and self-position estimation is performed. When the robot 1 enters the vehicle 50, the angles of the laser sensors 13 and 14 with respect to the horizontal plane are obtained by the output from the gyro sensor 17. Then, the self-position estimation unit 23 selects estimation map information corresponding to the angle of the measurement surface of the laser sensors 13 and 14. Then, self-position estimation is performed using the selected estimation map information. That is, the self-position estimation unit 23 selects estimation map information having an angle tag that is the same as or close to the angle of the measurement surface. Further, the self-position estimation unit 23 selects the estimation map information having the measurement position tag that is the same as or close to the position of the robot 1. As described above, the self-position estimation unit 23 selects the optimum map information for estimation according to the measurement point position and the measurement surface angle when measured. By doing so, the accuracy of self-position estimation can be improved.

Furthermore, in the present embodiment, the following processing is performed in order to prevent a reduction in processing speed. First, the self-position estimation unit 23 extracts a plurality of estimation map information that may be read out from the map information storage unit 22 in advance. Specifically, estimation map information close to the estimated self-location is extracted. Then, the self-position estimating unit 23 moves the extracted plurality of pieces of estimated map information to the buffer area 27 with an angle tag (FIG. 14). In FIG. 14, the estimation map information M ₃ to the estimation map information M ₇ are extracted from the map information storage unit 22 and stored in the buffer area 27. At the moment when the gyro sensor 17 is at a predetermined angle, one piece of estimation map information is moved to the calculation area. That is, the map information selection unit 26 selects map information for estimation having an angle tag that matches the angle of the measurement surface detected by the gyro sensor 17. By performing this process, a delay in the self-position estimation process can be avoided. Note that the position tag may be used to determine the possibility of reading. That is, the map information for estimation of the position tag closest to the current estimated self-position may be stored in the buffer area 27.

  Next, the self-position estimation method according to the present embodiment will be described with reference to FIG. FIG. 15 is a flowchart showing the self-position estimation method according to the present embodiment. Note that the self-position estimation method according to the present embodiment has a basic flow similar to that of the first and second embodiments, and thus description thereof will be omitted as appropriate.

  As in the second embodiment, a person moves to near the slope (step S301) and performs laser self-position estimation (step S302). When self-position estimation is completed (YES in step S303), the slope 54 and the inside of the vehicle 50 are moved while self-position estimation is performed with a laser (step S304). The processing up to this point is the same as the processing from step S201 to step S204 shown in the second embodiment, and a description thereof will be omitted.

  Then, in conjunction with the gyro sensor 17, a process of dynamically switching the map is performed (step S305). In step S 305, optimal angle tag estimation map information is selected from the estimation map information in the buffer area 27 in accordance with the measurement result of the gyro sensor 17. Thereafter, it is determined whether the connection position has been reached (step S306). The processing from step S306 to step S309 is the same as the processing from step S205 to step S208 shown in the second embodiment, and a description thereof will be omitted. Then, as in the second embodiment, when the estimation map information for self-position estimation is switched according to the self-position (step S209), the estimation map information with different measurement angles is developed in the buffer area 27 at the same position. To do. And it returns to step S304 and moves the inside of the slope 54 and the vehicle 50, estimating a laser self-position.

  Next, the process of dynamically switching the estimation map information in conjunction with the output of the gyro sensor will be described with reference to FIG. FIG. 16 is a flowchart showing this processing. First, the current robot angle is acquired from the gyro sensor 17 (step S401). That is, the angle of the main body 10 of the robot 1 is measured. Thereby, the angle of the measurement surface of the laser sensors 13 and 14 can be calculated | required. Here, the angle of the measurement surface is 0 ° on the horizontal plane. Then, it is determined whether or not the angle estimation map information acquired by the gyro sensor 17 exists in the buffer area 27 (step S402). That is, it is determined whether or not there is an angle tag that matches the measurement surface with reference to the angle tags of the plurality of estimation map information extracted in advance in the buffer area 27.

  When the map information for estimation of the matching angle tag exists (YES in step S402), the map information for estimation linked to the angle is extracted (step S403). Thereby, one piece of estimation map information moves to the calculation area. Then, self-position estimation is performed using the extracted map information for estimation and laser data (step S404). That is, the self-position is estimated by pattern matching between the distance image data measured by the laser sensors 13 and 14 and the estimation map information. On the other hand, if there is no estimation map information for the matching angle tag, self-position estimation is performed using the initially set estimation map information (step S405). By doing in this way, since self-position estimation can be carried out with the pattern of the same angle, estimation accuracy can be improved.

