A targeting robot confrontation system based on binocular recognition

Abstract

The present invention discloses a target shooting robot confrontation system based on binocular recognition, wherein the business layer includes a PC software system; the system core layer includes a GNSS base station, a GNSS mobile station, a microcontroller unit and a navigation system; the execution layer includes a bottom control panel and a laser simulation weapon system including an infrared transmitter and an infrared wireless receiving ring target and an infrared receiver, an emergency brake system including an ultrasonic device, a radar, an anti-collision bar and an emergency stop switch, a wireless communication system including a wireless transmission module and a remote controller, a binocular recognition and positioning system including a binocular high-speed camera and a marking component, a power system including a motor driver and a motor, and a battery management system including a power supply and a coulomb meter. The target shooting robot confrontation system of the present invention realizes intelligent confrontation shooting, evasion shooting and other functions, which can make the training with the target shooting robot closer to actual combat, and is conducive to improving the training efficiency of soldiers.

Description

Technical Field

[0001] The invention relates to the field of target shooting robot systems. Background Art

[0002] In order to continuously improve the military literacy of soldiers, shooting training is a common training subject for special combat personnel. With the improvement of the efficiency and safety of training projects, in order to improve some of the shortcomings of the traditional shooting series, at present, the country has developed shooting robots for some fixed targets. They have the advantages of unpredictable moving targets and automatic target reporting, which greatly improves the training efficiency of soldiers. Although using shooting robots to replace traditional target training can better improve the training effect of soldiers, it still has the disadvantages of being out of actual combat, lacking intelligent avoidance, and simulating human behavior functions. In view of this, the inventor of this case developed and improved the existing shooting robots, and after many tests and experimental training, this case came into being. Summary of the invention

[0003] The purpose of the present invention is to propose a target shooting robot confrontation system based on binocular recognition to address the shortcomings of current target shooting robots, such as being out of touch with actual combat, lacking intelligent evasion, and lacking the function of simulating human behavior.

[0004] To achieve the above-mentioned purpose, the technical solution of the present invention is: a target shooting robot confrontation system based on binocular recognition, characterized in that it includes a business layer, a system core layer and an execution layer;

[0005] The business layer includes a PC-side software system; the PC-side software system is connected and communicated with the system core layer and the execution layer through a wireless module;

[0006] The core layer of the system includes a GNSS base station, a GNSS mobile station, a micro control unit and a navigation system. The GNSS mobile station and the micro control unit are connected and communicated with the navigation system. The GNSS base station and the GNSS mobile station are connected and communicated with each other via a wireless module.

[0007] The execution layer includes a bottom control board and a laser simulation weapon system, an emergency braking system, a wireless communication system, a binocular recognition and positioning system, a power system and a battery management system. The laser simulation weapon system includes an infrared transmitter and an infrared wireless receiving ring target connected to the bottom control board and an infrared receiver worn on the trainee. The emergency braking system includes an ultrasonic device, a radar, an anti-collision strip and an emergency stop switch connected to the bottom control board. The wireless communication system includes a wireless transmission module and a remote controller connected to the bottom control board. The binocular recognition and positioning system includes a binocular high-speed camera connected to the bottom control board and a marking component worn on the trainee. The power system includes a motor driver connected to the bottom control board and a motor connected to the motor driver. The battery management system includes a power supply connected to the bottom control board and a coulomb meter connected to the power supply. The navigation system is connected to the bottom control board for communication.

[0008] The laser simulation weapon system also includes a display module and a voice broadcast module connected to the bottom control board. The output terminal of the display module is an LED display screen, and the output terminal of the voice broadcast module is a loudspeaker.

[0009] The above-mentioned targeting robot confrontation system based on binocular recognition is characterized in that the targeting confrontation working method is as follows: the battery management system supplies power to the entire targeting robot, detects the battery usage status in real time through a coulomb meter to obtain battery status information, and sends the battery status information to the terminal device through a wireless module;

[0010] The binocular recognition and positioning system calculates the position and distance information based on the marking component on the trainee, which is the current position information, and predicts the trainee's upcoming motion trajectory position information, which is the predicted position information. The current position information and the predicted position information are sent to the laser simulation weapon system and the navigation system through the bottom control board;

