Autonomous mobile vehicle control system, autonomous mobile vehicle control method, and program
The autonomous mobile vehicle control system addresses inaccuracies in measuring complex shapes by dynamically adjusting its path and sensor positions, ensuring efficient and accurate three-dimensional shape reconstruction.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- TOYOTA JIDOSHA KK
- Filing Date
- 2024-12-18
- Publication Date
- 2026-06-30
AI Technical Summary
Existing autonomous mobile vehicles struggle to accurately measure complex three-dimensional shapes of objects due to errors from insufficient distance between sensors and objects, leading to incomplete or inaccurate reconstructions, and require repeated measurements.
An autonomous mobile vehicle control system that adjusts its path and sensor positions based on the object's three-dimensional shape, allowing it to approach complex areas and measure with sensors more effectively, using path creation, driving control, and sensor control units to ensure accurate reconstruction.
The system efficiently and accurately reconstructs three-dimensional shapes by adjusting the vehicle's distance and orientation relative to the object, enabling precise measurement even with complex shapes.
Smart Images

Figure 2026106711000001_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to an autonomous movement control system, an autonomous movement control method, and a program.
Background Art
[0002] An autonomous mobile body equipped with a sensor for measuring the three-dimensional shape of an object is caused to travel, and the three-dimensional shape of the object in a specific region is restored.
[0003] Patent Document 1 discloses a three-dimensional modeling device that causes an autonomous mobile body equipped with a sensor that can rotate 360 degrees in the horizontal direction and move in height in the vertical direction to travel, and obtains a three-dimensional model represented by a set of point clouds based on the modeling sensor data acquired at a plurality of points and the three-dimensional position of the sensor when the data was acquired.
Prior Art Documents
Patent Documents
[0004]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0005] If the distance between the sensor for measuring the three-dimensional shape and the object is too close, an error is likely to occur in the measured three-dimensional shape. Therefore, it is considered that a distance of a certain amount or more is required between the sensor and the object. However, when measuring at a distance, if the three-dimensional shape of the object is complex, the three-dimensional shape cannot be measured, or the result that it cannot be restored to an accurate three-dimensional shape occurs, and it is necessary to cause the autonomous mobile body to travel again for measurement. The present disclosure solves such problems, and provides an autonomous movement control system, an autonomous movement control method, and a program that adjust the distance between the autonomous mobile body and the object according to the three-dimensional shape of the object, and efficiently and accurately restore the three-dimensional shape of the object. [Means for solving the problem]
[0006] This disclosure relates to an autonomous mobile vehicle control system that restores the three-dimensional shape of an object in a region by driving an autonomous mobile vehicle equipped with a sensor for measuring the three-dimensional shape of an object, comprising: a path creation unit that creates a path in the region on which the autonomous mobile vehicle will travel; a driving control unit that drives the autonomous mobile vehicle according to the created path in the region; an information acquisition unit that acquires the three-dimensional shape of the object measured by the sensor while the autonomous mobile vehicle is driving; and a path change unit that changes the path in the region on which the autonomous mobile vehicle will travel so that the autonomous mobile vehicle approaches the object based on the three-dimensional shape of the object acquired while the autonomous mobile vehicle is driving, wherein the driving control unit drives the autonomous mobile vehicle according to the changed path in the region, and the information acquisition unit acquires the three-dimensional shape of the object measured by the sensor while the autonomous mobile vehicle is traveling along the changed path in the region. This configuration allows the autonomous mobile unit to adjust the distance between itself and the object according to the object's three-dimensional shape, thereby efficiently and accurately restoring the object's three-dimensional shape.
[0007] The path changing unit changes the path of the autonomous mobile body in the area it travels through so that it approaches the space portion that is not visible from above in the three-dimensional range that overlaps with the object when viewed from above. With this configuration, if a space portion exists and the three-dimensional shape of the object is complex, the autonomous mobile body approaches the space portion and measures the three-dimensional shape of the object with a sensor, so that even if the object has a complex three-dimensional shape, the three-dimensional shape can be accurately reconstructed.
[0008] The system includes a sensor control unit that controls the position of the sensor, and the sensor control unit controls the position of the sensor so that the sensor enters the spatial portion while the autonomous mobile body is traveling along the modified path of the region. With this configuration, the sensor is brought close to the part of the object with a complex three-dimensional shape to measure the three-dimensional shape of the object, so that even if the object has a complex three-dimensional shape, the three-dimensional shape can be reconstructed more accurately.
