Autonomous Vehicles

Abstract

To track a tracking target while suppressing entry into an entry prohibited area.SOLUTION: The autonomous mobile body includes a vehicle, a distance measuring unit mounted on the vehicle, and a control device mounted on the vehicle. A storage unit of the control device stores map information including an environment map and information on an entry prohibited area. The control device uses an estimated local position and the map information stored in the storage unit to calculate a distance and azimuth from the local position to the entry prohibited area. The control device processes the measured value by replacing a measurement value, associated with that azimuth which is determined not to contain a measurement point between the vehicle and the entry prohibited area, with a calculated distance that is the distance calculated using the local position and the map information. The control device derives a movable range from measured values that have not been processed, and the calculated distance.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to an autonomous moving body that tracks a tracking target. [Background technology]

[0002] An autonomous mobile body that recognizes the surrounding environment by a distance measurement unit for detecting the direction and distance to a measurement point and moves autonomously is described in Patent Document 1. The autonomous mobile body described in Patent Document 1 includes a distance measurement unit, a storage unit in which an environmental map is stored, a self-position estimation unit that estimates its own position, a route generation unit, and a control device. In the environmental map, objects that can be detected by the distance measurement unit and objects that cannot be detected by the distance measurement unit are registered. The self-position estimation unit estimates its own position from the detection result of the distance measurement unit. The route generation unit grasps the position of the object from the environmental map and generates a route from the current position to the destination so as to avoid the object. That is, the position where the object is located is set as a no-entry area, and a route is generated so that the autonomous mobile body does not enter the no-entry area. The control device moves the autonomous mobile body according to the generated route. By registering an object that cannot be detected by the distance measurement unit in the environmental map and generating a route to avoid this object, the autonomous mobile body can move while avoiding the object that cannot be detected by the distance measurement unit. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2015-1820 A Summary of the Invention [Problem to be solved by the invention]

[0004] Incidentally, some autonomous mobile bodies track a tracking target. This type of autonomous mobile body includes a distance measurement unit, an object detection unit that detects the tracking target and obstacles from the detection result of the distance measurement unit, and a mobile body control unit that moves the autonomous mobile body so as to track the tracking target. Since the tracking target moves, a destination cannot be set. For this reason, the autonomous mobile body that tracks the tracking target tracks the tracking target within a movable range in which no obstacles are detected, without performing route generation. Therefore, the range in which the autonomous mobile body moves depends on the detection result of the distance measurement unit, and if the distance measurement unit cannot detect a no-entry area, there is a risk that the autonomous mobile body will enter the no-entry area.

[0005] An object of the present invention is to provide an autonomous moving body that can track a tracking target while preventing entry into a no-entry area. [Means for solving the problem]

[0006] The autonomous moving body that solves the above problem is an autonomous moving body that tracks a tracking target, and includes a moving body, a distance measurement unit for measuring a distance and a direction from the moving body to a measurement point, an object detection unit for detecting objects including the tracking target and an obstacle, a moving body control unit that moves the moving body so as to track the tracking target, a memory unit that stores map information including an environmental map and information about no-entry areas, a self-position estimation unit that estimates a self-position of the moving body on the environmental map, and a distance calculation unit that calculates a distance from the self-position to the no-entry area using the self-position estimated by the self-position estimation unit and the information about the no-entry area stored in the memory unit. The system includes a distance calculation unit that calculates distances for each of a plurality of orientations in the horizontal direction from the moving body's own position to the no-entry area, a processing unit that processes the measurement values ​​of the distance measurement unit that are associated with the same orientation as the orientation from the moving body to the no-entry area, the measurement values ​​that are associated with an orientation in which the measurement point does not exist between the moving body and the no-entry area, into a calculated distance calculated by the distance calculation unit, and a movable range derivation unit that derives the movable range of the moving body from the measurement values ​​of the distance measurement unit that have not been processed by the processing unit and the calculated distance, and the moving body control unit moves the moving body within the movable range.

[0007] The calculated distance is the distance to the no-entry area calculated from the self-position and map information. Therefore, the calculated distance can be said to be the distance measured by the distance measurement unit on the assumption that the no-entry area can be detected by the distance measurement unit. By processing the measurement value of the distance measurement unit into a calculated distance, the movable range derivation unit can recognize the no-entry area from the calculated distance. Therefore, the part excluding the no-entry area can be set as the movable range. Therefore, it is possible to track the tracking target while preventing the moving body from entering the no-entry area.

