Direction of change prediction system, movement system, direction of change prediction method, and program

Abstract

To predict a direction that a person makes a turnaround more accurately.SOLUTION: A turnaround direction prediction system includes: leg state detection means for detecting whether person's right and left legs are in an idling state or in a standing state; breast rotation detection means for detecting rotation information around a pitch axis, a yaw axis, and a roll axis of the person's breast; and direction prediction means for predicting a direction that the person makes a turnaround on the basis of the state of the right and left legs detected by the leg state detection means and the rotation information on the breast detected by the breast rotation detection means.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a turning direction prediction system, a movement system, a turning direction prediction method, and a program for predicting the direction in which a person turns.

Background Art

[0002] There is known a system that finds a template matching a person's pose and estimates that the pose is an intention pre-associated with the template (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the above system, it cannot always be said that a person's pose is an intention pre-associated with the template, and there is a risk that the direction in which a person turns cannot be accurately predicted.

[0005] The present invention has been made in view of such problems, and a main object thereof is to provide a turning direction prediction system, a movement system, a turning direction prediction method, and a program that can predict the direction in which a person turns with higher accuracy.

Means for Solving the Problems

[0006] One aspect of the present invention for achieving the above object is leg state detection means for detecting whether the left and right legs of a person are in a free leg state or a standing leg state, chest rotation detection means for detecting rotation information around the pitch axis, yaw axis, and roll axis of the person's chest, A direction prediction means predicts the direction in which the person will change direction based on the state of the left and right legs detected by the leg state detection means and the chest rotation information detected by the chest rotation detection means. A system for predicting the direction of change, equipped with That is the case. In this embodiment, the direction prediction means is Based on the rotational information of the chest around the pitch axis, yaw axis, and roll axis detected by the chest rotation detection means, if it is determined that the person's spine is in an extended state and the chest is rotating and laterally flexing in the opposite direction to the direction of movement, it is predicted that the person will change direction towards the swing leg direction. The direction in which the person will change direction may be predicted based on the predicted result of the change of direction in the swing leg direction and the state of the person's left and right legs detected by the leg state detection means. In this embodiment, the direction prediction means is Based on the rotational information of the chest around the pitch axis, yaw axis, and roll axis detected by the chest rotation detection means, if it is determined that the person's spine is in a flexed state and the chest is rotating and laterally flexed in the same direction as the direction of movement, it is predicted that the person will change direction toward the stance leg. The direction in which the person will change direction may be predicted based on the predicted result of the change of direction in the stance direction and the state of the person's left and right legs detected by the leg state detection means. In one embodiment, the neck direction detection means for detecting the rotational direction of the person's neck around the yaw axis is further provided. The direction prediction means is Based on the state of the left and right legs detected by the leg state detection means and the chest rotation information detected by the chest rotation detection means, the provisional direction in which the person will change direction is predicted. If it is determined that the predicted provisional direction and the rotational direction of the neck around the yaw axis detected by the neck direction detection means are the same, the provisional direction may be predicted as the direction in which the person will change direction. In one embodiment, the eye direction detection means for detecting the direction of the person's eyes is further provided, The direction prediction means is Based on the state of the left and right legs detected by the leg state detection means and the chest rotation information detected by the chest rotation detection means, the provisional direction in which the person will change direction is predicted. If it is determined that the predicted provisional direction and the direction of the eyes detected by the eye direction detection means are the same, the provisional direction may be predicted as the direction in which the person will change direction. In this embodiment, the direction prediction means may predict the direction in which the person will change direction based on the state of the left and right legs detected by the leg state detection means, the rotation information of the chest detected by the chest rotation detection means, the rotation direction of the neck detected by the neck direction detection means, and the direction of the eyes detected by the eye direction detection means. In this embodiment, the direction prediction means is Based on the state of the left and right legs detected by the leg state detection means and the chest rotation information detected by the chest rotation detection means, the provisional direction in which the person will change direction is predicted. If it is determined that the direction of the eyes, the direction of rotation of the neck around the yaw axis, and the direction of rotation of the chest around the yaw axis are all pointing in the same direction in this order, the provisional direction may be predicted to be the direction in which the person will change direction. One aspect of the present invention for achieving the above objective is: The above-mentioned change direction prediction system, A warning means that gives a warning to the person based on the direction of the person's change of direction predicted by the direction prediction means, It may also be a mobile system equipped with the following features. One aspect of the present invention for achieving the above objective is: The above-mentioned change direction prediction system, A control means that controls the moving body so as to avoid the person based on the direction of the person's change of direction predicted by the direction prediction means, It may also be a mobile system equipped with the following features. One aspect of the present invention for achieving the above objective is: A step of detecting whether the left and right legs of a person are in a swing position or a stance position, The step of detecting rotation information around the pitch axis, yaw axis, and roll axis of the person's chest; Based on the detected states of the left and right leg portions and the detected rotation information of the chest, the step of predicting the direction in which the person changes direction; A direction change prediction method including the above. It may be. One aspect of the present invention for achieving the above object is A process of detecting whether the left and right leg portions of a person are in a free leg state or a standing leg state; A process of detecting rotation information around the pitch axis, yaw axis, and roll axis of the person's chest; Based on the detected states of the left and right leg portions and the detected rotation information of the chest, a process of predicting the direction in which the person changes direction; A program for causing a computer to execute the above. It may be.

