Navigation device and program

Abstract

To provide a navigation device with which it is possible to respond to even a situation that, when a mobile entity is moving along a path, the path has become no longer optimal due to an event that has occurred during the movement.SOLUTION: The present invention executes: first processes 301, 308, 309, 310, 311 of monitoring, for a mobile entity that is in the middle of moving toward a destination along an optimal route having been searched within a field, the occurrence of an event that affects the movability of the mobile entity, on the optimal route; and a second process 306 of re-searching for an optimum route for the mobile entity to travel from the current point to the destination, taking into account a change of the movability when an event occurrence was encountered while monitoring in the first process.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a navigation device and program capable of setting a route for navigating a moving body such as a mobile robot, and a learning method for learning a cost calculation model for route search. [Background technology]

[0002] Navigation systems for searching routes for moving objects such as mobile robots are being considered.

[0003] Patent Document 1 discloses a mobile robot control system that centrally controls multiple mobile robots and searches for paths so that the robots do not interfere with each other's movements. Patent Document 2 discloses a path planning system that generates a global path plan for a robot according to a node designation principle defined based on the social norms of an indoor space in order to support human-friendly path searching for the robot in the indoor space. Patent Document 3 discloses a teaching method for redoing travel, in which a mobile robot stops traveling based on path information when it comes to a place under construction, and returns to traveling based on the original path information after going around the place under construction under the guidance of a worker or the like. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-115397 A [Patent Document 2] JP 2022-16331 A [Patent Document 3] JP 2012-187698 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, the conventional technologies described above did not anticipate a situation in which the optimal route searched for by a mobile robot would become no longer optimal during movement due to the occurrence of some event, such as rain starting to fall or an elevator malfunction, and were therefore unable to deal with such situations.

[0006] In view of the problems with the conventional techniques described above, a first object of the present invention is to provide a navigation device and a program that can handle situations in which a route becomes non-optimal due to an event that occurs while a moving object is already traveling along the route.

[0007] A second object of the present invention is to provide a method for learning a model for calculating route costs for route searches according to events. [Means for solving the problem]

[0008] In order to achieve the above-mentioned first object, the present invention is a navigation device that executes, for a mobile body moving toward a destination along an optimal route already searched in a field, a first process of monitoring the occurrence of an event that affects the mobility of the mobile body on the optimal route, and a second process of re-searching an optimal route for the mobile body to reach the destination from a current location taking into account a change in the mobility when an event occurs during the monitoring of the first process. The present invention is also characterized by a program that causes a computer to function as the navigation device.

[0009] Furthermore, in the navigation device, the optimal route is searched for as one that minimizes the cost on the route from map information of the field consisting of nodes, links connecting the nodes, and costs defined for the links, and as a fourth process for the navigation device to calculate the cost defined for the links, the second process calculates the cost for the re-search using a model that is pre-trained and that receives as input, as link status information, all or part of information on weather, temperature, status of facilities that may be passed through such as elevators, or population density, and outputs the cost of the link, In order to achieve the second object, the present invention is a learning method for learning the model, the learning method including: inputting, as input and correct answer values ​​in learning data, all or part of information on the state of a link, such as weather, temperature, the state of facilities that may be passed through, such as elevators, or information on population density; and outputting, as a correct answer value for the cost of the link in response to the input, a weighted value of the time required for the mobile body to travel along the link and the degree to which safety may be threatened by an incident when the mobile body travels along the link, A second feature is that the procedure for learning the model is executed by a computer. Effect of the Invention

[0010] According to the first feature, if an event occurs on the optimal route that affects the mobility of the moving body, the optimal route is re-searched taking into account the change in mobility due to the event, so that it is possible to deal with a situation in which the route is no longer optimal.

[0011] According to the second feature, it is possible to learn a model that calculates a route cost for route search according to an event. [Brief description of the drawings]

[0012] [Figure 1] 1 is a diagram showing an example of the configuration of a navigation system according to a first embodiment. [Diagram 2] 1 is a functional block diagram of each component of a navigation system according to a first embodiment. [Diagram 3] 4 is a flowchart of product delivery in the navigation system according to the first embodiment. [Figure 4] 3 is a flowchart showing a route update procedure of the mobile robot according to the first embodiment, showing a detailed example of step S13 in FIG. 2. [Diagram 5] FIG. 11 is a diagram showing an example of a movable range of two mobile robots in a navigation system according to a second embodiment. [Figure 6] FIG. 11 is a functional block diagram of a mobile robot according to a second embodiment. [Figure 7] FIG. 11 is a functional block diagram of a navigation device according to a second embodiment. [Figure 8] FIG. 11 is a functional block diagram of each component of a navigation system according to a second embodiment. [Figure 9] FIG. 5 is a diagram showing a procedure for updating a route and waypoints of a mobile robot according to a second embodiment, as a modification of the flowchart of FIG. 4 according to the first embodiment. [Figure 10] FIG. 11 is a diagram for explaining an example of a route update in the second embodiment. [Figure 11] FIG. 11 is a diagram for explaining an example of a route update in the second embodiment. [Figure 12] FIG. 11 is a diagram for explaining an example of a route update in the second embodiment. [Figure 13] FIG. 13 is a diagram showing a flow of calculating a link cost as a variable value. [Figure 14] FIG. 13 is a diagram illustrating an example of a neural network model for calculating a link cost. [Figure 15] FIG. 1 is a diagram illustrating an example of a hardware configuration of a general computer device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In this embodiment, a navigation system in a delivery service system that performs deliveries will be described as an example of an application that can be realized by using a robot that moves autonomously as a mobile body (hereinafter, referred to as a mobile robot).

[0014] 1 is a diagram showing an example of the configuration of a navigation system according to the first embodiment. The navigation system 100 includes a user terminal 1, a mobile robot 2 capable of moving in an indoor and / or outdoor field F, and a navigation device 3. These components 1, 2, and 3 can communicate with each other via a network NW configured by the Internet and / or a LAN (local area network), etc.

[0015] Although only one user terminal 1 and one mobile robot 2 are shown in Fig. 1, any number of each may exist, and Fig. 1 shows the user terminal 1 and the mobile robot 2 as any one of them. Also, the navigation device 3 may be realized by two or more server devices performing processing in cooperation with each other.

[0016] 2 is a functional block diagram of components 1, 2, and 3 of the navigation system 100 according to the first embodiment. The mobile robot 2 can communicate with the navigation device 3 via a wireless communication line.

