CN115903561A - Robot directional driving-away system and method - Google Patents

Abstract

The invention discloses a robot directional driving-away system and a method, wherein the system comprises a loading system, a strong sound dispersing system, a wireless control platform and a vehicle system; the uploading system comprises a video monitoring system, an integrated navigation system, a communication control system and a photoelectric tower; the strong sound dispersing system is used for directly steering the received target information to a target position and carrying out strong sound dispersing on the target; the vehicle system is used for driving the whole device to move, and the loading system and the strong sound dispersion system are both arranged on the vehicle system; and the wireless control platform is in wireless communication with the vehicle system and is used for remotely controlling the vehicle system. The unmanned platform adopts the characteristics of integration and modular design, has perfect self defense system, and can perform actions of maneuvering evasion, decoy interference and the like aiming at different threats. The invention can assist the fighter in the battlefield and form complementation with the fighter, thereby greatly reducing the casualties in the battle and improving the battle efficiency of the army.

Description

Robot directional driving-away system and method

Technical Field

The invention relates to the technical field of safety protection equipment, in particular to a robot directional driving-away system and a robot directional driving-away method.

Background

The directional sound wave dispersing device is a technical device which utilizes one or more high-directivity sound wave transmitting units to directionally generate sound waves, and generally generates high-pressure sound waves right in front for driving away people or birds and beasts.

Present sound wave dispel ware is mostly fixed mounting's form, chinese utility model patent as patent application number is CN201920834986.0 discloses an explosion-proof type sound wave dispel ware protector, though the device can avoid heavy rain weather to descend the rainwater direct impact to dispel the ware body, but the device sets up subaerial through triangle-shaped support snap-on, and it can't be directed against the sound wave of special complicated regions such as mountain region, jungle, beach, desert, snowfield, muddy and dispel. The existing acoustic wave dissipater has defects in the aspect of long-distance acoustic wave transmission, the acoustic wave dissipater cannot transmit signals in a long distance, and in a special regional environment, the surrounding environment is complex, and more unstable factors exist, so that the signals cannot be transmitted well.

Compared with the traditional weapon, the robot directional driving-away system is a non-lethal weapon, and can not directly cause equipment destruction and ecological environment destruction, but can be a humane weapon which can make the fighting ability of enemies lose and weaken through the technical means of soft killing and soft destruction. Aiming at the problems of the existing sound wave dissipater, the research and development of a robot directional expelling system is very important.

Disclosure of Invention

The technical problem to be solved by the invention is to provide a system and a method for directionally driving away a robot, so as to solve the problems in the background art.

In order to solve the technical problems, the technical scheme adopted by the invention is as follows.

A robotic directional drive-off system, comprising:

the uploading system comprises a video monitoring system, an integrated navigation system, a communication control system and a photoelectric tower; the video monitoring system is used for shooting a target position; the integrated navigation system is used for planning a path; the communication control system is used for transmitting shot target information; the photoelectric tower is used for searching, detecting, monitoring and tracking the target and inverting the geographic position information of the target;

the strong sound dispersing system is used for directly steering the received target information to a target position and carrying out strong sound dispersing on the target;

the vehicle system is used for driving the whole device to move, and the loading system and the strong sound dispersion system are both arranged on the vehicle system;

and the wireless control platform is in wireless communication with the vehicle system and is used for remotely controlling the vehicle system.

Further optimize technical scheme, strong sound dispels the system and includes:

a microphone unit for outputting sound waves of high decibels;

the power amplifier unit is used for providing the audio signal which is preprocessed and amplified for the loudspeaker unit;

the holder is used for driving the loudspeaker unit to rotate in pitching and azimuth directions;

and the multilingual module is used for providing a multilingual offline translation function.

According to the technical scheme, the power amplifier unit comprises a signal processing module, a network control module, a filtering power supply module and a power amplifier.

Further optimize technical scheme, the cloud platform includes:

the mechanical table body is used for pitching and rotating in direction;

the control system is used for controlling the action of the mechanical table body;

a load support.

Further optimize technical scheme, the facial make-up system still includes:

the power supply control system is used for providing electric energy for the upper mounting system;

the environment control system is used for controlling the environment around the vehicle body;

a light illumination system.

A robot directional driving-away method is carried out based on the robot directional driving-away system and comprises the following steps:

planning a path, shooting a target position, and performing a patrol task;

carrying out all-weather searching, investigation, monitoring and tracking on the target, inverting the geographic position information of the target, and sending the target information to a strong sound dispersion system and a command room in a wireless transmission mode;

and after receiving the target information, the strong sound dispersing system directly turns to the target position, and performs strong sound dispersion on the target by a non-fatal means.