[0009] This disclosure provides an autonomous mobile body control method for restoring the three-dimensional shape of an object in a region by driving an autonomous mobile body equipped with a sensor for measuring the three-dimensional shape of an object, the method comprising: creating a path in the region on which the autonomous mobile body will travel; driving the autonomous mobile body according to the created path in the region; acquiring the three-dimensional shape of the object measured by the sensor while the autonomous mobile body is driving; changing the path in the region on which the autonomous mobile body will travel so that the autonomous mobile body approaches the object based on the three-dimensional shape of the object acquired while the autonomous mobile body is driving; driving the autonomous mobile body according to the changed path in the region; and acquiring the three-dimensional shape of the object measured by the sensor while the autonomous mobile body is driving along the changed path in the region. This configuration allows the autonomous mobile unit to adjust the distance between itself and the object according to the object's three-dimensional shape, thereby efficiently and accurately restoring the object's three-dimensional shape.
[0010] This disclosure relates to a program for controlling an autonomous mobile vehicle control system that moves an autonomous mobile vehicle equipped with a sensor for measuring the three-dimensional shape of an object to restore the three-dimensional shape of an object in a region, the program causing the autonomous mobile vehicle control system to execute the following steps: creating a path in the region on which the autonomous mobile vehicle will travel; driving the autonomous mobile vehicle according to the created path in the region; acquiring the three-dimensional shape of the object measured by the sensor while the autonomous mobile vehicle is traveling; changing the path in the region on which the autonomous mobile vehicle will travel so that the autonomous mobile vehicle approaches the object based on the three-dimensional shape of the object acquired while the autonomous mobile vehicle is traveling; driving the autonomous mobile vehicle according to the changed path in the region; and acquiring the three-dimensional shape of the object measured by the sensor while the autonomous mobile vehicle is traveling along the changed path in the region. This configuration allows the autonomous mobile unit to adjust the distance between itself and the object according to the object's three-dimensional shape, thereby efficiently and accurately restoring the object's three-dimensional shape.
[0011] The route creation unit may use a learning model obtained through machine learning to create a route for the autonomous vehicle to travel through the area. [Effects of the Invention]
[0012] This disclosure provides an autonomous mobile vehicle control system, an autonomous mobile vehicle control method, and a program that efficiently and accurately restore the three-dimensional shape of an object by adjusting the distance between the autonomous mobile vehicle and the object according to the object's three-dimensional shape. [Brief explanation of the drawing]
[0013] [Figure 1] This is a system configuration diagram of an autonomous mobile vehicle control system according to an embodiment. [Figure 2] This is a schematic perspective view of an autonomous mobile robot. [Figure 3] This is a functional block diagram of the control unit and cloud server for an autonomous robot. [Figure 4] This is a flowchart of the control program executed by the first control unit and the second control unit. [Figure 5] Figure 1 shows a schematic perspective view of region a. [Modes for carrying out the invention]
[0014] The present disclosure will be explained below with reference to Figures 1 to 5. Figure 1 is a system configuration diagram of an autonomous mobile control system according to an embodiment. Figure 2 is a schematic perspective view of an autonomous mobile robot. Figure 3 is a functional block diagram of the control unit and cloud server of the autonomous mobile robot. Figure 4 is a flowchart of the control program executed by the first control unit and the second control unit. Figure 5 is a schematic perspective view of area a shown in Figure 1.
[0015] It should be noted that the right-handed XYZ Cartesian coordinate system shown in Figure 5 is merely a convenient representation for explaining the positional relationships of the constituent elements. In Figure 5, the positive Z-axis direction is the upward direction, and the XY plane is the horizontal plane.
[0016] Embodiment 1 Using FIG. 1, the system configuration of the autonomous movement control system S according to Embodiment 1 will be described. Region a is a region for restoring the three-dimensional shape of an object. An object A and an object B are arranged in region a, and FIG. 1 shows the planar shape viewed from above the object A and the object B. In the autonomous movement control system S, the autonomous mobile robot 1 is caused to travel along the path indicated by the solid arrow in region a to measure the three-dimensional shape of the object A or the like. The autonomous mobile robot 1 and the cloud server 50 are connected by a communication network N, and the cloud server 50 restores the three-dimensional shape of the object A or the like based on the three-dimensional shape of the object A or the like measured by the autonomous mobile robot 1.
[0017] Subsequently, the autonomous mobile robot 1 will be described using FIG. 2. The autonomous mobile robot 1 is an autonomous mobile body equipped with a sensor for measuring the three-dimensional shape of an object, and has a carriage part 10, a body part 20, and a head part 30.
