CN104656664A - Vehicle-mounted multi-rotor unmanned helicopter landing guidance control system and guidance control method - Google Patents

Abstract

The invention discloses a vehicle-mounted multi-rotor unmanned helicopter landing guidance control system and a guidance control method. The control system comprises a vehicle-mounted guidance subsystem and an airborne control subsystem, wherein the vehicle-mounted guidance subsystem mainly comprises two visual processing systems and a navigational computer. On the basis of the guidance control system and with the adoption of a binocular vision guidance method, a multi-rotor unmanned helicopter is accurately positioned according to data resolving of the two visual processing systems, a flying control computer controls the multi-rotor unmanned helicopter to land in the target position of a vehicle, the practicability of the multi-rotor unmanned helicopter is improved, and the application range of the multi-rotor unmanned helicopter is increased.

Description

Vehicle-mounted many rotor unmanned helicopters approaching and landing system control system and Guidance control method

Technical field

The invention discloses a kind of vehicle-mounted many rotor unmanned helicopters approaching and landing system control system and Guidance control method, belong to the autonomous flight control technical field of telecontrolled model (unmanned) multirotor helicopter.

Background technology

Many rotor unmanned helicopters be a kind of structure simple, be easy to manipulate, can the stable unmanned vehicle of vertical takeoff and landing, floating state.Many rotor unmanned helicopters roughly can be divided into according to support arm number difference: three axles, four axles, six axles and eight axles, and except three-axis structure, each support arm of other structures can adopt individual layer rotor-hub configuration or double-deck rotor-hub configuration.The support arm number that multi-rotor aerocraft is different and rotor quantity, can realize different load capacity.Many rotor unmanned helicopters have high controllability, maneuverability and stability, and have the features such as low noise, pollution-free, easy to carry, safety hazard is little, are very suitable for the aerial mission performing short-distance and medium-distance.It all has broad application prospects in military and civilian field, as reconnaissance and surveillance, communication relaying, searching rescue, target following, electric power overhaul, imaging etc. of taking photo by plane.

The independent landing mode of current many rotor unmanned helicopters adopts the mode of accuracy mostly.This causes many rotor unmanned helicopters can only land on ground or fixing immovable object, thus greatly limit way of recycling and the range of application of many rotor unmanned helicopters.

Therefore, realize technology that many rotor unmanned helicopters land on moving target in the urgent need to.

Summary of the invention

Goal of the invention: for above-mentioned prior art, proposes a kind of vehicle-mounted many rotor unmanned helicopters approaching and landing system control system and Guidance control method, for solving the engineering problem of many rotor unmanned helicopters independent landing on moving target automobile.

Technical scheme: a kind of vehicle-mounted many rotor unmanned helicopters approaching and landing system control system, comprises vehicle-mounted guiding subsystem and airborne control subsystem;

Described vehicle-mounted guiding subsystem comprises two cover vision processing systems, navigational computer, the first wireless data transmission module; Wherein, the described vision processing system that often overlaps comprises camera, diaxon The Cloud Terrace, two stepper motor drivers and data processing chip; Often overlapping in vision processing system, camera is arranged on diaxon The Cloud Terrace, and connects navigational computer; Two motors of diaxon The Cloud Terrace connect a stepper motor driver respectively; Described data processing chip connects navigational computer, connects two stepper motor drivers simultaneously; Described two cover vision processing systems are arranged on the both sides of target area that vehicle lands respectively;

Described airborne control subsystem comprises many rotor flyings platform, flight controller, the second wireless data transmission module; Wherein, described flight controller comprises flight-control computer, and connects the sensor assembly of described flight-control computer; Described sensor assembly comprises ultrasonic height meter, six axle attitude sensor and electronic compasss; Described many rotor flyings platform comprises symmetrical frame and topworks, and the control end of described topworks connects flight-control computer; Described second wireless data transmission module connects flight-control computer, and can connect the first wireless data transmission module by wireless link.

The Guidance control method of vehicle-mounted many rotor unmanned helicopters approaching and landing system control system, comprises the steps:

Step (1), the image transmitting of captured in real-time is constantly given the navigational computer of vehicle-mounted guiding subsystem by two cameras of vehicle-mounted guiding subsystem, navigational computer carries out identifying processing to image, after the many rotor unmanned helicopters being pasted with distinguishing mark image enter camera view, namely navigational computer recognizes many rotor unmanned helicopters;

Step (2), described navigational computer calculates the relative position relation of many rotor unmanned helicopters and each camera based on technique of binocular stereoscopic vision, then constantly adjust the corner of two diaxon The Cloud Terraces, make many rotor unmanned helicopters always be positioned at the center of each camera view;

Step (3), always be positioned at the prerequisite at the center of camera view at many rotor unmanned helicopters under, the number of revolutions count value of each diaxon The Cloud Terrace two motors that navigational computer obtains according to vision processing system, calculates the relative position (x of many rotor unmanned helicopters relative to fiducial cameras ₀, y ₀, z ₀), be specially:

If the camera wherein in a set of vision processing system is initial point O, the camera in another set of vision processing system is an O ₁, with OO ₁for x-axis, straight up for y-axis sets up right hand rectangular coordinate system, if the The Cloud Terrace crab angle of initial point O place camera is α ₁, the angle of pitch is β ₁, some O ₁the The Cloud Terrace crab angle of place's camera is α ₂, the angle of pitch is β ₂, OO ₁=s, then obtain:

$\left\{\begin{array}{c}{x}_0=\frac{s\tan {\mathrm{\α}}_1}{\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2}\\ y_0=\frac{s\tan {\mathrm{\β}}_1}{\cos {\mathrm{\α}}_1(\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2)}\\ z_0=-\frac{s}{\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2}\end{array}\right.$ (formula 1)

According to described positional information (x ₀, y ₀, z ₀) obtain unmanned plane relative vehicle landing point positional information (x, y, z):

$\left\{\begin{array}{c}x=x_0-\frac{s}{2}=\frac{s\tan {\mathrm{\α}}_1}{\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2}-\frac{s}{2}\\ y=y_0=\frac{s\tan {\mathrm{\β}}_1}{\cos {\mathrm{\α}}_1(\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2)}\\ z=z_0=-\frac{s}{\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2}\end{array}\right.$ (formula 2)

Then described positional information (x, y, z) is sent to the airborne control subsystem of unmanned plane by radio transmission apparatus;

Step (4), the airborne control subsystem of many rotor unmanned helicopters receives the positional information (x of described unmanned plane relative vehicle landing point, y, z) after, control many rotor unmanned helicopters by flight-control computer and drop to target area on vehicle, comprise the steps:

(41), the detection data of the sensor assembly Real-time Collection ultrasonic height meter of many rotor unmanned helicopters, six axle attitude sensors and electronic compass, flight-control computer resolves the altitude information, the attitude data that obtain many rotor unmanned helicopters, and according to described positional information (x, y, z) real time data resolve the speed data obtaining many rotor unmanned helicopters relative vehicle;

(42), the speed data of described altitude information, attitude data, station-keeping data and relative vehicle and expected data are compared, obtains error information;

(43), calculate the controlled quentity controlled variable of topworks according to described error information and unmanned aerial vehicle (UAV) control rule, and be sent to the topworks of many rotor unmanned helicopters; Described unmanned aerial vehicle (UAV) control rule comprises attitude control law, height-holding control law, horizontal direction control law; Wherein, described attitude control law comprises angle of pitch control rate, roll angle control rate, crab angle control rate;

Described height-holding control law is:

${\mathrm{\Δ\δ}}_c=K_{p_z}\mathrm{\Δz}+K_{p_z}^{\′}\stackrel{\·}{z}+K_{p_z}^{\′\′}\stackrel{\·\·}{z}+K_{i_z}^{\′\′}\∫\stackrel{\·\·}{z}+K_{d_z}^{\′\′}{\stackrel{\·\·}{z}}^{\′}$

Wherein, Δ z is height error data, short transverse speed, short transverse acceleration, Δ δ _cfor altitude channel controlled quentity controlled variable, it is right to represent differentiate, with it is controling parameters;

Many rotor unmanned helicopters horizontal x direction Landing Control is that described angle of pitch control rate is inner looping with the tracing control in horizontal x direction rule for external loop; Wherein, the tracing control rule in described horizontal x direction is:

$\left\{\begin{array}{cc}{\mathrm{\Δ\δ}}_{\mathrm{\θ}}=K_{p_x}\mathrm{\Δx}+K_{i_x}\∫\mathrm{\Δx}+K_{d_x}{\mathrm{\Δx}}^{\′}+K_{P_x}^{\′}\stackrel{\·}{x},& \left|\mathrm{\Δx}\right|\≤e\\ {\mathrm{\Δ\δ}}_{\mathrm{\θ}}=K_{p_x}\mathrm{\Δx}+K_{d_x}\mathrm{\Δx}+K_{P_x}^{\′}\stackrel{\·}{x},& \left|\mathrm{\Δx}\right|>e\end{array}\right.$

Many rotor unmanned helicopters horizontal y direction Landing Control is that described roll angle control rate is inner looping with the tracing control in horizontal y direction rule for external loop; Wherein, the tracing control rule in described horizontal y direction is:

$\left\{\begin{array}{cc}{\mathrm{\Δ\δ}}_{\mathrm{\φ}}=K_{p_y}\mathrm{\Δy}+K_{i_y}\∫\mathrm{\Δy}+k_{d_y}{\mathrm{\Δy}}^{\′}+K_{p_y}^{\′}\stackrel{\·}{y},& \left|\mathrm{\Δy}\right|\≤e\\ {\mathrm{\Δ\δ}}_{\mathrm{\φ}}=K_{p_y}\mathrm{\Δy}+K_{d_y}\mathrm{\Δ}{y}^{\′}+K_{p_y}^{\′}\stackrel{\·}{y},& \left|\mathrm{\Δy}\right|>e\end{array}\right.$

Wherein, Δ δ _θ, Δ δ _φbe respectively x, y direction controlling amount, be respectively the speed data of x, y direction relative vehicle, the site error that Δ x, Δ x ' are x-axis direction and error change rule, the site error that Δ y, Δ y ' are y-axis direction and error change rule, with the controling parameters in x-axis direction, with be the controling parameters in y-axis direction, e is the fixed error value of setting.