Further optimizing the technical scheme, adopting a wide dynamic algorithm technology to shoot the target position, wherein the wide dynamic algorithm technology comprises the following steps:

s1, extracting a brightness component of an image to be processed;

s2, calculating the environmental factor of each pixel point by using the pixel field information of the brightness component;

s3, obtaining a wide dynamic transformation formula by utilizing the environmental factors according to the characteristics of a human visual system;

s4, obtaining a wide dynamic range image by using a wide dynamic transformation formula;

and S5, gamma correction.

Further optimize technical scheme, carry out the modular design with megaphone unit and power amplifier unit in the strong sound disperse system, each sound source is a megaphone module and power amplifier module, can enough work alone, also can quick independent assortment, forms the sounder array.

Further optimizing the technical scheme, and designing the strong sound dispersion system by adopting a full digital power amplifier multi-channel dynamic balance technology.

Due to the adoption of the technical scheme, the technical progress of the invention is as follows.

The invention discloses a robot directional driving-away system, which is an operation platform integrating a directional sound source technology, a sound source array technology, a multi-language module technology, a task planning technology, a platform maneuvering technology, a navigation technology, a field communication technology and a natural cognition operation technology, and is mainly used for high-mobility sound wave striking, multi-language module off-line broadcasting and remote shouting, day and night patrol, fixed-point value guard, reconnaissance attack, information collection and communication relay in wide-range areas such as side defense, peak meeting security, military strategy rear warehouse, maintenance anti-terrorism, actual combat application and ground patrol.

The robot directional driving-away system has four working modes of remote control, target tracking, map prefabrication and autonomous driving under the conditions of no carrying of operators and no supporting environment. The unmanned platform has the advantages that the integrated and modular design characteristics are adopted, the self defense system is perfect, the overall appearance is low RCS stealth, low infrared characteristics and anti-hijack design, and the unmanned platform can perform actions such as maneuver evasion, decoy interference and the like aiming at different threats. Can assist the fighter in the battlefield and form complementation with the fighter, greatly reduce the casualties in the battlefield and improve the battle efficiency of the army.

The invention adopts a full digital power amplifier multi-channel dynamic balance technology to design a strong sound dispersion system: the full digital power amplifier with multiple channels, high power and high fidelity can perform dynamic load balance on each channel by applying a real-time protection technology. The phase of a multi-channel sound source signal is controlled to be consistent, a multi-channel high-efficiency power amplifier array is designed based on class D audio amplification technology, a single-stage power amplifier is adopted, a power supply and a power amplifier are directly combined, alternating current directly supplies power to the power amplifier stage through a rectifier module, and the efficiency is higher. And the later stage adopts an electric energy recovery device to recover the invalid power of the inductive load, thereby further improving the efficiency. Can meet the requirements of power drive and heat dissipation under severe conditions.

The invention relates to a power supply bidirectional isolation and self-adaptive redundancy management technology, which comprises the following steps: the self-adaptive redundant power supply management improves the reliability of the equipment, and simultaneously, the front and back bidirectional isolation is carried out to meet the requirements of electromagnetic compatibility standards on power line conducted emission and sensitivity.

The invention adopts the wide dynamic algorithm technology to shoot the target position, so that the image can cover a wider dynamic range, and the imaging quality of the scene is effectively improved.

The method for tracking the target is based on the contour tracking technology and comprises the following steps: the contour of the object is tracked in the binary edge image of the succeeding frame using the contour of the boundary of the object as a template, and the contour can be automatically and continuously updated. Compared with the tracking method based on the region, the contour tracking method can achieve better matching effect basically under the condition of small calculation amount, and if reasonably separating each moving object and realizing contour initialization are started, the tracking can be continuously carried out even under the condition that partial occlusion exists.

Drawings

FIG. 1 is a schematic block diagram of a robot directional dislodging system of the present invention;

FIG. 2 is a front view of a robotic directional dislodging system of the present invention;

FIG. 3 is a side view of a robotic directional dislodging system of the present invention;

FIG. 4 is a top view of a robotic directional dislodging system of the present invention;

FIG. 5 is a processing flow chart of a wide dynamic algorithm technique in a robot directional drive-away method of the present invention;

FIG. 6 is a schematic view of a single emitter array of the present invention;

FIG. 7 is a schematic view of a 1X 2 sounder array according to the present invention;

FIG. 8 is a schematic view of a 2X 2 sounder array according to the present invention;

FIG. 9 is a schematic view of a 2X 3 sounder array according to the present invention;

FIG. 10 is a schematic view of a 3X 4 sounder array according to the present invention;

FIG. 11 is a schematic view of the mounting of a single sound generator array and a pan/tilt head according to the present invention;

FIG. 12 is a schematic view of the installation of the 1X 2 sounder array and pan/tilt head of the present invention;

FIG. 13 is a schematic view of the installation of the 2X 2 sounder array and the pan/tilt head according to the present invention

FIG. 14 is a schematic view of the installation of the 2X 3 sounder array and the pan/tilt head according to the present invention

FIG. 15 is a schematic view of the installation of the 3X 4 sounder array and pan/tilt head of the present invention;

FIG. 16 is a schematic block diagram of a video surveillance system of the present invention;

FIG. 17 is a diagram of a moving object state during all-weather tracking of the object according to the present invention;

FIG. 18 is a schematic view of a multi-sensor fusion in accordance with the present invention;

fig. 19 is an electrical connection diagram of a power amplifier of the present invention;

FIG. 20 illustrates the signal processing board module concept of the present invention;

FIG. 21 is a flow chart of audio signal processing according to the present invention.