[0018] The carriage part 10 includes, for example, left and right drive wheels 12 and one caster (not shown) in a cylindrical housing, and the left and right drive wheels 12 are independently driven and controlled by motors. Thereby, the carriage part 10 can travel in an arbitrary direction on the floor surface. Note that instead of the drive wheels 12, it may be made to walk by a plurality of legs.
[0019] Obstacle detection sensors 11 are provided on the front side and the back side of the carriage part 10, respectively (the obstacle detection sensor on the back side is not shown). The obstacle detection sensor 11 is a sensor for detecting an obstacle in the path along which the autonomous mobile robot 1 travels, and is, for example, a Lidar sensor, a millimeter-wave radar sensor, a Radar sensor, an ultrasonic sensor, a depth sensor, or the like.
[0020] The body 20 has a body 21 and an arm unit 22. The body 21 is formed in a semi-cylindrical shape and extends in the vertical direction. A first joint portion (not shown) for connecting them is formed on the body 21 and the carriage portion 10, and the body 21 can rotate horizontally about the first connection axis AX1 of the first joint portion by a motor. Further, a first elevating portion 25 is provided between the arm unit 22 and the body 21, and the arm unit 22 can be elevated in the vertical direction by the first elevating portion 25. In the embodiment, there is one arm unit 22, but a plurality of them may be provided.
[0021] The arm unit 22 is disposed on the front side of the body 21 and has a first arm portion 23a, a second arm portion 23b, a wrist portion 23c, a first shape sensor 24, and a first camera 26.
[0022] The first arm portion 23a is provided on the body 21 side of the arm unit 22. A second joint portion (not shown) for connecting them is formed on the first arm portion 23a and the first elevating portion 25, and the first arm portion 23a can rotate vertically about the second connection axis AX2 of the second joint portion by a motor.
[0023] A second arm portion 23b is connected to the tip of the first arm portion 23a on the side opposite to the body 21. A third joint portion (not shown) for connecting them is formed on the second arm portion 23b and the first arm portion 23a, and the second arm portion 23b can rotate in the twisting direction about the third connection axis AX3 of the third joint portion by a motor.
[0024] A wrist portion 23c is connected to the tip of the second arm portion 23b on the side opposite to the first arm portion 23a. A fourth joint portion (not shown) for connecting them is formed on the wrist portion 23c and the second arm portion 23b, and the wrist portion 23c can rotate vertically or the like about the fourth connection axis AX4 of the fourth joint portion by a motor. Further, a fifth joint portion (not shown) for connecting them is formed on the wrist portion 23c and the second arm portion 23b, and the wrist portion 23c can rotate in the twisting direction about the fifth connection axis AX5 of the fifth joint portion by a motor.
[0025] A first shape sensor 24 is attached to the end of the wrist portion 23c opposite to the second arm portion 23b. The first shape sensor 24 is a sensor that measures the three-dimensional shape of objects around the autonomous mobile robot 1 as point cloud data, and is, for example, a Lidar sensor, millimeter-wave radar sensor, radar sensor, ultrasonic sensor, or depth sensor.
[0026] A first camera 26 is attached to the wrist portion 23c. The first camera 26 is a camera that takes pictures of objects in the vicinity of the autonomous mobile robot 1, and can be, for example, a monocular camera or a compound camera.
[0027] The torso 21 contains an IMU (Inertial Measurement Unit) 26. The IMU 26 is an inertial measurement device that measures the current position of the autonomous mobile robot 1.
[0028] The head 30 includes a display 31, a second shape sensor 32, and a second camera 33. The head 30 and the body 21 are connected by a sixth joint (not shown), and the head 30 can rotate horizontally around the sixth connecting axis AX6 of the sixth joint by a motor. The head 30 and the body 21 are also connected by a seventh joint (not shown), and the head 30 can rotate vertically around the seventh connecting axis AX7 of the seventh joint by a motor. Furthermore, a second lifting mechanism (not shown) is provided between the head 30 and the body 21, and the head 30 can move up and down by the second lifting mechanism.
[0029] For example, a touch panel LCD may be used as the display 31. The display 31 is used to display information about the autonomous mobile robot 1 to the user, or to input instructions from the user to the autonomous mobile robot 1, etc.
