JP4996453B2 - Sniper detection device - Google Patents

Description

  The present invention relates to a sniper hand detection device that identifies the direction of a sniper hand based on sniper sound in a defense ground robot system that can move by semi-autonomous driving that combines remote control and autonomous driving.

  Conventionally, reconnaissance and information gathering in international peace cooperation activities, etc. in humans and manned people in areas (particularly urban areas) where there are dangers such as the possibility of hiding by snipers, which are enemy molecules (terrorists, guerrillas, commando, etc.) Is done by the vehicle.

  In such reconnaissance, when a sniper sound (gunshot) is made, first of all, the top priority is to hide and hide, and then rely on the human senses to detect the sniper sound. They are wary of the direction, and trying to find a sniper.

  Conventionally, as described in Patent Document 1, the sound arrival direction of a gunshot collected by a microphone mounted on a vehicle is measured, and based on a distance measurement sensor, map information, or self-location information. Based on the principle of triangulation, an apparatus for calculating the position of a sniper instead of a person has been proposed.

JP 2005-308282 A

  By the way, in the case of reconnaissance or the like in the danger zone as described above, it is extremely dangerous for a person who performs reconnaissance or the like to be wary of the direction of the sniper sound by human sense when a sniper sound is made. In such a situation, reconnaissance and other activities should be avoided in the present age when the life of a reconnaissance person is at risk and the protection of human life is paramount. .

  In an emergency situation such as when a sniper is taken, it is very difficult to find a sniper because it is the first to avoid danger. If you try to find a sniper forcibly, the risk of life will increase.

  In the apparatus for calculating the position of the sniper as described above, there is a problem that the detection accuracy of the position of the sniper is not sufficiently high. The detection accuracy of the position of the sniper in such an apparatus is, for example, about 2 ° in a stationary state and about 5 ° in a moving state. That is, when the distance to the sniper is 300 m, the measurement accuracy is about 11 m in the stationary state and about 26 m in the moving state.

  With this level of detection accuracy, there is no problem if the approximate position of the sniper is detected and the heavy weapons are used to counterattack, but it is not possible to identify the room (window) where the sniper is located. It is enough. In particular, when such a case occurs in an urban area, it is important to first check whether the civilians are hostages, not only in general directions but also in collecting more detailed information. is necessary. In particular, in an operation that rushes and suppresses an enemy, the room in which the enemy is located is very important information.

  Therefore, the present invention has been made in view of the above circumstances, and in a ground robot system for defense that can be moved by semi-autonomous driving that combines remote control and autonomous driving, the position of the sniper is determined based on the sniper sound. An object of the present invention is to provide a sniper detection device that can be specified with high accuracy.

  In order to solve the above-described problems and achieve the above-described object, a sniper detection device according to the present invention has any one of the following configurations.

[Configuration 1]
Map by semi-autonomous driving that combines unmanned vehicle that can run, omnidirectional camera mounted on this unmanned vehicle, sniper sound sensor mounted on unmanned vehicle, remote control and autonomous driving mounted on unmanned vehicle And a control means for controlling the omnidirectional camera and the sniper sound sensor, enabling the movement of the unmanned vehicle based on the information, and the control means always records an image picked up by the omnidirectional camera by the recording means and at least The image recorded by the recording means is held for a predetermined time, and when the sniper sound is detected by the sniper sound sensor, the sniper sound generation position is specified, and when the sniper sound generation position is specified, the recording means The recorded position of the sniper sound identified from the images recorded for a predetermined period before and after the generation of the sniper sound and within the predetermined period It is characterized in that for reproducing the selected image to select the image in accordance with the travel distance of the unmanned vehicle.

[Configuration 2]
In the sniper detection device having the configuration 1, the generation position of the sniper sound is specified by specifying the direction and distance from the unmanned vehicle based on detection by the sniper sound sensor.

[Configuration 3]
In the sniper detection device having the configuration 1, the generation position of the sniper sound is specified by specifying the direction from the unmanned vehicle based on the detection by the sniper sound sensor and the direction of the image captured by the omnidirectional camera. The image processing is performed by image processing.

[Configuration 4]
In the sniper detection device having any one of Configurations 1 to 3, the omnidirectional camera and the sniper sound sensor are supported by lifting means mounted on the unmanned vehicle, and are lifted with respect to the unmanned vehicle by the lifting means. It is characterized in that it is moved to a position separated from each other.

