JP2020009107A - Evacuation guidance device, evacuation guidance system, and control program - Google Patents

Abstract

To provide an evacuation guidance device, evacuation guidance system, and control program which, on the basis of an action pattern of a moving entity after a notification, properly performs succeeding notification.SOLUTION: An evacuation guidance device 200 includes an acquisition unit 210 that acquires moving entity information obtained by measuring a moving entity in a predetermined area, an analysis unit 220 that analyzes an action pattern of the moving entity on the basis of the acquired moving entity information, and a decision unit 240 that determines the contents of notification on the basis of the action pattern, which is analyzed by the analysis unit 220, after a notification unit 300 has made the first notification, and implements control so that the notification unit 300 makes the second notification to the moving entity using the determined contents of notification.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an evacuation guidance device, an evacuation guidance system, and a control program.

  Conventionally, moving objects such as vehicles and pedestrians are detected by detection sensors such as ultrasonic sensors and light sensors that detect the inside of the detection area, and sound, voice, lighting or blinking of light, display of characters or images There is known a technique of notifying a moving object of entering a detection area by vibration or vibration.

  In Patent Document 1, in order to avoid unnecessary notification by detecting a moving object which is not originally a detection target, a plurality of Doppler radars using microwaves or the like are used to detect the absolute distance of the moving object, Measure the speed and the direction of movement respectively. Then, there is disclosed a technique of notifying only when the measured absolute distance, speed, and moving direction of the moving object are within a preset range.

JP-A-2004-192097

  However, in the technique of Patent Document 1, the notification is performed based on the absolute distance, the speed, and the moving direction of the moving object, and thus the notification may not function effectively. For example, in a dangerous area such as an expressway where pedestrians are prohibited from entering, if the pedestrian who has entered is not sufficiently recognizable, the notification may not function effectively even if it is notified.

  The present invention has been made to solve such a problem. That is, an object is to appropriately perform the next notification based on the behavior pattern of the moving body after the notification.

  The above object of the present invention is achieved by the following means.

(1) an acquisition unit that acquires moving body information obtained by measuring a moving body in a predetermined area;
Based on the obtained moving body information, an analysis unit that analyzes the behavior pattern of the moving body,
After the first notification by the notification unit, the notification content is determined based on the behavior pattern analyzed by the analysis unit, and the notification unit performs the second notification to the moving object with the determined notification content. And a determination unit that controls
Evacuation guidance device provided with.

  (2) The evacuation according to (1), wherein the determination unit retains a history of the action pattern and determines the content of the second notification based on the action pattern before and after the first notification. Guidance device.

(3) a type analysis unit that analyzes a type of the moving object based on the moving object information;
If the type of the moving object determined by the type analysis unit is a specific type of moving object that is not allowed to enter the predetermined area, the analysis unit analyzes the behavior pattern for the moving object. (1) The evacuation guidance device according to (2).

(4) the acquisition unit acquires, as the moving object information, distance measurement point group data indicating a distribution of distance values to a moving object existing in the predetermined area;
The behavior pattern determined by the analysis unit includes a moving locus of the moving object, a moving speed of the moving object, a falling state of the moving object, a position of the moving object, and a set dangerous area, which are obtained from distance measurement point group data. The evacuation guidance device according to any one of (1) to (3), wherein at least one of the distances of the moving body to the evacuation guidance device is included.

(5) the acquiring unit acquires image data obtained by photographing the inside of the predetermined area;
Further, from the image data acquired by the acquisition unit, comprises an attribute analysis unit for analyzing the attributes of the moving object,
The notification content of the second notification is determined based on the attribute of the moving object analyzed by the attribute analysis unit and the behavior pattern determined by the analysis unit. The evacuation guidance device according to item 1.

(6) The notification unit is one or more types of notification units,
The determination unit selects the notification unit together with the notification content according to the action pattern, and controls the selected notification unit to perform a second notification to the moving object. The evacuation guidance device according to any one of (5).

(7) The plurality of notification units are a plurality of notification units arranged at different positions,
The acquisition unit acquires, as the moving object information, distance measurement point group data indicating a distribution of distance values to an object in the predetermined region acquired from the distance measurement unit that measures the inside of the predetermined region,
The evacuation guidance device according to (6), wherein the determination unit selects the notification unit based on a movement locus of the moving object or a position value of the moving object obtained from the ranging point group data.

(8) a moving body information measuring unit that generates moving body information of a moving body in a predetermined area by measuring;
The reporting department,
The evacuation guidance device according to any one of (1) to (7), wherein the notification unit notifies the moving body using the moving body information acquired from the moving body information measurement unit,
Evacuation guidance system equipped with.

(9) A control program for controlling an evacuation guidance device that performs evacuation guidance for a moving body in a predetermined area,
(A) acquiring moving body information obtained by measuring a moving body in a predetermined area;
(B) analyzing an action pattern of the moving object based on the obtained moving object information;
(C) performing a first notification to the moving body by a notification unit;
(D) determining, after step (c), notification contents based on the action pattern determined in step (b);
(E) performing a second notification to the moving object based on the determined notification content;
A control program that causes a computer to execute processing that includes

  The evacuation guidance device according to the present invention, after performing the first notification by the notification unit, determines the notification content based on the behavior pattern of the moving object in the predetermined area analyzed by the analysis unit, and performs the notification based on the determined notification content. The second notification is made to the moving body by the unit. Thereby, the next notification can be appropriately performed based on the behavior pattern of the moving body after the notification.