Wherein: 1. vehicle system, 2, facial make-up system, 3, strong sound dissipation system, 4, megaphone unit, 5, megaphone unit compoboard, 6, cloud platform, 7, laser radar, 8, millimeter wave radar, 9, camera, 10, big dipper antenna.

Detailed Description

The invention will be described in further detail with reference to the following figures and specific examples.

A robot directional driving-away system is shown in a combined mode in fig. 1 and 15 and comprises a top-mounted system 2, a strong sound dispersing system 3, a vehicle system 1 and a wireless control platform.

The upper system comprises a video monitoring system, an integrated navigation system, a communication control system, a photoelectric tower, a power supply control system, an environment control system and a lighting illumination system.

And the video monitoring system is used for shooting the target position. The video monitoring system comprises a main shooting unit, a look-around unit, a vehicle body monitoring unit and a video coding unit, and is shown in a combined figure 16.

And the integrated navigation system is used for planning the path. The integrated navigation system comprises a positioning directional antenna, a satellite navigation module and an inertial navigation module.

The realization of the vehicle body state perception function is mainly based on equipment such as a BeiDou Navigation Satellite System (BDS), an Inertial Navigation System (INS) and the like, and aims to obtain information such as the running speed, the attitude and the direction of a vehicle and provide effective data for the positioning and Navigation of an intelligent driving vehicle. The invention adopts the double antennas, and judges the direction of the head of the vehicle through the double antennas, thereby realizing the direction judgment of the vehicle.

Generally, an IMU (Inertial measurement unit) is provided with a three-axis gyroscope and three-direction accelerometers to measure the angular velocity and acceleration of an object in a three-dimensional space and to solve the attitude of the object, and meanwhile, to improve reliability, each axis can be provided with more sensors, generally, the IMU is to be installed on the gravity center of the object to be measured, most IMUs are used in devices requiring motion control, such as automobiles and robots, and are also used in occasions requiring precise displacement estimation by attitude, such as Inertial navigation devices of submarines, airplanes, missiles, and spacecrafts.

The combined navigation system adopts two or more different navigation means, and fuses each navigation information data according to the characteristic advantage complementation of different navigation systems, thereby obtaining a more accurate navigation system. The information fusion technology adopted by the current integrated navigation system mainly comprises a track deduction method, a Kalman filtering method, a Bayesian estimation method, a neural network, a support vector machine and the like.

The most widely applied Kalman filtering algorithm is the classical method or the most widely applied Kalman filtering algorithm in the integrated navigation system, but the algorithm can only be applied to a linear system, and meanwhile, the state equation and the observation equation must be linear. Later researchers improved on the basis of the classic Kalman filtering algorithm, and excellent algorithms such as extended Kalman filtering and unscented Kalman filtering are derived. Classic Kalman filtering and extended Kalman filtering are all realized by combining a hidden Markov model and a Bayesian model, and the method is a method for predicting and updating the system state by observing information and an observation model.

And the communication control system is used for transmitting the shot target information. The communication control system comprises a main control unit and a radio station antenna feeder system. The unmanned vehicle is provided with the combined navigation system and the video monitoring system, so that a path can be freely planned, and a patrol task can be performed.

The photoelectric tower is provided with a high-performance multispectral detection unit, can perform searching, detection, monitoring, tracking and the like all weather, can invert geographical position information of a target, and sends the target information to a strong sound dispersion system and a command room.

The photoelectric tower mainly comprises a photoelectric load, a stable platform, a front shell assembly and an outer cable. The photoelectric load comprises an angle gauge, a black-and-white visible light camera, a color visible light camera, a laser camera, a non-refrigeration infrared camera, an optical interface board, an image tracking board, a power supply conversion board and an outer cable. The stabilized platform consists of a stabilized platform structure body, a voltage stabilizing power supply board, a communication servo control board, a gyro assembly, an electric locking assembly, a manual locking assembly and a platform cable. The front shell assembly mainly comprises a front shell structure body and various sensor windows. The photoelectric tower can realize the following functions:

1) The high-definition visible light and infrared video images can be provided, the infrared images have four fields of view, and the large infrared field of view can be used for target search;

2) The optical axis can be stabilized on the moving carrier;

3) The selected target can be automatically tracked;

4) The system can perform laser ranging and laser irradiation on a target and output distance information;

5) The information of the angle and the angular speed of the aiming line can be output;

6) Maintenance operations such as on-line zero calibration, drift correction and the like can be carried out;

7) The system can interact with external equipment through a communication interface and a video interface, and provides target position coordinates for a strong sound driving-away system.

