JP7623274B2 - Map utilization device and program - Google Patents

Description

The present invention relates to a map utilization device and a program that effectively utilizes map information in relation to a mobile robot and a map utilization environment.

Previously, technology has been considered that allows multiple robots to share a map for robot movement.

In Patent Document 1, in order to efficiently create a huge environmental map covering a wide area, multiple autonomous robots detect the relative positions of the robots and divide and define search areas for each robot, collect map information for the search area assigned to each robot using SLAM (Simultaneous Localization and Mapping) positioning, and then combine the map information for each search area to generate the entire environmental map.

In Patent Document 2, in response to the problem that "when an autonomous mobile robot moves autonomously while estimating its own position by referring to a created environmental map, if there are parts of the environmental map with low reliability, the mobile robot will not be able to accurately recognize its current location," the document attempts to change the environment to increase reliability by placing objects in parts with low reliability, for example by "sticking an opaque adhesive to a transparent glass door."

In Patent Document 3, in multiple robots that perform SLAM positioning, the map information that each robot possesses is shared between the robots via landmark information, thereby improving the positioning accuracy of each robot without each robot having to accumulate large amounts of map information individually.

JP 2013-109325 A JP 2018-185767 A JP 2019-57312 A

However, in the conventional technologies described above, no consideration was given to efficiently utilizing map information for robots as autonomous mobile devices in situations where the specifications for each robot's positioning and movement processing, as well as the driving environments, are diverse.

Normally, robots are taught how to understand the environment in which they will be operating by running the target robot thoroughly in advance and creating a map specific to each robot. However, by having multiple robots share a map, it is expected that the effort required to run each robot in advance can be reduced.

Because the shared map is not specific to the robot, there may be inconveniences when using the shared map. For example, robots vary in the type of sensors they use to recognize the outside world (3D LiDAR, 2D LiDAR, camera, etc.), the height at which the sensors are installed, and the shape and movement of the robot. As a result, the field of view in which obstacles can be detected differs depending on the robot, and some robots may not be able to detect the obstacle and may collide with it, or the robot may not be able to navigate over an obstacle depending on the uneven shape of the robot itself and the way it moves.

All of Patent Documents 1 to 3 concern the use of a shared map that assumes that robots operating in a factory, for example, are uniform robots with the same specifications. When there are multiple robots with various different specifications, the shared map cannot be used efficiently.

In view of the problems with the conventional technology described above, the present invention aims to provide a map utilization device and program that can efficiently utilize map information in situations where there are various possible specifications for mobile robots.

To achieve the above object, the present invention is a map utilization device or a program that causes a computer to function as such a device, which distributes map information along a specified route in order to move a mobile robot along the route, and is characterized in that it acquires 3D map information along the route as being in the vicinity of the route from the entire 3D map information constructed for the field in which the mobile robot can move, converts the 3D map information along the route into 2D map information taking into account the sensing range of an external recognition sensor when the mobile robot moves autonomously, complements map information outside the sensing range, and distributes it to the mobile robot.

According to the present invention, three-dimensional map information is converted into two-dimensional map information taking into account the individual sensing range of each mobile robot, and map information outside the sensing range is supplemented, making it possible to use a common map while still using appropriate map information according to the individual specifications of each mobile robot.

1 is a diagram illustrating a configuration of a map using system according to an embodiment. 1 is a functional block diagram of a user terminal, a mobile robot, and a map using device in a map using system according to an embodiment. FIG. 13 is a diagram showing example images of obstacles that may not be detectable by individual mobile robots. FIG. 4 is a functional block diagram of a complementing unit according to an embodiment. 10 is a flowchart showing information management of each mobile robot by a map using device according to an embodiment. 11 is a flowchart of arranging for product delivery and updating map information involving a map completion process by a map using device according to an embodiment. 10 is a diagram showing an example of device information stored in a device information storage unit; FIG. FIG. 13 is a diagram showing an example of an image in which passable obstacles are removed, as a schematic example of the complementation process. As a schematic example of the complementation process, an image example of changing the shape of an obstacle is shown. FIG. 13 is a diagram showing a schematic example of a flapper gate as a schematic example of the complementation process. 1A to 1C are diagrams showing examples of a two-dimensional map and an operation plan based on the map, as a schematic example of the complementation process. 1A to 1C are diagrams showing examples of a two-dimensional map and an operation plan based on the map, as a schematic example of the complementation process. FIG. 1 is a diagram illustrating a hardware configuration of a typical computer.

FIG. 1 is a diagram showing the configuration of a map usage system 4 according to one embodiment. In this embodiment, as an example of a use of the map usage system 4, a case where an autonomous robot moves through a field F, such as a city, to a destination specified by a delivery service user (user) and delivers goods will be described, but the map usage system 4 of this embodiment can be applied to any use that involves moving an autonomous robot along a specified route within the field F.

The map usage system 4 comprises multiple (n) user terminals 1-1, 1-2, ..., 1-n used to determine the destination, multiple (m) mobile robots 2-1, 2-2, ..., 2-m that travel to the destination by traveling through a field F such as a city, and a map usage device 3 that stores 3D maps for the mobile robots, determines whether to use the 3D map or a map converted to 2D for each mobile robot 2-1 to 2-m, and transmits the determined map, and performs operational monitoring and navigation to the destination. These elements that make up the map usage system 4 can communicate with each other via a network NW such as the Internet.

Figure 2 is a functional block diagram of a user terminal 1, a mobile robot 2, and a map utilization device 3 in a map utilization system 4 according to one embodiment. Note that the user terminal 1 and the mobile robot 2 in Figure 2 each represent the functional configuration of any one of the n user terminals 1-1 to 1-n and the m mobile robots 2-1 to 2-m in Figure 1, and represent the common functional configuration of these multiple user terminals 1-1 to 1-n and mobile robots 2-1 to 2-m.

However, although the functional configuration of the multiple user terminals 1-1 to 1-n and mobile robots 2-1 to 2-m is the same as that shown in FIG. 2, the hardware specifications (performance specifications) of the sensing mechanism for estimating the self-position and the mobile mechanism of the autonomous mobile robot may be different from each other. In particular, even if the specifications of the m mobile robots 2-1 to 2-m (each of the mobile robots 2) are variously different, the map utilization system 4 of this embodiment makes it possible to effectively utilize map information taking into account the resources of each mobile robot 2.

The user terminal 1 is, for example, a smartphone, into which a delivery service user inputs information about a destination when ordering a product. The user terminal 1 includes a product ordering unit 101 and a destination information input unit 102.

The mobile robot 2 is an autonomous robot that delivers packages to a destination specified by a delivery service user, for example. The mobile robot 2 includes a wireless communication unit 201, a current position acquisition unit 202, a status acquisition unit 203, an imaging unit 204, a device information storage unit 205, a route information storage unit 206, a map information storage unit 207, an operation planning unit 208, and an operation control unit 209.