[0008] For the above-mentioned autonomous moving body, the self-position estimation unit may estimate the self-position using the measurement values ​​of the distance measurement unit that have not been processed by the processing unit, and the measurement values ​​before processing by the processing unit.

[0009] This makes it possible to improve the accuracy of estimating the self-location. For the above-mentioned autonomous moving body, the object detection unit may detect the tracking target using the measurement values ​​of the distance measurement unit that have not been processed by the processing unit, and the measurement values ​​before being processed by the processing unit.

[0010] This makes it possible to recognize the tracking target even if the tracking target enters a no-entry area. The autonomous moving body may further include a notification unit that issues a notification when the self-position estimated by the self-position estimation unit is located within the no-entry area.

[0011] When the moving object is located within a prohibited area, the moving object may be unable to move. By providing a notification unit, the moving object can notify the tracking target that the moving object is located within the prohibited area.

[0012] For the above-mentioned autonomous moving body, when the self-position estimated by the self-position estimation unit is located within the no-entry area, the movable range derivation unit may derive the movable range without using the calculated distance.

[0013] According to this, the forbidden area can be recognized as a movable range, and therefore, when the mobile object is located within the forbidden area, the mobile object can be moved out of the forbidden area. Effect of the Invention

[0014] According to the present invention, it is possible to track a tracking target while preventing entry into a no-entry area. [Brief description of the drawings]

[0015] [Figure 1] FIG. 2 is a schematic diagram showing an autonomous moving body and a tracking target. [Diagram 2] FIG. [Diagram 3] FIG. 4 is a diagram showing a schematic diagram of a detection range of a distance measuring unit. [Figure 4] FIG. 2 is a diagram showing a schematic diagram of an environmental map and a no-entry area. [Diagram 5] 4 is a flowchart showing a process performed by a control device. [Figure 6] FIG. 4 is a diagram for explaining a process performed by a control device. [Figure 7] FIG. 4 is a diagram for explaining a process performed by a control device. [Figure 8] 1 is a schematic diagram showing a state in which a tracking target has entered a no-entry area. [Figure 9] FIG. 13 is a schematic diagram showing a state in which the vehicle's own position is located within a no-entry area. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] An embodiment of the autonomous moving body will be described below. 1 and 2, the autonomous moving body 10 includes a vehicle 20, a distance measurement unit 31 mounted on the vehicle 20, a control device 32 mounted on the vehicle 20, and a notification unit 35. The vehicle 20 includes a body 21, a plurality of wheels 22, and a drive mechanism 23 for driving the vehicle 20. The vehicle 20 is a moving body that moves autonomously so as to track a registered tracking target T by being controlled by the control device 32. The vehicle 20 performs tracking so that a distance L from the vehicle 20 to the tracking target T falls within a predetermined range.

[0017] 2, the drive mechanism 23 includes a motor 24 for rotating the wheels 22, a motor driver 25 for driving the motor 24, a travel control device 26, and an encoder 29. Although not shown in the figure, the number of motors 24 and motor drivers 25 provided is the same as the number of wheels 22. The encoder 29 is provided individually for each motor 24.

[0018] The cruise control device 26 includes a CPU 27 and a storage unit 28 in which programs for performing various controls are stored. Commands are input to the cruise control device 26 from the control device 32. The cruise control device 26 calculates a command rotation speed based on a speed command value which is a command for commanding the speed of the vehicle 20 and a traveling direction command which is a command for commanding the traveling direction of the vehicle 20.

[0019] The encoder 29 is, for example, an incremental type encoder that outputs a pulse signal based on the amount of rotation of the rotary shaft of the motor 24. The encoder 29 detects the number of rotations of the rotary shaft of the motor 24. The motor driver 25 performs feedback control so that the number of rotations of the motor 24 matches the command number of rotations. As a result, the vehicle 20 moves in the traveling direction according to the traveling direction command at a speed according to the speed command value.

[0020] The notification unit 35 is a member that issues a notification to the tracking target T. Examples of the notification unit 35 include a unit that issues a notification by sound and a unit that issues a notification by light. Next, the distance measurement unit 31 and the control device 32 will be described in detail.