It is a block diagram showing the composition of the evacuation guidance system concerning this embodiment. It is sectional drawing which shows the schematic structure of a rider. It is a schematic diagram which shows the state which scans the inside of a monitoring area by a rider. It is an example of the positional information data at each time of the moving object detected by the ranging point group data. It is a figure showing the circumference of an expressway entrance to which a monitoring field is applied. It is a schematic diagram which shows the monitoring area at the entrance of a highway. It is a flowchart which shows the alerting | reporting process in 1st Embodiment. It is a figure showing an example of a pedestrian attribute judged from image data of a camera. It is a figure showing an example of a locus of movement of pedestrians A-D in a monitoring area. It is a figure which shows the change of the moving speed of pedestrians A-D corresponding to FIG. 9A. It is a figure which shows the example of the effect determination result of the 1st notification to pedestrians A-D. It is a figure which shows the example of the notification content in the 2nd notification to pedestrians A-D. It is a figure which shows the example of the effect determination result of the 2nd notification to pedestrians A-D, and the example of the notification content in the 3rd notification. It is a figure showing the example which arranged a plurality of information parts in the monitoring area of the expressway entrance. It is a flowchart which shows the alerting | reporting process in 2nd Embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the description of the drawings, the same elements will be denoted by the same reference symbols, without redundant description. In addition, the dimensional ratios in the drawings are exaggerated for convenience of description, and may be different from the actual ratios.

  FIG. 1 is a block diagram showing a configuration of the evacuation guidance system 10. As shown in FIG. The evacuation guidance system 10 includes an input unit 100, an evacuation guidance device 200, and a notification unit 300. The evacuation guidance device 200 is, for example, a computer, and includes a CPU (Central Processing Unit), a memory (semiconductor memory, a magnetic recording medium (hard disk or the like)), an input / output unit (display, keyboard, or the like), a communication I / F (interface). ) Etc. The communication I / F is an interface for communicating with an external device. For communication, a network interface based on standards such as Ethernet (registered trademark), SATA, PCI Express, USB, and IEEE1394 may be used. For communication, a wireless communication interface such as Bluetooth (registered trademark), IEEE 802.11, or 4G may be used.

(Input unit 100)
The input unit 100 includes a rider (Light Detection and Ranging) 110 (LiDAR) and a camera 120. The input unit 100 may include only the rider 110. The camera 120 is constituted by, for example, a visible light camera. The camera 120 may be constituted by a thermal camera. Hereinafter, the input unit 100 will be described as being configured by the rider 110 and the camera 120.

  The input unit 100 measures a distance to a target (moving body) in a monitoring area (predetermined area) by a rider 110 as a distance measuring unit, and outputs ranging point group data indicating a distribution of distance values in the monitoring area. Generate. The rider 110 functions as a “moving object information measurement unit”, and the generated ranging point group data corresponds to moving object information obtained by measuring a moving object in a predetermined area. The monitoring area is photographed by the camera 120 to generate visible light image data.

(Rider 110)
Hereinafter, the configuration of the rider 110 will be described with reference to FIGS. FIG. 2 is a cross-sectional view illustrating a schematic configuration of the rider 110. FIG. 3 is a schematic diagram showing a state where the rider 110 scans the inside of the monitoring area. The rider 110 includes a light emitting / receiving unit 111 and a ranging point group data generation unit 112. The light emitting and receiving unit 111 has a semiconductor laser 51, a collimating lens 52, a mirror unit 53, a lens 54, a photodiode 55, a motor 56, and a housing 57 that houses these components. The light emitting and receiving unit 111 outputs a light receiving signal of each pixel obtained by scanning the monitoring space of the rider 110 with the laser spot light 500. The ranging point group data generator 112 generates ranging point group data based on the received light signal. The ranging point group data is also called a distance image or a distance map.

  The semiconductor laser 51 emits a pulsed laser beam. The collimator lens 52 converts divergent light from the semiconductor laser 51 into parallel light. The mirror unit 53 scans and projects the laser beam collimated by the collimating lens 52 toward the monitoring area by the rotating mirror surface, and reflects the reflected light from the object. The lens 54 collects the reflected light from the object reflected by the mirror unit 53. The photodiode 55 receives the light collected by the lens 54 and has a plurality of pixels arranged in the Z direction. The motor 56 drives the mirror unit 53 to rotate.

  The ranging point group data generator 112 obtains distance information (distance value) according to the time difference between the emission timing of the semiconductor laser 51 and the light reception timing of the photodiode 55. The ranging point group data generation unit 112 includes a CPU (Central Processing Unit) and a memory, and executes ranging programs by executing a program stored in the memory to obtain ranging point group data. And a dedicated hardware circuit for generating ranging point group data. In addition, the ranging point group data generation unit 112 may be integrated with an analysis unit 220 (described later) as one function of the above-described evacuation guidance device 200.