  In the present embodiment, optimal map information for estimation can be selected according to the measurement surface angle or the self-position. Thereby, the accuracy of self-position estimation can be improved. Therefore, it is possible to move safely and quickly in the well cab vehicle.

Embodiment 4 FIG.
In the present embodiment, an area where the position of the obstacle can be dynamically changed is provided in a partial area of the estimation map information. Then, the position of the obstacle in the dynamic region is determined based on the result of the laser measurement. For example, in the vehicle 50, the position and posture of the seat may change. In such a case, the position and orientation of the sheet differ from the information of the presumed map information stored in advance. In the present embodiment, there is provided a method for accurately estimating the self position by absorbing the difference in the position and posture of an obstacle such as a seat.

For example, as shown in FIG. 17, the regions R ₁ and R ₂ are dynamic regions in which the position of the obstacle changes. On the other hand, the regions R ₃ and R ₄ are static regions where the position of the obstacle does not change. Specifically, the movable range of the passenger seat and a region R _1, to set the movable range of the second seat as the region R _2. Since the positions and postures of these sheets change, the distance to the obstacle may change even when the robot 1 exists at the same position. Regions R ₃ and R ₄ correspond to, for example, the inner wall of the body of the vehicle 50.

The map information storage unit 22 stores in advance map information for estimation in which the regions R ₁ and R ₂ are dynamic regions. It is assumed that no obstacle exists in the dynamic area. That is, map information for estimation in which an obstacle exists in the moving environment excluding the areas R ₁ and R ₂ is created. Then, based on the first measurement results by the laser sensors 13 and 14, measures the position of the obstacle in the region _R 1, _{R 2.} Then, the estimation map information stored in advance, to synthesize the range image data of the obstacle in the region R _1, R _2. That is, the map information creation unit 21 cuts out a part of the distance image data and merges it with the map information for estimation already stored in the map information storage unit 22. Thereby, the estimation map information in which the coordinates of the obstacle in the dynamic region are fixed can be created.

Specifically, as shown in FIG. 18, estimation map information M ₈ to region R _1, R ₂ obstacle is no dynamic region is stored in the map information storage unit 22. The self-position estimating portion 23 is the range image data M ₉ of the laser measuring the first, matching the estimated map information M _8. Thereby, the self position in the estimation map information is estimated. Then, based on the estimated self-position, the distance image data of the obstacles in the regions R ₁ and R ₂ is extracted. That is, the distance image data of the regions R ₁ and R ₂ are extracted from the distance image data. Therefore, depending on the distance of the deployed obstacle region R1, R _2, the distance image data of the region R _1, R ₂ are extracted. Then, the distance image data of the regions R ₁ and R ₂ is combined with the estimation map information M ₈ . Thus, the estimation map information M ₁₀ a distance data of an obstacle in the dynamic region is established is created. Thereafter, a self position estimated by using the estimation map information M _10. That is, the estimation map information M ₈ on the basis of the R _1, R ₂ of the distance image data M ₉ is estimation map information M ₁₀ which are merged, the self-position is estimated. Thereby, even in a moving environment where there are movable obstacles, the accuracy of self-position estimation can be improved.

  Next, the self-position estimation method according to the present embodiment will be described with reference to FIG. FIG. 19 is a flowchart showing a self-position estimation method. Further, in the self-position estimation method shown in FIG. 19, the self-position estimation method shown in the second embodiment is used. That is, the estimation map information is switched according to the self position. Therefore, the description of the contents described in Embodiments 1 and 2 will be omitted as appropriate. Specifically, steps S501 to S503 shown in FIG. 19 correspond to steps S201 to S203 in FIG.

Here, in step S502, self-position estimation is performed using the estimation map information excluding obstacles in the dynamic region. Therefore, if self-position estimation is completed (YES in step S503), a seat position map is dynamically created (step S504). That is, the distance image data of the areas R ₁ and R ₂ are extracted from the distance image data M ₉ used for the self-position estimation and synthesized with the estimation map information M ₈ excluding the obstacles in the dynamic area. As a result, the estimation map information M ₁₀ is created. In the estimation map information ₁₀ , the position of the obstacle in the areas R ₁ and R ₂ is fixed.