Effect of the Invention

[0007] According to the present invention, it is possible to provide a direction change prediction system, a movement system, a direction change prediction method, and a program that can predict more accurately the direction in which a person changes direction.

Brief Description of the Drawings

[0008] [Figure 1] It is a block diagram showing a schematic system configuration of a direction change prediction system according to the present embodiment. [Figure 2] It is a diagram showing directions around the pitch axis, yaw axis, and roll axis of a person's chest. [Figure 3] It is a diagram showing a spin turn and a step turn. [[ID=~]] [Figure 4] It is a flowchart showing the flow of a direction change prediction method according to the present embodiment. [Figure 5] It is a block diagram showing a schematic system configuration of a direction change prediction system according to the present embodiment. [Figure 6] It is a flowchart showing the flow of a direction change prediction method according to the present embodiment. [Figure 7] This block diagram shows a schematic system configuration of the direction of change prediction system according to this embodiment. [Figure 8] This block diagram shows a schematic system configuration of the direction of change prediction system according to this embodiment. [Figure 9] This figure shows a schematic system configuration of the mobile system according to this embodiment. [Figure 10] This diagram shows a configuration in which a direction-changing prediction system is installed outside the moving body. [Modes for carrying out the invention]

[0009] The present invention will be described below through embodiments, but the claims are not limited to the following embodiments. Furthermore, not all of the configurations described in the embodiments are necessarily essential for solving the problem.

[0010] Embodiment 1 For example, when a moving object approaches a person from behind, or when a person and a moving object are moving side-by-side, the person may suddenly change direction without checking behind or to the side. In this case, a collision between the moving object and the person is possible.

[0011] In contrast, the turning direction prediction system according to this embodiment, as described later, can accurately recognize a person's intention to change direction by utilizing human biomechanics, thereby predicting the direction in which a person will change direction with high accuracy and speed. This makes it possible to reliably avoid collisions between moving objects and people, for example, as described above. Figure 1 is a block diagram showing a schematic system configuration of the turning direction prediction system according to this embodiment.

[0012] The turning direction prediction system 1 according to this embodiment includes a leg state detection unit 2 that detects whether the left and right legs of a person are in a swing leg state or a standing leg state, a chest rotation detection unit 3 that detects rotational information around the pitch axis, yaw axis, and roll axis of the person's chest, and a direction prediction unit 4 that predicts the direction in which the person will change direction.

[0013] The leg state detection unit 2 is a specific example of a leg state detection means. The leg state detection unit 2 detects whether the left and right legs of a person are in a swing leg state or a standing leg state, based on an image of the person acquired by, for example, a 3D camera.

[0014] The stance phase is the state in which a person supports their body by pressing their feet against the ground while walking. The swing phase is the state in which a person lifts their leg and swings it back and forth while walking. In human walking, the left and right legs alternate between the stance phase and the swing phase.