  In the present embodiment, the emission section 501 is constituted by the semiconductor laser 51 and the collimating lens 52, and the light receiving section 502 is constituted by the lens 54 and the photodiode 55. It is preferable that the optical axes of the light emitting unit 501 and the light receiving unit 502 are orthogonal to the rotation axis 530 of the mirror unit 53.

  The box-shaped housing 57 fixed and installed on the rigid wall 91 or the like includes an upper wall 57a, a lower wall 57b opposed thereto, and a side wall 57c connecting the upper wall 57a and the lower wall 57b. Have. An opening 57d is formed in a part of the side wall 57c, and a transparent plate 58 is attached to the opening 57d.

  The mirror unit 53 has a shape in which two quadrangular pyramids are joined in opposite directions to be integrated, that is, four pairs of mirror surfaces 531a and 531b that are inclined in a pair facing direction (however, not limited to four pairs) ) Have. The mirror surfaces 531a and 531b are preferably formed by depositing a reflective film on the surface of a resin material (for example, PC (polycarbonate)) in the shape of a mirror unit.

  The mirror unit 53 is connected to a shaft 56a of a motor 56 fixed to a housing 57, and is driven to rotate. In the present embodiment, for example, when installed on the wall 91, the axis (rotation axis) of the shaft 56a extends in the vertical Z direction, and is formed in the X direction and the Y direction orthogonal to the Z direction. Although the XY plane is a horizontal plane, the axis of the shaft 56a may be inclined with respect to the vertical direction.

  Next, the principle of detecting an object by the rider 110 will be described. In FIG. 2, divergent light emitted intermittently in a pulse form from a semiconductor laser 51 is converted into a parallel light beam by a collimator lens 52 and is incident on a first mirror surface 531 a of a rotating mirror unit 53. After that, the light is reflected by the first mirror surface 531a and further reflected by the second mirror surface 531b, and then transmitted through the transparent plate 58 and is projected toward an external measurement space, for example, as a laser spot light having a vertically long rectangular cross section. Be lighted. The direction in which the laser spot light is emitted and the direction in which the emitted laser spot light is reflected by the object and returned as reflected light overlap, and the two overlapping directions are referred to as the light emitting and receiving directions. Laser spot light traveling in the same light emitting and receiving direction is detected by the same pixel.

  FIG. 3 is a diagram illustrating a state in which the laser spot light 500 (indicated by hatching) emitted in accordance with the rotation of the mirror unit 53 scans the monitoring area. Here, in the combination of the pair of mirrors (first mirror surface 531a and second mirror surface 531b) of mirror unit 53, the four pairs have different crossing angles. The laser light is sequentially reflected by the rotating first mirror surface 531a and the rotating second mirror surface 531b. First, the laser light reflected by the first pair of the first mirror surface 531a and the second mirror surface 531b moves the uppermost region Ln1 of the measurement space horizontally from left to right according to the rotation of the mirror unit 53. Is scanned. Next, the laser light reflected by the second pair of the first mirror surface 531a and the second mirror surface 531b moves the second region Ln2 from the top in the measurement space horizontally from the left in accordance with the rotation of the mirror unit 53. Scanned to the right. Next, the laser light reflected by the third pair of the first mirror surface 531a and the second mirror surface 531b moves the third region Ln3 from the top in the measurement space horizontally from the left in accordance with the rotation of the mirror unit 53. Scanned to the right. Next, the laser beam reflected by the fourth pair of the first mirror surface 531a and the second mirror surface moves the lowermost region Ln4 of the measurement space horizontally from left to right in accordance with the rotation of the mirror unit 53. Scanned. This completes one scan of the entire measurement space where the rider 110 can measure. One frame 900 is obtained by combining the images obtained by scanning the areas Ln1 to Ln4. After one rotation of the mirror unit 53, the mirror unit 53 returns to the first pair of the first mirror surface 531a and the second mirror surface 531b again, and thereafter scans from the uppermost region Ln1 to the lowermost region Ln4 of the measurement space. Is repeated to obtain the next frame 900.

  In FIG. 2, a part of the laser beam that has been scanned and projected and reflected on the object passes through the transparent plate 58 again and is incident on the second mirror surface 531 b of the mirror unit 53 in the housing 57. Here, the light is reflected, further reflected by the first mirror surface 531a, collected by the lens 54, and detected by the light receiving surface of the photodiode 55 for each pixel. Further, the ranging point group data generation unit 112 obtains distance information according to a time difference between the emission timing of the semiconductor laser 51 and the light reception timing of the photodiode 55. As a result, the object is detected in the entire area in the monitoring space, and a frame 900 (see FIG. 3) as distance measurement point group data having distance information for each pixel can be obtained. Further, according to a user's instruction, the obtained ranging point group data may be stored as background image data in a memory in the ranging point group data generation unit 112 or a memory of the evacuation guidance device 200.

