JP7848876B2 - Autonomous mobile control device, autonomous mobile system, autonomous mobile control method, and program - Google Patents

Description

This disclosure relates to control technology for autonomously moving mobile bodies.

In indoor environments such as factories and warehouses, AGVs (Automatic Guided Vehicles) are used to autonomously navigate along predetermined paths without human intervention. For AGVs to operate, control is required to estimate the vehicle's own position and guide it along a predetermined path. Control methods for AGV operation include magnetic guidance, optical guidance, image recognition, laser guidance, and SLAM (Simultaneous Localization and Mapping).

The magnetic induction system guides a vehicle along a designated route by detecting the magnetic field emitted by magnetic rods or tapes laid on the floor along the route, using magnetic sensors mounted on the vehicle. While the magnetic induction system has the advantages of being simple and highly reliable, it has the disadvantage of incurring costs when changing the route. Furthermore, it may be affected by magnetic fields from nearby pipes and other structures.

The optical guidance system uses optical sensors mounted on the vehicle to detect a path formed on the floor by laying guidance tape, thereby guiding the vehicle along the designated route. The optical guidance system has the advantage of lower costs compared to the magnetic guidance system.

The image recognition method is a system in which a camera mounted on the vehicle reads markers such as QR codes (registered trademarks) installed on the floor or ceiling, allowing the vehicle to determine its own position and automatically drive to its destination. These three control methods are also called route guidance methods, as they allow the vehicle to travel only along predetermined, fixed routes.

The laser guidance system (reflector guidance system) is a system in which a laser beam irradiation device mounted on the vehicle irradiates a reflector installed on a wall, floor, or pillar with laser light, and a receiving device mounted on the vehicle detects the reflected wave from the reflector, allowing the vehicle to estimate its own position and drive without using a guide.

There are two types of SLAM guidance methods: Visual SLAM, which uses cameras, and LiDAR SLAM, which uses LiDAR (Light Detection and Ranging). Visual SLAM is a method where a camera mounted on the vehicle acquires continuous image data of the surrounding environment, such as walls, allowing the vehicle to estimate its own position and drive without the use of a guide. LiDAR SLAM is a method where a laser beam irradiator mounted on the vehicle irradiates the surrounding environment, such as walls, with laser beam receiving equipment mounted on the vehicle measures the time it takes for the reflected light to return, and by measuring the distance and direction to the surrounding environment, the vehicle can estimate its own position and drive without the use of a guide. Both laser guidance and SLAM methods are also called autonomous navigation methods because they do not require a guide.

Furthermore, AGVs that use SLAM are also called AMRs (Autonomous Mobile Robots). The advantage of the SLAM method is that it eliminates the need for guide sensors, allowing for flexible route setting. Also, while route guidance systems require the vehicle to temporarily stop when it detects an obstacle, the SLAM method allows the vehicle to avoid obstacles and continue driving by rerouting.

On the other hand, the SLAM method requires processing to eliminate the influence of changes in the surrounding environment, such as the opening and closing of doors, changes in the loading state of goods on shelves, and the movements of workers and other transport vehicles. Furthermore, it has the challenge that operation may be hindered in spaces without distinctive features.

To address the challenges of the SLAM method, an autonomous navigation system is being considered in which feature points are extracted from image data of the floor surface captured by a camera mounted on the vehicle, and the vehicle automatically navigates by tracing these points (Non-Patent Document 1).

Junpei Ogawa et al., "Accuracy Verification of a Positioning System Using Road Surface Image Sensors in Automated Guided Vehicles," June 12, 2021, Institute of Electrical Engineers of Japan, Systems Research Committee <https://www.bookpark.ne.jp/cm/ieej/detail/IEEJ-20210612C01101-002-PDF/>

However, while there are several methods for extracting feature points from surrounding environment image data, such as the feature-based method (extracting feature points from within the image) and the direct method (referencing the entire image), in either case, if there are various changes in the surrounding environment along the travel route, such as pedestrians, moving vehicles, and the opening and closing of doors in addition to static objects, generating map data and performing image matching processing becomes complex, increasing the amount of processing and thus the processing load.

