Information processing device, communication system, identification method, and program

Abstract

The purpose of the present invention is to provide an information processing device capable of generating control information for determining an appropriate travel route of an autonomously traveling mobile object. An information processing device (10) according to the present disclosure comprises: a communication unit (11) that receives, from at least one sensor, observation data obtained by observing obstacles in a plurality of areas; a determination unit (12) that determines whether or not there are obstacles in the plurality of areas by using the observation data; and a specification unit (13) that, when it is not clear whether there is an obstacle in a first area among the plurality of areas, uses the result of determining whether or not there is an obstacle in an area around the first area to specify a probability of existence of an obstacle in the first area, the specified probability varying depending on whether or not there is an obstacle in the area around the first area.

Description

[Technical field]

[0001] The present disclosure relates to an information processing device, a communication system, and an identification method. [Background technology]

[0002] In recent years, there are an increasing number of cases in which autonomously moving transport robots are used to carry out work in factories or warehouses. In factories, etc., multiple transport robots are moving and multiple people are working there. For this reason, there is a demand for control to prevent autonomously moving transport robots from colliding with other transport robots and people.

[0003] Patent Document 1 describes that dynamic map information is generated that reflects the positions of obstacles such as humans by using human position information obtained from a surveillance camera or a human presence sensor in addition to static map information previously configured in a two-dimensional or three-dimensional grid for each floor of a building. The autonomous mobile body control device uses the dynamic map information to control the autonomous mobile body so as to avoid collision with obstacles. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] International Publication No. 2018 / 189770 Summary of the Invention [Problem to be solved by the invention]

[0005] In Patent Document 1, dynamic map information is generated assuming that there is a certain probability that an obstacle exists in a space that is a blind spot from the surveillance camera. However, if it is assumed that there is a certain probability that an obstacle exists in a space that is a blind spot from the surveillance camera, even if there is no obstacle actually, the autonomous mobile body control device needs to perform control to avoid the autonomous mobile body traveling in the space that is a blind spot. Also, even if there is actually an obstacle in the space that is a blind spot from the surveillance camera, if the probability of the obstacle's existence is set low, there is a possibility that the autonomous mobile body will come into contact with the obstacle. Thus, there is a problem that it is not possible to determine an appropriate traveling route for the autonomous mobile body depending on the installation state of the surveillance camera.

[0006] One object of the present disclosure is to provide an information processing device, a communication system, and a specification method that can generate control information for determining an appropriate driving route for an autonomously traveling mobile body. [Means for solving the problem]

[0007] An information processing device according to a first aspect of the present disclosure includes a communication unit that receives observation data of an obstacle in a plurality of areas from at least one sensor, a determination unit that uses the observation data to determine the presence or absence of an obstacle in the plurality of areas, and an identification unit that, when the presence or absence of an obstacle in a first area of ​​the plurality of areas is unknown, determines a probability that an obstacle exists in the first area using a determination result indicating the presence or absence of an obstacle in areas surrounding the first area, wherein the probability that an obstacle exists in the first area varies depending on the presence or absence of an obstacle in the areas surrounding the first area.

[0008] A communication system according to a second aspect of the present disclosure includes a sensor that observes obstacles in a plurality of areas and transmits observation data indicating the observation results, a determination means that uses the observation data to determine the presence or absence of an obstacle in the plurality of areas, and an identification means that, when the presence or absence of an obstacle in a first area of ​​the plurality of areas is unknown, determines the probability that an obstacle exists in the first area using the determination result indicating the presence or absence of an obstacle in areas surrounding the first area, and the probability that an obstacle exists in the first area varies depending on the presence or absence of an obstacle in the areas surrounding the first area.

[0009] A method for identifying an obstacle according to a third aspect of the present disclosure includes receiving observation data from at least one sensor observing an obstacle in a plurality of areas, using the observation data to determine the presence or absence of an obstacle in the plurality of areas, and, when the presence or absence of an obstacle in a first area of ​​the plurality of areas is unknown, determining a probability that an obstacle exists in the first area using a determination result indicating the presence or absence of an obstacle in areas surrounding the first area, wherein the probability that an obstacle exists in the first area varies depending on the presence or absence of an obstacle in the areas surrounding the first area. Effect of the Invention

[0010] The present disclosure makes it possible to provide an information processing device, a communication system, and a specification method that can generate control information for determining an appropriate driving route for an autonomously traveling mobile body. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a configuration diagram of an information processing device according to a first embodiment. [Diagram 2] FIG. 1 is a configuration diagram of a communication system according to a first embodiment. [Diagram 3] FIG. 11 is a diagram showing a flow of a calculation process according to the first embodiment. [Figure 4] FIG. 11 is a configuration diagram of a control system according to a second embodiment. [Diagram 5]FIG. 11 is a diagram showing grid data reflecting information indicating the presence or absence of an obstacle according to the second embodiment. [Figure 6] 13 is a diagram for explaining a calculation example when calculating the probability that an obstacle exists in a grid where it is unclear whether an obstacle exists or not according to the second embodiment. FIG. [Figure 7] FIG. 11 is a diagram showing a flow of a process for determining a moving path of a moving object according to the second embodiment; [Figure 8] FIG. 13 is a diagram showing grid data taking into consideration time according to the third embodiment. [Figure 9] FIG. 13 is a diagram showing labels in each grid according to the third embodiment. [Figure 10] FIG. 13 is a configuration diagram of a control device according to a fourth embodiment. [Figure 11] FIG. 13 is a diagram showing a travel area of ​​a moving object according to a fourth embodiment. [Figure 12] FIG. 13 is a diagram showing a travel area of ​​a moving object according to a fourth embodiment. [Figure 13] FIG. 13 is a diagram showing a moving path of a moving object according to a fourth embodiment. [Figure 14] 13 shows a speed pattern of a moving object according to the fourth embodiment. [Figure 15] FIG. 2 is a configuration diagram of a control device according to each embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] (Embodiment 1) Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. A configuration example of an information processing device 10 according to the first embodiment will be described with reference to Fig. 1. The information processing device 10 may be a computer device that operates by a processor executing a program stored in a memory.

[0013] The information processing device 10 has a communication unit 11, a determination unit 12, and an identification unit 13. The communication unit 11, the determination unit 12, and the identification unit 13 may be software or a module that performs processing by a processor executing a program stored in a memory. Alternatively, the communication unit 11, the determination unit 12, and the identification unit 13 may be hardware such as a circuit or a chip.

[0014] The communication unit 11 receives observation data of obstacles observed in a plurality of areas included in a predetermined space from at least one sensor. The sensor may be, for example, a sensor that detects an object or a person regarded as an obstacle in the predetermined space. The object regarded as an obstacle may be, for example, a device, a facility, a shelf, a pillar, or the like disposed in the predetermined space. The object regarded as an obstacle may be a moving object or a non-moving object. The sensor may be, for example, a camera or an infrared sensor. The camera may be a 2D camera or a 3D camera. The sensor may be a 2D lidar or a 3D lidar. The camera or the sensor may be attached at a plurality of locations to observe the predetermined space.

[0015] The observation data may be, for example, depth data indicating a distance from a sensor to an object or person in a given space, or may be imaging data when a camera is used as the sensor. The depth data and imaging data may be referred to as image data.