[0030] A second shape sensor 32 is mounted at the very top of the head 30. The second shape sensor 32 has the same structure and function as the first shape sensor 24, so no explanation is provided. A second camera 33 is mounted between the display 31 of the head 30 and the second shape sensor 32. The second camera 33 has the same structure and function as the first camera 26, so no explanation is provided.
[0031] Next, using Figure 3, we will explain the function of the robot control unit 2 built into the autonomous mobile robot 1. The robot control unit 2 is built into the body 21 of the autonomous mobile robot 1.
[0032] The robot control unit 2 comprises a first control unit 3, a memory 4, and a communication unit 5. The first control unit 3 includes a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), etc., and controls the left and right drive wheels 12, etc. The memory 4 includes RAM (Random Access Memory), ROM (Read Only Memory), SSD (Solid State Drive), etc., and stores control programs for the autonomous mobile robot 1, such as the operation procedures of the body 21, etc., while the autonomous mobile robot 1 is moving. The communication unit 5 is connected to a communication network N and connects to a cloud server 50 via a mobile phone line (for example, 4G or 5G), Bluetooth (registered trademark), WIFI (Wireless Fidelity), etc.
[0033] The first control unit 3 includes a driving control unit 3a, a sensor control unit 3b, a point cloud data acquisition unit 3c, and an image data acquisition unit 3d.
[0034] The driving control unit 3a transmits control signals to the motors of the left and right drive wheels 12 to make the left and right drive wheels 12 move in any direction. The driving control unit 3a makes the autonomous mobile robot 1 move according to the path of the area created by the path creation unit 51a, which will be described later. The driving control unit 3a obtains the current position of the autonomous mobile robot 1 measured by the IMU 26 and controls the autonomous mobile robot 1 so that it can move according to the path of the area created by the path creation unit 51a. When the obstacle detection sensor 11 detects an obstacle, the driving control unit 3a controls the robot to avoid the obstacle.
[0035] The sensor control unit 3b transmits control signals to the first to seventh joints, the first lifting unit 25, and the second lifting unit to rotate and raise / lower the torso 21, the first arm 23a, the second arm 23b, the wrist 23c, and the head 30 in any direction. When the torso 21, etc., are rotated, the position and orientation of the first shape sensor 24 and the second shape sensor 32 change. In this way, the sensor control unit 3b controls the position and orientation of the first shape sensor 24 and the second shape sensor 32. At the same time, the sensor control unit 3b controls the position and orientation of the first camera 26 and the second camera 33.
[0036] The sensor control unit 3b operates the torso 21, etc., so that the first shape sensor 24, the first camera 26, etc., can measure the three-dimensional shape of objects around the autonomous mobile robot 1 while the autonomous mobile robot 1 is traveling along the path of the area created by the path creation unit 51a, which will be described later. Specifically, the sensor control unit 3b sends control signals to the first joints, etc., according to a procedure stored in the memory 4 in advance, so that the torso 21, etc., operates while the autonomous mobile robot 1 is traveling along the path of the area created by the path creation unit 51a.
[0037] The point cloud data acquisition unit 3c acquires point cloud data, which is the three-dimensional shape of an object measured by the first shape sensor 24 and the second shape sensor 32, at predetermined intervals while the autonomous mobile robot 1 is traveling along the path of the area created by the path creation unit 51a. The point cloud data acquisition unit 3c functions as an information acquisition unit that acquires the three-dimensional shape of an object measured by the first shape sensor 24 and the second shape sensor 32 while the autonomous mobile robot 1 is traveling along the path of the area created by the path creation unit 51a. In Embodiment 1, two shape sensors are provided, and the two sensors share the measurement range for measurement. Instead of two shape sensors, one or three or more shape sensors may be provided.
[0038] The timing at which the point cloud data acquisition unit 3c acquires point cloud data from the first shape sensor 24, etc., is synchronized with the position and orientation of the first shape sensor 24 and the second shape sensor 32, which are controlled by the sensor control unit 3b. In other words, the point cloud data acquisition unit 3c acquires point cloud data from the first shape sensor 24, etc., in accordance with the position and orientation of the first shape sensor 24 and the second shape sensor 32. By acquiring point cloud data in accordance with the position and orientation of the first shape sensor 24, etc., the second control unit 51, described later, integrates the multiple point cloud data acquired by the point cloud data acquisition unit 3c to reconstruct the three-dimensional shape of the object.