This invention has been made in view of the above circumstances, and aims to enable autonomous movement such as autonomous driving by suppressing the processing load.

To achieve the above objective, the invention according to claim 1 is an autonomous movement control device for controlling the movement of an autonomously moving autonomous mobile body, comprising: a self-position estimation unit that estimates the self-position and direction of travel of the autonomous mobile body in the quasi-periodic tiles by pattern matching data relating to an image obtained by photographing quasi-periodic tiles composed of a plurality of types of tile shapes with map data having a coordinate system corresponding to the quasi-periodic tiles; a control unit that controls the movement of the autonomous mobile body based on the self-position and direction of travel estimated by the self-position estimation unit and a preset movement route of the autonomous mobile body; a gyro sensor that detects the angle, angular velocity, or angular acceleration of the autonomous mobile body, and an acceleration sensor that measures the acceleration of the autonomous mobile body, wherein the self-position estimation unit calibrates the parameters of the gyro sensor and the acceleration sensor based on the estimated self-position and direction of travel .

As explained above, the present invention has the effect of enabling autonomous movement with reduced processing load.

This is an overall configuration diagram of the autonomous driving system according to the embodiment. This is a diagram showing the configuration of an autonomous vehicle. This is an electrical hardware configuration diagram of an information processing device. This is a plan view of a quasi-periodic tile. This diagram shows the shape of each tile. This is a plan view showing the travel pattern during the initial run. This is a flowchart showing the initialization process. This is a plan view showing the route of this trip. Flowchart of the process for this run. This diagram clarifies the 12-fold symmetry portion in the quasi-periodic tile.

The embodiments of the present invention will be described below with reference to the drawings.

[System configuration of the embodiment]
The general configuration of the autonomous driving system according to the embodiment will be explained using Figure 1. Figure 1 is an overall configuration diagram of the autonomous driving system according to the embodiment.

As shown in Figure 1, the autonomous driving system 1 is constructed from an autonomous vehicle (also referred to as "vehicle") 2, such as an autonomously driven unmanned transport vehicle, and quasi-periodic tiles 8 on the floor. The vehicle 2 estimates its own position and direction of travel and moves by reading the pattern composed of multiple tile shapes of the quasi-periodic tiles 8.

[Explanation of quasi-periodic tiles]
As shown in Figure 4A, the quasi-periodic tile 8 is composed of multiple types of tile shapes and is a tile on which a non-periodic geometric pattern is drawn, and is installed on the floor surface. As shown in Figure 4B, the multiple types of tile shapes are composed of three types of figures (squares with equal side lengths L and two types of rhombuses). Unlike plane tiling which has a periodic pattern composed of regular polygons, the quasi-periodic tile 8 is a tiling method that achieves a plane tiling without a periodic pattern (a non-periodic pattern that overlaps due to translation) using a finite number of polygons. As an example, Penrose tiling, which achieves plane tiling using two types of rhombuses, is known. The geometric pattern of the quasi-periodic tile 8 shown in Figures 4A and 4B is merely an example, and the geometric pattern may not have periodicity.

In this embodiment, the aperiodicity of the pattern of the quasi-periodic tile 8 is used to estimate the self-position of the vehicle 2. The quasi-periodic tile 8 is constructed on the floor surface and is imaged by the (described later) imaging unit 20 mounted on the vehicle 2. Note that, as long as the straight lines constituting the quasi-periodic tile 8 can be optically identified by the imaging unit 20, the tile pattern may be drawn on the floor surface where the vehicle 2 travels using paint or the like, or a sheet with the tile pattern printed on it may be used as carpet. Alternatively, the pattern of the quasi-periodic tile 8 may be constructed by applying tape or the like to the floor surface, or by laying tiles of multiple shapes on the floor surface. Alternatively, the pattern of the quasi-periodic tile 8 may be displayed on a display element such as a display installed on the floor surface, or the pattern of the quasi-periodic tile 8 may be projected onto the floor surface. In any case, the tiles are configured with a density sufficient to determine the position of the vehicle 2. That is, the quasi-periodic tiles are configured with a density (size) necessary to estimate the self-position of the vehicle 2, given the number of pixels of the imaging unit 20 mounted on the vehicle 2 and the distance from the floor surface.