[0011] The laser simulation weapon system starts the laser transmitter to conduct confrontation according to the current position information, and transmits an infrared signal beam with an encryption function to the current position of the trainee. During the confrontation, the infrared wireless receiving target and the infrared receiver on the trainee detect in real time whether the corresponding receiving area has received an infrared signal. If an infrared signal is received, the detected infrared signal is analyzed and decoded, and the actual coordinates of the infrared signal landing point are calculated, and sent to the display module and the voice broadcast module for display and broadcast;

[0012] The navigation system compares the current position information and predicted position information with the position information of the shooting robot, and uses the obstacle avoidance algorithm to plan and calculate the navigation path to avoid the trainee, and obtains the waypoint position information and navigation path information, and sends the waypoint position information and navigation path information to the bottom control board. The bottom control board calculates the motor driving information of the shooting robot to the waypoint position through the power system, and sends it to the motor driver, and the motor driver controls the motor driving the shooting robot to work; at the same time, the navigation system obtains the position sensor information of the ultrasonic device and radar of the emergency braking system and integrates it into the planning calculation of the navigation path, and continuously performs autonomous positioning and posture adjustment of the shooting robot;

[0013] The emergency braking system can perform emergency braking when the anti-collision strip is hit, or can perform emergency braking manually through an emergency switch.

[0014] The control flow method of the binocular recognition and positioning system is as follows:

[0015] Before training, the trainee is calibrated by a binocular high-speed camera. The trainee is completely placed in the shooting area. The marking component is worn on the trainee at a determined characteristic position. The trainee's image is captured and recorded by the binocular high-speed camera as a marking image. The marking component on the marking image is detected by an image processing algorithm to obtain a marking detection result of the marking component. Then, the spatial coordinates of the marking component in the three-dimensional world coordinates are calculated to complete the calibration.

[0016] When training, the marking component is worn in a conspicuous position on the trainee's body, and the following process is used:

[0017] First, stereoscopic video acquisition: During the movement of the target shooting robot, a binocular high-speed camera is used to capture and record the video image as the current video image.

[0018] Next, the corresponding matching points in the two frames of the video are obtained, and the corresponding matching points in the two frames before and after the current video image are obtained by feature point matching.

[0019] Next, the camera displacement is calculated by matching the coordinate changes of the points in the imaging space or establishing three-dimensional coordinates to calculate the camera displacement and obtain the camera displacement value.

[0020] Next, binocular vision positioning is used to obtain the position and rotation angle of the binocular high-speed camera at each moment during the movement process. Combined with Kalman filtering, the moving route of the binocular high-speed camera during the whole process is obtained, and the target shooting robot is positioned by binocular vision in real time.

[0021] Then, the data obtained by the above processing of the current video image is processed to obtain the information data required for calculating the trainee's current position information and predicted position information, and is sent to the central processing module for processing to calculate the trainee's current position information and predicted position.

[0022] By adopting the above technical scheme, the beneficial effects of the present invention are as follows: the system structure of the above-mentioned target shooting robot confrontation system constitutes a target shooting robot confrontation system that can realize avoidance confrontation, the system identifies and locates the trainees by utilizing the binocular recognition and positioning system, realizes confrontation training with the trainees by using the laser simulation weapon system, makes trajectory planning for avoiding the trainees by using the navigation system, and realizes the target shooting robot to avoid the trainees according to the trajectory planning hole group motor driver by the power system. In this way, the target shooting robot using the target shooting robot confrontation system of the present invention realizes intelligent confrontation shooting, as well as the function of avoiding shooting and other intelligent functions, which are functional effects not realized in the existing target shooting robots. Through the realization of these functions, the training with the target shooting robot can be closer to actual combat, which is beneficial to improving the training efficiency of soldiers.

[0023] When a target shooting robot using the target shooting robot confrontation system of the present invention encounters an emergency situation, emergency braking can be performed through several emergency systems. The operator can view various parameter indicators of the target shooting robot through the battery management system and the PC and observe the use status of the target shooting robot at any time. In addition, the target shooting robot of the present invention can also be controlled by a wireless remote control device to meet various needs of the target shooting robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a schematic block diagram of the structure of the confrontation system involved in the present invention;

[0025] Figure 2 It is a control flow chart of a binocular recognition and positioning system involved in the present invention;

[0026] Figure 3 It is a control flow chart of a laser simulation weapon system involved in the present invention;

[0027] Figure 4 The present invention relates to a navigation control flow chart of a target shooting robot. DETAILED DESCRIPTION

[0028] In order to further explain the technical solution of the present invention, the present invention is described in detail below through specific embodiments.