[0039] The image data acquisition unit 3d acquires image data of objects captured by the first camera 26 and the second camera 33 at predetermined intervals while the autonomous mobile robot 1 is traveling along the route of the area created by the route creation unit 51a. Since the image data includes the three-dimensional shape of the object, the image data acquisition unit 3d functions as an information acquisition unit that acquires the three-dimensional shape of the object measured by the first camera 26 and the second camera 33 while the autonomous mobile robot 1 is traveling along the route of the area created by the route creation unit 51a. In Embodiment 1, two cameras are provided, and the two cameras share the shooting range for shooting. Instead of two cameras, one or three or more cameras may be provided.
[0040] The timing at which the image data acquisition unit 3d acquires image data from the first camera 26, etc., is synchronized with the position and orientation of the first camera 26 and the second camera 33 controlled by the sensor control unit 3b. In other words, the image data acquisition unit 3d acquires image data from the first camera 26, etc., in accordance with the position and orientation of the first camera 26 and the second camera 33. By acquiring image data in accordance with the position and orientation of the first camera 26, etc., the second control unit 51, described later, integrates the multiple image data acquired by the image data acquisition unit 3d to reconstruct the three-dimensional shape of the object.
[0041] The timing at which the point cloud data acquisition unit 3c acquires point cloud data and the timing at which the image data acquisition unit 3d acquires image data may be the same or different. In Embodiment 1, the time interval for acquiring image data is set to be longer than the time interval for acquiring point cloud data.
[0042] Next, using Figure 3, the functions of the cloud server 50 that constitutes the autonomous mobile control system S according to Embodiment 1 will be described. The cloud server 50 includes a second control unit 51, a memory 52, and a communication unit 53.
[0043] The second control unit 51 includes a CPU, GPU, etc., and is responsible for creating paths in the area where the autonomous mobile robot 1 travels. The memory 52 includes RAM, ROM, SSD, etc., and stores the control program for the autonomous mobile robot 1 and data on the three-dimensional dimensions of the area used to reconstruct the three-dimensional shape of an object. The communication unit 53 is connected to the communication network N in a communication-enabled manner and connects to the autonomous mobile robot 1 via a mobile phone line, Bluetooth®, Wi-Fi, etc.
[0044] The second control unit 51 includes a path creation unit 51a, a path modification unit 51b, and an operation modification unit 51c. The path creation unit 51a creates a path in the area on which the autonomous mobile robot 1 will travel. Specifically, data regarding the three-dimensional dimensions of the area on which the three-dimensional shape of an object is reconstructed, as well as the planar shape and layout of equipment and machinery in the area as viewed from above, are pre-stored in the memory 52. Based on this data, the path creation unit 51a creates a path in the area on which the autonomous mobile robot 1 will travel. The path creation unit 51a creates the path in the area to be traveled using known methods such as Dijkstra's algorithm, Aester's algorithm, or Voronoi tessellation. If the distance between the first shape sensor 24, etc. and the object is too close, errors are likely to occur in the measured point cloud data. For example, if there is a path that gradually moves away from the object, it is created using a known method and selected.
[0045] The path changing unit 51b changes the path of the autonomous mobile robot 1 so that it approaches an object, based on point cloud data of an object acquired by the point cloud data acquisition unit 3c, while the autonomous mobile robot 1 is traveling along the path of the area created by the path creation unit 51a. In other words, the path changing unit 51b changes the path of the area on which the autonomous mobile robot travels so that it approaches an object, based on the three-dimensional shape of the object acquired while the autonomous mobile robot is traveling. Specifically, if there is a space that is not visible from above in the three-dimensional range that overlaps with the object in the vertical view while the autonomous mobile robot 1 is traveling along the path of the area created by the path creation unit 51a, the path of the area on which the autonomous mobile robot 1 travels is changed so that it approaches the space.
[0046] The operation modification unit 51c modifies the operation procedure of the body 21 of the autonomous mobile robot 1, etc., so that the first shape sensor 24 and the first camera 26 enter the space while the autonomous mobile robot 1 is traveling along the path of the area modified by the path modification unit 51b.
[0047] Next, we will describe the control program for the autonomous mobile robot 1 executed by the first control unit 3 of the autonomous mobile robot 1 and the second control unit 51 of the cloud server 50. Figure 4 shows a flowchart of the control program for the autonomous mobile robot 1 executed by the first control unit 3 and the second control unit 51.
[0048] Before the control program is executed by the first control unit 3 and the second control unit 51, data such as the three-dimensional dimensions of region a, where the three-dimensional shape of the object is to be reconstructed, and the planar shape and layout of equipment and machinery located in region a as viewed from above, are input into the memory 52 of the cloud server 50 using an input device (not shown). The control program is started after this data is input into memory 52.