[0021] The distance measurement unit 31 can cause the control device 32 to recognize objects present around the vehicle 20 and can measure the distance from the vehicle 20 to the objects. The objects include a tracking target T and an obstacle.

[0022] In this embodiment, a LIDAR (Laser Imaging Detection and Ranging) is used as the distance measurement unit 31. The LIDAR is a range finder that can recognize the surrounding environment by irradiating a laser to the surroundings and receiving the reflected light reflected from the part where the laser hits. As the LIDAR, a two-dimensional range finder that irradiates a laser while changing the irradiation angle in the horizontal direction is used.

[0023] As shown in Fig. 3, the laser irradiation range of the distance measurement unit 31 is, for example, a range of the horizontal irradiation possible angle θ1 = 270° and the irradiation possible distance d = 30 [m]. If the center of the irradiation possible angle θ1 is taken as the reference axis, the laser irradiation range is a range of the reference axis ± 135°. In other words, the laser irradiation range of the distance measurement unit 31 is a sector with a radius of 30 [m] and a central angle of 270°. This laser irradiation range becomes the detection range SA in which an object can be detected.

[0024] If the part where the laser hits is the measurement point, the LIDAR measures the distance to the measurement point in association with the irradiation angle. The irradiation angle is an angle at a fixed angle with respect to the horizontal direction, and is an angle according to the angular resolution θ2. The distance measurement unit 31 outputs the measurement value at fixed angles. In detail, when the LIDAR can receive the reflected light, it outputs detection information in which the measurement value of the distance to the measurement point is associated with the irradiation angle, and when the LIDAR cannot receive the reflected light, it outputs non-detection information in association with the irradiation angle. Therefore, the detection result of the LIDAR is information in which the irradiation angle and the measurement value are associated. The measurement value includes the measured distance and non-detection information indicating that the distance could not be detected. When an object exists around the vehicle 20, the laser hits the object and the distance to the measurement point is output in association with the irradiation angle. Therefore, the control device 32 can determine whether or not an object exists around the vehicle 20 from the detection result of the LIDAR, and when the object exists, it can grasp the distance to the object and the irradiation angle. If the angular resolution θ2 of the irradiation angle is 1.0°, 271 detection results are obtained in one period. The laser irradiation angle represents the direction from the vehicle 20. For example, if the reference axis faces the front of the vehicle 20, the irradiation angle on the clockwise side of the reference axis is the right side of the vehicle 20, and the irradiation angle on the counterclockwise side of the reference axis is the left side of the vehicle 20. Therefore, the distance measurement unit 31 measures the distance for each direction according to the angular resolution θ2.

[0025] The control device 32 includes a CPU 33 and a storage unit 34 including a RAM and a ROM. Various programs for controlling the vehicle 20 are stored in the storage unit 34. The control device 32 may include dedicated hardware for performing at least some of the various processes, such as an application specific integrated circuit (ASIC). The control device 32 may be configured as a circuit including one or more processors operating according to a computer program, one or more dedicated hardware circuits such as an ASIC, or a combination thereof. The processor includes a CPU and memory such as a RAM and a ROM. The memory stores program code or instructions configured to cause the CPU to execute processes. The memory, i.e., a computer-readable medium, includes anything that can be accessed by a general-purpose or special-purpose computer.

[0026] The control device 32 outputs various commands to the driving control device 26. This controls the rotation speed and the rotation direction of the wheels 22, and the vehicle 20 tracks the tracking target T. It can be said that the control device 32 controls the drive mechanism 23 to autonomously move the vehicle 20.

[0027] Map information is stored in the storage unit 34 of the control device 32. As shown in Fig. 4, the map information is information indicating an environmental map M and a no-entry area A. The environmental map M is information regarding the physical structure of the surrounding environment of the autonomous moving body 10, such as the shape and size of the environment in which the autonomous moving body 10 is used. The no-entry area A is a place where the autonomous moving body 10 is prohibited from entering, and is set within the environmental map M.

[0028] The environmental map M may be stored in advance in the storage unit 34 as long as the surrounding environment in which the autonomous moving body 10 is used can be grasped in advance. When the environmental map M is stored in advance in the storage unit 34, the coordinates of objects whose positions are unlikely to change, such as the walls W and pillars Pi of a building, are stored as the environmental map M.