The map utilization device 3 is, for example, a server device that exists on the cloud, which stores 3D maps for the mobile robot 2, converts them into 2D maps with appropriate content for each of the mobile robots 2-1 to 2-m, and also performs the process of complementing and transmitting obstacle information on the 2D maps, as well as performing operational monitoring and navigation to the destination. Note that the computational load when estimating the self-position is lower with a 2D map than when a 3D map is used, and therefore in this embodiment, the map is converted into a 2D map to enable self-position estimation even for mobile robots 2 that do not have abundant computational resources.

FIG. 3 shows an example of an image of an obstacle that may not be detected by an individual mobile robot 2. The detectable range determined by the height and vertical field of view of the camera (LiDAR) mounted on each mobile robot 2 for recognizing the surrounding environment may be exactly in the hollow part of the obstacle, and the obstacle may not be detected. For example, the rope 11 stretched on a pole or the barricade 12 indicating no entry shown in FIG. 3 may not be detected as an obstacle and may collide with the obstacle. That is, for the rope 11, one robot may have a detectable range R11a, which corresponds to the hollow range and therefore cannot be detected, while another robot may have a detectable range R11b and therefore can detect it. For the barricade 12, one robot may have a detectable range R12a, which corresponds to the hollow range and therefore cannot be detected, while another robot may have a detectable range R12b and therefore can detect it.

Therefore, in this embodiment, as described in detail below, obstacles that may not be detectable are virtually supplemented on the map in a manner that corresponds to the individual specifications of the mobile robot 2, allowing the mobile robot 2 to move while avoiding the obstacles.

The map device 3 includes an operation monitoring unit 301, an order information storage unit 302, a destination acquisition unit 303, a three-dimensional map information storage unit 304, a route search unit 305, a route distribution unit 306, a two-dimensional map conversion unit 307, a completion unit 300, and a map distribution unit 313.

Figure 4 is a functional block diagram of the completion unit 300 according to one embodiment, which further includes an obstacle identification unit 308, a detection range calculation unit 309, a difficult-to-detect obstacle extraction unit 310, a passable obstacle exclusion unit 311, and a virtual obstacle completion unit 312. As will be described in detail later, the completion unit 300 performs an interpolation process on the route obtained by the route search unit 305 and the map obtained by the two-dimensional map conversion unit 307 through the processing of each of these functional units, and outputs the interpolated route and map to the route distribution unit 306 and the map distribution unit 313, thereby realizing the distribution of this route and map to the corresponding mobile robot 2.

Fig. 5 is a flowchart showing how the map using device 3 manages information about each of the mobile robots 2 according to one embodiment, and Fig. 6 is a flowchart showing how the map using device 3 arranges for product delivery and updates map information with map completion processing according to one embodiment. The flowcharts in Figs. 5 and 6 can be executed in parallel by the map using device 3. According to the flow in Fig. 5, the map using device 3 periodically grasps and manages the position information, etc., of each of the mobile robots 2, and can collect the latest three-dimensional map information measured by the mobile robots 2 in a field F, such as a city. According to the flow in Fig. 5, the map using device 3 updates the three-dimensional map information to the latest state, and when a delivery request is received, determines a delivery route, etc., and determines which of the multiple mobile robots 2 will be in charge of the delivery task, and can instruct the mobile robots 2 to deliver the product along the delivery route by transmitting an appropriate two-dimensional map by interpolation processing.

Below, we will explain the details of the processing content of each functional block of the map utilization system 4 shown in Figures 2 and 4 while explaining each step in the flow in the order of Figures 5 and 6.

When the flow in FIG. 5 is started, in step S31, the mobile robot 2 determines whether the current time corresponds to the timing for transmitting its own position information, etc., and if so, proceeds to step S32, and if not, proceeds to step S33. This transmission timing may be, for example, every time a certain period of time has elapsed, and position information, etc. can be transmitted and received in real time at certain intervals.

<Step S32...Current position acquisition unit 202, status acquisition unit 203, operation monitoring unit 301>
In step S32, the current position acquisition unit 202 (and the status acquisition unit 203) processes the mobile robot 2 to transmit its own position information at the current time, etc. to the map using device 3, and the map using device 3 receives this position information and stores and manages it through the processing of the operation monitoring unit 301, after which the process proceeds to step S33.

That is, the current position acquisition unit 202 acquires the current position of the mobile robot 2 and transmits this position information to the operation monitoring unit 301. For this reason, the current position acquisition unit 202 acquires the current position using a positioning system based on any existing method, such as a GPS (Global Positioning System) or an indoor GPS.

The status acquisition unit 203 may also acquire the current status of the mobile robot 2 and transmit this status information to the operation monitoring unit 301. For this reason, the status acquisition unit 203 acquires status data such as remaining battery power, temperature, movement speed, etc. from various sensors equipped in the mobile robot 2.

Furthermore, the operation monitoring unit 301 monitors the operation of each mobile robot 2. The operation monitoring unit 301 can perform operation monitoring by linking the position information and status information received in real time from each mobile robot 2 with information that identifies the individual mobile robot 2. The operation monitoring unit 301 analyzes the received position information and status information, and may issue an alert to a system monitor if a malfunction occurs in a mobile robot 2.

Note that, not only in step S32 but also in each step of Figs. 5 and 6, the wireless communication unit 201 is responsible for transmitting and receiving information via wireless communication between the mobile robot 2 and the map utilization device 3. Here, the wireless communication unit 201 can perform wireless communication via a base station compatible with wireless communication methods such as 4G (fourth generation mobile communication system) and 5G (fifth generation mobile communication system) that are currently in practical use, and 6G (sixth generation mobile communication system) that is expected to be practical in the future.

Although not explicitly shown as a functional block, the map usage device 3 and the user terminal 1 also have functions similar to those of the wireless communication unit 201 in the mobile robot 2, and can transmit and receive information required in each embodiment of each step in Figures 5 and 6 between the map usage device 3 and the user terminal 1, and between the map usage device 3 and the mobile robot 2, via wireless communication.

<Steps S33, S34...Transmission of map information>
In step S33, the mobile robot 2 determines whether the current time corresponds to the timing for transmitting map information for itself, and if so, proceeds to step S34, and if not, returns to step S31. In step S34, the mobile robot 2 transmits map information to the map using device 3, and then returns to step S31. The map information may be obtained in the form of three-dimensional point cloud data or the like (a data format that can be obtained by the imaging unit 204) obtained by SLAM or the like when the mobile robot 2 performs autonomous movement, which will be described later, and can be transmitted to the map using device 3.

Note that any existing method for keeping 3D map information up to date may be used for acquiring this map information and updating the map in the map using device 3 in step S45 of FIG. 6 described later, and FIG. 2 does not show functional blocks for performing specific processes for these map updates. For example, the timing for transmitting the map information in step S33 may be when it is determined that the map information acquired by the mobile robot 2 contains information that has changed from the map information received from the map using device 3 (for example, when there is an object that is estimated to be a new object), etc.