[0029] The present invention discloses a target shooting robot confrontation system based on binocular recognition, such as Figure 1 , Figure 2 , Figure 3 and Figure 4 As shown, it includes a business layer, a system core layer and an execution layer. The difference between the application and the target shooting robot and the existing target shooting robot is that it realizes a more intelligent functional effect. The system working structure of its control system constitutes a confrontation system that can realize the functional effect of confronting and avoiding the trainees. The system structure of the system is described in detail below in conjunction with the accompanying drawings (such as Figure 1 shown) and control process.

[0030] The business layer includes a PC-side software system 11, and the PC-side software system 11 is connected and communicated with the system core layer and the execution layer through a wireless module;

[0031] The system core layer includes a GNSS base station 21, a GNSS mobile station 22, a micro control unit 23 and a navigation system 24. The GNSS mobile station 23 and the micro control unit 24 (MCU) are connected and communicated with the navigation system 24. The GNSS base station 21 and the GNSS mobile station 22 are connected and communicated via a wireless module. GNSS is a global navigation satellite system.

[0032] The execution layer includes a bottom control board 31 and a laser simulation weapon system 32, an emergency braking system 33, a wireless communication system 34, a binocular recognition and positioning system 35, a power system 36 and a battery management system 37; the laser simulation weapon system 32 includes an infrared transmitter 321 connected to the bottom control board 31, an infrared wireless receiving ring target 322 and an infrared receiver 323 worn on the trainee, the infrared wireless receiving ring target 322 or the ring target can automatically report the ring, is easy to use and has high ring reporting accuracy, the laser simulation weapon system 32 also includes a display module 324 and a voice broadcast module 325 connected to the bottom control board 31, the output terminal of the display module 324 can be an LED display screen; the output terminal of the voice broadcast module 325 can be a loudspeaker, the emergency braking system 33 includes an ultrasonic device 331, a radar 332, an anti-collision strip 333 and an emergency stop switch 334 connected to the bottom control board 31, which can realize the detection of obstacles. and emergency braking processing during collision; the wireless communication system 34 includes a wireless transmission module 341 and a remote controller 342 connected to the bottom control board 31, which can realize wireless transmission of the target shooting robot and remote control operation of the target shooting robot; the binocular recognition and positioning system 35 includes a binocular high-speed camera 351 connected to the bottom control board 31 and a marking component 352 worn on the trainee, the power system 36 includes a motor driver 361 connected to the bottom control board 31 and a motor 362 connected to the motor driver 361, which sets the corresponding number of motor drivers 361 and motors 362 according to the needs of the target shooting robot, such as controlling the servo motor of the chassis of the trolley, controlling the motor for lifting and adjusting the height of the target body on the fuselage, etc., the battery management system 37 includes a power supply 371 connected to the bottom control board 31 and a coulomb meter 372 connected to the power supply 371, which can detect the power usage in real time; the navigation system 24 is connected to the bottom control board 31 for communication.

[0033] From the above structure, it can be seen that the basic binocular recognition target shooting robot confrontation system of the present invention is composed of three architecture layers, such as Figure 1As shown, the execution layer, system core layer and business layer, the execution layer is located at the bottom of the entire architecture, which is mainly composed of the above-mentioned binocular recognition and positioning system, power system, battery management system, laser simulation weapon system, emergency braking system and wireless communication system. Of course, this layer can also include other system settings, which are not the main improvements of this case. Other system settings are specifically limited here. The bottom control board is the core component of the execution layer, which undertakes the tasks of upper layer (system core layer) data analysis and execution, data settlement and upload, power system control command issuance and motor encoder data collection. The system core layer mainly includes navigation system and GNSS base station, GNSS mobile station and MCU, which are mainly responsible for sensor data fusion, positioning solution, path planning, motion control and data transfer. In addition, the business layer is the control layer, which mainly includes terminal equipment, such as PC software system, mobile terminal, etc., and is mainly responsible for training task setting, robot status monitoring, bottom sensor data monitoring, training results storage and analysis.

[0034] The control flow of each part of the system in the target shooting robot confrontation system based on binocular recognition of the present invention is described in detail below.