Furthermore, the pattern of the quasi-periodic tile 8 does not necessarily need to be visible; for example, it may be constructed using invisible markers that emit light in response to wavelengths outside the visible light spectrum. In that case, the imaging unit 20 is composed of elements that detect the light rays of the invisible markers.

[Configuration of an autonomous vehicle]
Next, we will explain the configuration of vehicle 2 using Figure 2. Figure 2 is a diagram showing the configuration of autonomous vehicle 2.

Vehicle 2 consists of an autonomous driving control device 3 and a driving device 6. The autonomous driving control device 3 estimates the vehicle's own position and direction of travel by photographing quasi-periodic tiles 8 and controls the operation of the driving device 6. The driving device 6 consists of a motor, tires, battery, steering mechanism, etc., and is the device that drives vehicle 2.

<Autonomous Driving Control System>
The autonomous driving control device 3 consists of an information processing device 5, an imaging unit 20, a gyro sensor 26, an acceleration sensor 27, and an odometry device 28.

The imaging unit 20 is a camera that captures images of the floor surface, such as quasi-periodic tiles 8, and generates (obtains) image data. The imaging unit 20 is fixedly installed in the vehicle 2, facing the floor surface, and captures images of the floor surface. The offset (relative position) between any pixel on the image data obtained by the imaging unit 20 and the reference position of the vehicle 2 is assumed to have been measured and calibrated in advance. The imaging unit 20 may be a monocular camera, a stereo camera, or an omnidirectional imaging camera equipped with a fisheye lens. Furthermore, the imaging unit 20 may consist of a single camera, or multiple cameras may be used, each responsible for capturing different areas. In addition, the imaging unit 20 may be a remote sensing device other than a camera, such as a LiDAR.

The information processing device 5 is a computer such as a PC (Personal Computer). Based on the position and orientation of the camera unit 20 installed on the vehicle 2, and data related to the image acquired from the camera unit 20, it measures the relative position of the vehicle 2 with respect to feature points on the image (edges and intersections of the quasi-periodic tiles 8). It estimates the vehicle 2's own position (absolute position) and direction of travel, and also sends control signals to the running gear 6 to control its operation. The information processing device 5 will be described in detail later.

In this embodiment, the autonomous driving control device 3, having an imaging unit 20 and an information processing device 5, can perform autonomous driving by reading the quasi-periodic tiles 8. However, even if image data cannot be obtained temporarily due to a failure of the imaging unit 20, autonomous driving can continue using sensors, etc., through dead reckoning. Furthermore, the sensors, etc., provide robustness to the self-position and orientation estimation operations of the vehicle 2 during autonomous driving, in relation to the local periodicity and rotational symmetry of the pattern of the quasi-periodic tiles 8. However, the parameters of the gyro sensor 26, acceleration sensor 27, and odometry 28 must be calibrated before a failure of the imaging unit 20 occurs during the initialization driving stage (Figure 5), which will be described later.

Here, the gyro sensor 26 is a sensor that detects the angle, angular velocity, or angular acceleration of an object (in this case, the vehicle 2). The acceleration sensor 27 is a sensor that measures the acceleration of an object (in this case, the vehicle 2). The odometry device 28 is a device that, for example, measures the rotation speed of the vehicle 2's wheels using a rotary encoder, and from the rotation speed of the wheels, determines the distance traveled, speed, and turning angle of the vehicle 2. The odometry device 28 is a device for further improving the accuracy of self-position estimation and is a device that complements the gyro sensor 26 and acceleration sensor 27, so it does not necessarily have to be provided in the autonomous driving control device 3. In other words, the odometry device 28 allows the vehicle 2 to improve the accuracy and reliability of its self-position estimation.