(Evacuation guidance device 200)
Referring to FIG. 1 again, the configuration of the evacuation guidance device 200 will be described. The evacuation guidance device 200 includes an acquisition unit 210, an analysis unit 220, a database 230, and a determination unit 240. Mainly, the CPU of the above-described evacuation guidance device 200 functions as the analysis unit 220 and the determination unit 240, the memory functions as the database 230, and the communication I / F functions as the acquisition unit 210.

(Acquisition unit 210)
The acquisition unit 210 acquires the ranging point group data (also referred to as “moving object information”) arranged in time series generated by the rider 110 from the input unit 100, and sends the acquired ranging point group data to the analysis unit 220. Further, the acquisition unit 210 acquires the image data generated by the camera 120 from the input unit 100 and sends the image data to the analysis unit 220.

(Analysis unit 220)
The analysis unit 220 includes a moving object type analysis unit 221, a walking characteristic analysis unit 222, and a pedestrian attribute analysis unit 223.

  Here, an algorithm for detecting the measurement target of the rider 110 using the background subtraction method will be described. In the present embodiment, for example, a background difference method is employed. In the background subtraction method, a background image (also referred to as a reference image) generated and stored in advance is used. Specifically, as preparation for measurement (pre-processing), the laser spot light 500 is scanned from the rider 110 by a user's instruction in a state where no moving object such as a human or an animal exists. Thus, a background image can be obtained based on the reflected light obtained from the background object 92. At the time of actual measurement, if, for example, a pedestrian 93 appears as an object to be analyzed for behavior before the background target object 92, reflected light from the pedestrian 93 is newly generated.

  The moving object type analysis unit 221 has a function of detecting a moving object. The moving object type analysis unit 221 compares the background image data stored in the memory with the current distance measuring point group data, and if a difference occurs, any moving object (object) such as a pedestrian enters the monitoring space. You can recognize what has appeared. For example, the foreground data is extracted by comparing the background data with the current ranging point group data (distance image data) using the background subtraction method. Then, the pixels (pixel groups) of the extracted foreground data are divided into clusters according to, for example, distance values of the pixels. Then, the size of each cluster is calculated. For example, a vertical dimension, a horizontal dimension, a total area, and the like are calculated. Here, the “size” is an actual size, which is different from the apparent size (angle of view, that is, the spread of pixels), and a lump of pixel groups is determined according to the distance to the object. For example, the moving object type analysis unit 221 determines whether or not the calculated size is equal to or smaller than a predetermined size threshold value for specifying a moving object to be analyzed to be extracted. The size threshold can be arbitrarily set depending on the measurement place, the behavior analysis target, and the like. If the behavior is analyzed by tracking a pedestrian, the minimum value of the size of a normal person may be set as the size threshold for clustering. Conversely, if all moving objects are to be tracked, the size threshold may be a smaller value.

  The moving object type analysis unit 221 functions as a “type analysis unit”, and determines the type of the moving object by comparing the size, aspect ratio, and the like of the clustered moving object with reference data stored in a memory stored in advance. I do. The types of moving objects to be determined include a person (pedestrian) and a vehicle (four-wheeled, two-wheeled vehicle). Further, at this time, in the case of an object that is difficult to determine only by the size, the moving speed may be further calculated, and the type of the moving object may be determined by combining the moving speeds. For example, two-wheeled motorcycles that can pass on a highway and bicycles whose passage is prohibited are determined based on size and moving speed.

  The walking characteristic analysis unit 222 analyzes the behavior pattern (movement characteristic) of the moving object. FIG. 4 is an example of position information data at each time of a moving object detected by the ranging point group data. Each detected moving object is managed by a moving object ID (pedestrian ID). The walking characteristic analysis unit 222 analyzes a behavior pattern by analyzing time-series position information for each moving object. The action pattern includes at least one of the moving trajectory of the moving object, the moving speed of the moving object, the falling state of the moving object, the position of the moving object, and the distance of the moving object to the set dangerous area, obtained from the distance measurement point group data. . As the position of the moving object used in this analysis, the center value or the center of gravity of the clustered moving object may be used, and if the moving object is clustered by a rectangle, the center of the rectangle may be used as the position of the moving object. . The falling state of the moving object is used when the moving object is determined to be a pedestrian, and when the height of the center position of the moving object from the ground suddenly decreases, or the aspect ratio (the ratio of the height to the width) sharply changes. If it has fallen, or if that state has been maintained, it is determined that it has fallen. The setting of the dangerous area is set in advance, and for example, the main road of the highway is set as the dangerous area.

  In the present embodiment, in particular, when the moving object type analysis unit 221 determines the type of the moving object as the pedestrian 93, the moving speed (walking speed) and the trajectory of the pedestrian 93 are analyzed as an action pattern.

  The pedestrian attribute analysis unit 223 functions as an “attribute analysis unit”, and mainly analyzes attributes of a moving object from image data acquired from the camera 120. Specifically, the pedestrian is extracted by analyzing the two-dimensional image data, and the attribute of the pedestrian is analyzed. Analysis items include gender, age, and race. Further, as the attribute of the pedestrian, portable items such as a cane, a wheelchair, and a carry bag may be further determined. The determination of a portable item can be made by registering the shape and size of an object that can be a portable item in the pedestrian attribute analysis unit 223 in advance and matching the registered data.

