Control device, control method and program

Abstract

To generate an appropriate route for a moving object based on what characteristics a pedestrian has regarding a movement in future.SOLUTION: A control device that controls, at least temporarily, a moving object that autonomously moves a region where a pedestrian walks includes: a recognizing unit that recognizes the position of the pedestrian in a time-series manner; a calculating unit that calculates, based on the recognition result by the recognizing unit, an index value which represents a cooperation level for each pedestrian; a setting unit which sets a risk region around the pedestrian and which makes the setting in such a way that the smaller the index value is, the larger the risk region becomes; and a generating unit that generates a route through which the moving object should move in future so as to avoid the risk region.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a control device, a control method, and a program.

Background Art

[0002] In recent years, research and practical application have been advanced for a moving body capable of moving in the same space as a pedestrian. This type of moving body autonomously moves by generating a path so as not to approach an obstacle such as a pedestrian too closely. In relation to this, an invention of a device that detects a pedestrian's reaction due to the approach of the moving body and expands or contracts a non-entry area (personal space) according to the detected pedestrian's reaction has been disclosed (Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional technology, when a pedestrian makes a predetermined reaction that has been numerically determined in advance, such as stopping or sharply moving the upper body, the personal space is expanded. However, in such processing, there are cases where the path of the moving body cannot be appropriately generated based on the characteristics of how the pedestrian will move in the future.

[0005] The present invention has been made in consideration of such circumstances, and one of its objects is to provide a control device, a control method, and a program capable of appropriately generating the path of a moving body based on the characteristics of how a pedestrian will move in the future.

Means for Solving the Problems

[0006] ​The control device, control method, and program according to the present invention employ the following configurations. (1): A control device according to one aspect of the present invention is a control device that controls a moving body that autonomously moves in an area where a pedestrian walks, at least temporarily, and includes a recognition unit that recognizes the position of the pedestrian in time series, a calculation unit that calculates an index value indicating the degree of cooperation for each pedestrian based on the recognition result of the recognition unit, a setting unit that sets a risk area around the pedestrian, and the setting unit sets the risk area larger as the index value is smaller, and a generation unit that generates a path that the moving body should travel in the future so as to avoid the risk area.

[0007] (2): In the aspect of (1) above, the calculation unit predicts the position of the pedestrian at a second time point after the first time point at the first time point, and repeatedly executes in time series the calculation of the difference between the recognized position of the pedestrian and the predicted position of the pedestrian at the second time point, and the larger the value obtained by aggregating the differences, the larger the index value.

[0008] (3): A control device according to another aspect of the present invention is a control device that controls a moving body that autonomously moves in an area where a pedestrian walks, at least temporarily, and includes a recognition unit that recognizes the position of the pedestrian in time series, a calculation unit that calculates an index value for each pedestrian based on the recognition result of the recognition unit, a setting unit that sets a risk area around the pedestrian, and the setting unit sets the risk area larger as the index value is lower, and a generation unit that generates a path that the moving body should travel in the future so as to avoid the risk area. The calculation unit predicts the position of the pedestrian at a second time point after the first time point at the first time point, and repeatedly executes in time series the calculation of the difference between the recognized position of the pedestrian and the predicted position of the pedestrian at the second time point, and the larger the value obtained by aggregating the differences, the larger the index value is calculated.

[0009] (4) In any of the aspects (1) to (3) above, the calculation unit calculates the index value by narrowing down to pedestrians who are in a passing relationship with the moving object or other pedestrians.

[0010] (5) In the aspect (4) above, the setting unit sets the risk area for pedestrians for whom the index value has not been calculated, using a value near the upper limit of the range that the index value can take, instead of the index value.

[0011] (6) A control method according to another aspect of the present invention is such that at least temporarily, a control device that controls a moving object that autonomously moves in an area where pedestrians walk recognizes the positions of pedestrians in time series, calculates an index value indicating the degree of cooperation for each pedestrian based on the results of the recognition, sets a risk area around the pedestrians such that the smaller the index value, the larger the risk area, and generates a path that the moving object should travel in the future so as to avoid the risk area.