<Information Processing Device>
Next, we will explain the information processing device 5 in detail.

(Hardware configuration of information processing equipment)
The hardware configuration of the information processing device will be explained using Figure 3. Figure 3 is an electrical hardware configuration diagram of the information processing device 5.

The information processing device 5, as a computer, includes a CPU (Central Processing Unit) 501, ROM (Read Only Memory) 502, RAM (Random Access Memory) 503, SSD (Solid State Drive) 504, external device connection I/F (Interface) 505, network I/F 506, media I/F 509, and bus line 510, as shown in Figure 3.

Of these components, the CPU 501 controls the operation of the entire information processing unit 5. The ROM 502 stores programs used to drive the CPU 501, such as the IPL (Initial Program Loader). The RAM 503 is used as the work area for the CPU 501.

The SSD 504 reads or writes various types of data according to the control of the CPU 501. Alternatively, an HDD (Hard Disk Drive) may be used instead of the SSD 504.

The External Device Connection I/F 505 is an interface for connecting various external devices. These external devices include displays, speakers, keyboards, mice, USB (Universal Serial Bus) memory, and printers.

Network I/F 506 is an interface for data communication via the Internet and/or a communication network such as a LAN (Local Area Network). Alternatively, wireless communication equipment may be used instead of Network I/F 506.

The media interface 509 controls the reading or writing (storage) of data to or from the recording medium 509m, such as flash memory. The recording medium 509m includes DVDs (Digital Versatile Discs) and Blu-ray Discs (registered trademarks), among others.

The bus line 510 is an address bus, data bus, etc., for electrically connecting the various components, such as the CPU 501 shown in Figure 3.

(Functional configuration of information processing equipment)
Next, we will return to Figure 2 and explain the functional configuration of the information processing device 5.

The information processing device 5 includes a route data setting unit 31, a self-position estimation unit 33, and a control unit 35. Each of these units performs functions implemented by instructions from the CPU 501 according to a program stored in the RAM 503 or the like. Furthermore, the RAM 503 or SSD 504 of the information processing device 5 contains a route data storage unit 21, an image data storage unit 22, and a map data storage unit 23.

Of these, the route data storage unit 21 stores the data of the travel route set by the route data setting unit 31. The route data is given, for example, as a set of coordinate values for the start point, end point, and waypoints on the floor surface.

The image data storage unit 22 stores data relating to each image obtained by continuously capturing images by the imaging unit 20 while the vehicle 2 is in motion. The image data stored in the image data storage unit 22 may be the image data itself, or it may be data with a smaller data capacity (for example, polygon data obtained by extracting the shape of tiles from the image data, or data recording the arrangement of a finite number of identified tiles).

The map data storage unit 23 stores map data (information) with coordinate data of the floor drawing data constituting the quasi-periodic tiles 8, and is referenced during processing in the self-position estimation unit 33.

The route data setting unit 31 receives the vehicle 2's travel route setting from the user and stores the data of the set travel route in the route data storage unit 21.

The self-position estimation unit 33 estimates the vehicle's own position (current position) and direction of travel (direction of movement) by matching (pattern matching) image data stored in the image data storage unit 22 with map data stored in the map data storage unit 23. In the event of an emergency due to a malfunction of the imaging unit 20 or other components, as described above, the self-position estimation unit 33 acquires output signals from at least the gyro sensor 26 and acceleration sensor 27 among the gyro sensor 26, acceleration sensor 27, and odometry 28, and uses them in the emergency as data for the vehicle 2's direction of travel and speed.