[0016] The predetermined space may be, for example, an area monitored by at least one sensor. Or, the predetermined space may be, for example, an area in which a moving object travels. The multiple areas included in the predetermined space may be divided areas into which the predetermined space is divided. Also, each area included in the multiple areas may have an area overlapping with another area. The multiple areas may be shown using two-dimensional information or three-dimensional information. One sensor may observe multiple areas, or multiple sensors may observe one area. Or, one sensor may observe one area.

[0017] The determination unit 12 determines the presence or absence of an obstacle in each of the areas by using the observation data. For example, the determination unit 12 may determine the presence or absence of an obstacle in each of the areas by performing an image recognition process or an image analysis process using image data such as imaging data or depth data.

[0018] Here, a case where the presence or absence of an obstacle in a first area among the multiple areas is unknown will be described. In this case, the identification unit 13 calculates the probability that an obstacle exists in the first area by using the determination result of the presence or absence of an obstacle in the areas surrounding the first area.

[0019] For example, when a sensor detects an obstacle in a given space, the presence or absence of an obstacle may be determined to be unknown in a blind spot. Or, the presence or absence of an obstacle may be determined to be unknown depending on the arrangement of the obstacle. Or, the presence or absence of an obstacle may be determined to be unknown depending on the material of the obstacle, the environment around the obstacle, etc. The environment around the obstacle may be, for example, illuminance, humidity, temperature, etc.

[0020] The area surrounding the first area, in which the presence or absence of an obstacle is unknown, may be, for example, an area adjacent to the first area, or an area located a predetermined distance away from the first area. The probability that an obstacle exists in the first area varies depending on the presence or absence of obstacles in the surrounding areas. Here, the influence of the surrounding areas on the probability that an obstacle exists in the first area may be greater as the surrounding areas are closer to the first area.

[0021] Next, a configuration example of the communication system according to the first embodiment will be described with reference to FIG. 2. The communication system in FIG. 2 includes a sensor 30, a communication unit 11, a determination unit 12, and a specification unit 13. In FIG. 2, the communication unit 11 is shown communicating with one sensor 30, but the communication unit 11 may communicate with two or more sensors 30. The communication unit 11 may also communicate with a plurality of sensors 30 via a network. The communication unit 11, the determination unit 12, and the specification unit 13 may not be provided in one information processing device 10. For example, the communication unit 11, the determination unit 12, and the specification unit 13 may be distributed and arranged in different devices, and two of the communication unit 11, the determination unit 12, and the specification unit 13 may be arranged in the same device.

[0022] Next, the flow of the calculation process executed in the information processing device 10 according to the first embodiment will be described with reference to FIG. 3. First, the communication unit 11 receives observation data of an obstacle observed in a plurality of areas included in a predetermined space from at least one sensor 30 (S1). Next, the determination unit 12 determines the presence or absence of an obstacle in the plurality of areas using the observation data (S2). Next, when the presence or absence of an obstacle in a first area of ​​the plurality of areas is unknown, the identification unit 13 identifies the probability that an obstacle exists in the first area using a determination result indicating the presence or absence of an obstacle in areas surrounding the first area (S3). Here, the probability that an obstacle exists in the first area varies depending on the presence or absence of an obstacle in areas surrounding the first area.

[0023] As described above, the information processing device 10 determines the presence or absence of an obstacle using the observation data observed by the sensor, and when the determination result indicates that the presence or absence of an obstacle is unknown, calculates the probability that an obstacle exists in the area where the obstacle is unknown. The information processing device 10 calculates the probability that an obstacle exists in the area where the obstacle is unknown using the determination result indicating the presence or absence of an obstacle in the area surrounding the area where the obstacle is unknown. In this way, the information processing device 10 can calculate the probability that an obstacle exists in a predetermined space with high accuracy. As a result, when a mobile body autonomously travels in a predetermined space, the information processing device 10 can provide information on the presence of an obstacle as control information for determining an appropriate travel path for the mobile body. In addition, if the probability that an obstacle exists in an area where the presence or absence of an obstacle is unknown is set to a constant, it becomes difficult to set an appropriate value as the probability that an obstacle exists, and the accuracy of the value set as the probability that an obstacle exists deteriorates. On the other hand, the information processing device 10 calculates the probability that an obstacle exists in the area where the obstacle is unknown using the determination result indicating the presence or absence of an obstacle in the area surrounding the area where the obstacle is unknown. This allows the information processing device 10 to set an appropriate value for the probability that an obstacle exists in an area where the presence or absence of an obstacle is unknown, rather than setting the probability to a constant value.

[0024] (Embodiment 2) Next, a configuration example of the control system according to the second embodiment will be described with reference to Fig. 4. In the control system of Fig. 4, a control device 20, sensors 31 to 33, and a moving object 40 communicate with each other via a network 50. The control device 20 corresponds to the information processing device 10 of Fig. 1. The control device 20 controls the movement path of the moving object. The moving object 40 may be a robot that moves autonomously. The robot may be, for example, a transport robot (AGV (Automated Guided Vehicle)) that transports an object. The moving object 40 moves autonomously in a travel area so as to avoid obstacles based on obstacle information generated by the control device 20.

[0025] The sensors 31-33 detect objects or people that are regarded as obstacles in a predetermined space. The sensors 31-33 may be, for example, camera devices. The sensors 31-33, for example, periodically capture images of the travel area in which the mobile object 40 travels, and transmit the captured image data to the control device 20 via the network 50. The captured image data may be two-dimensional data or three-dimensional data. The two-dimensional data may be, for example, image data. The three-dimensional data may include image data and distance data from the sensor to the target object. The captured image data corresponds to observation data.

[0026] The control device 20 has a configuration in which a label setting unit 21 and a route control unit 22 are added to the information processing device 10 in Fig. 1. The label setting unit 21 and the route control unit 22 may be software or a module in which processing is performed by a processor executing a program stored in a memory. Alternatively, the label setting unit 21 and the route control unit 22 may be hardware such as a circuit or a chip.

[0027] The communication unit 11 receives the photographed data from the sensors 31 to 33 via the network 50. The sensors 31 to 33 generate photographed data of a part of an area in a travel area or the entire travel area. The determination unit 12 determines the presence or absence of an obstacle in the travel area photographed by the sensors 31 to 33. The determination unit 12 may generate map data displaying the entire travel area by synthesizing or combining the photographed data photographed by each of the sensors 31 to 33. The map data is image data displaying the entire travel area. Furthermore, the determination unit 12 generates grid data by dividing the map data into a plurality of grids, and determines or specifies whether an obstacle is present in each grid, whether an obstacle is not present, or whether the presence or absence of an obstacle is unknown. The grid data may be data divided such that, for example, 10 pixels of the image data constitute one grid. Alternatively, the grid data may be data divided at a predetermined interval (one meter interval) set by the user, or may be data appropriately divided by machine learning (divided according to the average size of the luggage). Alternatively, the grid data may be data divided so that the total number of grids is a predetermined number. Furthermore, the map data is not limited to being divided into a grid (lattice), and may be divided into a rectangle, a circle, an ellipse, or any other predetermined shape.