Each step in FIG. 5 has been described above. The flow in FIG. 5 is executed by each of the multiple mobile robots 2 (each of the m mobile robots 2-1 to 2-m in FIG. 1) for the map utilization device 3. In this way, the map utilization device 3 monitors all information on the m mobile robots 2-1 to 2-m in real time in the operation monitoring unit 301, and can keep the three-dimensional map information (stored in the three-dimensional map information storage unit 304, described later) up to date using map information sent appropriately from the m mobile robots 2-1 to 2-m as the latest measurement results of the field F. In this way, the map utilization device 3 grasps all of the multiple mobile robots 2 in real time through the flow in FIG. 5, and while keeping the three-dimensional map information up to date, it is possible to generate an appropriate two-dimensional map from the three-dimensional map according to each mobile robot 2 through the flow in FIG. 6, and the map utilization device 3 can arrange for product delivery.

<Step S41...3D map information storage unit 304, device information storage unit 205>
6 is started, in step S41, as a pre-processing required to enable product delivery, the map utilization device 3 constructs map information to store three-dimensional map information of the field F in the three-dimensional map information storage unit 304, and the complementing unit 300 receives and stores the equipment information stored in the equipment information storage unit 205 of each mobile robot 2, and then the flow proceeds to step S42. (Note that the complementing unit 300 uses the equipment information stored in step S41 to perform complementing processing according to the delivery route and the specifications of the mobile robot 2 in step S43, which will be described later. As will be described later with reference to the example of FIG. 7 regarding the equipment information storage unit 205, the equipment information is configured as information on the size of the mobile robot 2, such as the height and width, the range that the mobile robot 2 can detect with its sensor, and the mobility of the mobile robot 2, such as the height that the mobile robot 2 can overcome.)

The three-dimensional map information storage unit 304 stores nationwide road map data and associated facility data such as various facilities and stores in three dimensions as map information of the field F in which the mobile robot 2 operates. The three-dimensional map information storage unit 304 also stores indoor three-dimensional map data related to indoor facilities and stores to which the mobile robot 2 may move. The three-dimensional map information storage unit 304 may be, for example, three-dimensional point cloud map information created by the mobile robot 2 using SLAM (Simultaneous Localization and Mapping), a technology that simultaneously estimates the self-position and creates an environmental map. The three-dimensional map information may also be in a different format, such as an occupancy grid or a 3D model, or may be three-dimensional map information created by another moving body such as an automobile.

The three-dimensional map information storage unit 304 further links the stored three-dimensional map information to information on a graph (discrete graph) for enabling route search. As is well known, a graph for route search is defined as edges connecting nodes, and can be expressed in a data format such as an adjacency matrix. The nodes are linked to the point coordinates (three-dimensional spatial coordinates) of corresponding positions in the three-dimensional map information, and similarly, the edges can be linked to spatial coordinates in the three-dimensional map information as straight lines or curves connecting adjacent nodes. (Note that when the three-dimensional map information is given as a point cloud, for example, the spatial coordinates of the nodes in the graph to be associated may or may not match any of the point clouds. For example, the center of gravity of the point cloud in a certain local spatial range may be assigned to the spatial coordinates of one node as a representative point representing this local spatial range.)

This graph linked to the three-dimensional map information is a model of the route network (such as the road network along which the mobile robot 2 can move) in the field F represented by the three-dimensional map information.

The three-dimensional map information to which the above-mentioned graph is linked and stored in the three-dimensional map information storage unit 304 in step S41 may be obtained automatically or manually by an administrator from a server or the like that manages a map information database on the network NW. In step S45 described later, the three-dimensional map information stored in the three-dimensional map information storage unit 304 can also be updated to the latest state.

The device information storage unit 205 stores information on the sensing range of the mobile robot 2 as one piece of device information. This sensing range can be stored in the form of, for example, height information on the sensor viewpoint position of a camera or LiDAR mounted on the mobile robot 2 and used as a sensor input for self-position estimation, or information on the vertical field of view detectable by the sensor. The height information in the device information storage unit 205 may be input in advance by a person, or the height of the camera or LiDAR may be estimated by determining the distance to a characteristic object captured by the imaging unit 204 and the size of the object.

Figure 7 shows a schematic example of this device information. For a certain mobile robot 2, for example, the height h of the camera viewpoint position p from the ground and the upper limit height h1(d) and lower limit height h2(d) of the vertical field of view at a position a distance d ahead from this viewpoint position (which are generally functions of distance d, but may also be constants such as h1(d)=h1, h2(d)=h2, etc.) can be used as information on the sensing range R that the sensor can recognize. The upper limit of the recognizable distance d may also be included in the sensing range information. Note that, as mentioned above, such sensing range information can exist in a variety of ways for each individual mobile robot 2.

As shown in the example of FIG. 7, the device information storage unit 205 may further store, in the device information, information about the movement capability of the mobile robot 2 in the form of the height that the mobile robot 2 can overcome and move, which is a value that depends on the wheel size, for example, when the mobile robot 2 moves on wheels, and information about the size of the mobile robot 2 in the form of the height Hr (and width Wr) of the mobile robot 2, in order to enable the passable obstacle exclusion unit 311, which will be described later, to make an exclusion decision.

<Step S42...Determination of whether processing is required by the product ordering unit 101 and the destination information input unit 102>
In step S42, when a product delivery request is sent from any of the user terminals 1, it is determined whether or not the map using device 3 has received this request at the current time. If the request has been received, the process proceeds to step S43; if the request has not been received, the process proceeds to step S44.

When the following processing is performed by the input operation of the user in the product ordering unit 101 and destination information input unit 102 of any of the user terminals 1 used by the user requesting product delivery, a product delivery request is sent from the user terminal 1 and received by the map utilization device 3, and a positive judgment is obtained in step S42.

The product ordering unit 101 allows a user, who is a delivery service user, to select a product using a smartphone or the like owned by the user and place a manual order. The product ordering unit 101 may also manage the remaining amount of items owned by the delivery service user in real time, and automatically order replenishment products when the remaining amount falls below a preset threshold.

The destination information input unit 102 inputs information about the destination, such as an address, when a product order is placed by the product ordering unit 101 as described above. The destination information input unit 102 may reuse previous input and omit input each time. The destination information input unit 102 may also use a GPS positioning function provided in the smartphone to set the current location acquired by the GPS as the destination. The destination information input unit 102 may also use surrounding scenic information captured by the smartphone camera as a reference for the destination.

The information on the product that is the subject of the delivery request and the information on the destination to which the product is to be delivered, obtained by the product ordering unit 101 and the destination information 102, are transmitted from the user terminal 1 to the map utilization device 3 and stored as order information in the order information storage unit 302 of the map utilization device 3.

In step S43, the map utilization device 3 arranges for product delivery in accordance with the order information for which a positive judgment was obtained in step S42, and the process proceeds to step S44. In accordance with this arrangement, the mobile robot 2 can dispatch the product. Details of the product arrangement in step S43, such as route calculation, map preparation, and selection of the mobile robot 2 to be in charge of product delivery, will be described later.