[0012] (7) A control method according to another aspect of the present invention is such that at least temporarily, a control device that controls a moving object that autonomously moves in an area where pedestrians walk recognizes the positions of pedestrians in time series, calculates an index value for each pedestrian based on the results of the recognition, sets a risk area around the pedestrians such that the lower the index value, the larger the risk area, and generates a path that the moving object should travel in the future so as to avoid the risk area. The calculation includes predicting the position of the pedestrian at a second time point after the first time point at the first time point, and repeatedly executing in time series the calculation of the difference between the recognized position of the pedestrian and the predicted position of the pedestrian at the second time point, and calculating the index value to be larger as the value obtained by aggregating the differences is larger.

[0013] (8) A program according to another aspect of the present invention causes a processor of a control device that controls a moving body that autonomously moves in an area where a pedestrian walks to, at least temporarily, recognize the position of the pedestrian in time series, calculate an index value indicating the degree of cooperation for each pedestrian based on the result of the recognition, set a large risk area around the pedestrian so that the smaller the index value, the larger the risk area, and generate a path that the moving body should travel in the future so as to avoid the risk area.

[0014] (9) A program according to another aspect of the present invention causes a processor of a control device that controls a moving body that autonomously moves in an area where a pedestrian walks to, at least temporarily, recognize the position of the pedestrian in time series, calculate an index value for each pedestrian based on the result of the recognition, set a risk area around the pedestrian so that the lower the index value, the larger the risk area, and generate a path that the moving body should travel in the future so as to avoid the risk area. The calculating includes predicting the position of the pedestrian at a second time point after the first time point at the first time point, and repeatedly executing in time series calculating the difference between the recognized position of the pedestrian and the predicted position of the pedestrian at the second time point, and calculating the index value to be larger as the value obtained by aggregating the differences is larger.

Advantages of the Invention

[0015] According to the aspects (1) to (9) above, it is possible to appropriately generate the path of the moving body based on the characteristics of how the pedestrian will move in the future.

Brief Description of the Drawings

[0016]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Embodiments for Carrying Out the Invention

[0017] Hereinafter, with reference to the drawings, embodiments of the control device, control method, and program of the present invention will be described. The control device of the present invention controls the drive device of the moving body to move the moving body. The moving body in the present invention is one that autonomously moves in the area where pedestrians walk. The area where pedestrians walk includes sidewalks, public open spaces, floors in buildings, etc., and may include roadways. In the following description, it is assumed that no one rides on the moving body, but people may ride on the moving body.

[0018] The moving body leads a person to be led, follows a person to be followed, or moves toward a destination alone. The person to be led or the person to be followed may be, for example, a pedestrian, or may be a robot or an animal. When following a person to be followed, for example, the position around the person to be followed is treated as the destination. In the following description, it is assumed that the moving body, including the person to be followed, moves toward the destination. Note that such an operation is not always performed, and may be performed temporarily. For example, when the moving body is in a predetermined state, the control device of the moving body may temporarily perform an operation by executing the algorithm of the present invention.

[0019] FIG. 1 is a diagram showing the configuration of a moving body 1 equipped with a control device 100. The moving body 1 includes, for example, an HMI 10, a detection device 20, a position specifying device 30, a control device 100, a body unit 5 on which these are mounted, a moving mechanism 40 attached to the body unit 5, and a sensor 50 attached to the moving mechanism 40 or the like.

[0020] The HMI 10 presents various information to the follower F and accepts input operations by the user. The HMI 10 includes various display devices, speakers, buzzers, touch panels, switches, keys, and the like. For example, the HMI 10 accepts input of a destination (a predetermined place, oneself, etc.) by the user.

[0021] The detection device 20 is a device that generates data for recognizing objects and the follower F existing around the moving body 1. The detection device 20 includes, for example, sensors such as a camera, a radar device, a LIDAR (Light Detection and Ranging), and an ultrasonic sensor whose detection range is around the moving body 1, and an object recognition device that specifies an object by performing sensor fusion processing based on the outputs of these sensors.

[0022] The position specifying device 30 is a device that specifies the position of the moving body 1. The position specifying device 30 includes, for example, a GNSS receiver that specifies the position of the host vehicle M based on signals received from GNSS (Global Navigation Satellite System) satellites. The position specifying device 30 may specify or complement the position of the moving body 1 by an INS (Inertial Navigation System) that utilizes the output of the sensor 50 described later. Further, the position specifying device 30 has a function of receiving electromagnetic waves, and may specify or complement the position of the moving body 1 based on the intensity of electromagnetic waves arriving from surrounding electromagnetic wave transmission sources (whose positions are known).