Beneficial effect: (1) adopts the guidance method of binocular vision, resolved by the data of two cameras, the accurate location to many rotor unmanned helicopters can be realized, and then its landing above the mobile devices such as automobile can be realized, improve practicality and the range of application of many rotor unmanned helicopters.

(2) this control method adopts segmentation PID control method, achieves the degree of accuracy that many rotor unmanned helicopters follow moving target.The experimental result of final reality shows, by this control method, the tracking error of many rotor unmanned helicopters can ensure within the scope of 5cm.

(3) control mode that can adopt Non-follow control, automatically control and manually automatically combine, makes, when control is broken down automatically, to adopt Non-follow control, avoid the uncontrollable state of many rotor unmanned helicopters.

Accompanying drawing explanation

Fig. 1 is present system hardware block diagram;

Fig. 2 is software control FB(flow block) of the present invention;

Fig. 3 is binocular vision disposal system schematic diagram of the present invention;

Fig. 4 is the present invention's many rotor unmanned helicopters control system block diagram.

Embodiment

Below in conjunction with accompanying drawing the present invention done and further explain.

As shown in Figure 1, a kind of vehicle-mounted many rotor unmanned helicopters approaching and landing system control system, comprises vehicle-mounted guiding subsystem and airborne control subsystem.Vehicle-mounted guiding subsystem comprises two cover vision processing systems, navigational computer, the first wireless data transmission module, Switching Power Supply and telepilot.

Wherein, often overlap vision processing system and comprise camera, diaxon The Cloud Terrace, two stepper motor drivers and data processing chip.Camera: select USB camera in the present embodiment, hardware pixel is 100W720p, and angle is 170 °, and frame rate is 30fps/1280 × 720, and the storage format after Dynamical capture is AVI/WMV.The resolution of this camera can take into account image definition and image processing speed.

Diaxon The Cloud Terrace: diaxon The Cloud Terrace is by two step motor control, and a rotation of following many rotor unmanned helicopters yaw direction, is referred to as yaw motor; Another one follows the rotation of many rotor unmanned helicopters pitch orientation, is referred to as pitching motor.Stepper motor is opened loop control element electric impulse signal being changed into angular displacement or displacement of the lines.When non-overload, the rotating speed of motor, the position of stopping only depending on frequency and the umber of pulse of pulse signal, and not by the impact of load change, namely add a pulse signal to motor, motor then turns over a step angle.By size and the umber of pulse of recording pulse signal, the angle that motor turns over, the crab angle namely corresponding to The Cloud Terrace or the angle of pitch can be determined.Stepper motor has not by external interference, the error features such as long term accumulation, control performance be not good.

Stepper motor driver: stepping motor directly can not be received in direct current or AC power and work, and must use special driving power, i.e. stepping motor driver.Stepper motor driver is a kind of is the topworks of angular displacement by electroporation.When step actuator receives a pulse signal, it rotates a fixing angle with regard to Driving Stepping Motor by the direction of setting, and its rotation runs step by step with fixing angle.Pilot angle displacement can be carried out by gating pulse number, thus reach the object of accurately location.In the present embodiment, drive the stepping motor driver of yaw motor to be referred to as yaw drive, drive the stepping motor driver of pitching motor to be referred to as pitch drive.

Data processing chip: in the present embodiment, data processing chip uses and adopts ATmega328p-AU to be the Arduino Mini Pro microcontroller circuit plate of core.It has 14 digital I/O pins, 8 analog inputs, 1 pair of Transistor-Transistor Logic level serial ports transmitting-receiving port, and the resonator of a 16MHz, a reset button, supports that serial ports is downloaded, support external 3.3V ~ 12V DC power supply.ATmega328p-AU is the single-chip microcomputer with 32Kb flash memory that American ATMEL is produced.Utilize this data processing chip can complete the conversion of computerized information to pwm control signal, and implementation method is simple to operation.

Switching Power Supply is a kind of high frequency device for converting electric energy, and the present invention selects the specification of 36V, 8.3A, is used for powering to four stepper motor drivers.It has the features such as volume is little, lightweight, efficiency is high, voltage stabilized range is wide.

First wireless data transmission module passes mutually for the data realized between the navigational computer that is positioned on vehicle and the flight control system being positioned at aerial UAV flight, and its communication range determines unmanned plane and can to collect mail the scope of breath.In the present embodiment, the first wireless data transmission module selects 3DR Radio Telemetry digital transmission module, and transmission frequency is 433MHz, and maximum transmission distance is 700m, uses UART interface to export data.

This vehicle-mounted many rotor unmanned helicopters approaching and landing system control system can be applicable to many rotor unmanned helicopters such as four axles, six axles and eight axles, vehicle-mounted guiding subsystem is installed on landing target vehicle, camera often in cover vision processing system and The Cloud Terrace are installed on the side of roof landing target area, the line of centres of two cameras is vertical with the length direction of vehicle, other parts, navigational computer and the Switching Power Supply for powering in two cover vision processing systems are positioned over vehicle interior, and the astronomical cycle of the first wireless data transmission module is outside in car body.