  This analysis includes discrimination between a human and another object as the object type and discrimination of the determined attribute. For example, in recognition of gender, age, and race, the computer extracts a face from the obtained image data, and determines the attribute of the person from the extracted face feature amount. This discrimination algorithm can be machine-learned in advance by a known algorithm. This machine learning is performed by using a large amount of data on another high-performance computer in advance to determine parameters. The pedestrian attribute analysis unit 223 can analyze the attribute of the pedestrian using the parameters.

  Further, the pedestrian attribute analysis unit 223 may receive the pedestrian position information from the gait characteristic analysis unit 222, and may analyze the pedestrian attribute based on the image data and the position information. For example, the optical axis (monitoring area) of the rider 110 and the camera 120 are made to coincide, or calibration for positioning is performed in advance. Then, the attributes of the pedestrian are analyzed by referring to the size such as the height of the pedestrian, the walking speed, and the like from the image data corresponding to the pedestrian's position information obtained by analyzing the ranging point group data.

(Database 230)
The database 230 includes a pedestrian data storage unit 231 and a notification content storage unit 232. The pedestrian data storage unit 231 stores the walking characteristics for each pedestrian ID sent from the walking characteristics analysis unit 222 and the pedestrian attributes for each pedestrian ID sent from the pedestrian attribute analysis unit 223. The notification content storage unit 232 stores the notification content (history information regarding the notification content) and the effect for each pedestrian ID. In addition, the history of the notification contents and effects is managed. This notification content and effects will be described later.

(Determining unit 240)
The determination unit 240 includes a notification content determination unit 241 and a notification effect determination unit 242, determines the notification content, and controls the notification unit 300 to notify based on the determined content. The notification content determination unit 241 determines the notification content based on the walking characteristics (walking pattern) stored in the pedestrian data storage unit 231 and the determination result of the notification effect determination unit 242. The notification effect determination unit 242 determines the walking characteristics and pedestrian attributes for each pedestrian ID stored in the pedestrian data storage unit 231 and the notification content for each pedestrian ID stored in the notification content storage unit 232 ( The presence / absence of the notification effect is determined based on the history information regarding the notification content) and the effect.

(Notification unit 300)
The notification unit 300 includes two types of notification means, a speaker 310 and a signage 320. Each of the speaker 310 and the signage 320 may be plural. The notification unit 300 may be configured with only one of the speaker 310 and the signage 320. The speaker 310 emits a sound according to the notification content determined by the notification content determination unit 241. The signage 320 is a digital signage, and displays a character or an image according to the content of the notification on a liquid crystal display or a display unit in which a plurality of LEDs are arranged two-dimensionally.

(Example of application of the evacuation guidance system 10)
FIG. 5 is a diagram showing the vicinity of the entrance of an expressway to which the monitoring area of the evacuation guidance system 10 is applied. FIG. 6 is a schematic diagram showing a monitoring area at the entrance of an expressway.

  In recent years, entry of pedestrians and bicycles on expressways has become a social problem. In general, various measures are taken in danger areas such as expressways, such as installing signs near entrances and warning pedestrians and drivers to prevent entry of vehicles other than permitted vehicles. ing. Further, even in a conventional technique in which the moving direction and the moving speed of the moving body are in a predetermined range only when the moving direction and the moving speed of the moving body are within a set range as in the known document 1 (Japanese Patent Application Laid-Open No. 2004-192097), the speed is high. Similarly, when a pedestrian enters a road entrance, a notification for preventing the entry can be made.

However, according to the sign installation and the technology disclosed in the known document 1, a person who has accidentally entered the expressway due to the following reasons (1) to (5) cannot be evacuated based on an appropriate judgment. There is a risk that a new accident involving a person may be caused.
(1) There is no evacuation guidance on the highway.
(2) There is a problem with cognitive / thinking abilities due to panic, dementia, drunkenness, etc., making it impossible to recognize / understand the evacuation guidance on the expressway and appropriate evacuation. Can't take action.
(3) The evacuation guidance display is not written or notified in a language that the resident can understand because the resident's Japanese ability is insufficient, and the resident cannot understand it.
(4) It is necessary to evacuate quickly on the expressway, but there is no time for security guards and guides to go to the site.
(5) On an expressway, the degree of danger varies depending on the area before and after the tollgate, the roadway, the shoulder, the presence or absence of a curve, etc., but it is necessary to determine in detail whether a person approaching or approaching a more dangerous area. Since it is difficult to grasp, the evacuation guidance content is uniform for each area, and a more appropriate guidance effect cannot be obtained.

  Therefore, in the present embodiment, as shown in FIG. 6, the input unit 100 (the rider 110 and the camera 120) is arranged so that the approach road at the entrance of the highway becomes the monitoring area as the danger area. Then, an evacuation guidance instruction is issued from a dangerous area for a pedestrian such as an entrance of a highway by a notification process described below. Note that the rider 110 and the camera 120 are arranged so that the monitoring area (measurement area / photographing area) substantially matches and the optical axis substantially matches, and the positioning is performed. When the measurement area of the rider 110 and the shooting area of the camera 120 extend outside the approach road, the monitoring area may be set in advance so that only the approach road that is a no-go area is set as the monitoring area. .