[0035] The battery management system supplies power to the entire target shooting robot, detects the battery usage status in real time through a coulomb meter to obtain battery status information, and sends the battery status information to a terminal device through a wireless module.

[0036] The binocular recognition and positioning system calculates the position and distance information based on the marking component on the trainee, which is the current position information, and predicts the trainee's upcoming motion trajectory position information, which is the predicted position information. The current position information and the predicted position information are sent to the laser simulation weapon system and the navigation system through the bottom control board; the control process method of the binocular recognition and positioning system is as follows:

[0037] Before training, the trainee is calibrated by a binocular high-speed camera. The trainee is completely placed in the shooting area. The marking component is worn on the trainee at a determined characteristic position. The trainee's image is captured and recorded by the binocular high-speed camera as a marking image. The marking component on the marking image is detected by an image processing algorithm to obtain a marking detection result of the marking component. Then, the spatial coordinates of the marking component in the three-dimensional world coordinates are calculated to complete the calibration.

[0038] When training, the marking component is worn on the trainee's body at a certain characteristic position, and the following process is used:

[0039] First, stereoscopic video acquisition: During the movement of the target shooting robot, a binocular high-speed camera is used to capture and record the video image as the current video image.

[0040] Next, the corresponding matching points in the two frames of the video are obtained, and the corresponding matching points in the two frames before and after the current video image are obtained by feature point matching.

[0041] Next, the camera displacement is calculated by matching the coordinate changes of the points in the imaging space or establishing three-dimensional coordinates to calculate the camera displacement and obtain the camera displacement value.

[0042] Next, binocular vision positioning is used to obtain the position and rotation angle of the binocular high-speed camera at each moment during the movement process. Combined with Kalman filtering, the moving route of the binocular high-speed camera during the whole process is obtained, and the target shooting robot is positioned by binocular vision in real time.

[0043] Then, the data obtained by the above processing of the current video image is processed to obtain the information data required for calculating the trainee's current position information and predicted position information, and is sent to the central processing module for processing to calculate the trainee's current position information and predicted position.

[0044] The laser simulation weapon system starts the laser transmitter according to the current position information to launch an infrared signal beam with encryption function to the current position of the trainee. During the confrontation, the infrared wireless receiving target and the infrared receiver on the trainee detect in real time whether there is any infrared signal received in the corresponding receiving area. If an infrared signal is received, the detected infrared signal is analyzed and decoded, and the actual coordinates of the infrared signal landing point are calculated, and sent to the display module and voice broadcast module for display and broadcast.

[0045] The navigation system compares the current position information and predicted position information with the position information of the target shooting robot, and uses the obstacle avoidance algorithm to plan and calculate the navigation path to avoid the trainees, and obtains the waypoint position information and navigation path information, and sends the waypoint position information and navigation path information to the underlying control board. The underlying control board calculates the waypoint position information and navigation path information through the power system to obtain the driving information of the motor for the target shooting robot to move to the waypoint position and sends it to the motor driver, which controls the motor to drive the target shooting robot to move; at the same time, the navigation system obtains the position sensor information of the ultrasonic device and radar of the emergency braking system and integrates it into the planning calculation of the navigation path, and continuously performs autonomous positioning and posture adjustment of the target shooting robot.

[0046] The emergency braking system can perform emergency braking when the anti-collision strip is hit, or can perform emergency braking manually through an emergency switch.

[0047] The targeting robot using the targeting robot confrontation system based on binocular recognition according to the present invention can realize a variety of working modes, and can use corresponding training modes according to the training level of the trainees. The use of the targeting robot is more intelligent, the training is more efficient and accurate, and the training effect is improved rapidly.

[0048] The above embodiments and drawings do not limit the product form and style of the present invention. Any appropriate changes or modifications made by ordinary technicians in the relevant technical field should be deemed to be within the patent scope of the present invention.