In step S44, the map using device 3 determines whether the current time corresponds to the timing for updating the map information, and if so, proceeds to step S45, and if not, returns to step S42. In step S45, the map using device 3 updates the three-dimensional map information (and the graph information linked thereto) stored in the three-dimensional map information storage unit 304, for example by reflecting the latest map information transmitted from the mobile robot 2, and then returns to step S42. As mentioned above, any existing method may be used to update the map information to the latest state, and FIG. 2 omits the description of the specific functional blocks of this update process.

In this way, according to the flow of FIG. 6, the map utilization device 3 determines at each real-time instant whether a delivery request has been received in step S42 and whether a map update is necessary in step S44. If a delivery request has been received, delivery arrangements are made, and if a map update is necessary, the map is updated, thereby enabling delivery arrangements to be made in accordance with the latest map.

In the following, for step S43, which will be described in detail later, we will explain the details of the delivery arrangement involving map interpolation processing by the map utilization device 3, as well as the details of product delivery by the mobile robot 2 in accordance with this delivery arrangement.

<Delivery Arrangements by Map Using Device 3 in Step S43>
Specifically, the map utilization device 3 can make delivery arrangements by processing each of the functional blocks as follows.

The order information storage unit 302 receives and stores order information for product orders from the product ordering unit 101 and the destination information input unit 102. The order information storage unit 302 works in conjunction with a purchasing system (not shown) to process purchases and develop a product delivery plan.

For example, the product delivery plan can define, for each mobile robot 2, its current location (current point) as the start point, the location where the product to be delivered specified in the order information is automatically or manually loaded onto the mobile robot 2 as the relay location (relay point), and the destination (destination location, destination point) to which the product specified in the order information is to be delivered as the end point. If the current location of the mobile robot 2 is already in, for example, a warehouse or store where the product to be delivered is located, the start point and the relay location may be deemed to match, and the relay location may be omitted.

The destination acquisition unit 303 acquires destination information from the order information stored in the order information storage unit 302.

The route search unit 305 searches for a driving route from the starting point (current position) of the mobile robot 2 to the destination acquired from the destination acquisition unit 303 based on the map information stored in the three-dimensional map information storage unit 304. The route search unit 305 searches for a driving route from the starting point to the destination using an algorithm such as Dijkstra's algorithm, A* algorithm, or genetic algorithm. The shortest route can be searched for by providing a cost according to the length of the edge (the physical distance of the straight line or curve on the three-dimensional map information corresponding to the edge).

At this time, if the mobile robot 2 is not loaded with the product to be delivered, two route searches can be performed: "start point → relay position" and "relay position → destination." Also, for each of the three points of the start point, relay position, and destination, a corresponding node in a discrete graph linked to the three-dimensional map information can be determined, and a route search can be performed on the discrete graph. The current position of the mobile robot 2 can be obtained from information monitored and grasped in real time by the operation monitoring unit 301.

The route distribution unit 306 transmits the route guidance data searched by the route search unit 305 to the mobile robot 2. This route guidance data is configured as data that traces adjacent nodes on the graph, using each edge on the graph as a route from the starting point to the destination (via any relay points).

This route guidance data is linked to two-dimensional map information that is delivered to the mobile robot by the map delivery unit 313 after completion processing by the completion unit 300, which will be described later. The linked map information is referenced to estimate the mobile robot's own position when the mobile robot 2 moves according to the route guidance data. (In other words, map information of the surrounding environment when moving according to the route guidance data is delivered by the map delivery unit 313.)

The target for route distribution by the route distribution unit 306 (and map distribution by the map distribution unit 313, described later) may be one of the multiple mobile robots 2-1 to 2-m that has been determined to be in charge of delivery operations. This one robot may be determined automatically by estimating the time required for delivery from the route searched by the route search unit 305 and selecting the robot with the shortest delivery time, or may be manually selected by an administrator or the like from among multiple robots whose estimated delivery times are within a threshold value.

The 2D map conversion unit 307 reads map information about the area around the route from the 3D map stored in the 3D map information storage unit 304 for the mobile robot 2 delivering goods and its route, and converts it into a 2D map according to the height of the field of view of the mobile robot 2. This height information can be the height of the sensor position (height h in the example of FIG. 7) received from the device information storage unit 205 as the height of the sensing range of each mobile robot 2.

Each position on the 3D map is given in advance a distinction between a movable part such as the ground and an immovable part such as a wall, as well as information on the height of the ground (for example, in the case of a building, the information is the height of the first floor above ground, not the height of the building). For each position (x, y) on the 2D map, if the visual field height h of the mobile robot 2 corresponds to a movable part such as the ground (the three-dimensional position obtained by adding the visual field height h to the ground height is space and there is no object), then it is considered movable, and if it corresponds to the inside of a wall or other immovable obstacle (the three-dimensional position obtained by adding the visual field height h to the ground height is inside an object), then it is considered immovable. In this way, the conversion result to the 2D map can be obtained by distinguishing between movable and immovable for each position (x, y).

The complementing unit 300 complements the two-dimensional map obtained from the two-dimensional map conversion unit 307 through the processing of each unit in FIG. 4. The specific complementing process is described below, and after dividing the route searched by the route search unit 305 into multiple local routes, it is possible to perform complementing process on each of the two-dimensional map information located around these local routes. The relationship between the processing performed by the route search unit 305, the two-dimensional map conversion unit 307, and the complementing unit 300, etc., and the relationship with the route movement by the mobile robot 2 that has received the route and map delivered by these processes is as follows (Step 1) to (Step 3).

(Step 1) The route search unit 305 performs route search for product delivery (in other words, route search on the upper level) using the process described above.

(Step 2) The entire map related to the route searched in (Step 1) above is converted into a two-dimensional map by the two-dimensional map conversion unit 307 using the above-mentioned process (including the addition of various associated information as exemplified in Figs. 11 and 12 described later). Note that Figs. 11 and 12 depict two-dimensional maps of a narrow range for the sake of simplicity, but the actual two-dimensional map will be a two-dimensional map of a wide range related to the route searched in (Step 1). This conversion into a two-dimensional map is basically performed at the viewpoint height h of the mobile robot as in the existing method, but in areas where the processing of the complementation unit 300 of this embodiment is applied, the conversion result is overwritten and associated information is added. This two-dimensional map is distributed to the mobile robot 2 by the map utilization device 3.

(Step 3) The mobile robot 2 divides the above route (route corresponding to the upper side) obtained by the route search unit 305 to obtain multiple local routes L_k (k=1, 2, ..., K) (examples of each local route are shown in Figs. 11 and 12 described later) on the two-dimensional map, together with their start points S_k and end points G_k. In other words, the end point G_k of the kth route L_k becomes the start point S_k+1 of the next k+1th route L_k+1. The start point S_k and end point G_k may use the nodes of the discrete graph when the route search is performed by the route search unit 305, or may be determined separately by route division on the mobile robot 2 side.