[0017] The overall operation of the navigation system 100 in the first embodiment is as follows. The navigation device 3 provides the mobile robot 2 with route information indicating a route from a starting point (current location) to a destination. Here, the route of the mobile robot 2 may become inappropriate during movement due to some event such as rain, crowding, or elevator failure. Therefore, in this embodiment, the navigation device 4 provides the mobile robot 2 with updated route information for switching to a different route based on observation of an event on the route the robot is heading for by another robot system.

[0018] Below, we will explain the individual processing of each functional unit (excluding functional units 307 to 311) in Figure 2 to achieve the above-mentioned overall operation in the first embodiment, and then we will explain the processing flow performed by each functional unit (including functional units 307 to 311) in Figure 2 in the flowcharts of Figures 3 and 4 described later.

[0019] [User terminal 1] The user terminal 1 is a terminal used by a user of the delivery service system according to this embodiment (hereinafter referred to as a delivery service user). The delivery service user requests delivery by specifying a destination to which a package is to be delivered. The user terminal 1 may be a mobile communication terminal device such as a smartphone or a tablet computer (tablet PC (personal computer)), or may be a stationary communication terminal device (for example, a stationary personal computer).

[0020] The user terminal 1 includes a product ordering unit 101 and a destination information input unit 102. The product ordering unit 101 selects a product and places an order for the selected product in response to an operation by a delivery service user.

[0021] The product ordering unit 101 may manage the remaining amount of goods owned by the delivery service user in real time, and automatically order a replenishment product when the remaining amount of goods falls below a preset threshold value.

[0022] The destination information input unit 102 inputs destination information indicating a destination such as an address when the product ordering unit 101 orders a product. The destination information may be specified by the delivery service user each time, or the destination information input unit 102 may record destination information previously specified by the delivery service user and automatically reuse the recorded destination information.

[0023] Furthermore, the destination information input unit 102 may set as the destination the current location acquired by a GPS (Global Positioning System) using a positioning function using the GPS included in the user terminal 1. Furthermore, the destination information input unit 102 may add an image of the scenery around the user terminal 1 captured by a camera of the user terminal 1 to the destination information as reference information for the destination.

[0024] [Mobile robot 2] The mobile robot 2 is a robot capable of autonomously traveling as a moving body. The mobile robot 2 travels in a field F, such as a city, and delivers packages to a destination designated by a delivery service user.

[0025] 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 route information storage unit 205, an operation planning unit 206, an operation control unit 207, and a luggage compartment opening / closing unit 208.

[0026] The wireless communication unit 201 performs wireless communication with the navigation device 4. The wireless communication unit 201 supports wireless communication methods such as 4G (fourth generation mobile communication system) and 5G (fifth generation mobile communication system), and performs wireless communication via a base station of the wireless communication method that the wireless communication unit 201 supports. The wireless communication unit 201 may also support an unlicensed wireless communication method such as Wi-Fi (registered trademark), and perform wireless communication by the unlicensed wireless communication method.

[0027] The current position acquisition unit 202 acquires the current position (position information) by a positioning system such as GPS. The status acquisition unit 203 acquires status data (status information of the mobile robot 2) such as remaining battery level, temperature, moving speed, acceleration, and wireless communication quality information from various sensors equipped in the mobile robot 2. The imaging unit 204 can be configured as hardware such as a camera facing the traveling direction of the mobile robot 2, and captures an image of the traveling direction.

[0028] The route information storage unit 205 stores route information to the destination and utilizes it for determining the operation of the mobile robot 2. When moving along a route indicated by the route information stored in the route information storage unit 205, the operation plan unit 206 determines an operation suited to the surrounding environment of the mobile robot 2 by using the current position acquired by the current position acquisition unit 202, the captured image captured by the imaging unit 204, and the like.

[0029] The motion control unit 207 causes the mobile robot 2 to execute the motion determined by the motion planning unit 206. The motion control unit 207 controls the travel type and travel speed of the mobile robot 2, such as moving forward, backward, or turning right or left, as well as the type of motion other than travel, such as changing the imaging direction of the imaging unit 204.

[0030] The luggage compartment opening / closing unit 208 opens and closes the luggage compartment of the mobile robot 2 so that the delivery service user can take out the product. It is also preferable that the luggage compartment opening / closing unit 208 opens and closes the luggage compartment after authenticating the user by means of facial recognition, passcode input, or the like.

[0031] [Navigation device 3] The navigation device 3 navigates each of the multiple mobile robots 2 from a departure point to a destination on a field F. The navigation device 3 also monitors the operation of the multiple mobile robots 2. The navigation device 3 includes an operation monitoring unit 301, an order information storage unit 302, a destination acquisition unit 303, a robot allocation unit 304, a map information storage unit 305, a route search / update unit 306, a route distribution unit 307, a robot information management unit 308, a facility information storage unit 309, a population density information storage unit 310, and a weather / climate information storage unit 311.

[0032] Each function of the navigation device 3 is realized by the navigation device 3 including computer hardware such as a CPU (Central Processing Unit) and a memory, and the CPU executing a computer program stored in the memory. The navigation device 3 may be configured using a general-purpose computer device, or may be configured as a dedicated hardware device. For example, the navigation device 3 may be configured using a server computer connected to a communication network such as the Internet. Each function of the navigation device 3 may be realized by cloud computing. The navigation device 3 may be realized by a single computer, or may be realized by distributing the functions of the navigation device 3 among multiple computers. The navigation device 3 may be configured to set up a website using, for example, a WWW system or the like.

[0033] The operation monitoring unit 301 performs operation monitoring of a plurality of mobile robots 2. The operation monitoring unit 301 performs operation monitoring by associating position information acquired by the current position acquisition unit 202 and wireless communication quality information and status information acquired by the status acquisition unit 203, which are received in real time from the mobile robots 2, with information identifying the individual mobile robot 2 (mobile robot ID). The operation monitoring unit 301 performs operation monitoring in real time based on information (position information and status information) updated periodically, for example, at one-second intervals, and video received from the mobile robot 2. The operation monitoring unit 301 analyzes the position information, status information, and video received from the mobile robot 2, and when a fault in the mobile robot 2 is detected, for example, the operation monitoring unit 301 issues an alarm to a supervisor.

[0034] In addition, the position information, status information, and images received from the mobile robot 2 may be used to update the route of the mobile robot 2 acquired during movement by the route search and update unit 306, or may be used to update the routes of other mobile robots 2.