[0015] The leg state detection unit 2 may, for example, use a machine learning machine such as a neural network to learn images of the swing and stance states of a person's left and right legs, and use the learning results to detect the swing and stance states. The leg state detection unit 2 outputs the detected states of the left and right legs to the direction prediction unit 4.

[0016] The chest rotation detection unit 3 is a specific example of a chest rotation detection means. The chest rotation detection unit 3 detects rotational information around the pitch axis, yaw axis, and roll axis of a person's chest. The rotational information includes information such as the direction of rotation, the angle of rotation, and the amount of rotation.

[0017] The chest rotation detection unit 3 generates a human skeletal model based on an image of a person acquired by, for example, a 3D camera, and based on the generated human skeletal model, it detects rotation information around the pitch axis, yaw axis, and roll axis of the person's chest, as shown in Figure 2.

[0018] The chest rotation detection unit 3 may use a machine learning device such as a neural network to learn images of a person's chest and use the learning results to detect rotation information around the pitch axis, yaw axis, and roll axis of the person's chest. The chest rotation detection unit 3 outputs the detected rotation information around the pitch axis, yaw axis, and roll axis of the person's chest to the direction prediction unit 4.

[0019] The direction prediction unit 4 is a specific example of a direction prediction means. Based on the state of the left and right legs detected by the leg state detection unit 2 and the chest rotation information detected by the chest rotation detection unit 3, the direction prediction unit 4 predicts the direction in which a person will change direction. Specifically, the direction prediction unit 4 predicts whether the person will change direction to the right or to the left.

[0020] In the direction of direction prediction system 1 according to this embodiment, a method for determining the direction a person will change direction is set by focusing on the direction change strategy when a person walks and the kinetic chain when a person moves their body. By using this determination method, the direction of direction prediction system 1 can predict the direction a person will change direction with high accuracy and speed. The method for determining the direction a person will change direction will be described in detail below.

[0021] By focusing on the kinetic chain involved in human movement, it's possible to predict a person's next action based on their posture. For example, when a person changes direction, the following actions occur in this order: pushing out the center of gravity, controlling the direction of the center of gravity, controlling the acceleration of the center of gravity, controlling trunk balance, and deciding on a direction change strategy.

[0022] In the decision-making process for changing direction described above, for example, one of the following is decided: a spin turn or a step turn. A spin turn is a direction change movement in which the body is rotated like a spinning top around the pivot leg (standing leg), as shown on the left side of Figure 3, and the free leg is swung in the direction of travel. On the other hand, a step turn is a direction change movement in which, as shown on the right side of Figure 3, the pivot leg remains on the ground while the free leg is placed on the ground, and then the free leg is pushed off, causing the pivot leg (standing leg) to swing in the direction of travel.

[0023] As described above, before a spin turn or step turn is performed, core balance control is performed to maintain balance in the torso. In other words, this core balance control is a preparatory movement for performing a spin turn or step turn. Therefore, by focusing on this preparatory movement, it is possible to predict with high accuracy whether a person will perform a spin turn or a step turn. And, as will be explained later, by predicting spin turns and step turns, it is possible to predict with high accuracy the direction in which a person will change direction.

[0024] When a person performs a spin turn, as a preparatory movement to maintain balance, the spine extends, the chest rotates in the opposite direction of travel, and then the chest flexes laterally. On the other hand, when a person performs a step turn, as a preparatory movement to maintain balance, the spine flexes, the chest rotates in the same direction as travel, and then the chest flexes laterally. Note that the above method for determining direction changes assumes that the person is healthy and excludes elderly people with impaired balance abilities, etc.

[0025] The direction prediction unit 4 can detect, based on the rotation information of the chest around the pitch axis detected by the chest rotation detection unit 3, whether the person's spine is in an extended state or a flexed state.

[0026] The direction prediction unit 4 can detect, based on the rotation information of the chest around the yaw axis detected by the chest rotation detection unit 3, whether the chest is rotating in the same direction as the direction of travel, or whether the chest is rotating in the opposite direction to the direction of travel.