Often overlapping in vision processing system, camera is arranged on the top of diaxon The Cloud Terrace, and is connected with navigational computer by UART serial ports.The signal control end of two motors of diaxon The Cloud Terrace connects a stepper motor driver respectively.Data processing chip is connected with the input port of two stepper motor drivers; Meanwhile, data processing chip is connected with navigational computer by UART serial ports.First wireless data transmission module is connected with navigational computer by UART serial ports.

Airborne control subsystem comprises many rotor flyings platform, flight controller, the second wireless data transmission module.Wherein, flight controller comprises flight-control computer, sensor assembly and power module.Sensor assembly comprises ultrasonic height meter, six axle attitude sensor and electronic compasss.Many rotor flyings platform comprises symmetrical frame and topworks, and the control end of topworks connects flight-control computer.

The power device of multi-rotor aerocraft comprises brushless electric machine, electricity tune, screw propeller and electrical source of power, and wherein electric mechanical, electrical mediation screw propeller is referred to as topworks again.Power device determines maximum load ability, cruising time and the maneuverability of many rotors.The present embodiment adopts the brushless electric machine of T-motor 4014 model, the carbon fiber spiral oar of collocation 10000mAh 25C 22.2V lithium battery and 1555 models.This brushless electric machine arrange in pairs or groups 1555 type screw propellers and 22.2V lithium battery time, 100% throttle place corresponding current is 29.9A, thus electricity tune make good use of be full of company produce 30A electricity adjust.Based on above power device type selecting, this unmanned plane lifting capacity >=4.1kg, cruising time >=21min.

STM32F407 microcontroller selected by the flight-control computer of the present embodiment, and this microcontroller is produced by STMicroelectronics company, is 32 ARMCortex _-m4F kernel, 1MB Flash, 192KB RAM, LQFP100 encapsulate.STM32F407 has 140 railway digital I/O, 2 road UART, 3 road SPI, 3 road I2C interfaces and 2 tunnel CAN interfaces.It can be used as flight to control computer for controlling, its core frequency and peripheral hardware resource can be made full use of.

Flight-control computer mainly completes: collecting sensor signal process, control law are resolved, motor drives, land station communicates, remote information receives.Flight-control computer is control core, and all work completed constitutes the closed-loop control of band feedback.Its processes sensor information, for flight control system provides feed back input; Process steering order manually or automatically, adopt different control modes according to concrete instruction; Drive motor action.STM32F407 microcontroller demonstrates the advantages such as perfect in shape and function, speed is fast, exploitation is convenient with powerful order set and interface function, effectively can solve at a high speed and miniature contradiction, thus integrate the novel flight control modules of high speed, high precision and miniaturization for the design of unmanned plane amount body.

In sensor assembly, six axle attitude sensors: the MPU-6000 six axle attitude sensor selecting American I nvensense company to produce, it incorporates 3 axle gyroscopes, 3 axle accelerators, is the whole world the first conformability 6 axle motion process assembly.MPU-6000 exports the fusion calculation data of the rotation matrix of 6 axles or 9 axles, hypercomplex number, Eulerian angle form with numeral.Compared to multicompartment scheme, MPU-6000 eliminates the problem of between centers difference when combining gyroscope and accelerator, decreases a large amount of packaging spaces, has low-power consumption, low cost, high performance feature.The sensing range of angular velocity is ± 250 °, ± 500 °, ± 1000 ° and ± 2000 °/sec (dsp), can accurately follow the trail of fast and action at a slow speed; The sensing range of programmable accelerometer is ± 2g, ± 4g, ± 8g and ± 16g simultaneously.The measurement data of sensor finally exports by the IC bus of the highest 400kHz or the spi bus of the highest 20MHz.

Electronic compass: the HMC5883 electronic compass selecting Honeywell Corp. USA to produce, it is a kind of surface-pasted high integration module, and with the weak magnetic sensor chip of digital interface, is applied to low cost compass and magnetic field detection field.Compass precision controlling can be made at 12 analog to digital converters of 1 ° ~ 2 °, and with 16 pins, interface is easy I2C serial bus, is of a size of 3.0 × 3.0 × 0.9mm.Having in axial high sensitivity and linear high-precision feature, is the sensor that sensitivity is the highest and reliability is best.

Ultrasonic sensor: select the LV-MaxSonar-EZ1 sonic sensor that Pololu company of the U.S. produces.This ultrasonic distance measuring module has low in energy consumption, that precision is high, detection range is far away feature, and its sensing range is 0 to 6.45M (21.2ft), resolution 2.5cm, non-blind area.This sensor provides three output interfaces: the output of digit duration, analog signal output, and asynchronous serial exports, and uses more convenient.

Second wireless data transmission module arranges identical with the first wireless data transmission equipment of vehicle mounted guidance part, and both complete the transmission of information by the communication protocol defined.In the present embodiment, the second wireless data transmission module selects 3DR Radio Telemetry digital transmission module, and transmission frequency is 433MHz, and maximum transmission distance is 700m, uses UART interface to export data.