  And it moves on the slope 54 or the inside of the vehicle 50, estimating a self-position by the measurement result of the laser sensors 13 and 14 (step S505). Subsequent steps S506 to S509 are the same as steps S205 to S208 shown in the second embodiment, and thus description thereof is omitted.

  In this embodiment, map information for estimation is created from the result of laser measurement in step S504. Here, all the estimation map information used for switching is created from a single laser measurement result, but a part of the estimation map information may be created. In this case, the estimation map information is created from another laser measurement result.

  Next, creation of estimation map information to which obstacle information of a dynamic area is added will be described with reference to FIG. FIG. 20 is a flowchart illustrating a method of creating estimation map information. First, at the stage of step S504, matching processing between preliminarily created map information for estimation and laser data is performed (step S601). That is, the distance image data obtained by laser measurement is matched with map information for estimation excluding obstacles in the dynamic area.

  Next, laser data in the dynamic region is calculated (step S602). Here, for example, the coordinates of the obstacle in the dynamic region are calculated. Then, it is determined whether obstacle data exists in the dynamic area (step S603). If there is obstacle data in the dynamic area (YES in step S603), the target obstacle data is cut out from the memory (step S604). That is, the coordinates of the obstacle in the dynamic area are extracted. Then, the data extracted from the memory is overwritten on the estimation map information created in advance. Thereby, the information of the obstacle in the dynamic area is added to the estimation map information. Then, the self-position estimation shown in FIG. 19 is performed using the estimation map information to which the obstacle information is added. On the other hand, when there is no obstacle data in the dynamic area (NO in step S603), the merge process is not performed and the process proceeds to the next process (S606). That is, the estimation map information used for estimation in step S601 is used as it is.

  In this way, a region where the position / posture of the obstacle dynamically changes is defined as a dynamic region. In the dynamic area, the laser data of the obstacle is combined with the estimation map information. In this way, it is possible to use the estimation map information having accurate obstacle information. Therefore, self-position estimation can be performed with high accuracy. In the above description, the dynamic region is the movable range of the seat, but the movable range of other obstacles may be the dynamic region. Of course, the maximum range in which the obstacle can move may be set as the dynamic region.

  Note that Embodiments 1 to 4 described above may be implemented in appropriate combination or separately. For example, map information is created using the estimation map information created in the first embodiment, and self-position estimation is performed using the estimation map information according to the second, third, and fourth embodiments. Furthermore, two or more self-position estimation methods in the second, third, and fourth embodiments may be combined and may be used separately.

  The self-position estimation accuracy can be improved by the methods shown in the first to fourth embodiments. For example, in the case of a moving environment with large shaking, a moving environment with uneven floors, a moving environment where a slope is climbed, disturbance is likely to occur, but the robustness can be improved by the above method. Further, it is not necessary to use a special device or sensor for the mobile environment itself. As described above, the above method is suitable for a moving environment with a large disturbance such as entering a well cab vehicle. Furthermore, since it is not necessary to use a sensor that measures three dimensions, the size can be reduced. That is, the robot 1 can be reduced in size and weight by using a sensor that measures a two-dimensional measurement surface. Furthermore, since two-dimensional information is used, the calculation time can be shortened.

DESCRIPTION OF SYMBOLS 1 Robot 10 Main body 11 Front wheel 12 Rear wheel 13 Laser sensor 14 Laser sensor 15 Angle change mechanism 16 Control part 17 Gyro sensor 21 Map information preparation part 22 Map information storage part 23 Self-position estimation part 24 Drive control part 25 Tilt angle control part 26 Map information selection unit 27 Buffer area 50 Vehicle 51 Floor 52 Obstacle 53 Obstacle 54 Slope 55 Ceiling 61 3D position measuring device 91 Composite area

Claims (18)