When the mobile robot 2 actually moves autonomously, it follows multiple local routes L_k in order, with the constraint that it must pass through the starting point S_k and the ending point G_k. The robot is responsible for determining the specific local route from the starting point S_k to the ending point G_k and for autonomous movement.

As described above, the completion process in the completion unit 300, which is responsible for the detailed processing of (step 2), is explained, assuming (step 1) to (step 3) related to the overall processing. Figures 8 to 11 are schematic examples of the completion process by the completion unit 300, and these examples will be referred to as appropriate in the following explanation.

The obstacle identification unit 308 identifies obstacles from the three-dimensional map information stored in the three-dimensional map information storage unit 304 as three-dimensional spatial ranges on the three-dimensional map information along the movement path of the mobile robot 2 from the starting point to the destination obtained from the path search unit 305 (each of the multiple local paths L_k (k=1, 2, ..., K) described above). The obstacle identification unit 308 extracts objects from the three-dimensional map information and identifies them as obstacles using, for example, a method based on features such as DPM (Deformable Part Model) or a method of learning the features themselves by deep learning such as CNN (Convolutional Neural Network).

In this case, the three-dimensional map information is assumed to be composed of a point cloud, for example, and image features of the points in the point cloud are extracted from a camera image capturing the position of the point cloud (corresponding to the surface of some object) and linked, and the features are then linked to the three-dimensional map information.

The detection range calculation unit 309 calculates the detectable range of each of the mobile robots 2-1 to 2-m from the height information and vertical field of view of the mounted camera and LiDAR (sensor used for self-location estimation). (Note that this height and vertical field of view are acquired from the device information storage unit 205.) The detection range calculation unit 309 calculates the detectable range under the condition that the mobile robot 2 is a specific distance away from the obstacle, for example, a distance of 3 m. In addition, since the detectable range of the mobile robot 2 differs in places that are not horizontal but have an incline, the detectable range calculation unit 309 may calculate the detectable range taking into account the incline information for each place stored in the 3D map information storage unit 304.

This detectable range can be calculated as a spatial range along a route on the three-dimensional map information by constructing three-dimensional trajectory candidates on the three-dimensional map information that move at the height (and orientation) of the sensor viewpoint of the mobile robot 2 according to the route information. In other words, for each of the multiple local routes L_k, there are various candidate routes when specifically moving from its starting point S_k to its end point G_k, and the detectable range can be obtained by constructing a three-dimensional trajectory (the trajectory of the three-dimensional sensing range when the mobile robot 2 moves) for each of these candidate routes at the height of the sensor starting point of the mobile robot 2.

The difficult-to-detect obstacle extraction unit 310 extracts, on the two-dimensional map information, obstacles that may not be detectable within the detection range of each mobile robot 2-1 to 2-m calculated by the detection range calculation unit 309 for each obstacle identified by the obstacle identification unit 308, by comparing the three-dimensional coordinates of the obstacles. (Note that the extraction result may include the three-dimensional obstacle coordinates that were originally compared.) The difficult-to-detect obstacle extraction unit 310 compares the spatial coordinate (x, y, z) range R _{[detection range]} expressing the space of the detection range calculated by the detection range calculation unit 309 with the spatial coordinate (x, y, z) range R _{[obstacle range]} indicating the shape of the obstacle in the three-dimensional map identified by the obstacle identification unit, and determines that there is a possibility that the obstacle cannot be detected if the (x, y, z) coordinates of the detection range calculated by the detection range calculation unit 309 are not included in the (x, y, z) range of the obstacle's shape, as three-dimensional coordinates having the same two-dimensional coordinates (x, y).

The difficult-to-detect obstacle extraction unit 310 extracts obstacles that have a cavity or are low in height from among the obstacles identified by the obstacle identification unit 308, and determines whether or not there is a possibility that the obstacle cannot be detected within the detection range of each of the mobile robots 2-1 to 2-m. The difficult-to-detect obstacle extraction unit 310 may treat obstacles made of a material that is difficult to detect as an object, such as transparent glass, as equivalent to a cavity.

As shown in the example of FIG. 7, the lower and upper limits of the height of the sensing range are h2 and h1. (From the viewpoint of holding the device information as simple model information, the lower and upper limits may be constants that are independent of the distance d.) The height range occupied by the obstacle at each position (x, y) of the two-dimensional map is R _{[obstacle range]} . (If there is a part in this obstacle height range R _{[obstacle range]} that is not continuous with the ground height, it can be determined that the obstacle has a cavity or is hollow.) If R _{[obstacle range]} ≠ φ (empty set) and the sensing range [h2, h1] and the obstacle range R _{[obstacle range]} do not overlap at the position (x, y) where the obstacle exists, it can be determined that there is a possibility that the obstacle cannot be detected. Cases in which there is no overlap include (Case 1) when the upper limit of R _{[obstacle range]} (height of the obstacle) is lower than the lower limit h1 of the sensing range, causing the mobile robot 2 to overlook an obstacle at its feet; and (Case 2) when the lower limit of R _{[obstacle range]} (the lower limit height when at least part of the obstacle is in the air) is higher than the upper limit h2 of the sensing range and the lower limit height of this obstacle is lower than the height Hr of the mobile robot 2, causing the mobile robot 2 to overlook an obstacle that may collide with it overhead.

The passable obstacle exclusion unit 311 excludes obstacles that are difficult to detect as determined by the difficult-to-detect obstacle extraction unit 310, which the mobile robot 2 can pass through without any problem even if it cannot be detected as an obstacle, for example, by climbing over or passing under them.

Whether the mobile robot 2 can get over can be determined when, in the above (Case 1), the upper limit of R _{[obstacle range]} (height of the obstacle) is lower than the height (stored as device information) that the mobile robot 2 can get over and move to. Whether the mobile robot 2 can pass underneath can be determined when the lower limit of R _{[obstacle range]} (lower limit height when the obstacle is in midair) is higher than the upper limit h2 of the sensing range and the lower limit height of this obstacle is higher than the height Hr of the mobile robot 2, which is a case where the above (Case 2) does not hold.

In the above cases 1 and 2, the height Hr and vertical field of view [h2, h1] of the mobile robot 2 are taken into consideration, and the detectability and oversight judgments are made based on height information alone. However, the width Wr and horizontal field of view of the mobile robot 2 may also be taken into consideration to make the detectability and oversight judgments in the horizontal direction. Furthermore, as in the example of Figure 8 below, these judgments may also be made taking into consideration margins such as allowances M1 and M2.

8 is an image example of excluding passable obstacles. The passable obstacle excluding unit 311 excludes small objects 13, etc., whose height Hs is equal to or less than the height that the mobile robot 2 can overcome, as objects that can be passed through without any problems. The passable obstacle excluding unit 311 also compares the height Hr and width Wr of the mobile robot 2 with the height Ho and width Wo of the hollow area of an object with a wide hollow area, such as a desk 14,
Ho>(Hr+Margin M1) and
Wo＞(Wr + margin M2)
If the height and width are within the limits, the cavity is deemed to be passable without problems and is excluded. Note that the margins M1 and M2 are values provided as margins because if the height and width are at the very limit, it will be difficult to pass through.