[0027] The direction prediction unit 4 can detect that the chest is laterally flexed, as described above, based on the rotation information of the chest around the roll axis detected by the chest rotation detection unit 3.

[0028] Based on the direction change determination method described above, the direction in which a person will change direction is predicted by the direction prediction unit 4 as follows:

[0029] Based on the rotational information of the chest around the pitch axis, yaw axis, and roll axis detected by the chest rotation detection unit 3, the direction prediction unit 4 determines that the person's spine is in an extended state and the chest is rotating and laterally flexed in the opposite direction to the direction of travel, and predicts that the person will perform a spin turn and change direction in the swing leg direction. Then, based on the predicted change of direction in the swing leg direction and the state of the person's left and right legs detected by the leg state detection unit 2, the direction prediction unit 4 predicts the direction in which the person will change direction.

[0030] For example, the leg state detection unit 2 detects that the right leg is in a swing leg state and the left leg is in a standing leg state. The direction prediction unit 4 predicts that the person will change direction towards the swing leg direction based on the rotation information around the pitch axis, yaw axis, and roll axis of the chest detected by the chest rotation detection unit 3. In this case, the direction prediction unit 4 predicts that the person will change direction to the right based on the swing leg state of the right leg detected by the leg state detection unit 2 and the predicted result of changing direction towards the swing leg direction.

[0031] Furthermore, the direction prediction unit 4, based on the rotational information of the chest around the pitch axis, yaw axis, and roll axis detected by the chest rotation detection unit 3, predicts that the person will perform a step turn and change direction toward the stance leg if it determines that the person's spine is in a flexed state and the chest is rotating and laterally flexed in the same direction as the direction of travel. Then, the direction prediction unit 4 predicts the direction in which the person will change direction based on the predicted change of direction toward the stance leg and the state of the person's left and right legs detected by the leg state detection unit 2.

[0032] For example, the leg state detection unit 2 detects that the right leg is in a swing state and the left leg is in a stance state. The direction prediction unit 4 predicts that the person will change direction towards the stance state based on the rotation information around the pitch axis, yaw axis, and roll axis of the chest detected by the chest rotation detection unit 3. In this case, the direction prediction unit 4 predicts that the person will change direction to the left based on the detection result of the stance state of the left leg detected by the leg state detection unit 2 and the prediction result of the change in direction towards the stance state.

[0033] As described above, the turning direction prediction system 1 according to this embodiment can predict the direction in which a person will change direction (turning right, turning left) based on rotational information around the pitch axis, yaw axis, and roll axis of the person's chest, and the state of the person's left and right legs.

[0034] Next, the method for predicting the direction of change according to this embodiment will be described. Figure 4 is a flowchart showing the flow of the method for predicting the direction of change according to this embodiment. Note that the processing flow shown in Figure 4 may be executed repeatedly at predetermined time intervals.

[0035] The leg state detection unit 2 detects whether the left and right legs of a person are in a swing state or a standing state, and outputs the detection result to the direction prediction unit 4 (step S101).

[0036] The chest rotation detection unit 3 detects rotational information around the pitch axis, yaw axis, and roll axis of the person's chest and outputs the detection result to the direction prediction unit 4 (step S102).

[0037] Based on the rotational information of the chest around the pitch axis, yaw axis, and roll axis detected by the chest rotation detection unit 3, the direction prediction unit 4 predicts that the person will change direction towards the swing leg direction if it detects that the person's spine is in an extended state and the chest is rotating and laterally flexed in the opposite direction to the direction of travel (step S103). (step S104)

[0038] Based on the rotational information of the chest around the pitch axis, yaw axis, and roll axis detected by the chest rotation detection unit 3, the direction prediction unit 4 predicts that the person will change direction towards the stance leg direction if it detects that the person's spine is in a flexed state and the chest is rotating and laterally flexed in the same direction as the direction of travel (step S105). (step S106)

[0039] The direction prediction unit 4 predicts the direction in which the person will change direction based on the predicted results of the swing leg direction and the stance leg direction, and the state of the person's left and right legs detected by the leg state detection unit 2 (step S107).