  In the sniper detection device according to the present invention having the configuration 1, when the sniper sound is detected by the sniper sound sensor, the generation position of the sniper sound is specified, and the image is recorded by the recording means. Since the selected image is reproduced according to the generation position of the sniper sound identified from the images recorded in the predetermined period before and after the occurrence and the travel distance of the unmanned vehicle within the predetermined period, the selected image is reproduced. In a dangerous zone where snipers can be received, it is possible to conduct reconnaissance, information collection, monitoring and security operations while avoiding human damage. It is possible to specify in detail, and it is possible to perform detailed analysis on the enemy firepower, the number of people, and the like by the recorded image.

  In the sniper detection device having the configuration 2, the generation position of the sniper sound is specified by specifying the direction and distance from the unmanned vehicle based on the detection by the sniper sound sensor. Can be done quickly.

  In the sniper detection device having configuration 3, the generation position of the sniper sound is specified by specifying the direction from the unmanned vehicle based on the detection by the sniper sound sensor, and among the images captured by the omnidirectional camera. Since the image in the direction is processed by image processing, the generation position of the sniper sound can be specified quickly and accurately.

  In the sniper detection device having the configuration 4, the omnidirectional camera and the sniper sound sensor are moved and moved upwardly with respect to the unmanned vehicle by the elevating means. Can be detected.

  That is, the present invention relates to a sniper hand detection device capable of specifying the direction of a sniper with high accuracy based on a sniper sound in a defense ground robot system that can move by semi-autonomous driving that combines remote control and autonomous driving. Can be provided.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a perspective view showing an external configuration of a defense ground robot system which is a sniper detection device according to the present invention.

  The sniper detection device according to the present invention is, for example, a device that can perform reconnaissance, information collection, and the like in a dangerous zone where a sniper may be hiding, instead of a human or a manned vehicle. As shown in FIG. 1, the sniper detection device includes an unmanned vehicle (UGV) 1 that can be run, an omnidirectional camera 2 mounted on the unmanned vehicle 1, and a sniper sound mounted on the unmanned vehicle 1. And a sensor 3. The omnidirectional camera 2 and the sniper sound sensor 3 mounted on the unmanned vehicle 1 are controlled by control means mounted on the unmanned vehicle 1. Moreover, this control means enables the unmanned vehicle 1 to move based on the map information by semi-autonomous driving that combines remote control and autonomous driving.

  FIG. 2 is a block diagram showing the configuration of the sniper detection device according to the present invention.

  The unmanned vehicle 1 is equipped with a control means 4 as shown in FIG. This control means 4 has an unmanned vehicle overall control device 4a, and a sniper direction detection processing unit 4b, a self-localization processing unit 4c, an autonomous travel control unit 4d and a vehicle control unit which are controlled by this unmanned vehicle overall control device 4a. 4e. In addition, a map data management unit 4f is connected to the unmanned vehicle overall control device 4a. Furthermore, a wireless communication device 6 that performs wireless transmission and reception with the remote control device 5 is connected to the unmanned vehicle integrated control device 4a.

  Connected to the autonomous traveling control unit 4d are an external sensor such as a laser range finder and a camera, and an internal sensor 7 such as a gyroscope and an acceleration sensor. The vehicle control unit 4e controls the vehicle control device 8 such as a steering, an accelerator, and a brake. In addition, a GPS 9 that receives a radio wave from an artificial satellite and identifies its own position is connected to the self-position location processing unit 4c.

  The unmanned vehicle 1 specifies the self-location on the map based on the data from the self-location determination processing unit 4c and the map data management unit 4f under the control of the unmanned vehicle overall control device 4a, and the autonomous driving control unit 4d and the vehicle Autonomous traveling can be performed by performing sensing, environment recognition, and behavior control via the control unit 4e. That is, this unmanned vehicle 1 can reduce the load of the member who operates, and can drive | work safely also in the environment where an enemy exists.

  The unmanned vehicle 1 is equipped with a remote control camera 10. The remote control camera 10 captures an image in the traveling direction of the unmanned vehicle 1 and sends it to the image compression device 4g. The image signal sent to the image compression device 4g is recorded in the image recording unit 4h serving as a recording unit, is sent to the wireless communication device 6, and is sent to the remote operation device 5.