[0028] FIG. 5 shows grid data reflecting information indicating the presence or absence of an obstacle. FIG. 5 shows grid data generated based on image data obtained by photographing the driving area from above using sensors 31 to 33 installed on the ceiling or the like. FIG. 5 shows grids obtained by dividing a two-dimensional plane having an X direction and a Y direction, but the determination unit 12 may generate grid data obtained by dividing a three-dimensional space. In FIG. 5, a shaded grid indicates that an obstacle exists, a grid surrounded by a thick line indicates that it is unclear whether an obstacle exists, and other grids indicate that no obstacle exists.

[0029] Whether or not an obstacle exists in each grid may be determined by, for example, machine learning. For example, in a case where a learning model that has learned images, etc., related to obstacles and floor surfaces has been generated in advance, the determination unit 12 may determine whether or not an obstacle exists in each grid by applying the learning model to the grid data. For example, the determination unit 12 may determine that an obstacle does not exist in a grid in which a floor surface is displayed, and that an obstacle exists in a grid in which a floor surface is not displayed. The determination unit 12 may determine that it is unclear whether or not an obstacle exists in a grid in which a floor surface exists but the ratio of the floor surface area to the grid surface area is smaller or larger than a predetermined threshold value. In addition, in a case where the determination unit 12 applies the learning model to the grid data and the name, etc., of an object in the grid data cannot be identified, the determination unit 12 may determine that it is unclear whether or not an obstacle exists in a grid including an object whose name cannot be identified.

[0030] Alternatively, when the photographed data is three-dimensional data and the grid data has distance data from the sensor to the object, the determination unit 12 may determine that an obstacle exists in an area showing a different distance from the surrounding areas. Also, for an area whose difference from the distance of the surrounding areas is equal to or less than a predetermined threshold, it may be determined that it is unclear whether an obstacle exists. The determination unit 12 may determine whether an obstacle exists by performing an edge detection process on the grid data.

[0031] The identification unit 13 calculates the probability that an obstacle exists in a grid determined as being unknown as to whether an obstacle exists. Here, with reference to FIG. 6, a calculation example in which the identification unit 13 calculates the probability that an obstacle exists in a grid determined as being unknown as to whether an obstacle exists will be described. Here, for ease of explanation, a calculation example using one-dimensional grid data arranged in the X direction will be described. FIG. 6 shows that grid_0 to grid_5 are arranged in the X direction. Numbers shown in the grids in FIG. 6 are identification information for identifying each grid. Grid_0, grid_1, and grid_5 indicate that no obstacle exists. Grid_2 and grid_4 indicate that an obstacle exists. Grid_3 indicates that it is unknown as to whether an obstacle exists.

[0032] The identification unit 13 calculates the probability that an obstacle exists in grid_3 using the following Equations 1 and 2.

[0033] TIFF0007609285000001.tif15125

[0034] TIFF0007609285000002.tif10159

[0035] In Equation 2, E(x) is a formula for calculating the likelihood or weight of x occurring. Equation 1 is a formula for converting the weight E(x) into the probability q(x) that x will occur. x represents an instance of a probability event occurring, such as a number or vector. Z represents the sum of all possible exp(E(x)).

[0036] For example, assume that x is either 1 or 2. Also, know that E(1) = 1 and E(2) = 2, then Z = exp(E(1)) + exp(E(2)) = exp(1) + exp(2) = 10.107. In this case, the probability q(x=1) of x=1 is q(1) = exp(E(1)) / Z = 0.2689, or about 27 percent.

[0037] Here, there are 64 combinations in which grid_0 to grid_5 have or do not have obstacles. All E(x) for these 64 patterns are calculated. For example, in the case of a combination in which no obstacles exist in any of grid_0 to grid_5, E(x) is shown as E(none, none, none, none, none, none), and in the case of a combination in which an obstacle exists only in grid 3, it is shown as E(none, none, none, yes, none, none).

[0038] Here, P in the first term of Eq. i (x i |o i ) is the result of the judgment of grid_i (i is an integer between 0 and 5). i When x i The label indicates the weight for attaching a label such as "Obstacle Present". The determination result indicates whether an obstacle is present, not present, or whether it is unclear whether an obstacle is present. For example, if an obstacle is present, it is marked as "Present", if an obstacle is not present, it is marked as "Not present", and if it is unclear whether an obstacle is present, it is marked as "Unknown". The label indicates information that is set when it is determined whether an obstacle is present in each grid. For example, if it is determined that an obstacle is present in each grid, it is marked as "Present", and if it is determined that an obstacle is not present, it is marked as "Not present". A label of "Present" or "Not present" is set to a grid for which it is determined that it is unclear whether an obstacle is present.

[0039] P in the second term of Eq. ij (x i x j |o i o j ) is the result of the judgment of two adjacent grids i and j (j is an integer between 0 and 5 and is different from i) i and j When x i and x jHere, an example is shown in which two adjacent grids i and j are used, but in the second term of formula 2, the weight of labeling may be expressed using the judgment results of three or more grids that are arranged consecutively. Alternatively, a third term may be added to formula 2, and the weight of labeling may be expressed using the judgment results of three or more grids that are arranged consecutively in the third term. The case in which the judgment results of four or more grids that are arranged consecutively are used is similar to the case in which the judgment results of three or more grids that are arranged consecutively are used.

[0040] The weights are determined according to the following rules: i (x i |o i ), when the judgment result is "yes", the weight for labeling it as "yes" is P i Let (Present|Present) = 1. Also, when the judgment result is "Present", the weight for labeling it as "Absent" is P i (absence|presence) = 0. Also, P i (none|none) = 1, P i (Yes|No) = 0. Furthermore, when the judgment result is unknown, P i (Yes|Unknown)=1 and P i (none|unknown) = 1. Or, P i (Yes|Unknown)=1 and P i Alternatively, (absent|unknown)=0.5 may be used. In this way, when the determination result is unknown, the probability of determining that there is an obstacle can be increased by setting a difference between the weight for attaching a "present" label and the weight for attaching a "absent" label. By increasing the probability of determining that there is an obstacle, the safety of the moving path of the moving body 40 can be increased. Alternatively, P i (Yes|Unknown)=0.5 and P i (none|unknown)=1 may be used. This can increase the probability that it is determined that there is no obstacle. By increasing the probability that it is determined that there is no obstacle, a more efficient movement route for the moving body 40 can be set. An efficient movement route may be, for example, a route that reduces the movement distance of the moving body 40.

[0041] Alternatively, the weight value may be calculated as an optimal value by machine learning or the like. For example, data indicating the presence or absence of an obstacle in a grid as shown in FIG. 6 is prepared as test data for which the correct answer is known. The test data may be called, for example, correct answer data or teacher data. Furthermore, an evaluation value indicating an amount that quantifies the compatibility between the prediction result predicted from the probability field expressed by using Equation 1 and Equation 2, etc. and the correct answer data may be determined, and the weight may be determined so as to maximize the evaluation value. Here, a plurality of candidates for the weight may be prepared, the evaluation value when each weight is used may be calculated, and the weight with the maximum evaluation value may be selected.