And the power supply control system is used for providing electric energy for the upper mounting system. The power supply control system comprises an energy storage battery, a power supply switch unit and a power supply protection monitoring unit. The power control system adopts power bidirectional isolation and self-adaptive redundancy management technology: the self-adaptive redundant power supply management improves the reliability of the equipment, and simultaneously carries out front and back bidirectional isolation so as to meet the requirements of electromagnetic compatibility standards on power line conducted emission and sensitivity.

And the environment control system is used for controlling the environment around the vehicle body. The environment control system comprises a vehicle body environment control unit and a circuit module unit. The environmental control system comprises a laser radar 7, a millimeter wave radar 8, a camera 9, a Beidou antenna 10 and inertial navigation equipment. The laser radar is provided with 1. The millimeter wave radars are arranged in two numbers and are respectively arranged on the front side and the rear side of the vehicle system. The cameras are provided with 7 cameras, including a front-view camera and a blind-repairing camera. The Beidou antennas are arranged in two and are respectively arranged at the front part of the top end and the rear part of the top end of the vehicle system. And environmental perception is realized through a multi-sensor fusion processing system. A schematic of the location setting of the context awareness system is shown in fig. 4.

Lidar is considered an indispensable sensor for autonomous driving due to its strong environmental perception. The laser radar has 360-degree surrounding visual field and 15-degree vertical visual field, has the functions of a retro reflector and sunlight relief, has excellent remote detection performance, can detect dark or low-reflectivity objects such as tires, dark vehicles, low-reflectivity road surfaces and low-visibility pedestrians, has the highest resolution and strong reflectivity, simplifies the detection and tracking of vehicles, pedestrians and other obstacles, has an advanced inter-sensor interference suppression function, can ensure that the vehicles can safely run in a strange and dynamic environment, and improves the safety of driving the vehicles at night.

The image sensor-camera is capable of acquiring environmental color scene information, and is the second largest source for unmanned vehicles to acquire environmental information. The selectable models and types of cameras are very various and can be simply divided into a monocular camera, a binocular stereo camera and a panoramic camera.

The environment imaging of the unmanned vehicle is the application of machine vision on the vehicle, and needs to meet the requirements of the vehicle running environment and the self running condition. Weather changes, vehicle motion speed disturbances, camera mounting location, etc. all affect vehicle vision. The unmanned vehicle task has high requirements on image quality, and not only needs higher frame frequency in image output speed, but also has higher requirements on image quality.

The camera is mainly applied to: lane Departure Warning (LDW), lane Keeping Assist (LKA), forward Collision Warning (FCW), pedestrian Collision Warning (PCW), panoramic parking (SVP), traffic Sign Recognition (TSR). The front-view camera, which is generally a wide-angle lens, is installed at a higher position on the vehicle to realize a longer effective distance; the invention adopts 3 cameras in the foresight, so that the distance and the speed of the target can be calculated through the images of the 3 cameras, and the moving situation of the target can be estimated through an algorithm to realize image tracking.

The vehicle body of the invention uses 7 cameras in total, except that 3 front-view cameras are mainly used for monitoring the right front part of the vehicle at night in the daytime and in a redundant design, the other 4 cameras are positioned around the vehicle body and used for monitoring the surrounding environment of the vehicle. A schematic camera position is shown in fig. 4.

The three front-view cameras are respectively 1 color visible light camera and 2 low-illumination cameras, so that the unmanned vehicle can obtain clear video image information in the daytime and at night.

And finally, the pictures of all the cameras are synthesized into a path of video and transmitted to the command control platform through a wireless link, and the final synthesized picture is mainly a 1-path front view picture and 4-path blind complementing camera videos are positioned at four corners of the picture in consideration of the fact that too many pictures interfere with remote control operation.

Basic parameters of the front-view camera: 1920x1080 resolution, 30fps frame rate and delay time less than or equal to 50ms.

The millimeter wave radar sensor is a radar with the working frequency selected from a frequency domain of 30-300 GHz (the wavelength is 1-10 mm, namely a millimeter wave band). Its advantages are narrow wave beam, high angle resolution, wide frequency band, high hiding power, high anti-interference power, small size, light weight and long distance to be measured. Although the detection range is large without the laser radar, the good directivity and the good penetrating power of the laser radar still make the laser radar incapable of being replaced by the laser radar. According to different measurement principles, the millimeter wave radar sensor can be divided into a pulse mode and a frequency modulation continuous wave mode. Because the distance of millimeter wave radar is far away, the reliability is high, do not receive light, dust influence, compare the camera, its detection distance index is more than 150 meters, wins far away. Compared with laser radar, the price of about 1000 pieces of laser radar is very competitive.