In the present embodiment flight control use two pieces independently lithium battery power, use flying power that the lithium battery of 10000mAh, 25C, 22.2V is unmanned plane to power respectively and the lithium battery of 900mAh, 15C, 7.4V is powered for flight-control computer.Wherein, the power supply of flight-control computer need carry out step-down, provides these two kinds of direct supplys of 5V, 3.3V to meet the demand of main control chip.The UBEC-3A using Shenzhen Hao Ying scientific & technical corporation to produce realizes the high step-down of 7.4V → 5V, and 5V is as the VCC of digital circuit.UBEC is a kind of DC voltage-stabilizing system of Switching Power Supply mode, is operated in the switching frequency of 300KHz, and chip efficiency up to 92%, and provides overcurrent and overheat protective function.The MIC5219-3.3 using MICREL company of the U.S. to produce realizes the low pressure drop of 5V → 3.3V, and can reach very high efficiency, and cost is low, noise is low, and quiescent current is little, and circuit structure is simple.

Many rotor unmanned helicopters topworks is positioned on each end points of frame, is connected with flight-control computer by PWM output driving circuit.Six axle attitude sensors are connected with flight-control computer by spi bus level conversion.Electronic compass is connected with flight-control computer by I2C interface.Ultrasonic height meter is connected with flight-control computer by Transistor-Transistor Logic level interface.Second wireless data transmission module is connected with flight-control computer by UART serial ports.

In the present embodiment, the two kinds of modes that communicated of vehicle-mounted guiding subsystem and airborne control subsystem, mode one is communicated for being completed by the first wireless data transmission module and the second wireless data transmission module, and mode two be realize communicating by telepilot.First, second wireless data transmission module transmission frequency is 433MHz, and maximum transmission distance is 700m; First wireless data transmission module is called that Ground module (USB interface) connects navigational computer and uses on vehicle, and the first wireless data transmission module is called that Air module (serial ports) is carried for aircraft.In the present embodiment, FUTABA telepilot selected by telepilot, and the FUTABA telepilot of 2.4GHz is the general telepilots of a model plane that Japanese Futaba Denshi Kogyo Kabushiki Kaisha produces, supports the use with this brand receiver.Model plane operator can by stirring some driving levers on telepilot, and the diverse location residing for each driving lever corresponds to different strokes, can produce each passage remote control pwm signal with distinct pulse widths.

The system of being illustrated in figure 2 realizes the overall flow figure of many rotor unmanned helicopters at mobile roof Autonomous landing.Vehicle-mounted guiding subsystem mainly comprises 3 functional modules: (1) picture recognition module; (2) target tracking module; (3) position data resolves module.Airborne control subsystem mainly comprises 4 functional modules: (1) data acquisition module; (2) data resolve module; (3) control law resolves module; (4) motor control module.In fig. 2, be the control flow of vehicle-mounted guiding subsystem in the square frame of left-half, mainly complete to catch identifying target, tracking target, calculating deflection angle, calculating range deviation; Be the software control flow process of airborne control subsystem in the square frame of right half part, its performance period is 10ms, completes digital independent, data are resolved, calculates PID output quantity, arranges the tasks such as PWM output.

Vehicle-mounted guiding subsystem realizes many rotor unmanned helicopters descent and comprises the steps:

Step (11), image recognition: the image transmitting of captured in real-time is constantly given the navigational computer of vehicle-mounted guiding subsystem by two cameras of vehicle-mounted guiding subsystem, and navigational computer carries out identifying processing to image.After the many rotor unmanned helicopters being pasted with distinguishing mark image enter camera view, navigational computer can recognize many rotor unmanned helicopters.

Image recognition processing realizes in navigational computer, development environment is windows7 system, develop software as Microsoft Visual Studio 2010, use the storehouse OpenCV (Open Source Computer Vision Library) that increases income to develop.Wherein, OpenCV is a cross-platform computer vision library of issuing based on (increasing income), its lightweight and efficient---be made up of a series of C function and a small amount of C++ class, provide the interface of the language such as Python, Ruby, MATLAB simultaneously, achieve a lot of general-purpose algorithms of image procossing and computer vision aspect.

In the present embodiment, identify that the method for unmanned plane on its fuselage, pastes color for " flag is red " (RGB=176,22,4) and the fixing mark of shape.Camera passes the image of shooting back, and navigational computer constantly judges whether occur " flag is red " in each two field picture; When color meets default " flag is red ", then the profile of this part is mated; When this profile mates with preset shape, then can think that this target is unmanned plane, realizes the seizure to UAV targets and identification.According to unmanned plane position in the picture, resolve the relative position relation obtaining unmanned plane and each camera according to technique of binocular stereoscopic vision, and this information is sent in data processing chip Arduino mini pro by UART interface.