[0042] Next, P ij (x i x j |o i o j ), if the judgment result does not include "unknown", the weight for labeling as the judgment result is set to 1, and otherwise to 0. Also, if the judgment result includes "unknown", and the judgment results of two adjacent grids_i and j are (present, unknown) or (unknown, present), the weight for labeling as (present, present) is set to 1, and otherwise to 0. In other words, P ij (Yes, Yes | Yes, Unknown)=P ij (Present, Present | Unknown, Present) = 1. Furthermore, if the judgment result contains "unknown", and the judgment results of two adjacent grids_i and j are (absent, unknown) or (unknown, absent), the weight for labeling as (absent, absent) is set to 1, and otherwise to 0. In other words, P ij (none,none|none,unknown)=P ij Let (none, none | unknown, none) = 1.

[0043] For example, if the judgment results of grids 0 to 5 are (Absence, Absence, Yes, Unknown, Yes, Absence) and the labels are (Absence, Absence, Absence, Absence, Absence, Absence), then E(Absence, Absence, Absence, Absence, Absence) = 4 + 1 = 5. Also, if the observation data of grids 0 to 5 are (Absence, Absence, Yes, Unknown, Yes, Absence), and the labels are (Absence, Absence, Absence, Absence, Absence, Absence, Absence) then E(Absence, Absence, Absence, Absence, Absence, Absence) = 3 + 1 = 4. In this way, when the calculation is performed 64 times, which is the number of all label patterns, the label arrangement with the largest value of E(x) is calculated as E(Absence, Absence, Yes, Yes, Yes, Absence) = 6 + 5 = 11. In other words, for the judgment results of grids 0 to 5 being (Absence, Absence, Yes, Unknown, Yes, Absence), the label pattern (Absence, Absence, Yes, Yes, Yes, Absence) is the most likely, or in other words, the most likely.

[0044] Here, the probability q (grid_3=present) that grid 3 will be labeled as "present" when the judgment results for grids 0 to 5 are (absent, absent, present, unknown, present, absent), that is, the probability q (grid_3=present) that it will be judged that there is an obstacle in grid 3, is calculated using the following Equation 3.

[0045] TIFF0007609285000003.tif13170

[0046] The result of calculating Equation 3 is 0.8607, which means that there is a probability of about 86 percent that an obstacle exists in grid_3. Although one-dimensional grid data has been described in FIG. 6, the calculation can be performed in the same manner for two-dimensional grid data. In the case of two-dimensional grid data, the identification unit 13 calculates the probability that an obstacle exists in a grid where it is unclear whether an obstacle exists, using grid data arranged in the X direction and the Y direction. Specifically, the identification unit 13 uses grid data arranged in the X direction including grids where it is unclear whether an obstacle exists, and grid data arranged in the Y direction including grids where it is unclear whether an obstacle exists.

[0047] In addition to calculating the probability that grid_3, whose judgment result is unknown, is labeled as "present," the probability that grids whose judgment result is "absent" or "present" may also be calculated. In this way, even if there is an error in the judgment result, it is possible to correct it to an appropriate judgment result by calculating the probability that an obstacle exists. Specifically, when judging whether or not an obstacle exists in each grid, it may be judged that an obstacle exists in a grid where no obstacle actually exists due to an error in the photographed data, etc. In such a case, it is possible to detect an error in the judgment result by calculating the probability that an obstacle exists using the judgment results in the surrounding grids.

[0048] Returning to Fig. 4, the label setting unit 21 sets a label to each grid shown in Fig. 5. The label setting unit 21 sets a label indicating the presence of an obstacle to a grid in which it is determined that an obstacle exists. Furthermore, the label setting unit 21 sets a label indicating the absence of an obstacle to a grid in which it is determined that no obstacle exists.

[0049] For a grid where it is determined that it is unclear whether an obstacle exists, the label setting unit 21 determines a label to be set using the probability of the obstacle existing calculated by the identification unit 13. For example, the label setting unit 21 may set a label indicating that an obstacle exists when the probability of the obstacle existing exceeds a predetermined threshold, and may set a label indicating that an obstacle does not exist when the probability is equal to or less than the threshold. Alternatively, the label setting unit 21 may set the value of the probability of the obstacle existing as a label.

[0050] The route control unit 22 controls the moving route of the moving body 40 using the label set by the label setting unit 21. Specifically, the route control unit 22 may control the route of the moving body 40 so as to avoid traveling through a grid where an obstacle exists. The route control unit 22 determines the moving route of the moving body 40, and transmits information on the determined moving route to the moving body 40 via the network 50. The route control unit 22 may also transmit information on the control of the drive unit of the moving body 40 to the moving body 40 so that the moving body 40 travels along the determined moving route. The information on the control of the drive unit may be, for example, information on the moving speed including the angular velocity of each wheel. The route control unit 22 may also transmit information on the attitude of the moving body 40 so that the body of the moving body 40 does not tilt. The route control unit 22 may also transmit information on the angle at which the moving body 40 pushes the transported object so that the moving body 40 does not let go of or drop the transported object. Furthermore, when the moving body 40 transports an object in cooperation with another moving body, the route control unit 22 may transmit, to the moving body 40, information regarding the distance between the moving body 40 and the other moving body.

[0051] Next, a flow of a process for determining a moving path of the moving object 40 according to the second embodiment will be described with reference to Fig. 7. First, the communication unit 11 receives observation data generated by each of the sensors 31 to 33 from the sensors 31 to 33 (S11). The observation data may be, for example, photographed data generated by the sensors 31 to 33.

[0052] Next, the determination unit 12 generates grid data using the received observation data (S12). The determination unit 12 may generate map data indicating the entire travel area of ​​the moving object 40 by synthesizing multiple pieces of observation data, and may generate the grid data by dividing the map data into multiple grids.

[0053] Next, the determination unit 12 determines whether or not an obstacle exists in each grid shown in the grid data (S13). The determination unit 12 may determine whether or not an obstacle exists in each grid by using, for example, a learning model that has learned images related to obstacles and floor surfaces.

[0054] Next, the identification unit 13 calculates the probability that an obstacle exists in the grid where it is determined that the presence or absence of an obstacle is unclear (S14). For example, in the case of two-dimensional grid data, the identification unit 13 calculates the probability that an obstacle exists by using grid data arranged in the X direction and the Y direction.

[0055] Next, the label setting unit 21 sets a label to each grid using the result of the determination of whether an obstacle exists and the probability that an obstacle exists in the grid where it is determined that the existence of an obstacle is unknown (S15). The set label may be, for example, a label indicating the existence of an obstacle or a label indicating the absence of an obstacle.

[0056] Next, the route control unit 22 uses the labels set for each grid to determine a movement route for the moving object 40 (S16). For example, the route control unit 22 may determine the movement route for the moving object 40 so as to avoid a route on which an obstacle exists.

[0057] As described above, for a grid where it is unclear whether an obstacle exists, the control device 20 calculates the probability that an obstacle exists by using information indicating whether an obstacle exists in the grids surrounding the grid. In other words, the control device 20 calculates the probability that an obstacle exists for missing data. By considering the status of the grids surrounding the grid where it is unclear whether an obstacle exists, the weight of the existence of an obstacle in the grid is taken into account, and the accuracy of the calculated probability can be improved.

[0058] (Embodiment 3) Next, grid data considering time or time series will be described with reference to Fig. 8. Fig. 8 shows grid data arranged in the X direction and grid data arranged in time series. The numbers shown in Fig. 8 are identification information for identifying each grid. Also, t1 to t4 indicate times. Fig. 8 shows whether or not an obstacle exists in grid_0 to grid_4 from time t1 to t4.