[0035] The order information storage unit 302 receives order information for ordering a product from the user terminal 1, and stores the received order information. This order information includes product information specified by the product ordering unit 101 and destination information acquired by the destination information input unit 102. The order information storage unit 302 performs purchasing processing in cooperation with a purchasing system (not shown) and formulates a product delivery plan. The order information storage unit 302 stores scheduled delivery information such as the departure point, destination, and time period for the mobile robot 2 to deliver the product according to the formulated product delivery plan.

[0036] The destination acquisition unit 303 acquires destination information from the order information or the scheduled delivery information stored in the order information storage unit 302 .

[0037] The robot allocation unit 304 selects an appropriate mobile robot 2 for delivery based on a product delivery plan formulated by the order information storage unit 302, and allocates it to the delivery. For example, when delivering small, small quantities of products, the robot allocation unit 304 allocates a small, battery-efficient robot for delivery.

[0038] The map information storage unit 305 stores nationwide road map data and facility data associated therewith, such as various facilities and shops. The road map data is provided as link data in which roads on a map are divided into a plurality of parts with intersections and the like as nodes, and the parts between the nodes are defined as links. This link data includes data such as a link-specific identifier (link ID), link length, position information (longitude, latitude) of the start point and end point (node) of the link, angle (direction) data, road width, and road type. The road map data may include lane information indicating lanes that the mobile robot 2 can travel. In addition, the map information storage unit 305 desirably further stores indoor map data.

[0039] The route search and update unit 306 searches for a route from the starting point of the mobile robot 2 to the destination as a route that the mobile robot 2 is scheduled to travel, and updates the route information when the route of the mobile robot 2 becomes inappropriate during movement. As a route search method used by the route search and update unit 306, any existing method such as Dijkstra's algorithm, A* algorithm, genetic algorithm, etc. can be used for map data given with a graph structure of nodes and edges (links) connecting the nodes, and a cost defined for each link. The route search and update unit 306 sequentially calculates (accumulates) the cost of the road (link) from the starting point to the destination to the next reachable intersection (node), and selects the route from the starting point to the destination with the minimum cost.

[0040] The route search / update unit 306 updates the route when the route of the mobile robot 2 becomes inappropriate during movement due to some event such as rain, crowding, elevator failure, etc. As will be described later, when updating the route, the link costs are updated and the starting point is updated to the current point, and then, as in the initial search, an updated route can be obtained by searching for the minimum cost route using any existing method such as Dijkstra's algorithm for a graph structure.

[0041] 3 is a flowchart of product delivery in the navigation system 100 according to the first embodiment. In step S11, a request for order information is sent from the user terminal 1 at the current time t, and the navigation device 3 determines whether the order information has been newly received in the order information storage unit 302, and if new order information has been received, the process proceeds to step S12. If new order information has not been received, the process proceeds from step S11 to step S15, and the current time t is updated to the next current time t+1, and then the process returns to step S11, whereby the process of waiting in real time until new order information is received in step S11 continues. (Note that step S15 and similar step S16, described below, are steps for managing the processing timing (e.g., processing timing every second) for performing repeated processing in real time.)

[0042] In step S12, the robot allocation unit 304 determines a mobile robot 2 suitable for delivering the product in accordance with the order information, and the route from the current location to the destination for the mobile robot 2 to deliver the product is searched for and determined in the route search / update unit 306. The route distribution unit 307 distributes the searched route to the mobile robot 2. The mobile robot 2 starts moving according to the distributed route after the motion planning unit 206 makes a motion plan, and the process proceeds to step S13.

[0043] In step S13, the navigation device 3 determines whether or not an event affecting the movement of the mobile robot 2, which is currently delivering merchandise and moving at the current time t, has occurred on a route that has been distributed in advance (or recently redistributed) from the current position to the destination, and if it is determined that an influential event has occurred, it performs a route re-search and distributes the re-searched route to the mobile robot 2, and then proceeds to step S14. On the other hand, if it is determined in step S13 that an influential event has not occurred, it proceeds to step S14 without performing route re-search and distribution.

[0044] When the mobile robot 2 receives the re-searched route in step S13, it discards the route that it has been referring to up to the received current time t, and switches the movement control process so that it moves to the destination along the newly distributed and received re-searched route.

[0045] In step S14, the operation monitoring unit 301 confirms that the current position of the mobile robot 2 in charge of delivering the goods has reached the destination and that the cargo compartment has been opened by the cargo compartment opening / closing unit 208, meaning that delivery of the goods to the delivery service user has been completed. If the delivery is completed, the flow in FIG. 3 ends; if the delivery is not completed, the flow proceeds to step S16.

[0046] In step S16, the current time t is updated to the next current time t+1, and the process returns to step S13. In step S13, the system checks in real time whether there is any event that will affect the movement of the mobile robot 2 on its planned movement route, and if an affecting event is found, re-searches for the route and re-distributes the goods until delivery is completed.

[0047] FIG. 4 is a flowchart showing a route update procedure for the mobile robot 2 according to the first embodiment, showing a detailed example of step S13 in FIG.

[0048] In step S101, the operation monitor 301 of the navigation device 3 acquires current position information at the current time t from the current position acquisition unit 202 of the mobile robot 2, and the process proceeds to step S102.

[0049] In step S102, the navigation device 3 acquires various events (weather, facility status, population density, etc.) at the current time t on the planned movement route from the current position of the mobile robot 2 acquired in step S101 to the destination in the weather / temperature information storage unit 311, the facility status information storage unit 309, and the population density information storage unit 310, which will be described later, and then proceeds to step S103. At this time, the various events may be acquired by another mobile robot 2 (a mobile robot 2 other than the mobile robot 2 currently in charge of product delivery) or information held in another system.

[0050] In step S103, the navigation device 3 judges whether or not the event acquired in step S102 may affect the mobile robot 2, based on the characteristics (various performances described later) of the mobile robot 2 stored in the robot information management unit 308. This influence is judged with respect to the influence on the possibility of the mobile robot 2 moving along the planned movement route.

[0051] For example, when it starts to rain, the characteristics of the robot's movement mechanism are referred to, and for the mobile robot 2 whose movement mechanism is determined to be unsuitable for rainy weather due to low waterproof performance, etc., it becomes necessary to switch the route to a covered passage or the like in order to avoid rain outdoors. Also, when the degree of congestion of people increases significantly, it becomes necessary to refer to the characteristics of the robot's size and switch the route only for robots that are a certain size or larger and obstruct the passage. Also, when there is a puddle on the road, it becomes necessary to refer to the characteristics of the robot's movement mechanism and switch the route only for robots that move on wheels, such as small robots with small wheel sizes, that are determined to not have the ability to pass through puddles. Also, when one of the two passenger / freight elevators breaks down and it is expected that the remaining passenger / freight elevator will be used at a high rate, it becomes necessary to refer to the characteristics of the robot's size and switch the route for the large mobile robot 2 to a freight elevator that is separate from the passenger / freight elevator.