[0029] The environmental map M may be created by mapping using Simultaneous Localization and Mapping (SLAM). The mapping is performed, for example, by creating a local map from the coordinates of the measurement points obtained by the distance measurement unit 31 and combining the local map according to the self-position.

[0030] The no-entry area A is a place that cannot be detected by the distance measurement unit 31 or a place where it is difficult to detect by the distance measurement unit 31. Examples of places that cannot be detected by the distance measurement unit 31 include a place where members made of a material with high laser transmittance, such as colorless transparent glass, a place where members are placed at a height where the irradiated laser does not hit, and a place where the area is divided by the color of the floor, such as a walkway in a factory. In addition, any place without characteristics, such as a place where no members are placed or a place where the floor is not divided by color, can also be set as the no-entry area A. Examples of places where it is difficult to detect by the distance measurement unit 31 include a place where members are placed whose partial shape can be detected but whose overall horizontal shape is difficult to recognize, such as a shelf where a shelf board on which a load is placed is supported by a support.

[0031] The information indicating the forbidden area A is the coordinates of the forbidden area A on the environmental map M. The forbidden area A may be stored in advance in the storage unit 34 together with the environmental map M. In addition, the forbidden area A may be stored in the storage unit 34 by an external device after mapping is performed by the control device 32.

[0032] Next, a description will be given of the control performed by the control device 32 when tracking the tracking target T. As an example, a case will be described in which the autonomous moving body 10 is used in an environment shown by the map information in FIG.

[0033] 5 and 6, in step S1, the control device 32 acquires the detection result of the distance measurement unit 31, that is, the measurement values ​​of the direction and distance to the measurement point P1 caused by the surrounding environment of the autonomous moving body 10, from the distance measurement unit 31. As can be seen from Fig. 6, the measurement point P1 is generated by the laser hitting the wall W, the pillar Pi, and the tracking target T. The control device 32 can obtain the relative coordinates of the measurement point P1 with respect to the vehicle 20 from the detection result of the distance measurement unit 31.

[0034] Next, in step S2, the control device 32 performs self-location estimation to estimate the position of the vehicle 20 on the environmental map M. The self-location estimation is performed by integrating, by a Bayes filter, odometry to estimate the self-movement amount of the vehicle 20 using the detection result of the encoder 29 and a matching result between the characteristic points (= landmarks) around the vehicle 20 obtained from the detection result of the distance measurement unit 31 and the environmental map M. That is, the control device 32 performs probabilistic self-location estimation to estimate the self-location by correcting the self-location obtained by odometry with the relative position with the landmark. The landmark is a part whose shape changes among the shapes of the surrounding environment that can be grasped from the measurement point P1. The self-location is a coordinate indicating one point of the vehicle body 21, for example, the coordinate of the center of the vehicle body 21 in the horizontal direction. The control device 32 functions as a self-location estimation unit.

[0035] Next, as shown in FIG. 5 and FIG. 7, in step S3, the control device 32 calculates the distance and direction from the self-position to the forbidden area A using the self-position estimated in step S2 and the map information stored in the storage unit 34. The control device 32 calculates the distance to the forbidden area A for each of a plurality of directions in the horizontal direction according to the angular resolution θ2 of the distance measurement unit 31. Therefore, when the angular resolution θ2 of the distance measurement unit 31 is 1.0°, the distance calculated by the control device 32 is also associated with each direction at 1.0° intervals. In other words, the control device 32 calculates the distance to the virtual measurement point P2 that occurs when the laser hits the forbidden area A for each direction. The control device 32 functions as a distance calculation unit.

[0036] Next, in step S4, the control device 32 processes the measurement value of the distance measurement unit 31. First, the control device 32 determines whether or not a measurement point P1 exists between the vehicle 20 and the forbidden area A for each direction from the vehicle 20 to the forbidden area A. If a measurement point P1 exists between the vehicle 20 and the forbidden area A, it can be said that an object exists between the vehicle 20 and the forbidden area A. On the other hand, if a measurement point P1 does not exist between the vehicle 20 and the forbidden area A, it can be said that no object exists between the vehicle 20 and the forbidden area A.