The control unit 35 generates a control signal to control the running gear 6 based on the data acquired from the self-position estimation unit 33 (estimated values of the vehicle 2's self-position and direction of travel) and the travel route data acquired from the route data storage unit 21, and sends it to the running gear 6. The vehicle 2 may also be equipped with a camera unit (camera unit 20 or another camera unit), LiDAR, RADAR, etc., for detecting obstacles in the direction of travel of the vehicle 2. If an obstacle is detected, the route data setting unit 31 sets a new route in real time to avoid the obstacle and stores the new route data in the route data storage unit 21.

Furthermore, all or part of the functions of the information processing device 5 may be installed in a location other than the autonomous vehicle 2, for example, on a cloud infrastructure, via a communication network.

[Processing or operation of autonomous vehicles]
Next, the processing or operation of vehicle 2 will be explained using Figures 5 to 8. Figure 5 is a plan view showing the driving pattern of the initialization run. Figure 6 is a flowchart showing the processing of the initialization run. Figure 7 is a plan view showing the driving route of the main run. Figure 8 is a flowchart showing the processing of the main run.

The map data storage unit 23 stores map data having a coordinate system corresponding to a plan view of the floor surface having the pattern of quasi-periodic tiles 8. The imaging unit 20 is fixed to the vehicle 2 and continuously images the floor surface in the vicinity of the vehicle 2 while the vehicle 2 is in motion. Unlike the conventional technology, the vehicle 2's self-position estimation unit 33 does not extract feature points from the entire image data of the floor surface in the vicinity of the vehicle 2 that has been captured, but rather extracts each shape of the quasi-periodic tiles 8, which are composed of straight lines and intersections, from the floor surface image data, and uses image data stored in the image data storage unit 22 to identify and determine which of the three types of tile shapes it corresponds to through pattern matching.

First, before the vehicle 2 begins its actual run, it performs an initialization (preparation) run to determine its own position and orientation. For example, the user sets a running pattern for the initialization run using the route data setting unit 31. The running pattern for the initialization run is, for example, as shown in Figure 5, a pattern in which the vehicle 2 runs straight from the initial state, detects the boundary (edge) of the quasi-periodic tile 8, and then runs clockwise along the boundary. Alternatively, a running pattern in which the vehicle moves randomly on the quasi-periodic tile 8 during the initial run may be used. The data of the set running pattern for the initialization run is stored in the route data storage unit 21.

Furthermore, for the actual operation of vehicle 2, for example, the user sets the route for the actual operation, as shown by the arrow in Figure 7, as a movement route (path) on the floor surface where the quasi-periodic tiles 8 are constructed, using the route data setting unit 31. The data of the set route for the actual operation is stored in the route data storage unit 21.

After these preparations are complete, the following processes will be executed.

<Initialization process or operation>
First, vehicle 2 uses the data of an initialization driving pattern, with the initial state being an arbitrary position and orientation set by the user on the quasi-periodic tile 8, to perform the initialization driving process or operation shown below (Figure 6). As mentioned above, if vehicle 2 detects an obstacle, it may avoid the obstacle and reconfigure the route.

S11: As vehicle 2 moves, the imaging unit 20 captures images of the patterns of the quasi-periodic tiles 8 and stores the data related to the images in the image data storage unit 22. This allows vehicle 2 to store, in chronological order, the changes in the patterns composed of the three types of tiles on the image as vehicle 2 moves.

S12: The self-position estimation unit 33 extracts the pattern of the quasi-periodic tile 8 from the image data read from the image data storage unit 22.

S13: The self-position estimation unit 33 estimates the vehicle 2's own position (absolute position) and direction of travel by comparing the extracted pattern with the pattern of the map data stored in the map data storage unit 23 (pattern matching). The quasi-periodic tile 8 does not have periodicity in its pattern, but it has local periodicity and rotational symmetry. Therefore, by having the vehicle 2 travel a sufficient range during the initialization run, the ambiguity of the position can be resolved, and the vehicle 2's own position can be identified as an absolute position (coordinates) on the map. Furthermore, once the position on the map is identified, the vehicle 2 can continuously estimate its own position and direction of travel by continuously photographing and tracking the tile patterns while traveling.