[0040] Note that the process in (step S101) is executed at the beginning of this process, but is not limited to this. For example, the process in (step S101) may be executed after (steps 102) through (step S106), or simultaneously with them, and may be executed at any time as long as it is executed before (step S107).

[0041] As described above, the turning direction prediction system 1 according to this embodiment comprises: a leg state detection unit 2 that detects whether the left and right legs of a person are in a swing leg state or a standing leg state; a chest rotation detection unit 3 that detects rotational information of the person's chest around the pitch axis, yaw axis, and roll axis; and a direction prediction unit 4 that predicts the direction in which a person will change direction based on the state of the left and right legs detected by the leg state detection unit 2 and the rotational information of the chest detected by the chest rotation detection unit 3.

[0042] According to the direction of change prediction system 1 of this embodiment, by recognizing a person's intention to change direction using human biomechanics, it is possible to predict the direction a person will change direction in with high accuracy and speed.

[0043] Embodiment 2 For example, people tend to perform a preliminary action of turning their heads in the direction they are about to change direction. In this embodiment, the direction prediction unit 4 predicts the direction a person will change direction with higher accuracy by detecting preliminary actions related to the person's neck, in addition to the preliminary actions according to the above embodiment.

[0044] Figure 5 is a block diagram showing a schematic system configuration of the turning direction prediction system according to this embodiment. In this embodiment, the same parts as in the above embodiment are denoted by the same reference numerals and detailed descriptions are omitted. In addition to the configuration of Embodiment 1, the turning direction prediction system 20 according to this embodiment further includes a neck direction detection unit 5 that detects the rotational direction around the yaw axis of a person's neck.

[0045] The neck direction detection unit 5 is a specific example of a neck direction detection means. The neck direction detection unit 5 detects the rotation direction of a person's neck around the yaw axis based on an image of the person acquired by, for example, a 3D camera.

[0046] The direction prediction unit 4 predicts a provisional direction in which a person will change direction based on the state of the left and right legs detected by the leg state detection unit 2 and the rotation information of the chest detected by the chest rotation detection unit 3. Furthermore, if the direction prediction unit 4 determines that the predicted provisional direction is the same as the rotation direction of the neck detected by the neck direction detection unit 5, it predicts the provisional direction as the direction in which the person will change direction.

[0047] In this way, by detecting preparatory movements related to a person's neck in addition to preparatory movements related to their chest (spine), it is possible to predict the direction in which a person will change direction with greater accuracy.

[0048] Figure 6 is a flowchart showing the flow of the turning direction prediction method according to this embodiment. Note that the processing flow shown in Figure 6 may be executed repeatedly at predetermined time intervals. The leg state detection unit 2 detects whether the left and right legs of a person are in a swing state or a standing state, and outputs the detection result to the direction prediction unit 4 (step S201).

[0049] The chest rotation detection unit 3 detects rotational information around the pitch axis, yaw axis, and roll axis of the person's chest and outputs the detection result to the direction prediction unit 4 (step S202).

[0050] The direction prediction unit 4 predicts the provisional direction in which the person will change direction based on the state of the left and right legs detected by the leg state detection unit 2 and the chest rotation information detected by the chest rotation detection unit 3 (step S203).

[0051] The neck direction detection unit 5 detects the rotational direction of the person's neck around the yaw axis based on an image of the person acquired by a 3D camera (step S204).

[0052] The direction prediction unit 4 detects whether the predicted provisional direction is the same as the direction of neck rotation detected by the neck direction detection unit 5 (step S205).

[0053] If the direction prediction unit 4 detects that the predicted provisional direction and the direction of head rotation are the same (YES in step S205), it predicts the provisional direction as the direction in which the person will change direction (step S206).

[0054] The process in step S204 may be executed after or simultaneously with steps 201 through S203, for example, or before step S205, at any given time.

[0055] Embodiment 3 For example, people tend to perform a preliminary action of turning their eyes (gaze) in the direction they are about to change direction. In this embodiment, the direction prediction unit 4 predicts the direction a person will change direction with higher accuracy by detecting preliminary actions related to the person's eyes, in addition to the preliminary actions according to the above embodiment.