[0037] The control device 32 replaces the measurement value associated with the direction from the vehicle 20 to the forbidden area A, which is determined to be not present between the vehicle 20 and the forbidden area A, with the calculated distance, which is the distance calculated in step S3. In detail, the control device 32 processes the measurement value of the distance measurement unit 31 by replacing the measurement value of the distance measurement unit 31 with the calculated distance associated with the direction and the same direction to which the measurement value is associated. It can be said that the control device 32 replaces the measurement value longer than the calculated distance and non-detection information, which are associated with the direction and the same direction from the vehicle 20 to the forbidden area A, with the calculated distance. The calculated distance obtained by processing the measurement value is the distance to the virtual measurement point P2. The control device 32 can obtain the relative coordinates of the virtual measurement point P2 based on the self-position from the direction and the calculated distance associated with the direction. It should be noted that the measurement value is not processed for the measurement value associated with the direction different from the direction from the self-position to the forbidden area A, which is among the measurement values ​​of the distance measurement unit 31. Therefore, among the measurement values ​​of the distance measurement unit 31, the measurement values ​​associated with a direction different from the direction from the self-location to the forbidden area A are measurement values ​​that have not been processed. The control device 32 functions as a processing unit.

[0038] Next, in step S5, the control device 32 distinguishes between the tracking target T and an obstacle. First, the control device 32 detects an object by clustering the measurement points P1. The laser emitted from the distance measurement unit 31 is reflected when it hits an object, so that the measurement points P1 are interrupted at the boundary between the object and the space. Therefore, a set of the measurement points P1 represents a part of the contour of the object, and the object can be detected from the point cloud in which the measurement points P1 are clustered. In the example shown in FIG. 6, the wall W, the pillar Pi, and the tracking target T are each detected as an individual object by clustering the measurement points P1 generated by the wall W, the pillar Pi, and the tracking target T. The control device 32 can calculate the width of the object from the relative coordinates of the measurement points P1 constituting the clustered point cloud.

[0039] Information about the tracking target T is registered in the control device 32, and an object that matches this information is recognized as the tracking target T. In this embodiment, the tracking target T is a person. When a person is the tracking target T, the information about the tracking target T is, for example, the foot width. The tracking target T is registered by a person standing upright in front of the distance measurement unit 31, and operating an operation unit or an external device provided in the vehicle 20 to set the control device 32 to a registration mode.

[0040] Next, the control device 32 clusters the virtual measurement points P2, and regards the clustered points of the virtual measurement points P2 as objects. That is, the virtual measurement points P2 are also processed in the same manner as the measurement points P1. The control device 32 recognizes both the objects other than the tracking target T among the objects recognized by the measurement points P1 and the objects recognized by the virtual measurement points P2 as obstacles. It can be said that the control device 32 recognizes the tracking target T from the measurement points P1, and recognizes the obstacle from both the measurement points P1 and the virtual measurement points P2. That is, by processing the measurement values, the no-entry area A can be captured as a virtual obstacle. The control device 32 functions as an object detection unit.

[0041] Next, in step S6, the control device 32 derives the movable range MA. The control device 32 determines the movable range MA to be a range excluding the obstacle detected in step S5. In detail, the control device 32 controls the vehicle 20 so that the distance from the obstacle to the vehicle 20 is greater than a predetermined value, and therefore the movable range MA is determined by the obstacle. By recognizing the obstacle from the virtual measurement point P2, the no-entry area A is regarded as an obstacle. Therefore, the movable range MA is derived from the obstacle recognized by the measurement value not processed by the control device 32 and the obstacle recognized by the calculated distance. The control device 32 functions as a movable range derivation unit.

[0042] Next, in step S7, the control device 32 moves the vehicle 20 within the movable range MA to track the tracking target T. The control device 32 recognizes the distance L from the vehicle 20 to the tracking target T from the detection result of the distance measurement unit 31, and controls the vehicle 20 so that the distance L falls within a predetermined range. As shown in FIG. 8, when the tracking target T is located within the no-entry area A, the control device 32 moves the vehicle 20 within the movable range MA so that the distance to the tracking target T is the shortest. Note that since the tracking target T is detected using the measurement values ​​of the distance measurement unit 31 that have not been processed and the measurement values ​​before processing, it is possible to detect the tracking target T even if the tracking target T enters the no-entry area A. The control device 32 functions as a moving object control unit.