An autonomous mobile that moves within the mobile environment by estimating its own position based on estimation map information in the mobile environment,
A sensor unit that measures distance data according to the distance to surrounding obstacles;
A map information storage unit that stores a plurality of two-dimensional estimation map information according to the arrangement of obstacles in the moving environment, and stores map information for estimation on different measurement surfaces of the moving environment And
A map information selection unit that selects one estimation map information from a plurality of estimation map information stored in the map information storage unit based on the self-position or the measurement angle of the sensor unit;
An autonomous mobile body comprising: a self-position estimating unit that estimates a self-position based on map information for estimation selected by the map information selecting unit and distance data measured by the sensor unit. An angle sensor for measuring a measurement angle of the sensor unit;
The map information storage unit stores map information for estimation of measurement surfaces at different angles,
The autonomous mobile body according to claim 1, wherein the position information selection unit selects the estimation map information according to a measurement result of the angle sensor.   The autonomous mobile body according to claim 1 or 2, wherein when the self-position estimated by the self-position estimation unit exceeds a predetermined position, the map information for estimation selected by the map information selection unit is switched.   The autonomous mobile body according to claim 3, wherein a position for switching the estimation map information is preset according to the position of an obstacle in the moving environment.   The moving body according to claim 3 or 4, wherein the movement of the autonomous moving body stops when the map information for estimation is switched.   The autonomous mobile body according to any one of claims 1 to 5, wherein the self-position is estimated by pattern-matching distance image data corresponding to the distance data and the estimation map information. An autonomous mobile that moves within the mobile environment by estimating its own position based on estimation map information in the mobile environment,
A sensor unit that measures the distance to surrounding obstacles;
A map information storage unit for storing two-dimensional map information for estimation according to the arrangement of obstacles in the moving environment;
A map information creation unit that cuts out distance image data based on the distance measured by the sensor unit during the movement of the autonomous mobile body and synthesizes it with the estimation map information;
An autonomous mobile body comprising: a self-position estimation unit that estimates a self-position based on the estimation map information obtained by combining the distance image data and the measurement result of the sensor unit after the combination. The distance image data is cut out corresponding to a partial area of the moving environment,
The autonomous mobile body according to claim 7, wherein the partial area of the moving environment is set based on a movable range of the obstacle. A map information creation system for creating estimation map information used for self-position estimation in a mobile environment,
A sensor unit for measuring a distance to an obstacle in the moving environment on a predetermined measurement surface;
A position measuring unit for measuring the position of the sensor unit in the moving environment;
An angle changing mechanism for changing the angle of the measurement surface of the sensor unit;
A map information creation system comprising: a map information creation unit that creates the map information for estimation by synthesizing distance image data corresponding to the distance measured by the sensor unit on measurement surfaces at different angles. A self-position estimation method in a moving body that estimates self-position based on map information for estimation in a moving environment,
Measuring distance data according to the distance to surrounding obstacles;
One estimation map based on the position or orientation of the moving object from a map information storage unit storing two-dimensional estimation map information according to the arrangement of obstacles in the moving environment for each different measurement surface Selecting information,
A self-position estimation method comprising: estimating a self-position based on the selected map information for estimation and the distance data.
A self-position estimation method comprising: estimating a self-position based on Further comprising measuring a measurement angle of a sensor unit for measuring the distance data;
The map information storage unit stores map information for estimation of measurement surfaces at different angles,
The self-position estimation method according to claim 10, wherein the estimation map information is selected according to the measurement angle.   12. The self-position estimation method according to claim 10, wherein when the estimated self-position exceeds a predetermined position, switching to another estimation map information is performed.   The self-position estimation method according to claim 12, wherein a position for switching the estimation map information is preset according to the position of an obstacle in the moving environment.   14. The self-position estimation method according to claim 12, wherein the movement of the autonomous mobile body stops when the estimation map information is switched.   The self-position estimation method according to any one of claims 10 to 14, wherein the self-position is estimated by pattern matching the distance image data corresponding to the distance data and the estimation map information. A self-position estimation method in a moving body that estimates self-position based on map information for estimation in a moving environment,
Measuring the distance to surrounding obstacles;
Cutting out the distance image data based on the distance measured by the sensor unit during the movement of the autonomous mobile body and combining it with the estimation map information;
A self-position estimation method comprising: estimating the self-position based on the map information for estimation combined with the distance image data and the measurement result of the sensor unit after combining. The distance image data is cut out corresponding to a partial area of the moving environment,
The self-position estimation method according to claim 16, wherein the partial area of the moving environment is set based on a movable range of the obstacle. A map information creation method for creating map information for estimation used for self-location estimation in a mobile environment,
Measuring a distance to an obstacle in the moving environment on a predetermined measurement surface;
Measuring the position of the sensor unit in the mobile environment;
Changing the angle of the measurement surface of the sensor unit;
A map information creating method comprising: synthesizing distance image data corresponding to the distance measured by the sensor unit on measurement surfaces of different angles to create map information for estimation.

Images