[0052] If the navigation device 3 determines in step S103 that the event has affected the mobile robot 2, the process proceeds to step S104. On the other hand, if the navigation device 3 determines that the event has not affected the mobile robot 2, there is no need to change the route, and the process is terminated.

[0053] In step S104, the navigation device 3 updates the link costs used in the route search. In order to select a route with the smallest integrated value of the link costs to the destination, the route search / update unit 306 may, for example, change the cost of the link where the event has occurred to a predetermined value that is determined to be huge compared to the initially defined cost, so that a route that passes through that link will not be selected when re-searching for a route.

[0054] Note that changing the link cost to an enormous value when an event occurs as described above is an example of rule-based link cost setting. Other methods for setting the link cost values ​​before and after the change will be described later, but the following description will first use an example in which the link cost is changed to an enormous value when an event occurs.

[0055] In step S105, the route search / update unit 306 of the navigation device 3 searches again for a route from the current position of the mobile robot 2 to the destination, and the process proceeds to step S106. In the process of step S104, the cost of the link where the event has occurred is changed to a huge value so that the link can be avoided from the links constituting the route, and then an optimal route is searched for.

[0056] In step S106, the route distribution unit 307 of the navigation device 3 redelivers the route re-searched in step S105 to the mobile robot 2, and the flow of Fig. 4 ends. As a result, the mobile robot 2 switches to the re-searched route and moves to the destination. As described in the flowchart of Fig. 3, if no influencing event occurs on the planned route while the mobile robot 2 is delivering goods, the route is not re-searched and re-distributed. On the other hand, if influencing events occur multiple times on the planned route while the mobile robot 2 is delivering goods, the route may be re-searched and re-distributed not only once, but two or more times.

[0057] Hereinafter, the functional units 307 to 311 related to the flow of FIG. 4 will be described.

[0058] The route distribution unit 307 transmits route information indicating the route searched by the route search unit 306 to the mobile robot 2. As described above, the initial route search and distribution is performed in step S12 in Fig. 3, and if an influencing event occurs during delivery, re-search and re-distribution may be performed one or more times in step S13 (and the detailed flow in Fig. 4). Note that in the case of re-search, the search can be performed by an existing method of searching for the shortest route on a graph structure, similar to the initial search, except that the cost of the link where the influencing event occurs is updated.

[0059] The robot information management unit 308 stores information indicating the performance of each mobile robot 2. The robot information management unit 308 stores, for example, the overall length, overall width, overall height, weight, wheel size, maximum speed, climbing ability, driving time, communication method, luggage compartment size, waterproof / dustproof performance, slip resistance, and other performance (specifications (specs) related to various predetermined items, functional specifications) of each mobile robot 2. Such performance information may be stored in the robot information management unit 308 by accepting an input from an administrator or the like, or may be stored in the robot information management unit 308 by accessing a server or the like that holds such performance information in advance.

[0060] The facility status information storage unit 309 stores facility status information of a location where the mobile robot 2 delivers goods. The facility status information storage unit 309 stores, for example, the operating status and installation location of an elevator in an apartment building. The elevator is a facility that the mobile robot 2 can pass through on the delivery route, and the facility information can be stored as follows: if the elevator is operating, the elevator is passable (the link including the elevator is passable), and if it is not operating, the elevator is not passable (the link including the elevator is not passable). The facility information may include, not only elevators, but also information about the road (link) itself, such as whether it is under construction and temporarily closed, or whether it is not under construction and passable, in the form of the presence or absence of construction equipment. Real-time facility information (whether it is operating or not, etc.) may be acquired in the facility status information storage unit 309 by, for example, accessing a device or server that manages the facility.

[0061] The population density information storage unit 310 stores population density information on the movement route of the mobile robot 2. The population density information storage unit 310 may acquire the population density information by, for example, tallying up location information of smartphones subscribed to a mobile carrier, extracting people from images captured by other mobile robots 2, or by accessing a server or the like that manages population density information in real time at various positions on a map.

[0062] The weather and temperature information storage unit 311 stores weather forecasts, past history, and real-time weather and temperature information. The weather and temperature information storage unit 311 may, for example, periodically access a weather information server on the network NW to update and store the latest weather and temperature information.

[0063] In each of the information storage units 309-311, each piece of information may be stored as corresponding to each link (a predetermined range in the field F represented by the link) of the map information in the form of a graph structure stored in the map information storage unit 305. In addition, rules may be set so that the determination as to whether or not the value of the information stored in each of the information storage units 309-311 has an effect in relation to various characteristics of the mobile robot 2 managed by the robot information management unit 308, which is determined in step S103, can be determined based on the rules. Specific examples of the rules are as described in step S103, and various rules may be set in advance, such as setting the cost of a link to a huge value for a mobile robot 2 that has the characteristic of not having a waterproof function when a certain link corresponds to rainy weather.

[0064] [Second embodiment] Next, the processing contents of the navigation system 100 according to the second embodiment will be described, focusing on the differences from the first embodiment.

[0065] In the first embodiment, a case where one mobile robot 2 delivers goods from a current location to a destination by itself is taken as an example. In contrast, in the second embodiment, a use case is assumed in which multiple mobile robots 2-1, 2-2, ..., 2-n (n is an integer of 2 or more) cooperate with each other. Here, as an example of n=2 in a delivery service, an example in which a mobile robot 2-1 hands over a package to a mobile robot 2-2 will be described with reference to FIG. 5.

[0066] 5 is a diagram showing an example of the movable range of two mobile robots 2-1 and 2-2 in a navigation system 100 according to the second embodiment. In the description of the second embodiment, elements having the same reference numbers as those used up to that point are the same as the elements having the same reference numbers in the first embodiment, unless otherwise specified. In addition, when only a part of the configuration is described, the previously described embodiment can be applied to the other parts of the configuration.

[0067] As shown in Fig. 5, the mobile robot 2-1 is arranged so that it can only move within a first range R1 outside the apartment building, and the mobile robot 2-2 is arranged so that it can only move within a second range R2 inside the apartment building. When delivering a product from outside the apartment building to inside the apartment building, the product needs to be handed over from the mobile robot 2-1 to the mobile robot 2-2 at one of the two entrances ET1, ET2. In this way, the first area R1 and the second area R2 are configured so as not to overlap with each other within the field F, and the two entrances ET1, ET2 are provided at the boundary between the first area R1 and the second area R2.