Step (12), target following: two cover vision processing systems of vehicle-mounted guiding subsystem are respectively according to the camera of each system and the relative position of unmanned plane, the corner of its diaxon The Cloud Terrace of continuous adjustment, to ensure that unmanned plane is always positioned at the center of camera view.This process realizes in data processing chip Arduino mini pro, its development environment is windows7, develop software as Arduino IDE, this software is the computer program development environment supporting with Arduino, only need coding code in IDE, after program being uploaded to Arduino circuit board, can practical function demand.

Data processing chip Arduino mini pro obtains the relative position of camera and unmanned plane by UART interface, this relative position relation determines the yaw motor of diaxon The Cloud Terrace and turning to of pitching motor.When unmanned plane is not in camera center, Arduino mini pro, by four road I/O pins, sends pwm pulse signal and direction signal respectively to two motors; Wherein, pulse signal size is fixed, and direction signal is determined by relative position relation.Yaw motor often receives a pulse signal, and just according to direction signal, clockwise or be rotated counterclockwise a step angle, in like manner pitching motor also can according to signal rotation.When unmanned plane is in camera center, the four road I/O pins of Arduino mini pro no longer send signal, and two motors stop the rotation, thus achieve target tracking, ensure that unmanned plane is always positioned at the center of camera view.In addition, data processing chip Arduino mini pro to motor send signal time, the value of corresponding counter can increase progressively, and records the number of revolutions of two motors respectively, and timing pass navigational computer back by UART interface.

Step (13), position data is resolved: be always positioned at the prerequisite at the center of camera view at unmanned plane under, the number of revolutions count value of each diaxon The Cloud Terrace two motors that navigational computer obtains according to vision processing system, calculate the relative position of unmanned plane relative to fiducial cameras, and sent to the airborne control subsystem of unmanned plane by the first wireless data transmission module obtain the positional information of unmanned plane relative vehicle landing point according to this positional information after.

This partial routine realizes in navigational computer, and development environment is windows7 system, develops software as MicrosoftVisual Studio 2010.Be specially: set camera in wherein a set of vision processing system as initial point O, the camera in another set of vision processing system is a some O ₁, with OO ₁for x-axis, be y-axis straight up, set up right hand rectangular coordinate system M as shown in Figure 3.The number of revolutions count value of the motor that navigational computer is passed back by data processing chip Arduino mini pro, calculates the deflection angle of yaw motor and pitching motor in two cover vision processing systems; If the The Cloud Terrace crab angle of initial point place camera is α ₁, the angle of pitch is β ₁, O ₁the The Cloud Terrace crab angle of some place camera is α ₂, the angle of pitch is β ₂.Then at known OO ₁when=s, the coordinate (x in rectangular coordinate system being arranged in a unmanned plane at A place can be calculated ₀, y ₀, z ₀):

$\left\{\begin{array}{c}{x}_0=\frac{s\tan {\mathrm{\α}}_1}{\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2}\\ y_0=\frac{s\tan {\mathrm{\β}}_1}{\cos {\mathrm{\α}}_1(\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2)}\\ z_0=-\frac{s}{\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2}\end{array}\right.$ (formula 1)

The unmanned plane obtained according to formula 1 is relative to the station-keeping data of fiducial cameras.If the level point of unmanned plane is OO ₁mid point, then with this mid point for initial point, set up rectangular coordinate system N, three axles that its three axle is corresponding with M coordinate system are respectively parallel and direction is identical, then under rectangular coordinate system N, unmanned plane relative vehicle landing point positional information to be (x, y, z) be:

$\left\{\begin{array}{c}x=x_0-\frac{s}{2}=\frac{s\tan {\mathrm{\α}}_1}{\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2}-\frac{s}{2}\\ y=y_0=\frac{s\tan {\mathrm{\β}}_1}{\cos {\mathrm{\α}}_1(\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2)}\\ z=z_0=-\frac{s}{\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2}\end{array}\right.$ (formula 2)

Positional information (x, y, z) data in formula 2 are sent to airborne control subsystem by the first wireless data transmission equipment by navigational computer.Airborne control subsystem realizes the landing of many rotor unmanned helicopters and comprises the steps:

Step (21), data acquisition: the sensor assembly Real-time Collection of unmanned plane obtains attitude data, the height detection data of unmanned plane, and the data collected are sent to flight-control computer; Meanwhile, airborne control subsystem receives the positional information of unmanned plane relative vehicle landing point for (x, y, z) by the second wireless data transmission equipment, this relative position information (x, y, z) i.e. tracking error data of each control rate.Wherein, this relative velocity data is the speed of relative automobile, calculates obtain according to relative position information.

Step (22), data are resolved: the sensing data received is calculated attitude data, altitude information by flight-control computer, then the speed data of many rotor unmanned helicopters relative vehicle is calculated according to the real time data of positional information (x, y, z); Then the speed data of the altitude information calculated, attitude data, station-keeping data and relative vehicle and expected data are compared, obtain error information.Wherein, expected data comprises position data, the altitude information of expectation, also has speed, the attitude data expected.

Step (23), control law resolves: the error information obtained according to step (22) and unmanned aerial vehicle (UAV) control rule calculate the controlled quentity controlled variable of topworks; Wherein, unmanned aerial vehicle (UAV) control rule comprises attitude control law, height-holding control law, horizontal direction control law and speeds control rate.