[0043] As shown in FIG. 9, when the self-position is located within the forbidden area A, the control device 32 notifies the tracking target T by making a notification using the notification unit 35. When the self-position is located within the forbidden area A, the control device 32 may mistakenly recognize that the surroundings are surrounded by obstacles due to the virtual measurement point P2, and may become unable to move. Therefore, by making a notification using the notification unit 35, the control device 32 notifies that the vehicle 20 is in an immovable state. The cause of the self-position being located within the forbidden area A is a decrease in the accuracy of the self-position estimation. The accuracy of the self-position estimation may decrease depending on the position of the vehicle 20, such as when a landmark cannot be detected. In this case, the control device 32 may mistakenly recognize the forbidden area A as the movable range MA, and the vehicle 20 may enter the forbidden area A. Then, when the correct self-position is estimated after the vehicle 20 enters the forbidden area A, the self-position may be located within the forbidden area A. In addition, depending on the braking capability of the vehicle 20, there may be cases where the vehicle 20 enters the no-entry area A in an attempt to track the tracking target T that has entered the no-entry area A.

[0044] The operation of this embodiment will be described. The control device 32 moves within the movable range MA, and thus prevents the vehicle 20 from entering the no-entry area A. The control device 32 can prevent the vehicle 20 from entering the no-entry area A by determining the relative position between the vehicle's own position and the no-entry area A from map information and processing the measurement value of the distance measurement unit 31.

[0045] In the autonomous mobile body 10 that tracks the tracking target T, the tracking target T and an obstacle are recognized from the measurement value of the distance measurement unit 31, and the tracking target T is tracked while suppressing contact with the obstacle. As in the present embodiment, the measurement value of the distance measurement unit 31 is processed and the no-entry area A is recognized as an obstacle, and by adding a logic for processing the measurement value of the distance measurement unit 31 to the conventional control logic, it is possible to suppress the entry of the vehicle 20 into the no-entry area A.

[0046] If the coordinates of the tracking target T are set as the destination and a route from the vehicle's own position to the destination is generated to prevent the vehicle 20 from entering the no-entry area A, the route needs to be generated every time the tracking target T moves. In this case, the time required for route generation is long, and the maximum speed of the vehicle 20 may be slow. In contrast, by preventing the vehicle 20 from entering the no-entry area A without route generation as in this embodiment, the maximum speed can be made faster than when route generation is performed.

[0047] In addition, in order to prevent the vehicle 20 from entering the no-entry area A, it is also possible to detect the no-entry area A using a member other than the distance measurement unit 31. For example, if the no-entry area A is a place where an object that is transparent to a laser is placed, the no-entry area A can be recognized by providing an ultrasonic sensor. If the no-entry area A is a place where an object that is not hit by a laser is placed, the no-entry area A can be recognized by providing a three-dimensional distance meter. If the no-entry area A is a place where the area is divided by the color of the floor surface, the no-entry area A can be recognized by providing a stereo camera. However, in either case, it is necessary to increase the number of members for detecting the no-entry area A, which leads to an increase in manufacturing costs. Furthermore, if a place without features is set as the no-entry area A, the no-entry area A cannot be detected. In response to this, the measured value of the distance measurement unit 31 is processed to prevent the vehicle 20 from entering the no-entry area A, thereby suppressing an increase in manufacturing costs. Furthermore, even if a non-characteristic location is designated as the no-entry area A, the vehicle 20 can be prevented from entering the no-entry area A.

[0048] The effects of this embodiment will be described. (1) The control device 32 calculates the distance from its own position to the no-entry area A for each direction, and processes the measurement value of the direction corresponding to the calculated distance into the calculated distance. This allows the control device 32 to recognize the no-entry area A as an obstacle, and therefore the area excluding the no-entry area A can be set as the movable range MA. Since it is possible to prevent the vehicle 20 from entering the no-entry area A without generating a route, it is possible to prevent the vehicle 20 from entering the no-entry area A while tracking the tracking target T.

[0049] (2) The control device 32 performs self-location estimation using the unprocessed measurement values ​​of the distance measurement unit 31 and the measurement values ​​before processing. That is, the self-location estimation is performed without using the calculated distance. Compared to the case where self-location estimation is performed using only odometry, the estimation accuracy of the self-location can be improved. Also, compared to the case where self-location is estimated using both the measurement value of the distance measurement unit 31 and the calculated distance, the accuracy of self-location estimation can be improved. If the self-location is estimated using both the measurement value of the distance measurement unit 31 and the calculated distance, the control device 32 may misidentify its own location if a topography similar to the shape of the no-entry area A that can be recognized from the virtual measurement point P2 exists in the environmental map M. In contrast, by estimating the self-location using only the measurement value of the distance measurement unit 31, the control device 32 is less likely to misidentify its own location, and the estimation accuracy of the self-location can be improved.