Claims

1. A target shooting robot confrontation system based on binocular recognition, characterized in that: Includes business layer, system core layer and execution layer; The business layer includes a PC-side software system; the PC-side software system is connected and communicated with the system core layer and the execution layer through a wireless module; The core layer of the system includes a GNSS base station, a GNSS mobile station, a micro control unit and a navigation system. The GNSS mobile station and the micro control unit are connected and communicated with the navigation system. The GNSS base station and the GNSS mobile station are connected and communicated with each other via a wireless module. The execution layer includes a bottom control board and a laser simulation weapon system, an emergency braking system, a wireless communication system, a binocular recognition and positioning system, a power system and a battery management system. The laser simulation weapon system includes an infrared transmitter and an infrared wireless receiving ring target connected to the bottom control board and an infrared receiver worn on the trainee. The emergency braking system includes an ultrasonic device, a radar, an anti-collision strip and an emergency stop switch connected to the bottom control board. The wireless communication system includes a wireless transmission module and a remote controller connected to the bottom control board. The binocular recognition and positioning system includes a binocular high-speed camera connected to the bottom control board and a marking component worn on the trainee. The power system includes a motor driver connected to the bottom control board and a motor connected to the motor driver. The battery management system includes a power supply connected to the bottom control board and a coulomb meter connected to the power supply. The navigation system is connected to the bottom control board for communication; The target shooting confrontation working method is as follows: the battery management system supplies power to the target shooting robot, detects the battery usage status in real time through a coulomb meter to obtain battery status information, and sends the battery status information to a terminal device through a wireless module; The binocular recognition and positioning system calculates the position and distance information based on the marking component on the trainee, which is the current position information, and predicts the trainee's upcoming motion trajectory position information, which is the predicted position information. The current position information and the predicted position information are sent to the laser simulation weapon system and the navigation system through the bottom control board; The laser simulation weapon system starts the laser transmitter to conduct confrontation according to the current position information, and transmits an infrared signal beam with an encryption function to the current position of the trainee. During the confrontation, the infrared wireless receiving ring target and the infrared receiver on the trainee detect in real time whether the corresponding receiving area has received an infrared signal. If an infrared signal is received, the detected infrared signal is analyzed and decoded, and the actual coordinates of the infrared signal landing point are calculated, and sent to the display module and the voice broadcast module for display and broadcast; The navigation system compares the current position information and predicted position information with the position information of the shooting robot, and uses the obstacle avoidance algorithm to plan and calculate the navigation path to avoid the trainee, and obtains the waypoint position information and navigation path information, and sends the waypoint position information and navigation path information to the bottom control board. The bottom control board calculates the motor driving information of the shooting robot to the waypoint position through the power system, and sends it to the motor driver, and the motor driver controls the motor driving the shooting robot to work; at the same time, the navigation system obtains the position sensor information of the ultrasonic device and radar of the emergency braking system and integrates it into the planning calculation of the navigation path, and continuously performs autonomous positioning and posture adjustment of the shooting robot; The emergency braking system can perform emergency braking when the anti-collision strip is hit, or can perform emergency braking manually through an emergency switch; The control flow method of the binocular recognition and positioning system is as follows: Before training, the trainee is calibrated by a binocular high-speed camera. The trainee is completely placed in the shooting area. The marking component is worn on the trainee at a determined characteristic position. The trainee's image is captured and recorded by the binocular high-speed camera as a marking image. The marking component on the marking image is detected by an image processing algorithm to obtain a marking detection result of the marking component. Then, the spatial coordinates of the marking component in the three-dimensional world coordinates are calculated to complete the calibration. When training, the marking component is worn in a conspicuous position on the trainee's body, and the following process is used: First, stereoscopic video acquisition: During the movement of the target shooting robot, a binocular high-speed camera is used to capture and record the video image as the current video image. Next, the corresponding matching points in the two frames of the video are obtained, and the corresponding matching points in the two frames before and after the current video image are obtained by feature point matching. Next, the camera displacement is calculated by matching the coordinate changes of the points in the imaging space or establishing three-dimensional coordinates to calculate the camera displacement and obtain the camera displacement value. Next, binocular vision positioning is used to obtain the position and rotation angle of the binocular high-speed camera at each moment during the movement process. Combined with Kalman filtering, the moving route of the binocular high-speed camera during the whole process is obtained, and the target shooting robot is positioned by binocular vision in real time. Then, the data obtained by the above processing of the current video image is processed to obtain the information data required for calculating the trainee's current position information and predicted position information, and is sent to the central processing module for processing to calculate the trainee's current position information and predicted position.

2. A target shooting robot confrontation system based on binocular recognition as claimed in claim 1, characterized in that: The laser simulation weapon system also includes a display module and a voice broadcast module connected to the bottom control board. The output terminal of the display module is an LED display screen, and the output terminal of the voice broadcast module is a loudspeaker.

Images