[0049] In step S1, the path creation unit 51a of the second control unit 51 creates a path for the autonomous mobile robot 1 to travel through area a. For example, it creates a path for area a as shown by the solid arrow in Figure 1. The second control unit 51 transmits the path for area a created by the path creation unit 51a to the autonomous mobile robot 1 via the communication unit 53.
[0050] In step S2, the driving control unit 3a of the first control unit 3 transmits control signals to each motor of the left and right drive wheels 12 so that the autonomous driving robot 1 travels according to the path of region a created by the path creation unit 51a. The autonomous driving robot 1 travels according to the path of region a created by the path creation unit 51a.
[0051] In step S3, while the autonomous mobile robot 1 is traveling along the path in region a created by the path creation unit 51a, the sensor control unit 3b of the first control unit 3 operates the body 21, etc., so that the first shape sensor 24 and the first camera 26 can measure the three-dimensional shape of objects around the autonomous mobile robot 1. Specifically, the sensor control unit 3b transmits control signals to the first to seventh joints, the first lifting unit 25, and the second lifting unit in accordance with a procedure previously stored in the memory 4, so that the body 21, etc., operates while the autonomous mobile robot 1 is traveling along the path in region a created by the path creation unit 51a.
[0052] In step S4, while the autonomous mobile robot 1 is traveling along the path in region a created by the path creation unit 51a, the point cloud data acquisition unit 3c of the first control unit 3 acquires point cloud data of objects measured by the first shape sensor 24 and the second shape sensor 32. Also in step S4, while the autonomous mobile robot 1 is traveling along the path in region a created by the path creation unit 51a, the image data acquisition unit 3d acquires image data of objects captured by the first camera 26 and the second camera 33. The first control unit 3 transmits the point cloud data of objects acquired by the point cloud data acquisition unit 3c and the image data of objects acquired by the image data acquisition unit 3d to the cloud server 50 via the communication unit 5.
[0053] In step S5, based on the point cloud data acquired by the point cloud data acquisition unit 3c and the image data acquired by the image data acquisition unit 3d, the second control unit 51 determines whether there is a spatial portion that is not visible from above in the three-dimensional range that overlaps with the object in the vertical view. For example, as shown in Figure 5, it determines whether there is a spatial portion that is not visible from above in the three-dimensional range (area enclosed by dashed lines) that overlaps with object A in the vertical view (view along the Z axis). Note that in Figure 5, the dashed lines are shifted relative to the actual position to make the three-dimensional range easier to see. In the three-dimensional range that overlaps with object A in the vertical view, there is a spatial portion P that is not visible from above on the far side of region a (the positive Y-axis side in Figure 5).
[0054] As shown in Figure 5, the empty space includes not only cases where the entire area of object A is empty in one direction (the X-axis direction in Figure 5), but also cases where only a part of it is empty. Furthermore, since the first shape sensor 24 will be inserted into the empty space later, in step S5, if the empty space is large enough for the first shape sensor 24 to fit into, it is determined that there is an empty space. Therefore, in step S5, even if there is an empty space, if its size is not large enough for the first shape sensor 24 to fit into, it is determined that there is no empty space. If it is determined in step S5 that an empty space exists, the process proceeds to step S6; otherwise, the process proceeds to step S12.
[0055] In step S6, the path changing unit 51b of the second control unit 51 changes the path of region a created by the path creation unit 51a in step S1. Specifically, in order to allow the first shape sensor 24 and the first camera 26 to enter the spatial portion P, the path of region a that the autonomous mobile robot 1 travels is changed so that the autonomous mobile robot 1 approaches the spatial portion P of object A, as shown by the dashed line in Figure 1. The distance to approach is determined based on the size of the spatial portion P and the length of the extendable arm unit 22, etc. For example, if the length of the spatial portion P in the width direction (X-axis direction in Figure 5) is long, the autonomous mobile robot 1 is brought closer to the spatial portion P than when it is short. The second control unit 51 transmits the path of region a changed by the path changing unit 51b to the autonomous mobile robot 1 via the communication unit 53.