[0068] For example, when delivering a package from a starting point such as a store outside an apartment building to a destination in front of a room in the apartment building, the mobile robot 2-1 that has been moving around town to deliver the package cannot get past the autolocks of the entrances ET1 and ET2, so it is considered that the mobile robot 2-1 will hand over the package it is holding in front of the entrances ET1 and ET2 to the mobile robot 2-2 that is exclusively used within the apartment building, and the mobile robot 2-2 will deliver the package to the front of the room. In this case, if there are multiple entrances ET1 and ET2 in the apartment building, and some kind of event occurs during the mobile robot 2-1's delivery, such as an elevator breakdown or crowding, it will be necessary to switch the handover point to another entrance.

[0069] According to the second embodiment, for example, as shown in FIG. 5, before an event occurs, the routes are set so that the mobile robot 2-1 moves to the entrance ET1 via route 1A, and the mobile robot 2-2 moves to the entrance ET1 via route 2A, whereas after an event occurs, the routes can be updated and changed so that the mobile robot 2-1 moves to the entrance ET2 via another route 1B, and the mobile robot 2-2 also moves to the entrance ET2 via another route 2B.

[0070] FIG. 6 is a functional block diagram of a mobile robot 2 according to the second embodiment, which has a configuration common to that of the first embodiment shown in FIG. 2, except that a luggage delivery section 209 has been added as a configuration for enabling delivery of goods at entrances ET1, ET2, etc. in FIG. 4.

[0071] The baggage transfer unit 209 transfers the baggage held by the mobile robot 2-1 to the mobile robot 2-2. After the baggage transfer unit 209 opens the baggage compartment with the baggage compartment opening / closing unit 208, the baggage transfer unit 209 transfers the baggage using a mechanism such as a robot arm or a conveyor (not shown).

[0072] FIG. 7 is a functional block diagram of a navigation device 3 according to the second embodiment, and has a common configuration with the first embodiment, except that a delivery location calculation / update unit 312 is added to the configuration of the first embodiment shown in FIG. 2 as a configuration for setting a route including delivery of goods at entrances ET1, ET2, etc. in FIG. 4.

[0073] The delivery location calculation / update unit 312 calculates the delivery location between the mobile robot 2-1 and the mobile robot 2-2, and updates the delivery location when the delivery location becomes inappropriate during movement due to some event such as rain, crowding, or elevator failure.

[0074] Fig. 8 is a functional block diagram of each of the components 1, 2-1, 2-2, and 3 of the navigation system 100 according to the second embodiment. The functional block configurations of the mobile robots 2-1 and 2-2 are the same as those shown in Fig. 6, but -1 or -2 is added to the end of the reference numerals to distinguish between the mobile robots 2-1 and 2-2. For example, the imaging unit 204 provided in the mobile robot 2 in Fig. 6 is represented as imaging unit 204-1 in the mobile robot 2-1 and as imaging unit 204-2 in the mobile robot 2-2.

[0075] 8, due to space limitations, only the operation monitoring unit 301, the route distribution unit 307, and the delivery location calculation / update unit 312 are shown, and other functional units are omitted from the navigation device 3. As described in FIG. 7, the navigation device 3 of the second embodiment has a configuration in which the delivery location calculation / update unit 312 is added to all the components shown in FIG.

[0076] The route search and update unit 306 in the second embodiment updates the route, taking into consideration that not only the route but also the intermediate points (handover points) may change. Here, the procedure for updating the route and intermediate points of the mobile robot 2 in the second embodiment will be described with reference to the flowchart in FIG.

[0077] Fig. 9 is a diagram showing a procedure for updating a route and waypoints of a mobile robot 2 according to a second embodiment, as a modified example of the flowchart of Fig. 4 according to the first embodiment. Note that the flowchart of Fig. 3 as the overall flow on which the process is based is the same in the first and second embodiments, except that step S12 is changed as follows.

[0078] That is, in the second embodiment, the starting point of the order information received in step S11 is assumed to be in a first area R1 outside the apartment building shown in FIG. 5, and the destination is assumed to be a second area R2 inside the apartment building. In order to determine a route to deal with this in step S12, the following two shortest routes P1 and P2 are calculated. Route P1: Current location of mobile robot 2-1 → Entrance ET1 of relay point and Current location of mobile robot 2-2 → Entrance ET1 of relay point Route P2: Current location of mobile robot 2-1 → Entrance ET2 of relay point and Current location of mobile robot 2-2 → Entrance ET2 of relay point

[0079] For both routes P1 and P2, the sum of the costs of each route for the mobile robots 2-1 and 2-2 is set as the overall cost for routes P1 and P2. In step S12, the route search / update unit 306 determines which of routes P1 and P2 has the smaller overall cost as the route and distributes it to each of the mobile robots 2-1 and 2-2.

[0080] In the second embodiment, the entire route is shared and delivered by the mobile robots 2-1 and 2-2, so the decision to re-distribute the route in step S13 of FIG. 3 is made taking into account the routes of the two mobile robots 2-1 and 2-2, and the completion of step S14 is determined when both of the two mobile robots 2-1 and 2-2 have reached the relay point as their destination.

[0081] In the following, the second embodiment will be described up to the point where the product is handed over at a relay point as an intermediate destination. That is, the mobile robot 2-1 will be described up to the point where the mobile robot 2-1 reaches either of the entrances ET1 or ET2 as the destination from the current location, and the mobile robot 2-2 reaches either of the entrances ET1 or ET2 as the destination from the current location. Here, after both of the mobile robots 2-1 and 2-2 reach either of the entrances ET1 or ET2 as the intermediate destination, the mobile robot 2-2 can deliver the product from either of the entrances ET1 or ET2 as the current location to the front of a room in an apartment building as the final destination in the same manner as in the first embodiment after the delivery of the product is performed, and therefore the description thereof will be omitted.

[0082] Each step in FIG. 9 will be described below on the assumption that the flow in FIG. 3, which is the premise of the second embodiment, is changed as described above.

[0083] In step S201, the operation monitor 301 of the navigation device 3 acquires the current position information from the current position acquirers 202-1 and 202-2 of the mobile robots 2-1 and 2-2 that are moving toward a relay point as a destination, and then the process proceeds to step S202.