  The unmanned vehicle 1 can travel by remote control by the remote control device 5. The remote operation device 5 includes a general control device 5a and a wireless communication device 5b controlled by the general control device 5a. A remote control terminal 5c for performing an input operation is connected to the overall control device 5a. The overall control device 5a controls the wireless communication device 5b based on the operation input from the remote control terminal 5c, and transmits an operation signal to the wireless communication device 6 of the unmanned vehicle 1. An image restoration device 5d that restores an image signal sent from the wireless communication device 6 of the unmanned vehicle 1 is connected to the wireless communication device 5b. The image signal restored by the image restoration device 5d is recorded by the image recording unit 5e and displayed by the display unit 5f. A map data management unit 5g is connected to the overall control device 5a.

  Based on the image captured by the remote control camera 10 displayed on the display unit 5f and the map data supplied from the map data management unit 5g, the remote control device 5 performs an input operation on the remote control terminal 5c. The unmanned vehicle 1 can be remotely operated.

  The unmanned vehicle 1 is equipped with an omnidirectional camera 2 capable of simultaneously imaging the entire circumferential direction of the unmanned vehicle 1 and an acoustic sniper sound sensor 3. The omnidirectional camera 2 includes a visible camera for daytime and / or an infrared camera for nighttime, and can capture an image of the surrounding 360 °. As this omnidirectional camera 2, for example, a method of having a plurality of cameras and arranging them so that their orientations are in the entire imaging range, and superimposing the captured images, or arranging the cameras in the vertical direction Various types of systems can be used, such as a system in which a cone-shaped mirror is arranged at the tip, the entire circumference is imaged through this mirror, and a panoramic image is created by correcting distortion of the captured image. By using such an omnidirectional camera 2, since it is not necessary to turn the camera after receiving a sniper, it is possible to reliably capture an image of a sniper who is likely to hide immediately after the sniper. .

  The image captured by the omnidirectional camera 2 is sent to the image compression device 4g and recorded in the image recording unit 4h. The unmanned vehicle overall control device 4a always records an image captured by the omnidirectional camera 2 by the image recording unit 4h and holds the image recorded by the image recording unit 4h for at least a predetermined time. Even when the unmanned vehicle 1 moves under the direction of a remote operator and performs reconnaissance, surveillance, or security activities, the omnidirectional camera 2 always captures images, and the captured images are always captured by the image recording unit 4h. To be recorded.

  The sniper sound sensor 3 includes a plurality of microphones, and is configured to detect sniper sounds with these microphones. When the sniper sound sensor 3 detects the sniper sound, an acoustic signal is sent to the sniper direction detection processing unit 4b. The sniper direction detection processing unit 4b is configured to calculate the direction and distance in which the sniper sound is generated based on the sniper sound detected by the plurality of microphones of the sniper sound sensor 3. The calculation result is sent from the sniper direction detection processing unit 4b to the unmanned vehicle overall control device 4a.

  The unmanned vehicle overall control device 4a is connected to an extendable pole 11 serving as a lifting means. The telescopic pole 11 is mounted on the unmanned vehicle 1 with its base end supported, and is configured to expand and contract under the control of the unmanned vehicle overall control device 4a to raise and lower the distal end. The omnidirectional camera 2 and the sniper sound sensor 3 are mounted and supported on the distal end side of the telescopic pole 11, and when the telescopic pole 11 is extended, the omnidirectional camera 2 and the sniper sound sensor 3 are moved to a position separated upward from the unmanned vehicle 1 (raised )

  The omnidirectional camera 2 and the sniper sound sensor 3 can be improved in detection performance by being raised by the telescopic pole 11 within a range that does not hinder the traveling of the unmanned vehicle 1.

  FIG. 3 is a perspective view showing a state in which a sniper is received during security activities or movement.

  By the way, when analyzing the behavior patterns of terrorists and guerrillas commando hidden in the building, these enemies are unlikely to snip continuously. This is because, in view of the difference between the enemy and my strength (power ratio), the enemies are overwhelmingly strong, and the enemies are not active in engagement until they achieve their purpose. This is because there is a lack of merit in taking risks that can be bothered to understand the infiltration position. From this, it can be said that when performing reconnaissance activities while hiding behind a building and performing a reconnaissance activity, the activity of grasping the position of the enemy when there is an enemy sniper is unlikely to be performed. Rather, as shown in FIG. 3, it can be said that there is a high possibility of receiving a sniper S suddenly at an unexpected moment at the timing of entering an enemy's field of view during security activities or movement. At this time, there is a life risk for the member P who is conducting the security activities.