[0056] Next, in step S7, the operation modification unit 51c of the second control unit 51 modifies the operation procedure of the torso 21, etc., of the autonomous mobile robot 1 so that the first shape sensor 24 and the first camera 26 enter the spatial portion P while the autonomous mobile robot 1 is traveling along the path of region a modified by the path modification unit 51b. Specifically, as shown in Figure 5, the operation procedure of the torso 21, etc., of the autonomous mobile robot 1 is modified so that the arm unit 22 of the autonomous mobile robot 1 is extended so that the first shape sensor 24 and the first camera 26 enter the spatial portion P. Furthermore, after the first shape sensor 24 and the first camera 26 have entered the spatial portion P, the operation procedure of the torso 21, etc., of the autonomous mobile robot 1 is modified so that the second arm unit 23b and wrist unit 23c are rotated so that point cloud data and image data of objects can be acquired from the spatial portion P. The second control unit 51 transmits the operation procedure of the torso 21, etc., modified by the operation modification unit 51c to the autonomous mobile robot 1 via the communication unit 53.
[0057] In step S8, the driving control unit 3a of the first control unit 3 transmits control signals to the motors of the left and right drive wheels 12 so that the autonomous driving robot 1 travels according to the path of region a changed by the path changing unit 51b.
[0058] In step S9, the sensor control unit 3b of the first control unit 3 operates the body 21, etc., according to the procedure changed by the operation change unit 51c, so that the first shape sensor 24 and the first camera 26 enter the spatial portion P while the autonomous mobile robot 1 is traveling along the path of region a changed by the path change unit 51b.
[0059] In step S10, the point cloud data acquisition unit 3c of the first control unit 3 acquires point cloud data of objects measured by the first shape sensor 24 and the second shape sensor 32 while the autonomous mobile robot 1 is traveling along the path in region a changed by the path change unit 51b. Also in step S10, the image data acquisition unit 3d acquires image data of objects captured by the first camera 26 and the second camera 33 while the autonomous mobile robot 1 is traveling along the path in region a changed by the path change unit 51b. The first control unit 3 transmits the point cloud data of objects acquired by the point cloud data acquisition unit 3c and the image data of objects acquired by the image data acquisition unit 3d to the cloud server 50 via the communication unit 5.
[0060] In step S11, the first control unit 3 determines, based on the current position of the autonomous mobile robot 1 measured by the IMU 26, whether the robot has finished traveling along the route in area a that was changed by the route changing unit 51b. If it has not finished, the process returns to step S8 and is repeated; if it has finished, the process proceeds to step S12.
[0061] In step S12, the first control unit 3 determines whether the autonomous mobile robot 1 has finished traveling along the path in area a created by the path creation unit 51a, based on the current position of the autonomous mobile robot 1 measured by the IMU 26. If it has not finished, it returns to step S2 and repeats the process; if it has finished, it proceeds to step S13.
[0062] In step S13, the second control unit 51 reconstructs the three-dimensional shape of the object in region a. Specifically, while the autonomous mobile robot 1 is traveling along the path of region a created by the path creation unit 51a and the path of region a modified by the path modification unit 51b, the second control unit 51 reconstructs the three-dimensional shape of the object using a known method based on the point cloud data acquired by the point cloud data acquisition unit 3c and the image data acquired by the image data acquisition unit 3d. Once the reconstruction is complete, the reconstructed three-dimensional shape is displayed on a display device (not shown) or stored as data in the memory 52.
[0063] In this embodiment, while the autonomous mobile robot 1 is traveling along the path of region a created by the path creation unit 51a, the path modification unit 51b modifies the path of region a to which the autonomous mobile robot 1 is traveling, based on the acquired point cloud data and image data of the object, so that the autonomous mobile robot 1 approaches the object. Then, while the autonomous mobile robot 1 is traveling along the modified path of region a, the point cloud data acquisition unit 3c acquires the point cloud data of the object, and the image data acquisition unit 3d acquires the image data of the object. With this configuration, the distance between the autonomous mobile robot 1 and the object is adjusted according to the object's three-dimensional shape, and the three-dimensional shape of the object is measured by the first shape sensor 24 and the first camera 26, so that the three-dimensional shape of the object can be reconstructed efficiently and accurately.
[0064] In this embodiment, the path changing unit 51b changes the path of the autonomous mobile robot 1 in the region a that the autonomous mobile robot 1 travels through, so that the autonomous mobile robot 1 approaches the spatial portion P that is not visible from above in the three-dimensional range that overlaps with the object when viewed from above. With this configuration, if a spatial portion P exists and the three-dimensional shape of the object is complex, the autonomous mobile robot 1 approaches the spatial portion P and measures the three-dimensional shape of the object with the first shape sensor 24 and the first camera 26, so that even if the object has a complex three-dimensional shape, the three-dimensional shape can be accurately reconstructed.