[0050] (3) The control device 32 detects the tracking target T using unprocessed measurement values ​​among the measurement values ​​of the distance measurement unit 31 and the measurement values ​​before processing. Since the control device 32 can detect the tracking target T even if the tracking target T enters the forbidden area A, the control device 32 is unlikely to lose sight of the tracking target T. Furthermore, even if the tracking target T is located within the forbidden area A, the distance to the tracking target T can be shortened within the movable range MA, improving tracking performance.

[0051] (4) When the vehicle 20 is located within the no-entry area A, a notification is issued by the notification unit 35. Therefore, when the vehicle 20 is immobilized, a notification can be issued to the tracking target T.

[0052] The embodiment can be modified as follows: The embodiment and the following modifications can be combined with each other to the extent that there is no technical contradiction. When the vehicle's own position is located within the no-entry area A, the control device 32 may recognize the movable range MA only from the measurement value of the distance measurement unit 31. That is, the control device 32 may recognize an obstacle without using the calculated distance, and may set the range excluding the obstacle as the movable range MA. As a result, the control device 32 can recognize the no-entry area A as the movable range MA. This prevents the control device 32 from mistakingly determining that the vehicle is surrounded by obstacles due to the virtual measurement point P2, and prevents the vehicle 20 from becoming unable to move. When the vehicle's own position is located within the no-entry area A, the control device 32 can move the vehicle 20 outside the no-entry area A. In this case, the notification unit 35 does not need to be provided.

[0053] When the estimated own position is within the no-entry area A, the control device 32 does not need to perform the notification by the notification unit 35. The control device 32 may recognize the tracking target T using the calculated distance in addition to the measurement value of the distance measurement unit 31. In this case, if the tracking target T enters the no-entry area A, the control device 32 will lose sight of the tracking target T. The control device 32 moves toward the position where the tracking target T was last detected. Then, when the tracking target T leaves the no-entry area A and the control device 32 recognizes the tracking target T again, the control device 32 causes the vehicle 20 to track the tracking target T.

[0054] The control device 32 may estimate its own position using the measurement value of the distance measurement unit 31 and the calculated distance. The control device 32 only needs to detect the no-entry area A based on the calculated distance obtained by processing the measurement value of the distance measurement unit 31 and to obtain the movable range MA, and the process of FIG. 5 may be modified as appropriate.

[0055] The method of self-location estimation may be changed as appropriate. For example, the self-location estimation may be performed only by odometry. In addition, the odometry may be performed using an inertial measurement unit including a gyroscope, an acceleration sensor, or a combination of these, or may be performed using a camera.

[0056] The LIDAR used as the distance measurement unit 31 may be a three-dimensional range finder that emits a laser while changing the irradiation angle in both the horizontal and vertical directions. A stereo camera may be used as the distance measurement unit 31. The stereo camera allows the control device 32 to recognize the surrounding environment from a parallax image obtained by capturing images of the surrounding environment using multiple cameras. The parallax image indicates pixel differences that occur between cameras when multiple cameras capture images of the same feature point. The feature point is a portion where parallax is obtained, such as an edge of an object, that is, a pixel where the brightness of each pixel in the captured image changes. The control device 32 can obtain the distance from the vehicle 20 to the feature point and the direction from the vehicle 20 to the feature point using the interocular distance, focal length, parallax image, etc. of the stereo camera. When the stereo camera is used as the distance measurement unit 31, the feature point becomes the measurement point. Note that the distance measurement unit 31 may also be one that uses radio waves.

[0057] The distance measurement unit 31 may be a combination of multiple sensors, such as a LIDAR and a stereo camera. That is, the distance measurement unit 31 may be configured with a single sensor or multiple sensors.

[0058] The tracking target T may be something other than a person. For example, it may be a moving object other than the vehicle 20. In this case, the notification unit 35 may be a communication device that can communicate with the moving object that is the tracking target T. When the vehicle 20 enters the no-entry area A, the control device 32 notifies the moving object that is the tracking target T by communicating with the communication device.