[0084] In step S202, the navigation device 3 acquires various events (weather, facility status, population density, etc.) along the planned movement route from the current positions of each mobile robot 2-1, 2-2 acquired in step S201 to the destination in the weather / temperature information storage unit 311, the facility status information storage unit 309, and the population density information storage unit 310, and then proceeds to step S203.

[0085] In step S203, the navigation device 3 determines whether or not the event acquired in step S202 may affect each of the mobile robots 2-1, 2-2, based on the characteristics of each of the mobile robots 2-1, 2-2 stored in the robot information management unit 308. If it is determined in step S203 that the event affects at least one of the mobile robots 2-1, 2-2, the process proceeds to step S204. On the other hand, if it is determined that the event does not affect both of the mobile robots 2-1, 2-2, the process ends since there is no need to change the route or destination.

[0086] In step S204, the navigation device 4 updates the link costs of the mobile robots 2-1, 2-2 (either one or both) that are determined to be affected by the event.

[0087] In step S205, the navigation device 3 searches again for a route from the current location to the destination, depending on whether the destination has been changed or not, and then proceeds to step S206. (That is, the above-mentioned routes P1 and P2 are calculated based on the updated link costs for the current locations of the mobile robots 2-1 and 2-2.)

[0088] In step S206, the navigation device 3 compares the total link costs when there is a change to the destination and when there is no change to the destination. (That is, for example, if the previous destination was the entrance ET1, the entrance ET1 becomes the destination when there is no change to the destination, and the entrance ET2 becomes the destination when there is a change to the destination, and the total link costs of these are compared.) If the total link cost when there is no change to the destination is smaller, the process proceeds to step S207. If the total link cost when there is a change to the destination is smaller, the process proceeds to step S208.

[0089] In step S207, the navigation device 3 distributes the re-searched route without changing the destination to the mobile robots 2-1 and 2-2, and ends the flow. In step S208, the navigation device 3 distributes the re-searched route with the changed destination to the mobile robots 2-1 and 2-2.

[0090] Next, an example of route updating according to the second embodiment will be described. Figs. 10 to 12 are diagrams for explaining an example of route updating in the second embodiment. In Figs. 10 to 12, circles indicate nodes, and lines connecting nodes indicate links. Also, numbers written on links indicate link costs. For mobile robot 2-1, the current node is S1, and the destination node is f. For mobile robot 2-2, the current node is S2, and the destination node is f. While nodes and links are commonly managed by each mobile robot, link costs have different values ​​for each mobile robot. Also, the cost of links that are difficult for mobile robots to travel through is set to ∞ so that they are not selected as routes.

[0091] In Figures 10 to 12, node e corresponds to the first entrance ET1 in Figure 5, node f corresponds to the second entrance ET2, nodes S1, a, b, c, and d drawn to the left of nodes e and f correspond to the first area R1 outside the apartment building in which the mobile robot 2-1 can move, and nodes e and f and nodes e, f, g, and S2 drawn to the right of these correspond to the second area R2 inside the apartment building in which the mobile robot 2-2 can move.

[0092] Figure 10 shows an example of a route search before an event occurs. As a result of the route search, the mobile robot 2-1 In the example [1], the route "S1 → a → f" with the smallest total link cost is selected as the route from the starting point "node S1" to the destination "node f" in the first region R1, and the total link cost is 15. In the example [2] of the mobile robot 2-2, the route "S2 → g → f" with the smallest total link cost is selected as the route from the starting point "node S2" to the destination "node f" in the second region R2, and the total link cost is 11. The link costs between the mobile robots 2-1 and 2-2 are added together to get to 26.

[0093] FIG. 11 is an example of route search in the case where the destination is not changed after the occurrence of an event. Here, an event occurs that makes it difficult for the mobile robot 2-2 to pass through the link between "f" and "g", and the link cost between "f" and "g" for the mobile robot 2-2 is updated from 8 to "∞" (for example, 1000, which is a huge value compared to the usual link cost of the order of 1 to 10). In the example [1] of the mobile robot 2-1, since there is no effect on the event, there is no change in the route search result of the example [1] of FIG. 10. On the other hand, the mobile robot 2-2 selects "S2 → g → e → f", which is the route with the smallest total link cost, as the route from the starting point "node S2" to the destination "node f", and the total link cost is 38. As a result, the link cost between the mobile robot 2-1 and the mobile robot 2-2 is added up to 53.

[0094] FIG. 12 shows an example of route search when the destination is changed after an event occurs. The route is changed from "node f" to "node e". As a result, in the example [1] of the mobile robot 2-1, "S1 → a → e" is selected as the route from the starting point "node S1" to the destination "node e", which is the route with the smallest total link cost, and the total link cost is 22. In the example [2] of the mobile robot 2-2, "S2 → g → e" is selected as the route from the starting point "node S2" to the destination "node e", which is the route with the smallest total link cost, and the total link cost is 23. As a result, the total link cost between the mobile robot 2-1 and the mobile robot 2-2 is 45.

[0095] As described above, when comparing the total link cost 53 in the case where the destination shown in Fig. 11 is left unchanged at node f with the total link cost 45 in the case where the destination is changed from node f to node e shown in Fig. 12, the total link cost is smaller when the destination is changed. Therefore, in the examples of Fig. 10 to Fig. 12, it is preferable to change the destination, and a re-searched route with the changed destination is distributed to mobile robot 2-1 and mobile robot 2-2.

[0096] As described above, according to each embodiment of the present invention, even if a sudden event occurs while the mobile robot 2 is moving, it is possible to move along an optimal route. Various supplementary examples, alternative examples, additional examples, etc. will be described below.

[0097] (1) As an application example of the embodiment of the present invention, remote work involving movement using the mobile robot 2 in a field F, which is a work site or the like, can be performed more smoothly by remote control. That is, according to this embodiment, when the mobile robot 2 moves within the field F, if rain or the like occurs on the planned movement route, the route is searched again. This makes it possible to prevent trouble such as the mobile robot 2 being caught in rain and being stopped from moving if the mobile robot 2 has poor waterproofing, for example, and makes it possible to perform remote work more smoothly.

[0098] This will eliminate the need for staff to physically go to the work site, and by saving the energy resources required for human transportation, it will be possible to reduce carbon dioxide emissions, thereby contributing to Goal 13 of the United Nations-led Sustainable Development Goals (SDGs), which is to "Take urgent action to combat climate change and its impacts."

[0099] (2) In each of the above embodiments, when an influential event occurs on a planned travel route, the route search / update unit 306 changes the link cost of the relevant location from an initially given value to a value that is determined to be huge, and then searches for the route again. As another embodiment for changing this link cost (and for automatically calculating the cost of each link at the time of the initial route search in step S12 of FIG. 3 according to the state of the event of the link, rather than using a given value), the following is also possible.