Depending on the running mechanism of the mobile robot 2, it is considered that special transformations and operations may be required to overcome the obstacle. For example, in the case of a bipedal robot, it is necessary to raise one's legs high and straddle the obstacle to overcome it. Therefore, it is desirable for the passable obstacle exclusion unit 311 to further manage the running mechanism of each mobile robot 2, and in the case of an obstacle of a step height that requires special transformation or operation to overcome even if the step height is surmountable according to the specifications, to first transform the shape of the obstacle stored in the 3D map information storage unit 304 into another 3D object, and then create a 2D map in the 2D map conversion unit 307.

Figure 9 is an image example of changing the shape of an obstacle. As mentioned above, if the detectable range of the camera (LiDAR) is range R11a or the like, the rope 11 stretched between the poles shown in Figure 9 may end up in the hollow part of the obstacle, and the obstacle may not be detected. In this case, by changing the shape of the rope 11 to a rectangular parallelepiped 21 with width Wp, depth Dp and height Hp based on the width Wp between the poles, depth Dp of the poles and height Hp of the stretched rope, it is expected that the mobile robot 2 will be able to detect the obstacle and climb over it.

That is, although the location is undetectable in the two-dimensional map, the complementing unit 300 can add accompanying information so that the robot can overcome the wall by assuming that it is a wall of height Hp (and width Wp and depth Dp). Because the location is undetectable, when the mobile robot 2 estimates its own position for autonomous movement along a route, it can perform additional processing by not using the sensing results of the undetectable location, but using only the surrounding locations that can be sensed. As the device information of the mobile robot 2, information on a first height Hp1 that can be overcome without any special action, and information on a second height Hp2 that is higher than the first height and can be overcome by a special action such as a predetermined dedicated transformation can be stored.

If "Hp1<Hp<Hp2" and this height Hp is within an undetectable range, then additional information can be added to the two-dimensional map information to indicate that the obstacle can be overcome by special deformation or the like. On the other hand, if "Hp<Hp1", even if the height Hp is undetectable, the mobile robot 2 can overcome it without any special action, so there is no need to add additional information to the two-dimensional map information and the obstacle can simply be treated as a passable area.

In addition, if there is an obstacle that the mobile robot 2 cannot detect, such as a transparent flapper gate (a flapper gate whose gate shielding portion is made of a transparent material) but can be opened by issuing a specific command that the mobile robot 2 can transmit, the mobile robot 2 can pass through by issuing an open command in front of the obstacle. Figure 10 shows schematic examples of the closed state E1 and closed state E2 of the gate shielding portion of a flapper gate. In this case, too, the location is undetectable in the two-dimensional map, but the flapper gate corresponds, and the supplementing unit 300 can add accompanying information that allows passage by transmitting a special open command by radio or the like when the location is reached. Since the location is undetectable, when the mobile robot 2 estimates its own position, it can perform processing by not using the sensing results of the undetectable location, but using only the surrounding locations that can be sensed. In addition, when the moving path is searched for as passing through the flapper gate, etc., and the mobile robot 2 approaches a location composed of the flapper gate, etc., it can perform additional processing of issuing an open command in the mobile robot 2.

Obstacles that are impossible (or difficult) to detect because they are transparent regardless of their height, such as transparent flapper gates, and that can be opened by a specific command, can be identified by object recognition of the three-dimensional map information described above. Opaque flapper gates that can be detected if they are within the sensing range, and that need to be opened by a specific command and a record kept for security reasons, regardless of the height of the gate's shielding part and whether it is detectable within the sensing range of the mobile robot 2 or passable depending on the height, can also be identified by object recognition of the three-dimensional map information described above, and associated information can be added to the two-dimensional map information so that the mobile robot 2 can pass through by sending a specific command, just as in the case of transparent flapper gates.

As described above, the passable obstacle exclusion unit 311 can enable the autonomous movement of the mobile robot 2 by adding accompanying information to the distributed two-dimensional map so as to perform different processing, such as: (1) for an obstacle that is difficult for the mobile robot 2 to detect, if the step height is such that it is difficult for the mobile robot 2 to overcome, the unit treats the obstacle as a virtual obstacle; (2) if the step height is such that the mobile robot 2 can overcome it and the running mechanism does not require special deformation or movement to overcome the step, such as a wheeled robot, the unit excludes the obstacle from the virtual obstacles as one that can be passed through without any problems; (3) if the step height is such that the mobile robot 2 can overcome it and the running mechanism requires special deformation or movement to overcome the step, such as a bipedal robot, the unit changes the shape of the obstacle to another three-dimensional object so that the mobile robot 2 can detect it; and (4) if the obstacle can be opened by issuing a specific command, the unit does not treat the obstacle as a virtual obstacle and has the command issued in front of the obstacle. Alternatively, a normal two-dimensional map without accompanying information may be distributed, and when processing that uses the accompanying information becomes necessary during movement for product delivery, the map utilization device 3 may dynamically navigate the mobile robot 2.

The virtual obstacle complementation unit 312 adds associated information to the two-dimensional map converted by the two-dimensional map conversion unit 307 by superimposing, on the two-dimensional map converted by the two-dimensional map conversion unit 307, each obstacle that is ultimately determined to be difficult to detect based on the elimination result by the passable obstacle elimination unit 311, as a virtual obstacle that is undetectable but unmovable at the position where the obstacle is assumed to exist. The virtual obstacle complementation unit 312 complements the information of the virtual obstacle on the two-dimensional map so that the motion planning unit 208 (described later) of the mobile robot 2 can plan a motion to avoid the obstacle and move even if the imaging unit 204 cannot detect the obstacle. When complementing as a virtual obstacle, the virtual obstacle complementation unit 312 superimposes the obstacle as a virtual wall, so that the mobile robot 2 can perform additional processing to exclude the obstacle from the sensing target used for self-position estimation since it cannot be detected by the mobile robot 2, and can also prevent the mobile robot 2 from moving beyond that position.

In other words, when the 2D map conversion unit 307 first converts the map using the field of view height h, the locations that are determined to be actual physical obstacles are detectable for self-location estimation purposes and cannot be moved, whereas the obstacles virtually superimposed by the virtual obstacle complementation unit 312 are not detectable for self-location estimation purposes (and therefore require additional processing to prohibit use in self-location estimation) and cannot be moved, resulting in a difference in how they are used by the mobile robot 2.

In addition, when it is expected that the travel time on a route from a departure point to a destination searched by the route search unit 305 will increase by avoiding obstacles, the virtual obstacle completion unit 312 may make the route more difficult to select by increasing the cost map related to route selection.

In Figures 11 and 12, examples EA to EJ are shown, which show examples of images of a two-dimensional map delivered from the map utilization device 3 to the mobile robot 2 under the complementation process of the complementation unit 300, and an operation plan for the mobile robot 2 using the delivered two-dimensional map, to explain the effects obtained by this embodiment.