[0056] Figure 7 is a block diagram showing a schematic system configuration of the direction of change prediction system according to this embodiment. In this embodiment, the same parts as in the above embodiment are denoted by the same reference numerals, and detailed descriptions are omitted.

[0057] The direction of change prediction system 30 according to this embodiment further includes, in addition to the configuration of Embodiment 1, an eye direction detection unit 6 for detecting the direction of a person's eyes. The eye direction detection unit 6 is a specific example of an eye direction detection means. The eye direction detection unit 6 detects the direction of a person's eyes (line of sight) based on an image of a person acquired by a 3D camera, for example.

[0058] The direction prediction unit 4 predicts a provisional direction in which a person will change direction based on the state of the left and right legs detected by the leg state detection unit 2 and the chest rotation information detected by the chest rotation detection unit 3. Furthermore, if the direction prediction unit 4 determines that the predicted provisional direction is the same as the direction of the eyes detected by the eye direction detection unit 6, it predicts the provisional direction as the direction in which the person will change direction.

[0059] In this way, by detecting not only preparatory movements related to a person's chest (spine) but also preparatory movements related to a person's eyes, it is possible to predict the direction in which a person will change direction with greater accuracy.

[0060] Embodiment 4 In this embodiment, the direction in which a person will change direction is predicted with greater accuracy by detecting preparatory movements related to the person's chest (spine), preparatory movements related to the person's neck, and preparatory movements by the person's eyes.

[0061] Figure 8 is a block diagram showing a schematic system configuration of the direction of change prediction system according to this embodiment. In this embodiment, the same parts as in the above embodiment are denoted by the same reference numerals, and detailed descriptions are omitted.

[0062] The turning direction prediction system 40 according to this embodiment further includes, in addition to the configuration of Embodiment 1, a neck direction detection unit 5 for detecting the rotational direction around the yaw axis of a person's neck, and an eye direction detection unit 6 for detecting the direction of a person's eyes.

[0063] The direction prediction unit 4 predicts the direction in which a person will change direction based on the state of the legs detected by the leg state detection unit 2, the rotation information of the chest detected by the chest rotation detection means, the rotation direction of the neck detected by the neck direction detection means, and the direction of the eyes detected by the eye direction detection means.

[0064] This allows for more accurate prediction of the direction a person will change direction by detecting preparatory movements related to the chest (spine), preparatory movements related to the neck, and preparatory movements of the eyes.

[0065] Here, when a person changes direction, they tend to perform preparatory movements: first, they turn their gaze in the direction they are changing direction (the direction of travel), then their head (neck), and finally their chest. Therefore, when this series of preparatory movements is performed in the same direction, it is thought that they will change direction with a higher probability in the provisional direction predicted based on the rotational information of their chest.

[0066] Taking the above into consideration, in this embodiment, the direction prediction unit 4 predicts the provisional direction in which a person will change direction based on the state of the left and right legs detected by the leg state detection unit 2 and the rotation information of the chest detected by the chest rotation detection unit 3. The direction prediction unit 4 then determines that the direction of the eyes, the rotation direction of the neck around the yaw axis, and the rotation direction of the chest around the yaw axis are all facing in the same direction in this order, and predicts the provisional direction as the direction in which the person will change direction.

[0067] This allows us to capture preparatory movements related to a person's chest (spine), preparatory movements related to a person's neck, and preparatory movements of a person's eyes as a series of preparatory movements, enabling us to predict the direction a person will change direction with greater accuracy.

[0068] Embodiment 5 In the mobile system according to this embodiment, the direction of change prediction systems 1, 20, 30, and 40 according to the above embodiment are mounted on the mobile body. Figure 9 is a diagram showing a schematic system configuration of the mobile system according to this embodiment. In this embodiment, the same parts as in the above embodiment are denoted by the same reference numerals and detailed descriptions are omitted. The mobile body 50 according to this embodiment is configured as an autonomous mobile robot that moves autonomously by driving wheels or the like.