S14: The self-position estimation unit 33 continuously captures and tracks the patterns of the quasi-periodic tiles 8 to estimate its own position and direction of travel, and each time this is done, it performs calibration (adjustment, calibration, adjustment) of the parameters of the gyro sensor 26, acceleration sensor 27, and odometry unit 28. These parameters may be stored by the gyro sensor 26, acceleration sensor 27, and odometry unit 28, respectively, or they may be stored in a separate parameter storage unit constructed with RAM 503 or the like. Calibration of the gyro sensor 26 and acceleration sensor 27 includes initialization of parameters in the initial (stationary) state and optimization of parameters such as the extended Kalman filter in the moving state. Calibration of the odometry unit 28 includes optimization of the coefficient used to calculate the speed of the vehicle 2 from the rotation speed of the vehicle's wheels obtained by the encoder.

S15: The self-position estimation unit 33 determines whether the initialization run is complete based on the estimated state of the vehicle 2's self-position and direction of travel, as well as the calibration status. If the initialization run is not complete, the process returns to S11. On the other hand, if the initialization run is complete, the process proceeds to S21 below. Note that the calibration of each parameter may be performed after the initialization run is complete.

Furthermore, in steps S11 to S13, as an example of handling data related to the image rather than the image data itself, there is a method of estimating the vehicle 2's own position and direction of travel on the quasi-periodic tile 8 by focusing on the intersection points of each tile on the quasi-periodic tile 8 (projected by the image data), referring to map data (pattern matching) using the number of line segments constituting the intersection point and the pattern of angles formed by these line segments as clues, and identifying which intersection point on the quasi-periodic tile 8 it is. Point A in Figure 5 is an intersection point where nine line segments intersect, and the angles formed by each line segment at point A are a: 30°, b: 60°, c: 90°, d: 120°, e: 150°, resulting in baaabbaaa in a clockwise direction. Point C is the intersection point on the quasi-periodic tile 8 that has the same angle cyclically arranged, but by detecting point B following point A in the direction of travel of the vehicle 2, it is possible to identify that the intersection point is point A. By performing an initialization run with a characteristic intersection area on the quasi-periodic tile 8 as the initial position, the time required to complete the initialization run can be shortened.

The initialization process is now complete.

<Processing or operation during actual running>
Next, vehicle 2 uses the route data for the actual drive (Figure 7) to perform the following processing or operations for the actual drive (Figure 8). As mentioned above, if vehicle 2 detects an obstacle, it may avoid the obstacle and configure a new route.

S21: As vehicle 2 moves, the imaging unit 20 captures images of the patterns of the quasi-periodic tiles 8 and stores the data related to the images in the image data storage unit 22. This allows vehicle 2 to store, in chronological order, the changes in the patterns composed of the three types of tiles on the image as vehicle 2 moves.

S22: The self-position estimation unit 33 extracts a pattern from the image data read from the image data storage unit 22.

S23: The self-position estimation unit 33 estimates the vehicle 2's own position (absolute position) and direction of travel by comparing (or matching) the extracted pattern with the map pattern stored in the map data storage unit 23.

S24: The control unit 35 compares the vehicle's own position and direction of travel with the route data stored in the route data storage unit 21 to generate a control signal and outputs this control signal to the driving device 6. Even during this driving, the self-position estimation unit 33 may, at any time, perform calibration of the parameters of the gyro sensor 26, acceleration sensor 27, and odometry 28 based on the estimated self-position and orientation.

S25: The self-position estimation unit 33 determines whether the destination has been reached based on the route data for this journey. If the destination has not been reached, the process returns to S21. On the other hand, if the destination has been reached, the process for this journey ends.

This completes the process for this run.