[0059] At time t1, it is indicated that no obstacles exist in grid_0 and grid_1, and that obstacles exist in grid_2 and grid_3. At time t2, it is indicated that it is unclear whether an obstacle exists in grid_0, it is indicated that no obstacle exists in grid_1, and it is indicated that obstacles exist in grid_2 and grid_3. At time t3, it is indicated that no obstacles exist in grid_0 and grid_2, and it is indicated that obstacles exist in grid_1 and grid_3. At time t4, it is indicated that no obstacles exist in grid_1 and grid_2, and it is indicated that obstacles exist in grid_0 and grid_3.

[0060] The communication unit 11 receives the observation data generated at t1 to t4 from the sensors 31 to 33, and the determination unit 12 generates grid data using the received observation data and arranges the respective grid data in chronological order to generate the grid data shown in FIG. 8. The grid data is not limited to being arranged in chronological order, but may be generated in consideration of the chronological order. The sensors 31 to 33 generate observation data periodically. The sensors 31 to 33 may set a timestamp indicating the timing or time when the observation data was generated to the generated observation data. The determination unit 12 may check the timestamp and arrange the grid data in chronological order. The sensors 31 to 33 generate observation data at substantially the same timing. In other words, the sensors 31 to 33 synchronize the timing of generating the observation data.

[0061] The identification unit 13 calculates the probability that an obstacle exists in grid_0 at time t2, where it is indicated that it is unknown whether an obstacle exists or not. For example, there are 256 combinations in which an obstacle exists or does not exist in a total of 16 grids from grid_0 to grid_3 from time t1 to time t4. The identification unit 13 calculates all E(x) related to these 256 patterns using Equation 2. Here, when calculating a weight using two adjacent grids in the second term of Equation 2, the identification unit 13 uses adjacent grids vertically, horizontally, and diagonally in FIG. 8.

[0062] Furthermore, the specification unit 13 uses Equations 1 and 3 to calculate the probability q (time t2 and grid_0=present) that the label "present" is applied to grid_0 at time t2.

[0063] The label setting unit 21 sets labels for grid_0 to grid_3 at times t1 to t4 using the probability calculated by the identification unit 13 and the determination result of whether or not an obstacle exists in each grid. FIG. 9 shows the labels for each grid set by the label setting unit 21. In FIG. 8, the label setting unit 21 sets a label indicating the presence of an obstacle to a grid where it is determined that an obstacle exists, and sets a label indicating the absence of an obstacle to a grid where it is determined that an obstacle does not exist. Furthermore, FIG. 9 shows that the label setting unit 21 sets a label indicating the absence of an obstacle to grid_0 at time t2 where it is determined that it is unclear whether an obstacle exists. For example, the label setting unit 21 may set a label indicating the presence of an obstacle when the probability of the presence of an obstacle exceeds a threshold, and may set a label indicating the absence of an obstacle when the probability of the presence of an obstacle is equal to or less than the threshold.

[0064] The route control unit 22 controls the movement route of the moving object 40 by using the label set by the label setting unit 21. Here, the label setting unit 21 may use the set label to determine whether an obstacle is moving or stationary.

[0065] For example, the label setting unit 21 increases the number of types of labels to be set for each grid. For example, the label setting unit 21 uses a label in which the label is "stop" when a stationary obstacle exists, the label is "absent" when an obstacle does not exist, the label is "left" when an obstacle moving to the left exists, and the label is "right" when an obstacle moving to the right exists. In this case, there are 4^16 (=4294967296) patterns of combinations in which a stationary obstacle exists, an obstacle does not exist, an obstacle moving to the left exists, and an obstacle moving to the right exists in the 16 grids. The identification unit 13 calculates all E(x) related to these 4^16 patterns using Equation 2. Furthermore, the identification unit 13 increases the number of weight patterns to be used in the second term of Equation 2. For example, when the determination results of grid_x at time t and grid_x-1 at time t+1, which are diagonally adjacent to each other, are (present, present), the identification unit 13 sets the weight for attaching the label (left, left) to 1. That is, the specification unit 13 determines P ij (left, left | present, present) = 1. Furthermore, the specification unit 13 specifies P ij (right, right | present, present) = 1. In addition, the specification unit 13 specifies P ij A weight may be set for the case where the determination result includes "unknown", such as (left, left | yes, unknown) = 1. Also, when the determination results of grid_x at time t and grid_x at time t+1 are (yes, yes), the weight for attaching the label "stop, stop" may be set to 1. That is, the identification unit 13 calculates the weight for P ijIt may be set as (Stop, Stop | Present, Present) = 1. The identification unit 13 uses Equation 1 and Equation 3 to calculate the probability that there is a stationary obstacle in each grid, the probability that there is no obstacle, the probability that there is an obstacle moving to the right, or the probability that there is an obstacle moving to the left. Furthermore, if the number of labels increases, the calculation load on the identification unit 13 increases. Therefore, the identification unit 13 may perform an approximation calculation using belief propagation.

[0066] The label setting unit 21 uses the respective probabilities calculated by the specification unit 13 to determine the labels to be set for each grid.

[0067] The route control unit 22 sets a route for the mobile unit 40 so as to avoid moving and stationary obstacles.

[0068] The identification unit 13 may also calculate the probability that an obstacle exists at a time when no observation data exists, and the label setting unit 21 may set a label according to the calculation result. For example, the determination unit 12 determines whether or not an obstacle exists for each grid up to time t4 using the observation data. For time t5 and thereafter, the identification unit 13 calculates the probability that an obstacle exists in each grid using the grid data up to time t4 and the determination result. For example, the identification unit 13 assumes that grid_0 at time t5 is grid data indicating that it is unclear whether an obstacle exists, and calculates the probability that an obstacle exists in grid_0 at time t5 using grid data from time t1 to time t5. The identification unit 13 may also calculate the probability that an obstacle exists in grid_0 at time t5 using grid data of grid_1 at time t4, which is grid data diagonally opposite to grid_0 at time t5.

[0069] As described above, the label setting unit 21 can update the labels set for each grid over time. This allows the label setting unit 21 to determine whether a stationary obstacle or a moving obstacle is present in each grid. The route control unit 22 sets a route taking into consideration stationary and moving obstacles, thereby preventing the moving body 40 from colliding with the obstacle.

[0070] In addition, the identification unit 13 can calculate the probability that an obstacle exists in each grid at a future time when no observation data is received. This allows the route control unit 22 to set a long-term movement route for the moving object 40.

[0071] (Embodiment 4) Next, a configuration example of the control device 60 according to the fourth embodiment will be described with reference to Fig. 10. The control device 60 has a configuration in which an evaluation unit 61 and a speed control unit 62 are added to the control device 20 in Fig. 4. In the following, differences between the control device 60 and the control device 20 will be mainly described, and detailed descriptions of functions or configurations similar to those of the control device 20 will be omitted.

[0072] The evaluation unit 61 evaluates the safety when the moving body 40 moves along the route set in the route control unit 22. The moving body 40 may meander without moving accurately along the route set in the route control unit 22. For example, the moving body 40 may meander due to factors such as the condition of the ground on which the moving body 40 is in contact, external environmental conditions such as weather conditions, or internal environmental conditions such as aging deterioration of the moving body 40. The evaluation unit 61 calculates a score indicating whether the moving body 40 can travel safely, taking into account the probability that the moving body 40 does not move accurately along the set route. The evaluation unit 61 calculates a higher score as the possibility of the moving body 40 colliding with an obstacle decreases.