[0023] The movement mechanism 40 is a mechanism for moving the moving body 1 including the body portion 5 in an arbitrary direction. The movement mechanism 40 includes, for example, a plurality of wheels, a drive motor attached to one or more wheels, and a steering device attached to one or more wheels. There are no particular restrictions on the configuration of the movement mechanism 40, and the movement mechanism 40 may include pseudo legs for bipedal walking.

[0024] The sensor 50 is a sensor for detecting the behavior of the moving body 1. The sensor 50 includes, for example, a wheel speed sensor for detecting the speed of the wheels, an acceleration sensor for detecting the acceleration acting on the moving body 1, a yaw rate sensor attached near the center of gravity in the horizontal direction of the body portion 5, a steering angle sensor for detecting the steering angle of the steered wheels (steering wheels), a direction sensor for detecting the direction of the moving body 1 in the horizontal direction, and the like.

[0025] FIG. 2 is a diagram showing an example of the configuration of the control device 100. The control device 100 includes, for example, a recognition unit 110, a calculation unit 120, a setting unit 130, a generation unit 140, and a movement control unit 150. These components are realized, for example, by a hardware processor such as a CPU (Central Processing Unit) executing a program (software). Some or all of these components may be realized by hardware (including a circuit unit; circuitry) such as an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a GPU (Graphics Processing Unit), or may be realized by the cooperation of software and hardware. The program may be stored in advance in a storage device (a storage device having a non-transitory storage medium) such as an HDD (Hard Disk Drive) or a flash memory, or may be stored in a removable storage medium (a non-transitory storage medium) such as a DVD or a CD-ROM, and may be installed in the storage device by mounting the storage medium on a drive device. Note that the control device 100 may hold map information including at least a local map of the location where the moving body 1 operates in the storage unit.

[0026] The processes of the recognition unit 110, the calculation unit 120, the setting unit 130, the generation unit 140, and the movement control unit 150 described below are repeatedly executed (executed in time series) for each control cycle that arrives at a predetermined period. Therefore, until the moving body 1 reaches the destination, it does not mean that it moves along the path generated at a certain point in time. Instead, the path is updated to a new one due to changes in the surrounding environment, etc., and the control content of the moving body 1 is updated based on the updated latest path.

[0027] Based on the information input from the detection device 20, the recognition unit 110 recognizes the objects existing around the moving body 1. The objects include pedestrians including leading subjects and following subjects if any, and static obstacles. The recognition unit 110 recognizes the states of the objects such as the position, velocity, and acceleration. The position of the object is recognized as, for example, the relative position seen from the moving body 1, and is converted into a position on an assumed plane S represented by a two-dimensional plane when viewing the space around the moving body 1 from above, and is used for subsequent processing. In the following description, it is assumed that the position refers to a single point.

[0028] Based on the recognition result of the recognition unit 110, the calculation unit 120 calculates an index value α for each pedestrian. The index value α indicates the degree of cooperation of the pedestrian. This will be described later.

[0029] The setting unit 130 sets a risk, which is an index value indicating the degree to which the follower F should not enter or approach, in the above-mentioned assumed plane S. The larger the value of the risk, the less the follower F should enter or approach, and the closer the value is to zero, the more preferable it is for the follower F to pass. However, this relationship may be reversed. When the follower F can fly by boarding a flying object, the setting unit 130 may perform similar processing in a three-dimensional space instead of the assumed plane S. An example of a "risk area" is a group of areas where the risk regarding the object is not zero.

[0030] The setting unit 130 sets the risk in the assumed plane S not only for the current time t but also for each future time point defined at regular time intervals such as after Δt (time t + Δt), after 2Δt (time t + 2Δt),.... The setting unit 130 predicts the risk at each future time point based on the change in the position of the moving object continuously recognized by the recognition unit 110.