(A notification process according to the first embodiment)
The notification process executed by the evacuation guidance device 200 according to the first embodiment will be described with reference to FIGS. FIG. 7 is a flowchart showing the notification processing.

(Step S110)
The moving body information in a predetermined area (monitoring area) is acquired. Specifically, the acquisition unit 210 acquires, from the rider 110, ranging point group data arranged in a time series in which the monitoring area is measured. Further, the acquisition unit 210 acquires image data (moving image) obtained by photographing the monitoring area from the camera 120.

(Step S120)
The analysis unit 220 analyzes the moving body information acquired in step S110. Here, the moving object type analysis unit 221 analyzes the type of the moving object in the monitoring area. Types to be determined include pedestrians, bicycles, four-wheeled vehicles, and two-wheeled vehicles.

(Step S130)
The determination unit 240 determines whether the type of the moving object is a specific type that is not allowed to enter (enter) the monitoring area. For example, in the case of the approach road shown in FIGS. 5 and 6, the pedestrian 93 is a specific type of moving object not allowed to enter (YES), and the process proceeds to step S140. On the other hand, if the vehicle is a four-wheeled vehicle or a two-wheeled vehicle and is an object of a permitted type (NO), the process ends (END). At this time, the signage 320 of the notification unit 300 may display that the vehicle can pass.

(Step S140)
Here, the notification content determination unit 241 of the determination unit 240 causes the notification unit 300 to perform notification based on the notification content set in advance or the notification content determined in step S170 described below. If the pedestrian 93 who has entered has not been notified in the past, the first (first) notification is performed using the preset notification contents.

  For example, the preset notification content for the first notification is “please go back” (Japanese), which is uniformly determined according to the type of moving object (person: pedestrian). The first notification (evacuation instruction) is immediately performed according to the determination that the pedestrian is present in the monitoring area, without waiting for the result of the pedestrian behavior pattern determination to be performed below. This is because prompt evacuation instructions and start of evacuation are important for accident prevention on the approach road of the highway. The notification performed here is continued until it is determined that there is no need for notification, or until notification is performed with new notification content. For example, the speaker 310 repeats and outputs the sound of the notification content, and the signage 320 continuously displays the sound.

(Step S150)
After performing the first notification, the analysis unit 220 continuously acquires the moving object information and determines the behavior pattern of the moving object. Here, the walking characteristic analysis unit 222 analyzes the walking speed and the trajectory. In addition, the pedestrian attribute analysis unit 223 analyzes the pedestrian attribute.

  Hereinafter, a case where pedestrians A to D are assumed as examples of the pedestrian 93 will be described. Note that it is assumed that these pedestrians A to D enter the expressway entrance at different timings. However, a situation in which a plurality of persons enter the monitoring area at the same time may be assumed. In that case, each pedestrian 93 is individually tracked in parallel by the moving object ID (see FIG. 4). FIG. 8 is a diagram illustrating an example of pedestrian attributes of pedestrians A to D determined by the pedestrian attribute analysis unit 223 from image data of the camera 120. FIG. 9A and FIG. 9B are diagrams illustrating changes in the movement trajectory and movement speed of the same pedestrians A to D, respectively. FIG. 9A is a bird's-eye view (X, Y coordinates), showing that the lower side is the entrance of the expressway, the upper side is the main road side of the expressway, and has proceeded in the direction of the arrow. In FIG. 9A, the position of the pedestrian at the time of performing the first notification is indicated by a black circle. The position of the black circle corresponds to time t1 in FIG. 9B. The walking speed, the trajectory, and the behavior pattern configured by the behavior pattern classification described later are stored in the database 230 in a state where the history is managed before and after the notification is performed.

  FIG. 8 is a diagram illustrating determination results of gender, age (classification), portable goods, and race for pedestrians A to D performed by the pedestrian attribute analysis unit 223. For gender, the likelihood (%) of the determination calculated by the pedestrian attribute analysis unit 223 is shown on the right side of the male / female determination result.

  In the example shown in FIGS. 9A and 9B, since the pedestrian A is traveling straight ahead, the behavior pattern is classified as “straight ahead”. Since the pedestrian B has returned to the entrance, the behavior pattern is classified as “U-turn”. Since the pedestrian C is moving at a relatively slow speed while swaying left and right, the behavior pattern is classified as “wandering”. Since the pedestrian D has stopped halfway, the behavior pattern is classified as “staying”. These classifications are performed by the walking characteristic analysis unit 222.

(Step S160)
Here, the notification effect determination unit 242 determines an effect from the notification in step S140 and the behavior pattern determined in step S150. In particular, the effect is determined from the history of the action patterns before and after the notification.

  FIG. 10 is a diagram illustrating an example of the effect determination result of the notification (the first (first notification)) in step S140. The figure shows the recognition level, understanding level, evacuation achievement rate level, and state prediction result of each of the pedestrians A to D. Each level is displayed in five levels from 1 to 5 (1 is the lowest and 5 is the highest).