[0073] Fig. 11 shows the travel area of ​​the moving object 40. Specifically, Fig. 11 shows the travel area of ​​the moving object 40 viewed from above in a direction perpendicular to the two-dimensional area which is the travel area of ​​the moving object 40. Fig. 11 shows that the moving object 40 moves in a straight line in the X direction along the route set in the route control unit 22.

[0074] On the other hand, Fig. 12 shows that the moving object 40 is traveling in a meandering manner, not along the route set by the route control unit 22. Fig. 12 shows a number of routes along which the moving object 40 meanders, that is, route variation patterns. A probability model showing that the moving object 40 does not necessarily travel along the route set as shown in Fig. 12 is expressed by the following Equation 4.

[0075] TIFF0007609285000004.tif11137

[0076] r=(x, y, θ) is a vector representing the state of the robot. f is a function describing the dynamics of the robot. u is the control input. S is the covariance matrix. dB t is the increment of Brownian motion in a small time. Equation 4 expresses the SdB , which indicates the stochastic effect of the moving object 40 meandering, on dr = f (r, u) dt, which indicates the position of the grid according to the time change dt, when a function f (r, u) is given. t is added. That is, Equation 4 indicates the fluctuation of the position of the grid in which the mobile object 40 exists at a certain time or period. In other words, Equation 4 indicates the fluctuation pattern of the route shown in FIG.

[0077] Here, an example of the calculation of the score executed by the evaluation unit 61 will be described with reference to Figs. 13 and 14. Fig. 13 shows that the moving body 40 moves so as not to collide with stationary obstacles 71 and 72 and a moving obstacle 73. The stationary obstacles 71 and 72 may be, for example, a wall, and the moving obstacle 73 may be, for example, a walking person. An arrow added to the moving obstacle 73 indicates the moving path of the obstacle 73. Also, an arrow added to the moving body 40 indicates the moving path of the moving body 40 set to pass between the stationary obstacles 71 and 72.

[0078] 14 shows the speed patterns of the moving body 40. For example, speed pattern 1 indicates that the moving body 40 moves at the speed Vm without decelerating. Speed ​​pattern 2 indicates that the moving body 40 decelerates until it approaches the moving path of the moving obstacle 73, and accelerates when it reaches the moving path of the moving obstacle 73 or when it exceeds the moving path of the moving obstacle 73. For example, speed pattern 2 takes into consideration a case where the moving obstacle 73 is passed before the moving body 40 reaches the moving path of the moving obstacle 73, and the moving body 40 passes the moving path of the moving obstacle 73 after the moving obstacle 73 has passed.

[0079] The speed pattern is not limited to the pattern shown in Fig. 14. For example, the speed pattern may indicate that the speed of the moving body 40 increases monotonically, or may indicate that the speed of the moving body 40 decreases monotonically. Alternatively, the speed pattern may indicate that the speed of the moving body 40 increases monotonically and then decreases monotonically. In addition, the speed pattern in Fig. 14 may be a speed determined based on the moving speed determined by the route control unit 22, or may be a speed set from outside the control device.

[0080] The evaluation unit 61 calculates the safety score of the moving object 40 that moves along the movement route as shown in FIG. 12 in accordance with Equation 4 and further moves at the speed pattern shown in FIG. 13 in accordance with the following Equation 5.

[0081] TIFF0007609285000005.tif10170

[0082] J indicates the cost function. The smaller the value of the cost function, the higher the security. i indicates the time step. v i indicates a predetermined speed in FIG. 13 at the time or period indicated by time step i. m indicates the speed that is kept constant in speed pattern 1. m may be the maximum speed of the moving body 40. i (x, y, i) indicates the probability that an obstacle exists in a grid including the two-dimensional coordinates of the moving object 40 at the time or period indicated by time step i.

[0083] The first term on the right hand side, λv i p i There is a high possibility that an obstacle exists in the grid through which the moving body 40 passes, and the faster the speed of the moving body 40, the larger the value of (x, y, i). In other words, the slower the speed of the moving body 40, the smaller the value of the first term on the right-hand side. Therefore, the slower the speed of the moving body 40, the smaller the value of the cost function can be. However, even if the value of the cost function is reduced by slowing the speed of the moving body 40 as much as possible, thereby increasing safety, the transportation efficiency of the moving body 40 will decrease. Therefore, it is necessary to reduce the value of the cost function while taking into consideration the balance with the second term on the right-hand side. The second term on the right-hand side, (v i -v m ) 2 is the velocity that is kept constant, v mThe value becomes larger as the speed becomes slower compared to . Here, a constant λ is used as a parameter indicating how much importance is placed on safety. Specifically, the larger the value of λ, the higher the ratio of the value of the first term to the value of the cost function compared to the value of the second term, and therefore the higher the safety. Also, in places where the risk is low, such as where p of the first term is zero, the speed becomes faster due to the action of the second term, which contributes to improving the conveyance efficiency. λ may be set in advance by an operator, for example, or may be changed depending on the risk of the conveyed goods (fragile goods, weight, etc.), or may be changed depending on environmental conditions. Also, λ may be changed depending on the shape of the cart on which the conveyed goods are loaded and the height of the loaded goods. Alternatively, λ may be set in advance as a candidate value by an operator, or a value that was found to be safe by accumulating (learning) past conveyance data may be adopted.

[0084] The evaluation unit 61 takes into consideration the uncertainty of the moving path of the moving object 40 and calculates a cost function in the case where the moving object 40 moves along the multiple paths shown in Fig. 12 with the same speed pattern. For example, the evaluation unit 61 may use the maximum value of the multiple cost function values ​​calculated for the same speed pattern as the cost function for that speed pattern. Alternatively, the evaluation unit 61 may use another value, such as the average value, minimum value, or median value, instead of the maximum value, as the cost function for that speed pattern.

[0085] 14 , and determines the speed pattern with the lower cost function, i.e., the safest speed pattern, as the speed pattern for the moving body 40. The speed control unit 62 transmits the speed pattern determined by the evaluation unit 61 to the moving body 40, and controls the speed of the moving body 40.

[0086] As described above, the control device 60 according to the fourth embodiment considers the uncertainty of the moving path of the moving object 40 and calculates a cost function in the case where the moving object 40 moves along a plurality of paths for each speed pattern of the moving object 40. The moving object 40 can reduce the possibility of colliding with an obstacle by moving along a speed pattern determined based on the calculated cost function.

[0087] FIG. 15 is a block diagram showing a configuration example of an information processing device 10, a control device 20, and a control device 60 (hereinafter, referred to as information processing device 10, etc.). Referring to FIG. 15, the information processing device 10, etc. include a network interface 1201, a processor 1202, and a memory 1203. The network interface 1201 may be used to communicate with a network node (e.g., eNB, MME, P-GW, etc.). The network interface 1201 may include, for example, a network interface card (NIC) conforming to the IEEE 802.3 series. Here, eNB stands for evolved Node B, MME stands for Mobility Management Entity, and P-GW stands for Packet Data Network Gateway. IEEE stands for Institute of Electrical and Electronics Engineers.