[0100] In other words, when changing the link cost, the route search / update unit 306 learns weightings for each parameter such as population density, road width, road surface condition, model of the mobile robot 2 (climbing ability, driving time, slip resistance, etc.), weather (rainfall, wind speed, etc.), temperature, goods delivered by the mobile robot 2, and communication quality, and by linking these with the link cost, which is a one-dimensional value, it is possible to change the link cost to an appropriate one taking into account other parameters when an event occurs, rather than necessarily choosing a route that goes through that link.

[0101] The route search and update unit 306 may use, for example, backpropagation to calculate the weighting of each parameter by setting the output value as the link cost and the input value as each of the above parameters. Here, the link cost may be determined as a variable value according to the situation (event) taking into consideration travel time and safety (probability of occurrence of various accidents). The route search and update unit 306 can calculate the link cost as a variable value according to the flow shown in FIG. 13.

[0102] In the flow of FIG. 13, step S301 is carried out as a learning phase in which a large amount of learning data is used to calculate the weighting (model) of each parameter for the link cost. Then, in step S302, an execution phase can be carried out in which the link cost is updated by changing the parameters related to an event when the event occurs based on the model resulting from the learning phase, and the updated link cost is used to enable route re-searching.

[0103] The learning phase in step S301 is performed using a large amount of learning data before actually delivering goods, and the execution phase in step S302 can be performed each time the link costs are updated in step S104 in FIG. 4 or step S204 in FIG. 9 (as well as when the initial link costs are calculated to search for the initial optimal route in step S12 in FIG. 3).

[0104] For example, when the travel time is increased due to the degree of population density, the weight of the population density is calculated in the learning phase, and the link cost is updated based on the change in population density at the time of the occurrence of the event in the execution phase. For example, even if the mobile robot 2 is adapted to handle rainy weather, when the travel time is increased due to the deterioration of visibility caused by rainfall, the weight of the amount of rainfall is calculated in the learning phase, and the link cost is updated based on the change in the amount of rainfall at the time of the occurrence of the event in the execution phase. For example, when the travel time is increased due to strong winds, the weight of the wind speed is calculated in the learning phase, and the link cost is updated based on the change in wind speed at the time of the occurrence of the event in the execution phase. For example, when a fragile item is broken in a pothole in the road when delivering the item, the weight of the item to be delivered is calculated in the learning phase, and the link cost is updated based on the observation of the pothole in the road at the time of the occurrence of the event in the execution phase. For example, when ice cream melts due to moving along a route exposed to sunlight when delivering the ice cream, the weight of the item to be delivered is calculated in the learning phase, and the link cost is updated based on the temperature observation at the time of the occurrence of the event in the execution phase.

[0105] Here, the process of calculating the link cost from each parameter will be described with reference to Fig. 14. Fig. 14 shows an example of a neural network model for calculating the link cost.

[0106] In Fig. 14, n pieces of input data x1, x2, ... x i , x n For example, road width x1 = 4.0m, temperature x2 = 35℃, rainfall per hour x i = 3 mm, and are examples of data whose real-time values ​​are stored in the facility status information storage unit 309, population density information storage unit 310, and weather and temperature information storage unit 311. j ,…,y k are the link costs of each road, such as road A and road B, which are output when input data is input to the network in Fig. 14, and k is the total number of links in the entire map information. The network in Fig. 14 configured in this way can be used, for example, as follows in the estimation phase of step S302, which will be described later. For example, if only temperature is input, all the links (all the map information) are town A and town B, and the temperature in town A is 20 degrees and the temperature in town B is 15 degrees, then 20 degrees is input and the costs at 20 degrees are calculated for all links between town A and town B, so the link cost of town A is obtained, and 15 degrees is input and the costs at 15 degrees are calculated for all links between town A and town B, so the link cost of town B is obtained.

[0107] Correct data t1, t2, ... t j ,…,t kare correct answer data for the link cost of each of the k roads, such as road A, road B, etc. Here, the correct answer data may be the travel time T for each road (the time required for travel by the mobile robot 2). The correct answer data may also be calculated by taking into consideration the travel time and safety. Here, safety is calculated from the degree to which safety may be threatened by an incident, and the degree of the threat may be quantified, for example, from the occurrence probability of individual accidents, near misses, etc. that constitute the incident, and / or the criticality C of the accidents, near misses, etc. (for example, three levels of criticality C may be set depending on the severity of the accident or near miss problem).

[0108] The weighted sum of travel time T and criticality C may be used as the correct answer data, and in this case, the balance between travel time T and safety (criticality C) may be adjustable. Specifically, during learning in step S301, the travel time and the number of near misses that occurred for each criticality level are obtained each time various mobile robots 2 pass through a route, and the input data are the model of mobile robot 2, weather, road surface, and other conditions, and the travel time T and the number of near misses for each criticality level are learned as correct answer data.

[0109] In the estimation phase of step S302, the travel time T and the occurrence probability C of each near miss are obtained by inputting the model of the mobile robot 2, the weather, the road surface, etc. For example, the model has been trained in such a way that the link cost is increased by 1 for every 1 minute increase in the travel time T, the link cost is increased by 1000 for every 1% increase in the occurrence probability of a large critical near miss, the link cost is increased by 100 for every 1% increase in the occurrence probability of a medium critical near miss, and the link cost is increased by 10 for every 1% increase in the occurrence probability of a small critical near miss, and the link cost is calculated in accordance with the model.

[0110] If a mobile robot 2 takes 10 minutes to travel a link (road), and the probability of a large critical event occurring on that link is 0.001%, the probability of a medium critical event occurring is 0.01%, and the probability of a small critical event occurring is 0.1%, the link cost for that road is as follows: 10+1000×0.001+100×0.01+10×0.1=13

[0111] In the neural network of FIG. 14, weights w1 to w18 and m hidden layers f(u1), ..., f(u m ) can be calculated by the backpropagation method based on the input data, output data, and correct answer data. As long as the output data can be obtained from the similar input data, the neural network is not limited to the one shown in FIG. 14, and learning may be performed using an existing deep learning network model or machine learning model that is configured with more layers.