[0031] Figure 3 is a diagram showing an overview of the risks set by the setting unit 130. The setting unit 130 sets risks with ellipses or circles based on the traveling direction and speed as contour lines on the assumed plane S for objects excluding the follower F and the moving body 1, and sets a fixed value of risk for immovable areas BD such as walls. In the figure, DF is the traveling direction of the follower F, and DM is the traveling direction of the moving body 1. R(OB1) is the risk of the stationary object (a person standing still) OB1, R(OB2) is the risk of the moving object (pedestrian) OB2, R(OB3) is the risk of the moving object (pedestrian) OB3, and R(OB4) is the risk of the moving object (pedestrian) OB4. Since the pedestrians are continuously moving, risks are set at different positions from the current time for each future time point. R(OB2)_t is the risk of the object OB2 in a certain control cycle, R(OB2)_t+Δt is the risk of the object OB2 in the next control cycle, and R(OB2)_t+Δ2t is the risk of the object OB2 in the control cycle after that. R(BD) is the risk of the immovable area BD. In the figure, the darkness of the hatching indicates the value of the risk, showing that the darker the hatching, the greater the risk.

[0032] The setting unit 130 may set an ideal path connecting the moving body 1 to the destination of the moving body 1, and add to the risk as the distance from the ideal path increases. Figure 4 is a diagram showing an example of the distribution of risks set considering the deviation from the ideal path IP. In the figure, R(K) is the risk based on the deviation from the ideal path IP. In the example of Figure 4, the ideal path IP is set linearly, but depending on the structure of the place where the follower F moves, the ideal path IP may be set as a polyline or a curve at the turning angle where the follower F turns. In this case, the setting unit 130 (or other components) performs processes such as dividing the moving range into links and determining detailed positions for each link to generate the ideal path IP.

[0033] Based on the risk according to the recognition result of the recognition unit 110, the generation unit 140 generates a path that the mobile body 1 should travel in the future so as to pass through a location with a low risk (in other words, to avoid the risk area). For example, the generation unit 140 generates a path by sequentially connecting positions where the risk corresponding to each future time point does not exceed a threshold value. The path that satisfies this condition is not uniquely determined, and there may be cases where a plurality of path candidates are generated. The generation unit 140 may calculate a score for each of the plurality of path candidates and select the path candidate with the highest score as the path. For example, the generation unit 140 calculates the score so that the score becomes higher as the degree of turning (evaluated by, for example, the angle formed by the vector from a past trajectory point to a target trajectory point and the vector from the target trajectory point to a future trajectory point) is small and the total risk for each passing point is small.

[0034] The movement control unit 150 controls the movement mechanism 40 so that the mobile body 1 moves along the path. The movement control unit 150 controls the drive motor and the steering device so that the position and behavior of the mobile body 1 obtained from the output of the sensor 50 approach the path.

[0035] [Setting of the index value and the risk area based thereon] The relationship between the index value α and the risk area will be described below. The calculation unit 120 repeatedly predicts the position of the pedestrian at the second time point after the first time point in time series, and calculates the difference between the recognized position of the pedestrian at the second time point and the predicted position of the pedestrian at the first time point. The larger the aggregated value of the differences, the larger the calculated index value α. Since the index value α represents the degree and frequency of the pedestrian changing the path, it represents the degree of changing the path by looking at the moving body 1 or other pedestrians, that is, the degree of cooperation. FIG. 5 is a diagram comparing the assumed behavior (path R1) in the passing scene with the moving body 1 for the pedestrian P1 with high cooperation (large index value α) and the assumed behavior (path R2) in the passing scene with the moving body 1 for the pedestrian P2 with low cooperation (small index value α). As shown in the figure, a pedestrian with high cooperation changes the path early to avoid approaching the moving body 1, while a pedestrian with low cooperation is assumed to change the path after approaching the moving body 1 as close as possible.

[0036] In addition, the index value α can be calculated in the same way not only when the pedestrian changes the path by looking at the moving body 1, but also when the pedestrian changes the path by looking at other pedestrians. Even when the pedestrian and the moving body 1 are separated, the index value α may be calculated based on the degree to which the pedestrian changes the path by looking at other pedestrians. Thereby, the index value α can be calculated for a wider range of pedestrians.

[0037] FIG. 6 is a diagram for explaining the content of the processing by the calculation unit 120. In the figure, t, t-1, t-2, … represent control timings (hereinafter referred to as times) that come at regular intervals, Lp(t) represents the position of the pedestrian at time t, and #Lp(t) represents the position of the pedestrian at time t predicted at time t-1. The calculation unit 120 predicts the position of the pedestrian at the next (or a predetermined time ahead) control timing at the control timing that comes at regular intervals. The calculation unit 120 predicts the position of the pedestrian, for example, by assuming a constant speed or a constant acceleration state and extending the movement vector between the previous time and the current time. Then, the calculation unit 120 calculates the index value α based on, for example, Equation (1). In the equation, q is the number of backward steps when aggregating the differences.