The map completion device 3 generates a two-dimensional map, for example, as shown in example EA, and distributes it to the mobile robot 2. Here, the white grids in example EA indicate areas that the mobile robot 2 can pass through, and the black grids indicate areas such as obstacles and walls that the mobile robot 2 cannot pass through. The mobile robot 2 performs grid-based local path planning, as shown in example EB, from the two-dimensional map shown in example EA to determine a movement path.

As mentioned above in steps 1 to 3, the route search unit 305 plans the global movement route, such as which route to select until arriving at the destination, and the motion planning unit 208 of the mobile robot 2 plans the local movement route along that movement route, such as avoiding obstacles, and the contents of the local path plan in step 3 are the subject of this discussion. (Note that the movement process of the mobile robot 2 itself can use existing methods, and details will be described later.)

As a grid-based path search method, for example, Dijkstra's algorithm or A* algorithm can be used. For example, the mobile robot 2 searches for a path that minimizes the total link cost of each link traversed from the current point to the destination point. Link cost indicates, for example, the length of travel time. Here, in example EB, the link cost related to obstacles, walls, etc. (gray areas) is set to ∞, and the link cost of other areas (white areas) is set to 1. In the case of example EB, there are no obstacles in the straight line to the destination point, so the mobile robot 2 plans a travel path that goes straight to the destination point.

Next, an example will be described in which there is an obstacle with a step height that the mobile robot 2 cannot detect and that is difficult to overcome. In this case, the virtual obstacle complementation unit 312 superimposes the obstacle as a virtual wall on the two-dimensional map as indicated by the reference numeral 51, as shown in example EC. Then, the mobile robot 2 that receives the map of example EC treats the link cost connected to the virtual wall 51 as ∞, as shown in example ED. As a result, the mobile robot 2 plans a movement route that avoids the obstacle, as shown in example ED.

Next, an example of delivery to the mobile robot 2 where the step is surmountable by the mobile robot 2, even though the mobile robot 2 cannot detect it, and no special transformation or movement is required to surmount it, will be described. In this case, the virtual obstacle complementation unit 312 treats the obstacle as if there is nothing, as shown in 52, as shown in example EE. Then, the mobile robot 2 that has received the map of example EE plans a travel route that will take it straight to the target point, as shown in example EF.

Next, an example of a delivery to the mobile robot 2 is sent for a step that the mobile robot 2 cannot detect but can overcome, and that requires special deformation or movement to overcome. In this case, the virtual obstacle complementation unit 312 marks the step position 53 on the two-dimensional map as shown in example EG. This marking may be done, for example, by marking obstacles that are less than the height that can be overcome when converting from a three-dimensional map to a two-dimensional map. As a result, the mobile robot 2 is able to make a special deformation or movement to overcome the obstacle at the step position. Also, in order to enable the mobile robot 2 to determine how to deform, it is desirable to transmit information on the length, width, and height of the marked obstacle in addition to the position information, or to accompany information on how to deform on the two-dimensional map.

As shown in example EH, if there is a step that can be overcome, the link cost may be increased (e.g., the link cost may be increased from the initial 1 to 2). This is because, for example, a dedicated transformation may require additional time for travel, and the travel route may no longer be the shortest.

Finally, an example will be described in which there is an object that the mobile robot 2 cannot detect, such as a transparent flapper gate, but which can be opened by issuing a specific command. In this case, the virtual obstacle complementation unit 312 marks the obstacle position 54 on the two-dimensional map, as shown in example EI. At this time, it is also desirable to transmit identification information that the obstacle is a flapper gate and API information for opening the flapper gate, or to attach identification information that the obstacle is a flapper gate and a command for opening the flapper gate to the two-dimensional map. Note that, as shown in example EI, the link cost may be increased (here, the link cost is set to 2). This is because, for example, it may take time to stop temporarily in front of the flapper gate to open the gate, and therefore the travel route may no longer be the shortest.

As described above, in this embodiment, the mobile robot 2 can perform appropriate local path planning by using different methods of complementing the 2D map depending on the relationship between a wide variety of robots (wheeled, bipedal, etc.) and a wide variety of obstacles.

<Delivery process of the mobile robot 2 that has received the delivery instruction in step S43>
The above has described in detail the processing content of each functional block in charge of arranging delivery by the map using device 3 in step S43. Below, the processing of the delivery process performed by the mobile robot 2 in charge of delivery operations, which has received route information and two-dimensional map information as delivery instructions in accordance with the delivery arrangements made by the map using device 3, will be described as detailed processing of each functional block. This delivery process can be realized by moving along a specified route while estimating its own position using an existing method, referring to the received two-dimensional map.

The imaging unit 204 uses a camera mounted on the mobile robot 2 to capture images of the moving direction, etc. of the mobile robot 2 as input necessary for estimating the self-position of the mobile robot 2 when it moves. The imaging unit 204 may further use a LiDAR mounted on the mobile robot 2 to measure the moving direction, etc.

The route information storage unit 206 stores route information to the destination received from the route distribution unit 306 of the map usage determination device 3, and uses this information to determine the operation of the mobile robot 2.

The map information storage unit 207 stores map information of places where the mobile robot 2 may move (i.e., places around the route) received from the map distribution unit 311 of the map-in-use determination device 3, and uses the map information to determine the operation of the mobile robot 2. The map information stored in the map information storage unit 207 is a two-dimensional map received from the map distribution unit 313 of the map-in-use determination device 3.

The motion planning unit 208 plans the next motion of the mobile robot 2 (motion at each time when moving along a route in real time). When the motion planning unit 207 moves along a route stored in the route information storage unit 206 by referring to the map stored in the map information storage unit 207, it plans motions that match the surrounding environment using the current position acquired by the current position acquisition unit 202, the state of the mobile robot 2 acquired by the state acquisition unit 203, and the image acquired by the image acquisition unit 204. For example, when there is an obstacle ahead, the motion planning unit 208 plans motions for each local environment in which the mobile robot 2 is located so that the robot moves along a route that avoids the obstacle at an appropriate moving speed.

The details of the current position acquisition unit 202 and the status acquisition unit 203 are as described above in the explanation of step S32, and the current position and status of the mobile robot 2 can be acquired by each sensor or the like when the mobile robot 2 moves in accordance with a delivery command.

The motion control unit 209 executes the motion determined by the motion planning unit 208. The motion control unit 209 controls the driving type, such as moving forward, backward, or turning right or left, the driving speed, and the motion type other than driving, such as changing the camera direction.

The mobile robot 2 that receives a command for delivery work can move to the destination and deliver the goods as described above. The mobile robot 2 may move alone to make the delivery, or may periodically communicate with the map using device 3 along the way and move according to navigation from the map using device 3. For example, instead of receiving delivery route information and two-dimensional or three-dimensional map information all at once for the entire delivery route, the delivery route may be divided into a number of partial routes, and each time a partial route is passed, route and map information for the next partial route may be received from the map using device 3.

Below, we explain various supplementary, additional, and alternative examples.