  FIG. 4 is a perspective view showing a state in which a sniper is received during security activities or movement using the sniper detection device according to the present invention.

  In the sniper detection device according to the present invention, as shown in FIG. 4, when a sniper S is suddenly received during movement, the sniper sound is detected by the sniper sound sensor 3 at that moment, and sniper sound is generated. A location is identified. When the generation position of the sniper sound is specified, the specified generation position of the sniper sound from the images recorded by the image recording unit 4h and recorded for a predetermined period before and after the generation of the sniper sound And an image is selected according to the travel distance of the unmanned vehicle 1 within a predetermined period, and the selected image is reproduced. That is, an image in the direction detected by the sniper sound sensor 3 among the recorded images captured by the omnidirectional camera 3 can be selected (cut out), zoomed up, and provided to the operator of the unmanned vehicle 1. . At this time, the member P can run the unmanned vehicle 1 in advance and can hide behind the building until it is confirmed that there is no danger, so there is no danger of life.

  The operator of the unmanned vehicle 1 immediately removes the unmanned vehicle 1 that has received the sniper from the scene, then observes a zoom-up image in the sniper direction over a predetermined time including the moment of the sniper, and the flash recorded in the image. It is possible to specify the detailed position of the sniper by using a figure or a figure.

  FIG. 5 is a plan view showing the relationship between the movement path of the unmanned vehicle and the sniper sound detection position.

In the reproduction of the image after detecting the sniper sound, as shown in FIG. 5, the selection of the image is corrected according to the travel distance and turning state of the unmanned vehicle 1 within the duration of the reproduced image. That is, when the unmanned vehicle 1 is moving on a predetermined movement route, it is assumed that a sniper is received at the position A ₀ and the sniper sound generation position is specified as the direction θ ₀ and the distance d ₀ . At this time, when reproducing an image for a predetermined period before and after the sniper moment, the position of the unmanned vehicle 1 when the reproduction start image is imaged is the position A ₁ , and the reproduction end image is imaged. the position of the unmanned vehicle 1 as the position a _2. At position A ₁ , the specified sniper sound generation position is in direction θ ₁ and distance d ₁ , and at position A ₂ , the specified sniper sound generation position is in direction θ ₂ and distance d _2. . Accordingly, the reproduced image, the direction theta ₁ at position _{A 1,} the distance _{d 1} of the image, the position _A direction theta ₀ in _0, the image distance _{d 0,} the direction theta ₂ at position _{A 2,} so that the distance _{d 2} of the image By sequentially reproducing the images whose direction and distance are changed, it is possible to reproduce an image in which the sniper sound generation position is always captured at the center.

  In this sniper detection device, the sniper sound generation position can be specified by specifying the direction and distance from the unmanned vehicle 1 based on the detection by the sniper sound sensor 3. The direction from the unmanned vehicle 1 can be specified based on the detection, and the image in the direction among the images captured by the omnidirectional camera 2 can be processed. That is, a fixed object (such as a window frame of a building or a handrail) among subjects captured in the direction of sniper sound generation at the time of sniper sound generation is extracted by image processing, and from images captured before and after the sniper sound is generated. By specifying a fixed object extracted by image processing, a sniper sound generation position can be specified.

  In addition, the distance to the subject can be calculated from the in-focus position of the image in the direction specified by the sniper sound sensor 3 in the image captured by the omnidirectional camera 2. That is, when the captured image is in focus, the distance from the imaging lens to the subject can be calculated from the focal length of the imaging lens of the omnidirectional camera 2 and the distance from the imaging lens to the focal plane. it can.

  Further, the position of the shortest obstacle (building) in the direction specified by the sniper sound sensor 3 may be specified as the sniper sound generation position based on the map data.

  FIG. 6 is a flowchart showing the operation of the sniper detection device according to the present invention.

  In this sniper detection device, as shown in FIG. 6, the operation is started at step st1, and the operation proceeds to step st2, where the target value and the travel route (reconnaissance route) are set on the map, and the unmanned vehicle 1 is An operation start command is issued.