  The notification effect determination unit 242 determines that the evacuation has been completed for the pedestrian B because the behavior pattern indicates that the first notification functioned effectively. The notification effect determination unit 242 determines (estimates) that the pedestrian A is “language difference” and determines that the pedestrian C is “drunkenness or cognitive impairment” from the behavior pattern and the pedestrian attribute. Since the pedestrian D has stopped after the first notification, the pedestrian D is determined to be “panic or language difference” from the history before and after.

(Step S170)
The notification content determination unit 241 determines the necessity of the next notification and the notification content. FIG. 11 is a diagram illustrating an example of notification content in the second notification determined based on the effect determination result of FIG. 10 with respect to the first notification or based on the action pattern and the pedestrian attribute. These notification contents are stored in the database 230 in advance and selected from them, but the notification content determination unit 241 may automatically create the notification contents. For the pedestrian A, the achievement level is low and it is determined to be "language difference". Therefore, it is determined that the second notification is necessary, and the notification content is determined to be "Stop and U turn". For the pedestrian B, it is determined that the second notification is unnecessary because the achievement level is sufficiently high. For the pedestrian C, the achievement level is low and it is determined that "drunkenness or cognitive impairment". Therefore, it is determined that the second notification is necessary, and the notification content is determined to be "Grandmother, go back". For pedestrian C, the achievement level was low and it was judged as "panic or language difference", so it was judged that the second notification was necessary, and the content of the notification was "calm down, check left and right, Please stop by. "

(Step S180)
The determination unit 240 determines whether the next notification is necessary based on the determination in step S170. As shown in FIG. 11, if it is a pedestrian B, there is no need for notification (NO), the notification up to that point is stopped, and the process ends. On the other hand, if it is pedestrians A, C, and D, notification is necessary (YES), the process returns to step S140, and the notification process is performed using the second notification content shown in FIG. Then, the processing after step S150 is repeated.

  FIG. 12 is a diagram illustrating an example of the effect determination result of the second notification to the pedestrians A to D, and an example of the notification content in the third notification. FIGS. 12A and 12B correspond to FIGS. 10 and 11, respectively.

  As shown in FIG. 12, after the second notification is performed, the history of the pedestrian's action patterns before and after the previous notification (first notification) and the pedestrian's The necessity of the next notification (second notification) and the content of the notification are determined from the pedestrian attribute.

  As described above, in the present embodiment, after the previous notification (first notification), the determination unit 240 determines the next notification (second notification) based on the pedestrian's behavior pattern analyzed by the analysis unit 220. Notification) and the content of the notification. If the notification is required, the notification unit 300 is controlled to notify the pedestrian with the determined notification content.

  In this way, it is possible to perform appropriate notification based on the behavior pattern of the moving body after the notification in response to the notification from the evacuation guidance device 200. Specifically, if there is no change in the behavior of the target pedestrian, is the subject too old to notice the voice because it is too old or cannot understand the contents because the language used for notification cannot be used? , Etc., can be determined (guessed) with high accuracy why the notification does not function effectively. Then, based on the result of the determination, by selecting an appropriate notification content according to the target person, the success rate of evacuation guidance can be increased. Further, in the present embodiment, when used on a highway, the above-mentioned problems (1) to (5) can be solved.

(Second embodiment)
Hereinafter, the notification processing executed by the evacuation guidance device 200 according to the second embodiment will be described with reference to FIGS. FIG. 13 is a schematic diagram showing a monitoring area at the entrance of an expressway. FIG. 14 is a flowchart showing the notification processing. In the second embodiment, the notifying unit 300 includes a plurality of speakers 310a, 310b, 310c, and a plurality of signages 320a, 320b, 320c, and uses the notifying unit according to the position and behavior pattern of the pedestrian 93. Select 300 (speaker 310 or signage 320). Each position of these notification units 300 is stored in the memory of the evacuation guidance device 200 in advance. As for the signage 320, information on the direction and direction of the display surface of the display is stored in the memory. The second embodiment is the same as the first embodiment unless otherwise specified.

(Steps S210 to S230)
The same processing as in steps S110 to S130 in FIG. 7 is performed, and if the type of the moving object is a pedestrian, the processing proceeds to step S235.

(Steps S235, S240)
Here, according to the current position of the moving object (pedestrian), the type (speaker 310 or signage 320) and position of the notification unit 300 to be used for the next notification are selected. As shown in FIG. 13, if the pedestrian 93 is moving at a position far from the entrance of the highway, the speaker 310 c and the signage 320 c which are closest to the pedestrian 93 and / or ahead in the traveling direction according to the movement locus are selected. The notification is made by the selected speaker 310c and signage 320c.

  The notification content at this time is the notification content set in advance or the notification content determined in step S270, as in the first embodiment. At this time, when it is determined that any signage 320 cannot be visually recognized from the position of the pedestrian 93, only the speaker 310 may be selected as the type of the notification unit 300. Further, as another example, when the evacuation guidance device 200 determines that there is no effect as a result of performing the effect determination after the first notification, the evacuation guidance device 200 selects both the speaker 310 and the signage 320. Is also good. The speaker 310 and the signage 320 selected at this time may be the one closest to the pedestrian 93 and / or the one in front of the traveling direction.