The unmanned patrol car disclosed by the invention uses two products in common and is respectively arranged in the front direction and the rear direction. Scanning is performed for a relatively distant region.

Millimeter wave radar parameters: 1. coverage angle: 18 degrees to 120 degrees; 2. detection range: 70m to 250m.

The environmental perception in the invention is mainly realized by means of multi-sensor fusion, which is a data processing method and tries to acquire real data of measured parameters by combining incomplete, defected and complementary multi-sensor data or multiple sensor data with proper processing algorithm. In a system including multiple sensors, the types of information provided by the various sources, data characteristics, may be different, linear or non-linear, time-varying or non-time-varying, stochastic, or deterministic. The multi-sensor fusion is to fully utilize observation data of a plurality of information sources and process complementary or redundant information of the data in time and space according to a certain optimization criterion, so as to obtain unbiased description of the consistency of an external environment. Fig. 18 shows a schematic diagram of multi-sensor fusion. The multi-sensor fusion can be divided into data-level fusion, feature-level fusion and decision-level fusion.

And the lamplight illuminating system is used for illuminating the whole device at night. The light illuminating system comprises an infrared illuminating lamp, a front lamp and a back lamp.

The directional robot driving-away system further comprises an electric lifting rod, and the strong sound driving-away system is arranged on the electric lifting rod and can perform lifting action.

The strong sound dispersing system is used for directly turning the received target information to a target position, and performing strong sound drive on the target by a non-lethal means, so that the function of integrating the directional drive system of the robot and the inspection of the target is realized.

The strong sound dispersing system comprises a loudspeaker unit 4, a power amplifier unit, a holder 6 and a multi-language module.

And the loudspeaker unit is used for outputting sound waves with high decibels. The loudspeaker unit comprises a shell, an electromagnetic compression medium-high frequency transducer, a horn, an electric connector and the like.

And the power amplifier unit is used for providing a preprocessed and amplified audio signal for the electromagnetic type compressed medium-high frequency transducer in the loudspeaker unit, wherein the amplifier adopts a D-type digital power amplifier.

And the holder is used for driving the loudspeaker unit to rotate in pitching and azimuth directions, so that automatic balance and intelligent control are realized. The cloud platform includes mechanical stage body, control system and load support.

A mechanical table body: the mechanical platform body of the holder is a vertical two-shaft holder. Consists of a frame, a base and internal hardware. The internal hardware comprises two rotation shafting of pitch and azimuth, and also comprises hardware parts such as a driving motor and a position feedback element. The frame structure is a size-adjustable structure, and is adapted to the array mode of the sounder according to the sound pressure level use requirement, so as to provide an action platform for the sounder.

As shown in fig. 12, the loudspeaker unit 4 is disposed on the loudspeaker unit combination board 5, and the loudspeaker unit combination board 5 is disposed on the pan/tilt head, and the pan/tilt head drives the loudspeaker unit 4 to rotate in pitch and in horizontal. A plurality of megaphone units are quickly installed on the megaphone unit combination plate through screws to form a loud sound array, the megaphone unit combination plate is connected with a flange plate of a side arm of a holder and fixed by bolts, a horizontal motor inside the holder rotates to drive the whole side arm of the holder to rotate horizontally, a pitching motor inside the side arm rotates to drive the flange plate of the side arm to rotate, and then the megaphone unit combination plate 5 and the megaphone unit 4 are driven to rotate in a pitching mode.

And the control system is arranged in the table body and is designed into a mechanical control integrated structure.

The sounder array and the pan-tilt head of the invention are mounted as shown in fig. 11 to fig. 15.

And the multi-language module is used for providing a multi-language offline translation function. The directional strong-sound-driving weapon is embedded with a multi-language module, has low distortion and high language definition, can realize the functions of off-line translation and remote broadcasting, and can directionally install a language package aiming at a user area.

The power amplifier unit comprises a signal processing module, a network control module, a filtering power supply module and a power amplifier.

Dynamic balance, according to the response curve of the horn loudspeaker, the digital power amplifier increases power supply to the low-response frequency, and decreases power supply to the high-response frequency, so that the frequency response curves of the loud sound dispeller are close to the same.

Fig. 19 shows the electrical connection diagram of the power amplifier, and the power amplifier module amplifies the dispersed audio to a suitable amplitude to drive the high-pitch driver.