[0112] (3) Figure 15 is a diagram showing an example of the hardware configuration of a general computer device 70. Each of the user terminal 1, mobile robot 2, and navigation device 3 constituting the navigation system 100 can be realized as one or more computer devices 70 having such a configuration. When a partial configuration of the navigation system 100 is realized by two or more computer devices 70, information required for processing may be transmitted and received via a network. The computer device 70 comprises 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 GPU 72), a ROM 74 as an auxiliary memory device, a communications interface 75, a display 76, an input interface 77 that accepts user input via a mouse, keyboard, touch panel, etc., a camera 81 and a sensor 82 that can be used as hardware to realize the imaging unit 204, current position acquisition unit 202, and status acquisition unit 203 of the mobile robot 2, an actuator 83 such as a motor and a moving mechanism 84 such as wheels (including a transmission mechanism such as axles and gears) that provide the autonomous moving function of the mobile robot 2, and a bus BS for exchanging data between these.

[0113] Each functional unit of the navigation system 100 (including the configuration for learning the network in FIG. 14) 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 arithmetic device (processor). Here, when display-related processing is performed, a display 76 configured as a monitor such as a liquid crystal display or a lighting device such as an LED also operates in conjunction with the above, and when communication-related processing related to data transmission and reception is performed, a communication interface 75 also operates in conjunction with the above.

[0114] In the mobile robot 4, an actuator 83 such as a motor is driven by a control output from the operation control unit 44, and a moving mechanism 84 such as wheels is operated, thereby enabling the mobile robot 4 to move. The actuator 83 and the moving mechanism 84 may be provided to realize a luggage compartment opening / closing unit 208 and a luggage delivery unit 209 in addition to the above-mentioned processes for moving the mobile robot 4. [Explanation of symbols]

[0115] 100...navigation system, 1...user terminal, 2...mobile robot, 3...navigation device, 101...product ordering unit, 102...destination information input unit, 201...wireless communication unit, 202...current position acquisition unit, 203...status acquisition unit, 204...imaging unit, 205...route information storage unit, 206...motion planning unit, 208...luggage compartment opening / closing unit, 209...luggage delivery unit, 301...operation monitoring unit, 302...order information storage unit, 303...destination acquisition unit, 304...robot allocation unit, 305...map information storage unit, 306...route search / update unit, 307...route distribution unit, 308...robot information management unit, 309...facility status information storage unit, 310...population density storage unit, 311...weather / temperature information storage unit, 312...delivery location calculation / update unit

Claims

1. a first process for monitoring, for a moving object moving toward a destination along an optimum route already searched for in a field, the occurrence of an event that affects the moving possibility of the moving object along the optimum route; a second process for re-searching an optimal route for the moving object to reach the destination from a current location thereof while taking into account a change in the mobility possibility, when an event occurs during monitoring of the first process; the optimal route is searched for as a route that minimizes the cost on the route from map information of the field that is composed of nodes, links connecting the nodes, and costs defined for the links; In the second process, a cost of a link in which an event is determined to have occurred by monitoring in the first process is updated to a larger value, and an optimal route for the moving body to reach the destination from a current location is re-searched so as to minimize the cost on the route; A navigation device characterized in that whether or not to update the value to a larger value and perform a search again is determined according to the type of the event that has occurred and the specifications of the moving object that correspond to that type.

2. The navigation device of claim 1, characterized in that the type of the event that has occurred includes the start of rain, and whether or not to update to the larger value and re-search is determined depending on whether or not the mobile mechanism is capable of handling rain as a specification of the corresponding mobile body.

3. The navigation device of claim 1, characterized in that the type of the event that has occurred includes a significant increase in the level of congestion of people, and whether or not to update to the larger value and re-search is determined depending on whether the specifications of the corresponding moving object are a size above a certain level and there is a risk of obstructing passage.

4. The navigation device of claim 1, characterized in that the type of the event that has occurred includes the presence of a puddle on the road, and whether or not to update to the larger value and re-search is determined depending on whether or not the specifications of the corresponding moving object include whether or not the wheel size is large enough to enable it to pass through puddles.

5. The navigation device described in claim 1, characterized in that the event that occurs includes a failure of some of multiple passenger / freight elevators, and in order to deal with the concentrated use of some of the passenger / freight elevators due to the failure, whether or not to update to the larger value and re-search is determined depending on whether or not the size of the corresponding moving object is large as a specification.

6. 2. The navigation device according to claim 1, further comprising a third process of distributing the re-searched optimum route to the mobile unit.

7. The navigation device according to claim 1, characterized in that in the second process, a cost of a link in which it is determined that an event has occurred through monitoring in the first process is updated to a value determined to be enormous, thereby re-searching for an optimal route for the moving body from its current location to the destination such that the optimal route does not include the link in which it is determined that the event has occurred.

8. 2. The navigation device according to claim 1, wherein the first process monitors the occurrence of an event that affects the mobility of the mobile object as being related to a given performance of the mobile object.

9. The navigation device according to claim 1, characterized in that in the first process, at least one of weather, temperature, the state of facilities that the moving object may pass through, such as elevators, and population density is monitored as information on events that may affect the mobility of the moving object.

10. the moving bodies are configured to include a first moving body and a second moving body that are moving along a first optimum route and a second optimum route, respectively, toward a common destination; The common destination is set to be selectable from a plurality of destinations, In the first process, an occurrence of an event affecting a mobility of at least one of the first moving body and the second moving body is monitored; The navigation device according to claim 1, characterized in that, in the second process, when an event occurs as a result of monitoring the first process, the first moving body and the second moving body re-search for an optimal route from their respective current locations to a common destination, with the common destination being reselected from among the multiple destinations as a candidate.

11. The navigation device according to claim 10, characterized in that the first moving body and the second moving body each move within a first area and a second area that do not overlap with each other within the field to make a delivery, and the common destination is set as a point on the boundary between the first area and the second area for handing over the item to be delivered from the first moving body to the second moving body.

12. The navigation device according to claim 1, characterized in that as a fourth process for calculating the cost defined for the link, a pre-trained model is used which inputs all or part of information on the state of the link, such as weather, temperature, the state of facilities that the user may pass through, such as elevators, or information on population density, and outputs the cost of the link, to calculate the cost for re-searching in the second process.

13. As a fifth process for learning the model, as an input in the learning data and a correct answer value for the input, As link status information, all or part of information on weather, temperature, status of facilities that may be passed through such as elevators, or population density is input, As a correct answer value of the output of the cost of the link for the input, a weighted value of the time required for the mobile object to move on the link and the degree to which safety may be threatened by an incident when the mobile object moves on the link is used, 13. The navigation device of claim 12, further comprising: training the model.

14. 14. A program for causing a computer to function as the navigation device according to any one of claims 1 to 13.

Images