[0069] The moving system 10 includes the turning direction prediction systems 1, 20, 30, and 40 according to the above embodiment, a warning unit 7 that warns people, and a control unit 8 that controls the movement of the moving body 50.

[0070] The direction change prediction systems 1, 20, 30, and 40 have a hardware configuration similar to that of a normal computer, comprising, for example, a processor 11 such as a CPU (Central Processing Unit) or GPU (Graphics Processing Unit), internal memory 12 such as RAM (Random Access Memory) or ROM (Read Only Memory), storage devices 13 such as an HDD (Hard Disk Drive) or SSD (Solid State Drive), input / output I / F 14 for connecting peripheral devices such as displays, and communication I / F 15 for communicating with devices outside the device.

[0071] The warning unit 7 is one specific example of a warning means. The warning unit 7 provides a warning to the user, for example, by outputting sound or light.

[0072] The warning unit 7, for example, based on the direction of the person's turn predicted by the turning direction prediction systems 1, 20, 30, and 40, issues a warning to the person if it determines that the person and the moving object 50 are in danger of colliding. This warning allows the person to recognize the approach of the moving object 50 and avoid a collision between them.

[0073] The control unit 8 is a specific example of a control means. For example, if the control unit 8 determines that a collision is imminent between a person and the moving object 50 based on the direction of a person's turn predicted by the turning direction prediction systems 1, 20, 30, and 40, it controls the moving object 50 to avoid the person. The control unit 8 performs actions such as deceleration control and steering control of the moving object 50 to avoid the person. This makes it possible to avoid a collision between a person and the moving object 50.

[0074] In this embodiment, the mobile body 50 may be configured to include either the warning unit 7 or the control unit 8. Alternatively, at least one of the warning unit 7 or the control unit 8 may be located outside the mobile body 50.

[0075] Furthermore, in this embodiment, the direction of change prediction systems 1, 20, 30, and 40 are mounted on the moving body 50, but are not limited to this configuration. For example, the direction of change prediction systems 1, 20, 30, and 40 may be configured to be provided outside the moving body 50, as shown in Figure 10.

[0076] The turning direction prediction systems 1, 20, 30, and 40 transmit the predicted direction of a person's turning to the mobile body 50 via wireless communication or the like. The warning unit 7 of the mobile body 50 issues a warning to the user based on the direction of the person's turning transmitted from the turning direction prediction systems 1, 20, 30, and 40. The control unit 8 of the mobile body 50 controls the mobile body 50 to avoid the person based on the direction of the person's turning transmitted from the turning direction prediction systems 1, 20, 30, and 40.

[0077] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents.

[0078] The present invention can also be implemented, for example, by having a processor execute a computer program to perform the processes shown in Figures 4 and 6.

[0079] Programs can be stored and supplied to a computer using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memory (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, RAMs (random access memory)).

[0080] Programs may be supplied to a computer by various types of transient computer-readable medium. Examples of transient computer-readable medium include electrical signals, optical signals, and electromagnetic waves. Transitory computer-readable medium can be supplied to a computer via wired communication channels such as electric wires and optical fibers, or via wireless communication channels.

[0081] Each component of the direction of change prediction systems 1, 20, 30, and 40 according to the above-described embodiments can be implemented not only by program, but also, in whole or in part, by dedicated hardware such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field-Programmable Gate Array). [Explanation of symbols]

[0082] 1. Direction prediction system, 2. Leg state detection unit, 3. Chest rotation detection unit, 4. Direction prediction unit, 5. Neck direction detection unit, 6. Eye direction detection unit, 7. Warning unit, 8. Control unit, 11. Processor, 12. Internal memory, 13. Storage device, 20. Direction prediction system, 30. Direction prediction system, 40. Direction prediction system, 50. Mobile body

Claims

1. A leg state detection means for detecting whether a person's left and right legs are in a swing position or a standing position, A chest rotation detection means for detecting rotational information around the pitch axis, yaw axis, and roll axis of the person's chest, A direction prediction means predicts the direction in which the person will change direction based on the state of the left and right legs detected by the leg state detection means and the chest rotation information detected by the chest rotation detection means. A system for predicting the direction of change, equipped with the necessary features.