[0059] The tracking target T may be detected by providing the tracking target T with a mark or the like that can be detected by the distance measurement unit 31. In addition, the tracking target T may be detected by a detection unit other than the distance measurement unit 31.

[0060] The moving object is not limited to the vehicle 20, and may be a multi-legged robot or an autonomously flying unmanned aerial vehicle. The drive mechanism 23 may have any configuration as long as it can control the speed of the vehicle 20 so as to follow the speed command value.

[0061] The control device 32 may have the function of the driving control device 26. The control device 32 may output a command rotation speed to the motor driver 25, and the motor driver 25 may drive the motor 24 in accordance with the command rotation speed. [Explanation of symbols]

[0062] A...no entry area, M...environmental map, MA...movement range, T...tracking target, 10...autonomous moving body, 20...vehicle (moving body), 32...control device (object detection unit, moving body control unit, self-position estimation unit, distance calculation unit, processing unit, and movable range derivation unit), 34...memory unit, 35...alarm unit.

Claims

1. An autonomous moving body that tracks a tracking target, A moving body, A distance measurement unit for measuring a distance and a direction from the moving object to a measurement point; an object detection unit that detects objects including the tracking target and obstacles using the measurement result of the distance measurement unit; A moving object control unit that moves the moving object so as to track the tracking target; a storage unit in which map information including an environmental map and information on no-entry areas previously set in the environmental map is stored; a self-position estimation unit that estimates a self-position of the moving object on the environmental map; a distance calculation unit that calculates a distance from the self-location to the no-entry area for each of a plurality of directions in a horizontal direction from the self-location to the no-entry area, using the self-location estimated by the self-location estimation unit and information on the no-entry area stored in the storage unit; a processing unit that processes the measurement value associated with a direction in which the measurement point does not exist between the moving body and the no-entry area, among the measurement values ​​of the distance measurement unit associated with the same direction as the direction from the moving body to the no-entry area, into a calculated distance calculated by the distance calculation unit; a movable range deriving unit that derives a movable range of the moving body from the measurement value of the distance measurement unit that has not been processed by the processing unit and the calculated distance; a notification unit that issues a notification when the self-position estimated by the self-position estimation unit is located within the no-entry area, the distance measurement unit measures a distance and a direction to the measurement point by irradiating a laser, the no-entry area includes a location that cannot be detected by the distance measurement unit, The moving body control unit is an autonomous moving body that moves the moving body within the movable range.

2. An autonomous moving body that tracks a tracking target, A moving body, A distance measurement unit for measuring a distance and a direction from the moving object to a measurement point; an object detection unit that detects objects including the tracking target and obstacles using the measurement result of the distance measurement unit; A moving object control unit that moves the moving object so as to track the tracking target; a storage unit in which map information including an environmental map and information on no-entry areas previously set in the environmental map is stored; a self-position estimation unit that estimates a self-position of the moving object on the environmental map; a distance calculation unit that calculates a distance from the self-location to the no-entry area for each of a plurality of directions in a horizontal direction from the self-location to the no-entry area, using the self-location estimated by the self-location estimation unit and information on the no-entry area stored in the storage unit; a processing unit that processes the measurement value associated with a direction in which the measurement point does not exist between the moving body and the no-entry area, among the measurement values ​​of the distance measurement unit associated with the same direction as the direction from the moving body to the no-entry area, into a calculated distance calculated by the distance calculation unit; a movable range deriving unit that derives a movable range of the moving body from the measurement value of the distance measurement unit that has not been processed by the processing unit and the calculated distance, the distance measurement unit measures a distance and a direction to the measurement point by irradiating a laser, the no-entry area includes a location that cannot be detected by the distance measurement unit, The moving object control unit moves the moving object within the movable range, The movable range derivation unit derives the movable range without using the calculated distance when the self-position estimated by the self-position estimation unit is located within the no-entry area.

3. The autonomous moving body according to claim 1 or claim 2, wherein the self-position estimation unit estimates the self-position using the measurement values ​​of the distance measurement unit that have not been processed by the processing unit, and the measurement values ​​before being processed by the processing unit.

4. The object detection unit detects the tracking target using the measurement values ​​of the distance measurement unit that have not been processed by the processing unit, and the measurement values ​​before processing by the processing unit. An autonomous moving body as described in any one of claims 1 to 3.

Images