Furthermore, it is not necessarily required to perform the initialization run before the main run. After the main run, if vehicle 2 is parked at any parking position on the quasi-periodic tile 8 and the position and orientation data of vehicle 2 is stored, the system can immediately move to the starting point of the main run and begin the main run without performing the initialization run at startup.

As described above, the vehicle 2 of this embodiment can estimate its own position with high accuracy while suppressing processing load by extracting a combination pattern of simple-shaped tiles from image data, pattern matching it with the stored quasi-periodic tile 8 data, and obtaining absolute coordinates from the tile positions. In this case, the vehicle 2 can trace the relative displacement, i.e., the direction of movement and distance of movement, with high accuracy using the tile shape even when moving in any direction. Therefore, it can continuously estimate its own position and direction of travel, and by controlling it by referring to route data, it can drive accurately along a virtual path. Furthermore, it can construct a route to avoid obstacles in real time and avoid them accurately.

Furthermore, the quasi-periodic tile 8 has local two-dimensional rotational symmetry. For example, as shown in Figure 9, the portion within the dashed line of the quasi-periodic tile 8 has 12-fold symmetry. Therefore, if the region near the center of rotational symmetry in the radial direction is included in the vehicle 2's travel route, directional ambiguity must be considered. This ambiguity can be resolved by using directional data from the gyro sensor 26 in conjunction with the quasi-periodic tile 8.

[Examples of application of this embodiment]
As in this embodiment, the method in which the vehicle 2 moves while reading the quasi-periodic tiles 8 can be used for composite positioning in areas where GNSS (Global Navigation Satellite Systems) signals cannot be received, such as under elevated structures, when used in an outdoor environment, or for estimating the absolute position in an urban canyon reception environment. Specifically, in areas where it is difficult to receive GNSS signals, when performing composite positioning by fusing relative positioning means such as the imaging unit 20, LiDAR, and Radar with GNSS positioning, a pattern of quasi-periodic tiles 8 can be constructed on a part of the road (for example, an intersection) to eliminate the cumulative error of the relative positioning means. In the case of an automobile, when the vehicle 2 passes by, the imaging unit 20 can capture an image of the tile pattern of the quasi-periodic tiles 8 to estimate the absolute position of the vehicle 2.

[Effects of the Embodiment]
As described above, according to this embodiment, even when there are various changes in the surrounding environment such as pedestrians, moving vehicles, and the opening and closing of doors along the vehicle 2's travel path, the vehicle 2 can estimate its own position with high accuracy while suppressing the processing load, and autonomous driving to the destination is possible by reading the pattern of the quasi-periodic tiles 8 on the floor surface as it moves.

Furthermore, according to this embodiment, the vehicle 2 can be operated with flexible route setting without using a derivative. Therefore, even when the vehicle 2 temporarily avoids an obstacle during autonomous driving, it can quickly return to the route from any position.

Furthermore, in this embodiment, where the vehicle 2 moves while reading the quasi-periodic tiles 8, unlike the Visual SLAM method, brightness is not required to recognize the entire environment in which the vehicle 2 travels; local brightness sufficient to identify the floor pattern near the vehicle 2 is sufficient. Also, unlike the SLAM method, it is not affected by changes in the surrounding environment, and therefore processing and operation are not affected by changes in the positions of other unmanned vehicles or workers. Moreover, it can be applied to indoor or outdoor environments with large floor areas and few feature points. Furthermore, objects (such as luggage) placed on the floor surface in which the vehicle 2 travels can be moved at any time, and the floor layout can be dynamically changed.

〔supplement〕
As described above, the present invention is not limited to the embodiments described above, and various modifications and applications are possible, for example, as shown below.

(1) The information processing device 5 can be implemented using a computer and a program, but this program can also be recorded on a (non-temporary) recording medium or provided via a communication network such as the Internet.

(2) The CPU 101 (microprocessor) may be a single unit or multiple units.