The embedded software of the power amplifier box mainly comprises two parts, namely DSP software and ARM control software. The DSP software mainly has the function of realizing audio signal processing in two input modes, namely player input and network input. The ARM control software mainly has the functions of task scheduling, IO control, state monitoring and state telemetering information return to a console.

The signal processing board module is used for receiving the control command and the audio input signal, switching according to different audio input modes and amplitudes, processing the audio signal, outputting the processed audio signal to the power amplifier module, amplifying the processed audio signal and then sending the amplified audio signal to the sound producing body for sound production; and state remote measuring information of the sounding body and the power amplifier module is collected at the same time, and the state remote measuring information is returned to the network control panel module, and is displayed or warned correspondingly on the upper computer.

The principle of the signal processing board module is as shown in fig. 20: the signal processing board module comprises a DSP minimum system, an ARM, a power supply conversion circuit, a reset circuit, a filtering amplification circuit, a temperature pre-processing circuit, a high pitch head on-off detection circuit and the like.

The audio signal of the invention is two paths of input signals, and in order to realize the consistency of output, normalization processing needs to be carried out on the two paths of input signals. The normalized signal is subjected to sound mixing, the signal after sound mixing is divided into two paths, and one path is subjected to dynamic range control, voice detection and frequency equalization; and the other path of energy detection is carried out, and the detection value is compared with the threshold value, is larger than the threshold value selection gain 1 and is smaller than the threshold value selection gain 2. The gain is multiplied by the signal output by the equalizer and then output.

And the vehicle system is used for integrally moving the driving device, and the loading system and the strong sound dissipation system are both arranged on the vehicle system.

The vehicle platform has 4 operation modes: remote control mode, tracking mode, pre-map mode, and autonomous driving mode. Wherein the autonomous mode of travel has the ability to proceed in a completely unsupported environment. The internal design of the vehicle platform adopts a multi-bus redundancy design, 3 physical buses are distributed at different positions of a vehicle body, parallel communication is realized, and the condition that the use of the system is not influenced by the failure or war loss of 2 buses is met.

The vehicle platform system comprises a chassis system, a vehicle body assembly and a wire control system. The chassis system comprises a brake pipeline unit, an air filtering unit, a heat dissipation unit, a suspension and speed ratio unit, a generator unit and the like. The wire control system comprises control units of direction, accelerator, brake, gear, parking, ignition and the like. The vehicle chassis is of a four wheel drive configuration and the system is comprised of a power source subassembly, a drive and control subassembly, a transmission subassembly, an appliance subassembly, a brake subassembly, a travel subassembly, a body subassembly, a appearance subassembly, and accessories. The chassis of the unmanned vehicle adopts a gasoline-electric hybrid wheeled light all-terrain vehicle chassis, and has the capability of maneuvering in special complex regions such as mountainous regions, jungles, beaches, deserts, snowfields, muddiness and the like.

The wire control system comprises a direction control unit, an accelerator control unit, a brake control unit, a gear control unit, a parking control unit and an ignition control unit.

The wireless control platform and the unmanned vehicle can be remotely controlled or automatically planned to run in a field wireless communication technology mode.

The robot directional driving-away system has four working modes of remote control, target tracking, map prefabrication and autonomous driving under the conditions of no carrying of operators and no supporting environment. The unmanned platform has the advantages that the integrated and modular design characteristics are adopted, the self defense system is perfect, the overall appearance is low RCS stealth, low infrared characteristics and anti-hijack design, and the unmanned platform can perform actions such as maneuver evasion, decoy interference and the like aiming at different threats. Can assist the fighter in the battlefield and form complementation with the fighter, greatly reduce the casualties in the battlefield and improve the battle efficiency of the army.

A robot directional driving-away method is carried out based on a robot directional driving-away system and comprises the following steps:

planning the path, shooting the target position, and performing patrol tasks.

The target is searched, detected, monitored and tracked all weather, the geographic position information of the target is inverted, and the target information is sent to the strong sound dispersion system and the command room in a wireless transmission mode.

1. All-weather searching: for example, people or animals are used as a search target, a vehicle system can autonomously run according to a specified route, a monitoring system can monitor and identify the surroundings, and after the target is identified, information can be uploaded through a communication system for further processing.

The scout mission area (about 100) was divided into roughly 4 25 sector areas and scout was developed from left to right for these areas one by one. In each 25-degree area, a zigzag operation mode is adopted, and scouting is carried out from far to near through a multispectral scouting means, so that information processing is carried out on the targets in the found and identified fan-shaped areas.

2. Detection and monitoring: the optical devices such as visible light, infrared and the like are arranged, and the optical aperture of the visible light meets the requirements that the effective pixel of the farthest detection target on the CCD target surface is larger than 6 multiplied by 6 pixels (detection) and 12 multiplied by 12 pixels (identification). The infrared optical aperture satisfies that the effective pixel of the farthest detection target on the infrared thermal imaging target surface is larger than 4 multiplied by 4 pixels (detection) and 8 multiplied by 8 pixels (identification).