Be a under-actuated systems and unmanageable feature for unmanned plane, adopt classical PID controller to control multirotor helicopter, control system block diagram as shown in Figure 4:

(1) attitude control law

The control law of unmanned plane three axis attitude keeping system is all fed back by attitude angular rate and angular displacement and forms, and improves the stability of unmanned plane by this electrical feedback mode.Concerning vertical/horizontal gesture stability and Heading control, attitude keeping system adopts classical PID to control.Attitude keeping system is in innermost ring, all needs to connect in whole flight course.

Attitude keeps the attitude angle reference value in loop to refer to the trim value of unmanned plane under hovering mode, and obtain by experiment, usually differing is decided to be zero; The actual measured value of attitude and attitude data are recorded by six airborne axle attitude sensors.The difference of actual value and reference value is error, the source that Error Feedback that Here it is controls; Loop is kept for attitude, attitude angular velocity and attitude angle displacement two errors can be obtained.

The attitude retentive control rule of unmanned plane pitch channel is:

${\mathrm{\Δ\δ}}_e=k_{p_{\mathrm{\θ}}}\mathrm{\Δ\θ}+k_{i_{\mathrm{\θ}}}\∫\mathrm{\Δ\θ}+k_{d_{\mathrm{\θ}}}{\mathrm{\Δ\θ}}^{\′}$

The attitude retentive control rule of unmanned plane roll passage is:

${\mathrm{\Δ\δ}}_a=k_{p_{\mathrm{\φ}}}\mathrm{\Δ\φ}+k_{i_{\mathrm{\φ}}}\∫\mathrm{\Δ\φ}+k_{d_{\mathrm{\φ}}}{\mathrm{\Δ\φ}}^{\′}$

The attitude retentive control rule of unmanned plane jaw channel is:

${\mathrm{\Δ\δ}}_r=k_{p_{\mathrm{\ψ}}}\mathrm{\Δ\ψ}+k_{i_{\mathrm{\ψ}}}\∫\mathrm{\Δ\ψ}+k_{d_{\mathrm{\ψ}}}{\mathrm{\Δ\ψ}}^{\′}$

Wherein, Δ δ _e, Δ δ _a, Δ δ _rbe respectively the controlled quentity controlled variable of pitching, roll, yaw direction, Δ θ and Δ θ ' is the error of angle of pitch error and rate of pitch respectively, Δ φ and Δ φ ' is the error of roll angle error and roll angle speed respectively, and Δ ψ and Δ ψ ' is the error of crab angle error and yaw rate respectively with be the coefficient of angle of pitch displacement, angle of pitch displacement anomalous integral pitch rate, with be the coefficient of roll angle displacement, roll angle displacement anomalous integral rolling angle rate, with be the coefficient of yaw displacement, yaw displacement anomalous integral yawrate.

Restrain for the attitude retentive control of pitch channel, with constitute a kind of PID control law.Δ δ _ebe converted into the pwm signal of corresponding ratio, act on aviation mechanism.When unmanned plane is bowed, corresponding control law can make two, front motor speed increase generation nose-up pitching moment, allows unmanned plane come back, and to offset the trend of bowing, make unmanned plane keep balance, vice versa. with the acquisition of three parameters takes the scheme of " first according to modelling, test adjustment again " to debug out, and object makes the fast response time of unmanned plane, overshoot little.

Experiment proves, this control program is simple, Be very effective, the ability of the anti-interference in air flow that unmanned plane is possessed is certain.

(2) height-holding control law

Change the method for height, namely change motor speed, thus improve lift.Altitude control is the size being adjusted motor speed by the difference DELTA z of the positional information z under many rotor unmanned helicopters rectangular coordinate system N and Desired Height 0.Altitude control adopts classical PID to control.

The height of landing mission is the command signal on class slope, and the control law of altitude channel is:

${\mathrm{\Δ\δ}}_c=K_{p_z}\mathrm{\Δz}+K_{p_z}^{\′}\stackrel{\·}{z}+K_{p_z}^{\′\′}\stackrel{\·\·}{z}+K_{i_z}^{\′\′}\∫\stackrel{\·\·}{z}+K_{d_z}^{\′\′}{\stackrel{\·\·}{z}}^{\′}$

Wherein, Δ z is height error, short transverse speed, short transverse acceleration, Δ δ _cfor altitude channel controlled quentity controlled variable, it is right to represent differentiate, with be controling parameters, constitute PID control law.The output Δ δ of control law _cbe superimposed upon on motor, like this, when unmanned plane height is less than Desired Height, motor speed can strengthen, lift-rising; When unmanned plane height exceedes Desired Height, motor speed reduces, and produces a kind of degenerative automatic control effect.The scheme that these three controling parameters also adopt " first according to modelling, test adjustment again " debugs out.

(3) horizontal direction control law

Position control and unmanned plane TRAJECTORY CONTROL.TRAJECTORY CONTROL mainly constantly revises the position of unmanned plane to ensure that unmanned plane drops to the landing point on vehicle exactly by the positional information of vehicle-mounted guiding subsystem feedback.As shown in Figure 4, pitch channel and roll channel control position deviation and P to control rate information to feed back to inner looping respectively by PID.