(3) The vehicle 2 of the above embodiment is an example of an autonomous mobile body capable of autonomous movement. Autonomous mobile bodies also include ships, aircraft, and underwater vehicles (for example, a mobile body that detects cracks in a swimming pool). The autonomous driving control device 3 is similarly an example of an autonomous mobile control device. That is, in this specification, the word "driving" can be replaced with "moving". In addition, the driving device 6 may be moved by a jet engine or propeller, etc., instead of tires.

1. Autonomous driving system (an example of an autonomous mobility system)
2. Autonomous vehicles (an example of an autonomous mobile device)
3. Autonomous Driving Control System (An example of an autonomous mobility control system)
5. Information processing device 6. Traveling device (an example of a mobile device)
8 Quasi-periodic tile 20 Imaging unit 21 Route data storage unit 22 Image data storage unit 23 Map data storage unit 26 Gyro sensor 27 Acceleration sensor 28 Odometry 31 Route data setting unit 33 Self-position estimation unit 35 Control unit

Claims (7)

An autonomous mobile control device that controls the movement of an autonomous mobile body capable of moving autonomously,
A self-position estimation unit estimates the self-position and direction of movement of the autonomous mobile body on the quasi-periodic tiles by pattern matching data related to an image obtained by photographing quasi-periodic tiles composed of multiple types of tile shapes with map data having a coordinate system corresponding to the quasi-periodic tiles.
A control unit controls the movement of the autonomous mobile body based on the self-position and direction of travel estimated by the self-position estimation unit, and the predetermined movement route of the autonomous mobile body.
A gyro sensor for detecting the angle, angular velocity, or angular acceleration of the autonomous mobile body, and an acceleration sensor for measuring the acceleration of the autonomous mobile body,
It has,
The self-position estimation unit is an autonomous mobile control device that calibrates the parameters of the gyro sensor and the accelerometer based on the estimated self-position and direction of travel . The autonomous mobile control device according to claim 1, wherein, when the self-position estimation unit cannot acquire data relating to the image , it estimates the self-position and direction of travel of the autonomous mobile body using the gyro sensor and the acceleration sensor after calibration, instead of pattern matching. The autonomous mobile entity is an autonomous vehicle having wheels,
The autonomous mobile control device has an odometry function that determines the distance traveled, speed, and rotation angle from the rotation speed of the wheels.
The autonomous mobile control device according to claim 1 , wherein the self-position estimation unit calibrates the odometry parameters based on the estimated self-position and direction of travel. The autonomous mobile control device according to claim 1, wherein the self-position estimation unit patterns-matches the map data based on the number of line segments constituting the intersection points of each quasi-periodic tile and the pattern of angles formed by those line segments, as displayed in the image data, and identifies which intersection point it is on the quasi-periodic tile, thereby estimating the self-position and direction of travel of the autonomous mobile object on the quasi-periodic tile. The autonomous mobile body equipped with the autonomous mobile control device according to any one of claims 1 to 4 ,
The aforementioned quasi-periodic tile and,
An autonomous mobile system having An autonomous movement control method performed by an autonomous movement control device that controls the movement of an autonomously moving autonomous mobile body,
The autonomous mobile control device is
A self-position estimation process is performed to estimate the self-position and direction of movement of the autonomous mobile body on the quasi-periodic tiles by pattern matching data related to an image obtained by photographing quasi-periodic tiles composed of multiple types of tile shapes with map data having a coordinate system corresponding to the quasi-periodic tiles.
A control process that controls the movement of the autonomous mobile body based on the self-position and direction of travel estimated by the self-position estimation process, and the predetermined movement route of the autonomous mobile body,
The process involves detecting the angle, angular velocity, or angular acceleration of the autonomous mobile object using a gyro sensor, and measuring the acceleration of the autonomous mobile object using an acceleration sensor.
Execute,
The self-position estimation process is an autonomous movement control method that calibrates the parameters of the gyro sensor and the acceleration sensor based on the estimated self-position and direction of travel . A program that causes a computer to perform the method described in claim 6.