3. Tracking: the existing mature tracker is used, a product-related target tracking algorithm is built in, an intelligent target transient loss recapture algorithm is combined, stable target tracking is achieved, a user is supported to display custom characters, a gate cross and tracking information are displayed, and the target-to-target deviation updating rate is 30Hz;

target to target offset delay < 15ms;

track speed 12 pixels/frame;

minimum target contrast 5%;

the tracking of moving objects was modeled with one MDP (fig. 17).

State of moving object: s ∈ S = S _ active ≦ S _ tracked ≦ S _ lost ≦ S _ inactive, and these several subspaces each contain an infinite number of target states. The recognized target firstly enters an active state, if the target is recognized by mistake, the target enters an inactive state, and otherwise, the target enters a tracked state. A target in the tracked state may enter the lost state, and a target in the lost state may return to the tracked state, or remain in the lost state, or enter the inactive state after a sufficiently long time.

The actions a ∈ A, all of which are deterministic.

S × A → S defines the change of the target state to S' under state S and action a.

The reward function R S A → R defines the immediate reward for reaching state S after action a, which is learned from training data.

The rule π S → A determines the effect a adopted in state S.

The state space variation of this MDP is as follows:

in the active state, an object candidate proposed by an object recognition algorithm passes through an offline-trained Support Vector Machine (SVM) to judge whether the next action is a _1 or a _2, and the input of the SVM is the feature vector, the spatial position size and the like of the candidate object, which determine the MDP rule pi _ active in S _ active.

In the tracked state, an object-line appearance model based on a tracking-learning-detection tracking algorithm is used to determine whether the target object remains in the tracker state or enters the lost state. This appearance model (appearance model) uses a rectangle (bounding box) where the target object is located in the current frame as a template (template), and all the object appearance templates collected in the tracked state are used in the lost state to determine whether the target object returns to the tracked state. In addition, in the tracked state, the tracking of the object uses the appearance model template, the Optical Flow in the rectangular range and the coincidence proportion of the candidate object and the target object provided by the object recognition algorithm to decide whether to keep in the tracked state, and if so, the appearance template of the target object is automatically updated.

In the lost state, if an object keeps the lost state exceeding a threshold frame number, the inactive state is entered; whether the object returns to the tracked state is determined by a classifier based on the similarity feature vectors of the target object and the candidate object, and corresponds to pi _ lost in S _ lost.

This MDP-based algorithm achieves the industry leader level in object tracking evaluation of KITTI datasets.

2. And (3) inversion of target position information: the target geographic position information (longitude and latitude and altitude) obtained by the GPS can be deduced through the conversion of geodetic longitude and latitude coordinates and Gaussian projection coordinates.

And after receiving the target information, the strong sound dispersing system directly turns to the target position, and performs strong sound dispersion on the target by a non-fatal means.

The wide dynamic algorithm technology is adopted to shoot the target position, and the wide dynamic technology is a technology which enables a camera to see the image clearly under the condition that the brightness conversion range is very large. The method comprises the steps of framing a scene which simultaneously comprises a highlight area and a backlight area, wherein an output image becomes a white area due to overexposure in the highlight area, and becomes a black area due to underexposure in the dark area, so that the scenes in the areas cannot be seen clearly, and the image quality cannot meet the requirements of practical application at all. Compared with the common image, the high dynamic range image has richer expressed picture levels and more vivid and realistic light and shadow effect. In order to enable the image to cover a wider dynamic range and effectively improve the imaging quality of a scene, the invention adopts a wide dynamic imaging method based on a vision system.

The wide dynamic algorithm technique includes the following steps:

s1, extracting a brightness component of an image to be processed;

s2, calculating an environment factor of each pixel point by using the pixel field information of the brightness component;

s3, obtaining a wide dynamic transformation formula by utilizing the environmental factors according to the characteristics of a human visual system;

s4, obtaining a wide dynamic range image by using a wide dynamic transformation formula;

and S5, gamma correction is performed, so that the contrast of the image is improved.

The method for tracking the target is based on the contour tracking technology and comprises the following steps: the contour of the object is tracked in the binary edge image of the succeeding frame using the contour of the boundary of the object as a template, and the contour can be automatically and continuously updated. Compared with the tracking method based on the region, the contour tracking method can achieve better matching effect basically under the condition of small calculation amount, and if reasonably separating each moving object and realizing contour initialization are started, the tracking can be continuously carried out even under the condition that partial occlusion exists.