[0038] α = Σ k=t-q t {Lp(k)-#LP(k)} …(1)

[0039] The setting unit 130 sets the risk area by reflecting the index value α. Hereinafter, it is assumed that the setting unit 130 sets the risk area in a circular shape, but the risk area is set based on the same principle even when it is set in an elliptical or other shape. For example, the setting unit 130 determines the radius r of the risk area based on Equation (2). In the equation, D is the distance between the moving body and the pedestrian, V is the speed of the pedestrian, and α is the index value. The function f is a function that returns a smaller value as D is larger, a larger value as V is larger, and a smaller value as α is larger.

[0040] r = f(D, V, α) …(2)

[0041] FIG. 7 is a diagram showing an example in which the transition of the risk area set according to the above principle is applied to the pedestrians P1 and P2 shown in FIG. 5. In this figure, it is assumed that the speeds of the pedestrians P1 and P2 are the same. In the figure, Path1(1) and Path1(2) are the paths of the moving object 1 generated in each case. As shown in the figure, both the risk area Rp1 corresponding to the pedestrian P1 and the risk area Rp2 corresponding to the pedestrian P2 increase as they approach the moving object 1. However, the risk area Rp2 is set to be larger at an earlier stage than the risk area Rp1 and will be set to be larger constantly thereafter. From the processing characteristics of the generation unit 140 described above, since the path of the moving object 1 is set to avoid the pedestrian P2 greatly, it is possible to take an avoidance trajectory earlier for a pedestrian with a relatively high contact possibility and a small index value α.

[0042] Thus, according to the embodiment, based on the characteristics of how the pedestrian will move in the future, the path of the moving object can be appropriately generated.

[0043] Note that the calculation unit 120 may perform the process of calculating the index value α described above only for pedestrians who are in a passing relationship with the moving object 1 or other pedestrians. FIG. 8 is a diagram for explaining pedestrians who are in a passing relationship with the moving object 1. As shown in the figure, the calculation unit 120 sets a monitoring area WA centered on the traveling direction DM of the moving object 1, and narrows down to pedestrians whose angle θ formed by the movement vector P→ of the pedestrian with respect to the straight line connecting the pedestrian and the moving object 1 within the monitoring area is within a predetermined range, and calculates the index value α, and may not calculate the index value α for other pedestrians. In this case, since the pedestrians excluded from the narrowing-down are pedestrians with low importance in the path generation of the moving object 1, instead of the index value α, a value near the upper limit value of the range that the index value α can take may be used to determine the radius r of the risk area. Pedestrians who are in a passing relationship with other pedestrians can also be extracted by the same method.

[0044] The embodiment described above can be expressed as follows. A control device that at least temporarily controls a moving body that autonomously moves in an area where a pedestrian walks, One or more storage media that store computer-readable instructions, A processor connected to the one or more storage media, and comprising: By executing the computer-readable instructions, the processor: Recognizes the position of the pedestrian over time, Based on the result of the recognition, calculates an index value indicating the degree of cooperation for each pedestrian, Sets a risk area around the pedestrian so that the smaller the index value, the larger the area, Generates a path that the moving body should follow in the future to avoid the risk area, Control device.

[0045] Also, the above-described embodiment can be expressed as follows. A control device that at least temporarily controls a moving body that autonomously moves in an area where a pedestrian walks, One or more storage media that store computer-readable instructions, A processor connected to the one or more storage media, and comprising: By executing the computer-readable instructions, the processor: Recognizes the position of the pedestrian over time, Based on the result of the recognition, calculates an index value for each pedestrian, Sets a larger risk area around the pedestrian so that the lower the index value, the larger the area, Generate a path that the moving object should follow in the future so as to avoid the risk area, The calculating includes, At a first time point, predicting the position of the pedestrian at a second time point after the first time point, and at the second time point, repeatedly executing in time series calculating the difference between the recognized position of the pedestrian and the predicted position of the pedestrian, and calculating the index value to be larger as the value obtained by aggregating the differences is larger, Control device.

[0046] As described above, the embodiments for implementing the present invention have been described using the embodiments, but the present invention is not limited to such embodiments at all, and various modifications and substitutions can be made without departing from the gist of the present invention.