2. A change direction prediction system according to claim 1, The direction prediction means is Based on the rotational information of the chest around the pitch axis, yaw axis, and roll axis detected by the chest rotation detection means, if it is determined that the person's spine is in an extended state, the chest is rotating in the opposite direction to the direction of the change of direction, and the chest is laterally flexed, it is predicted that the person will change direction in the direction of the swing leg. Based on the predicted result of the change of direction in the swing leg direction and the state of the person's left and right legs detected by the leg state detection means, the direction in which the person will change direction is predicted. A system for predicting the direction of change.

3. A change direction prediction system according to claim 1 or 2, The direction prediction means is Based on the rotational information of the chest around the pitch axis, yaw axis, and roll axis detected by the chest rotation detection means, if it is determined that the person's spine is in a flexed state, that the chest is rotating in the same direction as the direction of the change of direction, and that the chest is laterally flexed, it is predicted that the person will change direction towards the stance direction. Based on the predicted result of the change of direction in the stance direction and the state of the left and right legs of the person detected by the leg state detection means, the direction in which the person will change direction is predicted. A system for predicting the direction of change.

4. A change direction prediction system according to claim 1, The system further includes neck direction detection means for detecting the rotational direction of the person's neck around the yaw axis, The direction prediction means is Based on the state of the left and right legs detected by the leg state detection means and the chest rotation information detected by the chest rotation detection means, the provisional direction in which the person will change direction is predicted. If it is determined that the predicted provisional direction and the rotational direction of the neck around the yaw axis detected by the neck direction detection means are the same, the provisional direction is predicted to be the direction in which the person will change direction. A system for predicting the direction of change.

5. A change direction prediction system according to claim 4, The system further comprises eye direction detection means for detecting the direction of the person's eyes, The direction prediction means is Based on the state of the left and right legs detected by the leg state detection means and the chest rotation information detected by the chest rotation detection means, the provisional direction in which the person will change direction is predicted. If it is determined that the predicted provisional direction and the direction of the eyes detected by the eye direction detection means are the same, the provisional direction is predicted to be the direction in which the person will change direction. A system for predicting the direction of change.

6. A change direction prediction system according to claim 5, The direction prediction means predicts the direction in which the person will change direction based on the state of the left and right legs detected by the leg state detection means, the rotation information of the chest detected by the chest rotation detection means, the rotation direction of the neck detected by the neck direction detection means, and the direction of the eyes detected by the eye direction detection means. A system for predicting the direction of change.

7. A change direction prediction system according to claim 6, The direction prediction means is Based on the state of the left and right legs detected by the leg state detection means and the chest rotation information detected by the chest rotation detection means, the provisional direction in which the person will change direction is predicted. If it is determined that the direction of the eyes, the direction of rotation of the neck around the yaw axis, and the direction of rotation of the chest around the yaw axis are all pointing in the same direction in this order, the provisional direction is predicted to be the direction in which the person will change direction. A system for predicting the direction of change.

8. A change direction prediction system according to claim 1, A warning means that gives a warning to the person based on the direction of the person's change of direction predicted by the direction prediction means, Equipped with, A mobile system.

9. A change direction prediction system according to claim 1, A control means that controls the moving body so as to avoid the person based on the direction of the person's change of direction predicted by the direction prediction means, Equipped with, A mobile system.

10. A step of detecting whether the left and right legs of a person are in a swing position or a stance position, The steps include detecting rotational information about the pitch axis, yaw axis, and roll axis of the person's chest, A step of predicting the direction in which the person will change direction based on the detected state of the left and right legs and the detected chest rotation information, A method for predicting the direction of change, including the method described above.

11. A process for detecting whether a person's left and right legs are in a swing position or a standing position, A process for detecting rotational information about the pitch axis, yaw axis, and roll axis of the person's chest, A process to predict the direction in which the person will change direction based on the detected state of the left and right legs and the detected chest rotation information, A program that causes a computer to execute something.

Images