In order to meet the requirements of different sound pressure levels, the loudspeaker unit and the power amplifier unit in the strong sound dispersion system are modularly designed, each sound source is a loudspeaker module and a power amplifier module, and the sound sources can work independently and can be quickly and freely combined to form a sounder array: a combination of a 1 × 2 array, a 2 × 3 array, a 3 × 4 array, etc., as shown in fig. 7 to 10.

Each module is provided with a signal input end, audio signals are input and then amplified through the power amplification module, and then the amplified signals are output through the loudspeaker module, namely signal input → signal amplification → signal output. When the formed sounding array works independently, audio signals are independently input to a single module, and the signal input of each module is connected in parallel for quick combination, so that the use of multiple modules can be realized.

Each module is provided with a signal input end, audio signals are input and then amplified through the power amplification module, and then the amplified signals are output through the loudspeaker module, namely signal input → signal amplification → signal output. When the formed sounding array works independently, the audio signals are independently input to a single module, and the signal inputs of all the modules are connected in parallel for quick combination, so that the use of multiple modules can be realized.

The strong sound dispersion system is designed by adopting a full digital power amplifier multi-channel dynamic balance technology: the full digital power amplifier with multiple channels, high power and high fidelity can perform dynamic load balance on each channel by applying a real-time protection technology. The phase of a multi-channel sound source signal is controlled to be consistent, a multi-channel high-efficiency power amplifier array is designed based on the class D audio amplification technology, a single-stage power amplifier is adopted, a power supply and a power amplifier are directly combined, alternating current directly supplies power to the power amplifier stage through a rectifier module, and the efficiency is higher. And the later stage adopts an electric energy recovery device to recover the invalid power of the inductive load, thereby further improving the efficiency. Can meet the requirements of power drive and heat dissipation under severe conditions.

Claims (9)

1. A robotic directional drive-off system, comprising:

the uploading system comprises a video monitoring system, an integrated navigation system, a communication control system and a photoelectric tower; the video monitoring system is used for shooting a target position; the integrated navigation system is used for planning a path; the communication control system is used for transmitting shot target information; the photoelectric tower is used for searching, detecting, monitoring and tracking the target and reflecting the geographical position information of the target;

the strong sound dispersing system is used for directly steering the received target information to a target position and carrying out strong sound dispersing on the target;

the vehicle system is used for driving the whole device to move, and the loading system and the strong sound dispersion system are both arranged on the vehicle system;

and the wireless control platform is in wireless communication with the vehicle system and is used for remotely controlling the vehicle system.

2. A robotic directional drive-away system according to claim 1, wherein the megasonic drive-away system comprises:

a microphone unit for outputting sound waves of high decibels;

the power amplifier unit is used for providing a preprocessed and amplified audio signal for the loudspeaker unit;

the holder is used for driving the loudspeaker unit to rotate in pitching and azimuth directions;

and the multi-language module is used for providing a multi-language offline translation function.

3. The directional robot driving-away system according to claim 2, wherein the power amplifier unit comprises a signal processing module, a network control module, a filtering power supply module and a power amplifier.

4. A robotic directional drive-off system according to claim 2, wherein the head comprises:

the mechanical table body is used for pitching and rotating in direction;

the control system is used for controlling the action of the mechanical table body;

a load support.

5. A robotic directional drive-off system according to claim 1, wherein the mounting system further comprises:

the power supply control system is used for providing electric energy for the upper mounting system;

the environment control system is used for controlling the environment around the vehicle body;

a lighting system.

6. A robot directional driving-away method, characterized in that the method is based on a robot directional driving-away system of any one of claims 1 to 5, and comprises the following steps:

planning a path, shooting a target position, and performing a patrol task;

the method comprises the steps of carrying out all-weather searching, investigation, monitoring and tracking on a target, inverting the geographic position information of the target, and sending the target information to a strong sound dispersion system and a command room in a wireless transmission mode;

and after receiving the target information, the strong sound dispelling system directly turns to the target position and carries out strong sound dispelling on the target by a non-fatal means.

7. A robot directional driving-off method according to claim 6, characterized in that a wide dynamic algorithm technology is adopted to shoot the target position, and the wide dynamic algorithm technology comprises the following steps:

s1, extracting a brightness component of an image to be processed;

s2, calculating the environmental factor of each pixel point by using the pixel field information of the brightness component;

s3, obtaining a wide dynamic transformation formula by utilizing the environmental factors according to the characteristics of a human visual system;

s4, obtaining a wide dynamic range image by using a wide dynamic transformation formula;

and S5, gamma correction.

8. The method of claim 6, wherein the microphone unit and the power amplifier unit of the loud sound dissipation system are designed in a modular manner, and each sound source is a microphone module and a power amplifier module, which can work independently and can be combined freely and rapidly to form a sounder array.

9. The directional robot driving method according to claim 6, wherein the megasonic driving system is designed by using a full digital power amplifier multi-channel dynamic balancing technology.

Images