Description of Reference Numerals

[0047] 1 Moving object 10 HMI 20 Detection device 30 Position specifying device 40 Moving mechanism 50 Sensor 100 Control device 110 Recognition unit 120 Calculation unit 130 Setting unit 140 Generation unit 150 Movement control unit

Claims

1. A control device that controls a moving body that autonomously moves in an area where a pedestrian walks, at least temporarily, comprising: a recognition unit that recognizes the position of the pedestrian over time; a calculation unit that calculates an index value indicating the degree of cooperation for each pedestrian based on the recognition result of the recognition unit; a setting unit that sets a risk area around the pedestrian, wherein the smaller the index value, the larger the risk area is set; a generation unit that generates a path that the moving body should follow in the future so as to avoid the risk area; A control device comprising the above.

2. The calculation unit: At a first time point, predicts the position of the pedestrian at a second time point after the first time point, and at the second time point, repeatedly executes in time series the calculation of the difference between the recognized position of the pedestrian and the predicted position of the pedestrian, and the larger the aggregated value of the differences, the larger the index value is set. The control device according to Claim 1.

3. A control device that controls a moving body that autonomously moves in an area where a pedestrian walks, at least temporarily, comprising: a recognition unit that recognizes the position of the pedestrian over time; a calculation unit that calculates an index value for each pedestrian based on the recognition result of the recognition unit; a setting unit that sets a risk area around the pedestrian, wherein the lower the index value, the larger the risk area is set; a generation unit that generates a path that the moving body should follow in the future so as to avoid the risk area; Comprising: The calculation unit: At a first time point, predicts the position of the pedestrian at a second time point after the first time point, and at the second time point, repeatedly executes in time series the calculation of the difference between the recognized position of the pedestrian and the predicted position of the pedestrian, and the larger the aggregated value of the differences, the larger the index value is calculated. A control device.

4. The calculation unit calculates the index value by narrowing down to pedestrians who are in a passing relationship with the moving body or other pedestrians. The control device according to any one of Claims 1 to 3.

5. For pedestrians for whom the index value has not been calculated, the setting unit sets the risk area using a value near the upper limit of the possible range of the index value instead of the index value. The control device according to Claim 4.

6. A control device that controls a moving body that autonomously moves in an area where a pedestrian walks, at least temporarily, recognizes the position of the pedestrian over time; Based on the result of the recognition, calculating an index value indicating the degree of cooperation for each pedestrian; Setting a large risk area around the pedestrian such that the smaller the index value, the larger the risk area; Generating a path that the moving object should follow in the future so as to avoid the risk area; A control method for executing the above.

7. A control device that controls a moving object that autonomously moves in an area where a pedestrian walks, at least temporarily, Recognizing the position of the pedestrian over time; Calculating an index value for each pedestrian based on the result of the recognition; Setting a risk area around the pedestrian such that the lower the index value, the larger the risk area; Generating a path that the moving object should follow in the future so as to avoid the risk area; Executing the above, The calculating includes: At a first time point, predicting the position of the pedestrian at a second time point after the first time point, and at the second time point, repeatedly calculating the difference between the recognized position of the pedestrian and the predicted position of the pedestrian over time, and calculating the index value to be larger as the aggregated value of the differences is larger; A control method.

8. A program for causing a processor of a control device that controls a moving object that autonomously moves in an area where a pedestrian walks, at least temporarily, To recognize the position of the pedestrian over time; To calculate an index value indicating the degree of cooperation for each pedestrian based on the result of the recognition; To set a large risk area around the pedestrian such that the smaller the index value, the larger the risk area; To generate a path that the moving object should follow in the future so as to avoid the risk area; To execute the above.

9. A program for causing a processor of a control device that controls a moving object that autonomously moves in an area where a pedestrian walks, at least temporarily, To recognize the position of the pedestrian over time; To calculate an index value for each pedestrian based on the result of the recognition; To set a risk area around the pedestrian such that the lower the index value, the larger the risk area; To generate a path that the moving object should follow in the future so as to avoid the risk area; To execute the above, and The calculating includes: At a first point in time, predict the position of the pedestrian at a second point in time that is after the first point in time, and at the second point in time, calculate the difference between the recognized position of the pedestrian and the predicted position of the pedestrian, and repeatedly execute this in time series, and the greater the value obtained by aggregating the differences, the greater the calculated value of the index value, including Program.

Images