(Steps S250 to S280)
Hereinafter, the same processes as those in steps S150 to S180 in FIG. 7 are performed, and the process ends (END).

  As described above, also in the second embodiment, the same effect as in the first embodiment can be obtained.

  The configurations of the evacuation guidance device 200 and the evacuation guidance system 10 described above are the main components in describing the features of the above-described embodiment, and are not limited to the above-described configuration, but fall within the scope of the claims. Can be variously modified. Further, the configuration provided in the general evacuation guidance system 10 is not excluded.

  For example, in the above-described embodiment, an example in which a pedestrian is determined to be a specific type of moving object and evacuation guidance is performed has been described. However, the present invention is not limited to this, and a bicycle (and a driver) may be used as a specific type of moving object. This may be a target for evacuation guidance. In this case, the first notification content and the second and subsequent notification content may be configured to be different from the pedestrian notification content.

  The means and method for performing various processes in the evacuation guidance device according to the above-described embodiment can be realized by either a dedicated hardware circuit or a programmed computer. The program may be provided by a computer-readable recording medium such as a USB memory or a DVD (Digital Versatile Disc) -ROM, or may be provided online via a network such as the Internet. In this case, the program recorded on the computer-readable recording medium is usually transferred to and stored in a storage unit such as a hard disk. Further, the program may be provided as independent application software, or may be incorporated in software of the device as one function of the device.

Reference Signs List 10 Evacuation guidance system 100 Input unit 110 Rider 120 Camera 200 Evacuation guidance device 210 Acquisition unit 220 Analysis unit 221 Moving object type analysis unit 222 Walking characteristics analysis unit 223 Pedestrian attribute analysis unit 230 Database 231 Pedestrian data storage unit 232 Notification content storage unit 240 judgment unit 241 notification content determination unit 242 notification effect determination unit 300 notification unit 310, 310a to 310c speaker 320, 320a to 320c signage

Claims (9)

An acquisition unit that acquires moving body information that measures a moving body in a predetermined area,
Based on the obtained moving body information, an analysis unit that analyzes the behavior pattern of the moving body,
After the first notification by the notification unit, the notification content is determined based on the behavior pattern analyzed by the analysis unit, and the notification unit performs the second notification to the moving object with the determined notification content. And a determination unit that controls
Evacuation guidance device provided with.   2. The evacuation guidance device according to claim 1, wherein the determination unit retains a history of the behavior pattern and determines the content of the second notification based on the behavior pattern before and after the first notification. 3. Based on the moving object information, comprising a type analysis unit for analyzing the type of the moving object,
If the type of the moving object determined by the type analysis unit is a specific type of moving object that is not allowed to enter the predetermined area, the analysis unit analyzes the behavior pattern for the moving object. The evacuation guidance device according to claim 1 or 2. The acquisition unit acquires, as the moving object information, distance measurement point group data indicating a distribution of distance values to a moving object existing in the predetermined area,
The behavior pattern determined by the analysis unit includes a moving locus of the moving object, a moving speed of the moving object, a falling state of the moving object, a position of the moving object, and a set dangerous area, which are obtained from distance measurement point group data. The evacuation guidance device according to any one of claims 1 to 3, wherein at least one of the distances of the moving body is included. The obtaining unit obtains image data of the inside of the predetermined area,
Further, from the image data acquired by the acquisition unit, comprises an attribute analysis unit for analyzing the attributes of the moving object,
5. The notification content of the second notification is determined based on an attribute of the moving object analyzed by the attribute analysis unit and the behavior pattern determined by the analysis unit. 6. Evacuation guidance device. The notification unit is a plurality of notification units of one or more types,
The said judgment part selects the said alerting | reporting part with the said alert | report content according to the said action pattern, and controls so that the said alerting | reporting part may perform 2nd alerting | reporting to the said moving body. 5. The evacuation guidance device according to any one of 5. The plurality of notification units are a plurality of notification units arranged at different positions,
The acquisition unit acquires, as the moving body information, distance measurement point group data indicating a distribution of distance values to an object in the predetermined area acquired from the distance measurement unit that measures the inside of the predetermined area,
The evacuation guidance device according to claim 6, wherein the determination unit selects the notification unit based on a moving trajectory of the moving object or a position of the moving object obtained from the distance measurement point group data. By measuring, a moving body information measuring unit that generates moving body information of a moving body in a predetermined area,
The reporting department,
The evacuation guidance device according to any one of claims 1 to 7, wherein the notification unit notifies the moving body using moving body information obtained from the moving body information measurement unit.
Evacuation guidance system equipped with. A control program for controlling an evacuation guidance device that performs evacuation guidance for a moving body in a predetermined area,
(A) acquiring moving body information obtained by measuring a moving body in a predetermined area;
(B) analyzing an action pattern of the moving object based on the obtained moving object information;
(C) performing a first notification to the moving body by a notification unit;
(D) determining, after step (c), notification contents based on the behavior pattern determined in step (b);
(E) performing a second notification to the moving object based on the determined notification content;
A control program that causes a computer to execute processing that includes