[0088] The processor 1202 reads out and executes software (computer programs) from the memory 1203 to perform the processing of the information processing device 10 and the like described using the flowcharts in the above-mentioned embodiments. The processor 1202 may be, for example, a microprocessor, an MPU, or a CPU. The processor 1202 may include multiple processors.

[0089] The memory 1203 is configured by a combination of volatile memory and non-volatile memory. The memory 1203 may include storage located away from the processor 1202. In this case, the processor 1202 may access the memory 1203 via an I / O (Input / Output) interface (not shown).

[0090] 15, the memory 1203 is used to store a group of software modules. The processor 1202 reads out and executes the group of software modules from the memory 1203, thereby performing the processing of the information processing device 10 and the like described in the above-mentioned embodiment.

[0091] As described with reference to FIG. 15, each of the processors of the information processing device 10 in the above-described embodiment executes one or more programs including a group of instructions for causing a computer to execute the algorithm described with reference to the drawings.

[0092] In the above examples, the program includes instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more functions described in the embodiments. The program may be stored in a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, the computer-readable medium or tangible storage medium includes random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technology, CD-ROM, digital versatile disc (DVD), Blu-ray® disk or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The program may be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, the transitory computer-readable medium or communication medium includes electrical, optical, acoustic, or other forms of propagated signals.

[0093] The present disclosure is not limited to the above-described embodiment, and can be modified as appropriate without departing from the spirit and scope of the present disclosure.

[0094] A part or all of the above-described embodiments can be described as, but is not limited to, the following supplementary notes. (Appendix 1) A communication unit that receives observation data of obstacles in a plurality of areas from at least one sensor; a determination unit that determines the presence or absence of an obstacle in the plurality of areas using the observation data; a determination unit that, when it is unclear whether an obstacle exists in a first area of ​​the plurality of areas, determines a probability that an obstacle exists in the first area by using a determination result indicating the presence or absence of an obstacle in an area surrounding the first area, The information processing device, wherein the probability that an obstacle exists in the first area varies depending on the presence or absence of an obstacle in an area surrounding the first area. (Appendix 2) The information processing device according to claim 1, further comprising a label setting unit that sets a label indicating whether or not an obstacle exists in the multiple areas using the judgment results for each of the multiple areas and the probability that an obstacle exists in the first area. (Appendix 3) The plurality of areas include: A plurality of grids are generated by dividing a travel area of ​​a moving object, The identification unit is 3. The information processing device according to claim 2, further comprising: calculating a probability that an obstacle exists in a first grid where the observation data indicates that the presence or absence of an obstacle is unknown. (Appendix 4) The identification unit is The information processing device according to claim 3, further comprising: a weight value, which changes depending on a determination result of a second grid adjacent to the first grid, used when the labels are set to the first grid and the second grid, to calculate a probability that an obstacle exists in the first grid. (Appendix 5) The communication unit is receiving observation data of the plurality of grids at a first time and then receiving observation data of the plurality of grids at a second time; The label setting unit 5. The information processing device according to claim 3, further comprising: determining whether or not a moving obstacle is present based on a transition of a determination result of each of the grids at the first time point and the second time point. (Appendix 6) The label setting unit determining whether or not a moving obstacle is present, using weight values ​​when the labels are set to the third grid at the first time and the third grid or a grid adjacent to the third grid at the second time, according to a determination result of the third grid at the first time and a determination result of the third grid or a grid adjacent to the third grid at the second time; 6. The information processing device according to claim 5. (Appendix 7) an evaluation unit that calculates a cost function indicating the safety of traveling of the moving body using a speed pattern that indicates the speed of the moving body, a movement pattern that indicates a movement route of the moving body, the movement pattern being expressed using a probabilistic model, and a probability that an obstacle exists in each grid on the movement route shown in the movement pattern; and 7. The information processing device according to claim 3, further comprising: a speed control unit that controls a speed of the moving object based on the cost function. (Appendix 8) A sensor that observes obstacles in a plurality of areas and transmits observation data indicating the observation results; a determination means for determining the presence or absence of an obstacle in the plurality of areas using the observation data; and a determination means for, when it is unclear whether an obstacle exists in a first area of ​​the plurality of areas, determining a probability that an obstacle exists in the first area by using a determination result indicating the presence or absence of an obstacle in an area surrounding the first area, A communication system, wherein the probability that an obstacle exists in the first area varies depending on the presence or absence of obstacles in areas surrounding the first area. (Appendix 9) The communication system described in claim 8, further comprising a label setting means for setting a label indicating whether or not an obstacle exists in the multiple areas using the judgment results for each of the multiple areas and the probability that an obstacle exists in the first area. (Appendix 10) The plurality of areas include: A plurality of grids are generated by dividing a travel area of ​​a moving object, The identification means is 10. The communication system of claim 9, further comprising: calculating a probability that an obstacle exists in a first grid where the observation data indicates that the presence or absence of an obstacle is unknown. (Appendix 11) The identification means is The communication system according to claim 10, further comprising: calculating a probability that an obstacle exists in the first grid by using a weight value when the labels are set to the first grid and the second grid, the weight value changing depending on a determination result of a second grid adjacent to the first grid. (Appendix 12) a communication means for receiving observation data of the plurality of grids at a second time after receiving the observation data of the plurality of grids at a first time, The label setting means includes: The communication system according to claim 10 or 11, further comprising: determining whether or not a moving obstacle is present based on a transition of the determination results of each of the grids at the first time and the second time. (Appendix 13) The label setting means includes: Label each grid to indicate whether it contains a stationary obstacle, a moving obstacle, or no obstacle. The identification means is determining whether or not a moving obstacle is present, using weight values ​​when the labels are set to the third grid at the first time and the third grid or a grid adjacent to the third grid at the second time, according to a determination result of the third grid at the first time and a determination result of the third grid or a grid adjacent to the third grid at the second time; 13. The communication system of claim 12. (Appendix 14) an evaluation means for calculating a cost function indicating the safety of traveling of the moving body, using a speed pattern indicating the speed of the moving body, a movement pattern indicating a movement route of the moving body, the movement pattern being expressed using a probabilistic model, and a probability that an obstacle exists in each grid on the movement route indicated in the movement pattern; and 14. The communication system according to any one of claims 10 to 13, further comprising: a speed control means for controlling a speed of the moving object based on the cost function. (Appendix 15) receiving observation data from at least one sensor that observes obstacles in a plurality of areas; Using the observation data, determining the presence or absence of obstacles in the plurality of areas; when it is unclear whether an obstacle exists in a first area of ​​the plurality of areas, determining a probability that an obstacle exists in the first area by using a determination result indicating the presence or absence of an obstacle in an area surrounding the first area; A method for identifying, wherein the probability that an obstacle exists in the first area varies depending on the presence or absence of obstacles in areas surrounding the first area. (Appendix 16) The method of claim 15, further comprising setting labels indicating whether or not an obstacle exists in the multiple areas using the judgment results for each of the multiple areas and the probability that an obstacle exists in the first area. (Appendix 17) The plurality of areas include: A plurality of grids are generated by dividing a travel area of ​​a moving object, In determining the probability that the obstacle is present, 17. The method of claim 16, further comprising calculating a probability that an obstacle is present in a first grid where the observation data indicates that the presence or absence of an obstacle is unknown. (Appendix 18) In determining the probability that the obstacle is present, The method of claim 17, further comprising calculating a probability that an obstacle exists in the first grid by using a weight value when the labels are set to the first grid and the second grid, the weight value changing depending on a determination result of a second grid adjacent to the first grid. (Appendix 19) When receiving the observation data, after receiving observation data of the plurality of grids at a first time, receiving observation data of the plurality of grids at a second time; The identification method according to claim 17 or 18, further comprising determining whether or not a moving obstacle is present based on a transition of the determination results of each of the grids at the first time and the second time. (Appendix 20) When setting the label, determining whether or not a moving obstacle is present, using weight values ​​when the labels are set to the third grid at the first time and the third grid or a grid adjacent to the third grid at the second time, according to a determination result of the third grid at the first time and a determination result of the third grid or a grid adjacent to the third grid at the second time; The method of identification described in Appendix 19. (Appendix 21) receiving observation data from at least one sensor that observes obstacles in a plurality of areas; Using the observation data, determining the presence or absence of obstacles in the plurality of areas; a program for causing a computer to execute, when it is unknown whether an obstacle exists in a first area among the plurality of areas, determining a probability that an obstacle exists in the first area by using a determination result indicating the presence or absence of an obstacle in an area surrounding the first area, The probability that an obstacle exists in the first area varies depending on the presence or absence of obstacles in areas surrounding the first area.