(1) As an application example of an embodiment of the present invention, it is possible to more smoothly perform tasks using robots in a certain environment, such as a work site, by remote control (centralized management by a remote manager using the map utilization device 3). This makes it unnecessary for a worker to actually go to the work site, and it is possible to reduce carbon dioxide emissions by saving energy resources required for human movement, which makes it possible to contribute to Goal 13 of the Sustainable Development Goals (SDGs) led by the United Nations, which is to "take urgent action to combat climate change and its impacts."

(2) Figure 13 is a diagram showing an example of the hardware configuration of a typical computer device 70. Each of the user terminal 1, mobile robot 2, and map usage device 3 constituting the map usage system 4 can be realized as one or more computer devices 70 having such a configuration. Note that when a partial configuration of the map usage system 4 is realized by two or more computer devices 70, information required for processing may be sent and received via a network. The computer device 70 includes a CPU (Central Processing Unit) 71 that executes predetermined instructions, a GPU (Graphics Processing Unit) 72 as a dedicated processor that executes some or all of the instructions executed by the CPU 71 in place of or in cooperation with the CPU 71, a RAM 73 as a main memory device that provides a work area for the CPU 71 (and the GPU 72), a ROM 74 as an auxiliary memory device, a communication interface 75, a display 76, an input interface 77 that accepts user input via a mouse, keyboard, touch panel, or the like, a camera 81 and a sensor 82 used for the self-position estimation function of the mobile robot 2, an actuator 83 such as a motor and a movement mechanism 84 such as wheels (including a transmission mechanism such as axles and gears) that provide the autonomous movement function of the mobile robot 2, and a bus BS for transmitting and receiving data between these.

Each functional unit of the map utilization system 4 can be realized by a CPU 71 and/or a GPU 72 that reads from a ROM 74 a predetermined program corresponding to the function of each unit and executes it. Both the CPU 71 and the GPU 72 are a type of computing device (processor). Here, when display-related processing is performed, a display 76 configured as a monitor such as an LCD or a lighting device such as an LED also operates in conjunction with the display, and when communication-related processing related to data transmission and reception is performed, a communication interface 75 also operates in conjunction with the display.

In the mobile robot 2, the imaging unit 204 and current position acquisition unit 202 are realized using inputs from the camera 81 and sensor 82, and the actuator 83 such as a motor is driven by the control output from the operation control unit 209, and the mobile mechanism 84 such as wheels is operated, thereby enabling the mobile robot 2 to move. The wireless communication unit 201 of the mobile robot 2 and the communication functions of the user terminal 1 and map utilization device 3 can be realized by the communication interface 75.

4...map utilization system, 1...user terminal, 2...mobile robot, 3...map utilization device
101...Product ordering section, 102...Destination information input section
201: wireless communication unit, 202: current position acquisition unit, 203: status acquisition unit, 204: imaging unit, 205: device information storage unit, 206: route information storage unit, 207: map information storage unit, 208: operation planning unit, 209: operation control unit
301: operation monitoring unit, 302: order information storage unit, 303: destination acquisition unit, 304: three-dimensional map information storage unit, 305: route search unit, 306: route distribution unit, 307: two-dimensional map conversion unit, 300: completion unit, 313: map distribution unit
308: obstacle identification unit, 309: detection range calculation unit, 310: difficult-to-detect obstacle extraction unit, 311: passable obstacle removal unit, 312: virtual obstacle complement unit

Claims (8)

A map utilization device that distributes map information along a specified route to a mobile robot so that the mobile robot can move along the specified route,
acquiring three-dimensional map information along the route as information in the vicinity of the route from the entire three-dimensional map information constructed for a field in which the mobile robot can move;
converting three-dimensional map information along the route into two-dimensional map information based on a sensing range of an external environment recognition sensor when the mobile robot moves autonomously, complementing the map information outside the sensing range, and distributing the two-dimensional map information to the mobile robot ;
The map utilization device recognizes obstacles from three-dimensional map information along the route, and superimposes obstacles that are located outside the height range of the sensing range when the mobile robot moves autonomously but whose two-dimensional position overlaps with the sensing range as virtual obstacles that cannot be sensed and are therefore impassable, thereby complementing the converted two-dimensional map information . A map utilization device that distributes map information along a specified route to a mobile robot so that the mobile robot can move along the specified route,
acquiring three-dimensional map information along the route as information in the vicinity of the route from the entire three-dimensional map information constructed for a field in which the mobile robot can move;
converting three-dimensional map information along the route into two-dimensional map information based on a sensing range of an external environment recognition sensor when the mobile robot moves autonomously, complementing the map information outside the sensing range, and distributing the two-dimensional map information to the mobile robot ;
The map utilization device recognizes obstacles from three-dimensional map information along the route, and among the obstacles that are located outside the height range of the sensing range when the mobile robot moves autonomously but whose two-dimensional positions overlap with the sensing range, those obstacles whose height range overlaps with the height range of the mobile robot are superimposed as virtual obstacles that cannot be sensed and are therefore impassable, thereby complementing the converted two-dimensional map information . The map utilization device according to claim 1 or 2, further characterized in that among obstacles that are located outside the height range of the sensing range when the mobile robot moves autonomously but whose two- dimensional positions overlap with the sensing range, those whose heights are within a range that the mobile robot can overcome when moving in normal operation are not superimposed as the virtual obstacles. The map utilization device according to claim 1 or 2, further comprising: a map utilization device that supplements the converted two-dimensional map information by providing additional information that distinguishes an obstacle from a virtual obstacle as one that requires a special action, the obstacle being located outside the height range of a sensing range when the mobile robot moves autonomously but whose two-dimensional position overlaps with the sensing range, the obstacle having a height that the mobile robot cannot overcome when moving in a normal manner, but is within a range that the mobile robot can overcome by moving in a special manner. A map utilization device that distributes map information along a specified route to a mobile robot so that the mobile robot can move along the specified route,
acquiring three-dimensional map information along the route as information in the vicinity of the route from the entire three-dimensional map information constructed for a field in which the mobile robot can move;
converting three-dimensional map information along the route into two-dimensional map information based on a sensing range of an external environment recognition sensor when the mobile robot moves autonomously, complementing the map information outside the sensing range, and distributing the two-dimensional map information to the mobile robot ;
The map utilization device further comprises a map utilization device which complements the converted two-dimensional map information with respect to locations of objects which the mobile robot cannot sense when moving autonomously due to the presence of transparent components or the like, but which can be passed over by giving a specific command . The map utilization device according to claim 4 or 5, further comprising: a step of increasing a cost of route search for the mobile robot when the mobile robot moves autonomously for the location requiring the special operation or the location that becomes passable by giving the specific command, and defining the location. The map utilization device according to any one of claims 1 to 6, further comprising: converting the three-dimensional map information along the route, other than the portion obtained by the complementation, into two-dimensional map information in accordance with a sensing height of an external environment recognition sensor when the mobile robot moves autonomously. 8. A program for causing a computer to function as the map utilization device according to claim 1.

Images