[0065] In this embodiment, the sensor control unit 3b controls the positions of the first shape sensor 24 and the first camera 26 so that they enter the spatial portion P while the autonomous mobile robot 1 is traveling along the modified path of region a. With this configuration, the first shape sensor 24 and the first camera 26 are brought closer to the part of the object with a complex three-dimensional shape to measure the three-dimensional shape of the object, so that even if the object has a complex three-dimensional shape, the three-dimensional shape can be reconstructed more accurately.
[0066] This disclosure is not limited to the embodiments described above, and may be modified as appropriate without departing from its spirit.
[0067] For example, an autonomous mobile device could be a mobile device that can move through the air, such as a drone. In addition to controlling the position and orientation of sensors in the joints and lifting mechanisms of the autonomous mobile unit, it is also permissible to separate sensor-equipped parts, such as sub-drones, from the autonomous mobile unit. In this embodiment, the autonomous mobile unit was controlled by dividing it into a first control unit and a second control unit, but it may also be combined into a single control unit. In this embodiment, the three-dimensional shape of an object was measured using both a shape sensor and a camera, but either one or the other may be used. In this embodiment, a cloud server was used, but other servers such as dedicated servers may also be used. [Explanation of symbols]
[0068] 1. Autonomous mobile robot 50. Cloud Server S... Autonomous Mobile Control System
Claims
1. In an autonomous mobile vehicle control system that uses sensors to measure the three-dimensional shape of objects to move an autonomous mobile vehicle and reconstruct the three-dimensional shape of objects in a given area, A path creation unit that creates a path in the area on which the autonomous mobile body travels, A driving control unit that drives the autonomous mobile body according to the path of the created area, An information acquisition unit that acquires the three-dimensional shape of the object measured by the sensor while the autonomous mobile body is moving, The system includes a path changing unit that changes the path of the autonomous mobile body in the region to which the autonomous mobile body travels so as to approach the object, based on the three-dimensional shape of the object acquired while the autonomous mobile body is traveling. The driving control unit causes the autonomous mobile body to travel according to the modified path of the region. The information acquisition unit acquires the three-dimensional shape of the object measured by the sensor while the autonomous mobile body is traveling along the modified path of the region. Autonomous mobile vehicle control system.
2. The path changing unit changes the path of the autonomous mobile body in the area it travels through so that the autonomous mobile body approaches the spatial portion that is not visible from above in the three-dimensional range that overlaps with the object when viewed from above. The autonomous mobile control system according to claim 1.
3. The system includes a sensor control unit that controls the position of the aforementioned sensor, The sensor control unit controls the position of the sensor so that the sensor enters the spatial portion while the autonomous mobile body is traveling along the modified path of the region. The autonomous mobile control system according to claim 2.
4. In an autonomous mobile vehicle control method that uses sensors to measure the three-dimensional shape of an object to move an autonomous mobile vehicle and reconstruct the three-dimensional shape of an object in a given area, The steps include creating a route in the area on which the autonomous mobile vehicle will travel, The steps include: driving the autonomous mobile body according to the path of the created area, The steps include acquiring the three-dimensional shape of the object measured by the sensor while the autonomous mobile body is moving, The steps include: changing the path of the autonomous mobile body in the region on which it travels so as to approach the object, based on the three-dimensional shape of the object acquired while the autonomous mobile body is traveling; The steps include: driving the autonomous mobile body according to the modified route of the region, The step includes acquiring the three-dimensional shape of the object measured by the sensor while the autonomous mobile body is traveling along the modified path of the region, A method for controlling autonomous mobile vehicles.
5. In a program for controlling an autonomous mobile vehicle control system that moves an autonomous mobile vehicle equipped with sensors to measure the three-dimensional shape of an object and reconstructs the three-dimensional shape of an object in a given area, The steps include creating a route in the area on which the autonomous mobile vehicle will travel, The steps include: driving the autonomous mobile body according to the path of the created area, The steps include acquiring the three-dimensional shape of the object measured by the sensor while the autonomous mobile body is moving, The steps include: changing the path of the autonomous mobile body in the region on which it travels so as to approach the object, based on the three-dimensional shape of the object acquired while the autonomous mobile body is traveling; The steps include: driving the autonomous mobile body according to the modified route of the region, The steps include acquiring the three-dimensional shape of the object measured by the sensor while the autonomous mobile body is traveling along the modified path of the region, A program to be executed by an autonomous mobile vehicle control system.