[0095] Although the present invention has been described above with reference to the embodiment, the present invention is not limited to the above. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the invention.

[0096] This application claims priority based on Japanese Patent Application No. 2021-140626, filed on August 31, 2021, the disclosure of which is incorporated herein in its entirety. [Explanation of symbols]

[0097] 10. Information processing device 11 Communications Department 12 Judgment section 13 Specific section 20 Control device 21 Label setting section 22 Route control section 30 Sensors 31 Sensors 32 Sensors 33 Sensors 40 Mobile 50 Network 60 Control device 61 Evaluation Department 62 Speed ​​control section 71 Obstacles 72 Obstacles 73 Obstacles

Claims

1. a communication means for receiving, from at least one sensor, first observation data in which an obstacle is observed at a first time in a plurality of areas generated by dividing a travel area of ​​the mobile object, and second observation data in which the obstacle is observed at a second time in the plurality of areas; a determination means for determining the presence or absence of an obstacle in the plurality of areas by using the first observation data or the second observation data; a determination means for, when it is unclear whether an obstacle exists in a first area of ​​the plurality of areas, determining a probability that an obstacle exists in the first area by using a determination result indicating the presence or absence of an obstacle in an area surrounding the first area; a label setting means for setting a label indicating whether or not an obstacle exists in each of the plurality of areas by using the determination results for each of the plurality of areas and the probability that an obstacle exists in the first area, The label setting means includes: determining whether or not a moving obstacle is present in the third area at the first time and in the third area or an area adjacent to the third area at the second time, using a weight value related to the label, according to a determination result of the third area at the first time and a determination result of the third area or an area adjacent to the third area at the second time; Information processing device.

2. The plurality of areas include: A plurality of grids are generated by dividing a travel area of ​​a moving object, The identification means is The information processing apparatus according to claim 1 , further comprising: a calculation unit for calculating a probability that an obstacle exists in a first grid in which the first observation data or the second observation data indicates that the presence or absence of an obstacle is unknown.

3. The identification means is 3. The information processing device according to claim 2, further comprising: a weight value, which changes depending on a determination result of a second grid adjacent to the first grid, used when the labels are set to the first grid and the second grid, to calculate a probability that an obstacle exists in the first grid.

4. The communication means includes: receiving observation data of the plurality of grids at a first time and then receiving observation data of the plurality of grids at a second time; The label setting means includes: The information processing apparatus according to claim 2 , further comprising: determining whether or not a moving obstacle is present based on a transition of determination results for each of the grids at the first time point and the second time point.

5. an evaluation means for calculating a cost function indicating the safety of traveling of the moving body, using a speed pattern indicating the speed of the moving body, a movement pattern indicating a movement route of the moving body, the movement pattern being expressed using a probabilistic model, and a probability that an obstacle exists in each grid on the movement route indicated in the movement pattern; and The information processing apparatus according to claim 1 , further comprising: a speed control unit that controls a speed of the moving object based on the cost function.

6. A sensor that transmits first observation data observing an obstacle at a first time in a plurality of areas generated by dividing a driving area of ​​a moving body, and second observation data observing the obstacle at a second time in the plurality of areas; a determination means for determining the presence or absence of an obstacle in the plurality of areas by using the first observation data or the second observation data; a determination means for, when it is unclear whether an obstacle exists in a first area of ​​the plurality of areas, determining a probability that an obstacle exists in the first area by using a determination result indicating the presence or absence of an obstacle in an area surrounding the first area; a label setting means for setting a label indicating whether or not an obstacle exists in each of the plurality of areas by using the determination results for each of the plurality of areas and the probability that an obstacle exists in the first area, The label setting means includes: determining whether or not a moving obstacle is present in the third area at the first time and in the third area or an area adjacent to the third area at the second time, using a weight value related to the label, according to a determination result of the third area at the first time and a determination result of the third area or an area adjacent to the third area at the second time; Communication systems.

7. receiving, from at least one sensor, first observation data in which an obstacle is observed at a first time in a plurality of areas generated by dividing a travel area of ​​the moving object, and second observation data in which the obstacle is observed at a second time in the plurality of areas; determining the presence or absence of an obstacle in the plurality of areas using the first observation data or the second observation data; when it is unclear whether an obstacle exists in a first area of ​​the plurality of areas, determining a probability that an obstacle exists in the first area by using a determination result indicating the presence or absence of an obstacle in an area surrounding the first area; setting labels indicating whether or not an obstacle exists in each of the plurality of areas using the determination results for each of the plurality of areas and the probability that an obstacle exists in the first area; determining whether or not a moving obstacle is present in the third area at the first time and in the third area or an area adjacent to the third area at the second time, using a weight value related to the label, according to a determination result of the third area at the first time and a determination result of the third area or an area adjacent to the third area at the second time; Specific method.

8. receiving, from at least one sensor, first observation data in which an obstacle is observed at a first time in a plurality of areas generated by dividing a travel area of ​​the moving object, and second observation data in which the obstacle is observed at a second time in the plurality of areas; determining the presence or absence of an obstacle in the plurality of areas using the first observation data or the second observation data; when it is unclear whether an obstacle exists in a first area of ​​the plurality of areas, using a determination result indicating the presence or absence of an obstacle in areas surrounding the first area, determining a probability that an obstacle exists in the first area, the probability varying depending on the presence or absence of an obstacle in areas surrounding the first area; setting labels indicating whether or not an obstacle exists in each of the plurality of areas using the determination results for each of the plurality of areas and the probability that an obstacle exists in the first area; A program that causes a computer to determine whether or not a moving obstacle is present in the third area at the first time and in the third area or an area adjacent to the third area at the second time, using a weight value related to the label, depending on a judgment result of the third area at the first time and a judgment result of the third area or an area adjacent to the third area at the second time.

Images