  Next, in step st3, the unmanned vehicle 1 starts traveling and travels along the commanded route while avoiding obstacles by autonomous traveling.

  In step st4, in the unmanned vehicle 1, the surrounding environment is always imaged by the omnidirectional camera 2 and an image is recorded, and the sniper sound sensor 3 constantly detects the presence or absence of sniper sound. At this time, the omnidirectional camera 2 and the sniper sound sensor 3 are raised by the telescopic pole 11 within a range that does not hinder the traveling of the unmanned vehicle 1. Further, the image captured by the omnidirectional camera 2 is displayed on the display unit 5 f of the remote operation device 5.

  Next, in step st5, it is determined whether or not a sniper sound has been detected. If a sniper sound is not detected, the process returns to step st3, and if a sniper sound is detected, the process proceeds to step st6.

  In step st6, the direction of the sniper sound generation position is detected. At this time, the distance to the sniper sound generation position is also specified, and the sniper sound generation position is specified.

  In the next step st7, an image in the direction specified as the sniper sound generation direction is selected from the images captured and recorded by the omnidirectional camera 2, and is zoomed up and reproduced. This image is displayed on the display unit 5 f of the remote control device 5. At this time, the selection of the image is corrected according to the travel distance and turning state of the unmanned vehicle 1 within the duration of the reproduced image.

  In the next step st8, the emergency escape mode is set, and the unmanned vehicle 1 retreats at a high speed to the nearest place in the shielding space (such as the shadow of the building) with respect to the sniper sound generation direction based on the map data.

  In the next step st9, based on the GPS information and the self-position of the unmanned vehicle 1 that is always located by an internal sensor (gyroscope or acceleration sensor), information is generated by plotting the sniper sound generation position on the map data, Send to remote control device 5.

  In the next step st10, an image captured and recorded by the omnidirectional camera 2 and selected as an image in a direction specified as the sniper sound generation direction, and the state before and after the moment of sniper sound generation, A more detailed analysis can be performed.

  In step st11, the operation is terminated, and the process returns to step st1.

It is a perspective view which shows the external appearance structure of the ground robot system for defense which is a sniper detection apparatus which concerns on this invention. It is a block diagram which shows the structure of the sniper detection apparatus which concerns on this invention. It is a perspective view which shows the state which received the sniper during the security activity or movement. It is a perspective view which shows the state which received the sniper during the guard activity using the sniper hand detection apparatus which concerns on this invention, or movement. It is a top view which shows the relationship between the movement path | route of an unmanned vehicle and a sniper sound detection position. It is a flowchart which shows operation | movement of the sniper detection apparatus which concerns on this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Unmanned vehicle 2 Omnidirectional camera 3 Sniper sound sensor 4 Control means 11 Lifting means

Claims (4)

An unmanned vehicle that can run,
An omnidirectional camera mounted on the unmanned vehicle;
A sniper sound sensor mounted on the unmanned vehicle;
Control means mounted on the unmanned vehicle, enabling the unmanned vehicle to move based on map information by semi-autonomous driving that combines remote control and autonomous driving, and controls the omnidirectional camera and the sniper sound sensor; With
The control means includes
The image captured by the omnidirectional camera is always recorded by the recording means, and the image recorded by the recording means is held for at least a predetermined time,
When the sniper sound is detected by the sniper sound sensor, the generation position of the sniper sound is specified,
When the generation position of the sniper sound is specified, the image of the specified sniper sound is selected from the images recorded by the recording unit and recorded during a predetermined period before and after the generation of the sniper sound. A sniper detection device, wherein an image is selected according to a generation position and a travel distance of the unmanned vehicle within the predetermined period, and the selected image is reproduced. 2. The sniper detection device according to claim 1, wherein the generation position of the sniper sound is specified by specifying a direction and a distance from the unmanned vehicle based on detection by the sniper sound sensor. The generation position of the sniper sound is specified by specifying the direction from the unmanned vehicle based on detection by the sniper sound sensor, and image processing in the direction of the images taken by the omnidirectional camera is performed. The sniper detection device according to claim 1, wherein: The omnidirectional camera and the sniper sound sensor are supported by lifting and lowering means mounted on the unmanned vehicle, and are moved and operated to a position spaced apart from the unmanned vehicle by the lifting and lowering means. The sniper detection device according to any one of claims 1 to 3.

Images