Many rotor unmanned helicopters horizontal x direction Landing Control is with the tracing control in horizontal x direction rule for external loop, with angle of pitch control rate for inner looping; Wherein, the tracing control rule in horizontal x direction is:

$\left\{\begin{array}{cc}{\mathrm{\Δ\δ}}_{\mathrm{\θ}}=K_{p_x}\mathrm{\Δx}+K_{i_x}\∫\mathrm{\Δx}+K_{d_x}{\mathrm{\Δx}}^{\′}+K_{P_x}^{\′}\stackrel{\·}{x},& \left|\mathrm{\Δx}\right|\≤e\\ {\mathrm{\Δ\δ}}_{\mathrm{\θ}}=K_{p_x}\mathrm{\Δx}+K_{d_x}\mathrm{\Δx}+K_{P_x}^{\′}\stackrel{\·}{x},& \left|\mathrm{\Δx}\right|>e\end{array}\right.$

Many rotor unmanned helicopters horizontal y direction Landing Control is with the tracing control in horizontal y direction rule for external loop, with roll angle control rate for inner looping; Wherein, the tracing control rule in horizontal y direction is:

$\left\{\begin{array}{cc}{\mathrm{\Δ\δ}}_{\mathrm{\φ}}=K_{p_y}\mathrm{\Δy}+K_{i_y}\∫\mathrm{\Δy}+k_{d_y}{\mathrm{\Δy}}^{\′}+K_{p_y}^{\′}\stackrel{\·}{y},& \left|\mathrm{\Δy}\right|\≤e\\ {\mathrm{\Δ\δ}}_{\mathrm{\φ}}=K_{p_y}\mathrm{\Δy}+K_{d_y}\mathrm{\Δ}{y}^{\′}+K_{p_y}^{\′}\stackrel{\·}{y},& \left|\mathrm{\Δy}\right|>e\end{array}\right.$

Wherein, Δ δ _θ, Δ δ _φbe respectively x, y direction controlling amount, be respectively the speed data in x, y direction, the site error that Δ x, Δ x ' are x-axis direction and error change rule, the site error that Δ y, Δ y ' are y-axis direction and error change rule, with the controling parameters in x-axis direction, with it is the controling parameters in y-axis direction, Δ x equals the difference of the positional information x and 0 under many rotor unmanned helicopters rectangular coordinate system N, Δ y equals the difference of the positional information y and 0 under many rotor unmanned helicopters rectangular coordinate system N, e is a fixed error value, this value is determined by actual conditions, is generally 10cm.

For the tracing control rule of unmanned plane in horizontal x direction, with form PID control law.The output Δ δ of control law _θas the input of pitch channel inner looping.Like this, when unmanned plane is positioned at vehicle landing point rear, the angle of pitch expected is on the occasion of increase, acts on inner looping and strengthens unmanned plane forward flight speed; Otherwise when unmanned plane is positioned at vehicle landing point front, the angle of pitch expected is that negative value increases, and acts on inner looping and reduces unmanned plane forward flight speed.The scheme that these four controling parameters also adopt " first according to modelling, test adjustment again " debugs out.Experiment proves, this control program is simple, Be very effective.

Each loop is all adopt classical pid control algorithm from inside to outside above, need choose rational controling parameters, make the command signal of each channels track controller, improve its stability.The process of choosing of parameter is, first picks out each channel transfer function according to the result of modelling by mechanism and identification modeling, for the method successively design control law of single channel root locus; The parameter of design is transplanted to actual flying to go in the control algolithm of control plate, constantly tune is participated in the experiment and is flown, and finally obtains optimum parameter value.

Adjust ginseng through theoretical modeling and actual taking a flight test, achievement of the present invention reaches the object of expection.This method can make the many rotor unmanned helicopters of vehicle-mounted guiding subsystem tenacious tracking, for many rotor unmanned helicopters provide effective guidance information in the landing of vehicle landing point, there is stronger antijamming capability, solve the engineering problem of many rotor unmanned helicopters independent landing on moving target automobile.Meanwhile, design of hardware and software of the present invention can apply to, on multiple many rotor unmanned helicopters, to possess certain practicality and versatility.

In addition to the above steps, the manual control that telepilot realizes can also be added with.Remote-control receiver receives remote controller signal and transfers to flight-control computer, and flight-control computer controls topworks's work to exporting to respectively after the signal transacting received, and realizes Non-follow control unmanned plane and drops on roof.

The above is only the preferred embodiment of the present invention; it should be pointed out that for those skilled in the art, under the premise without departing from the principles of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.

Claims (2)

1. vehicle-mounted many rotor unmanned helicopters approaching and landing system control system, is characterized in that: comprise vehicle-mounted guiding subsystem and airborne control subsystem;

Described vehicle-mounted guiding subsystem comprises two cover vision processing systems, navigational computer, the first wireless data transmission module; Wherein, the described vision processing system that often overlaps comprises camera, diaxon The Cloud Terrace, two stepper motor drivers and data processing chip; Often overlapping in vision processing system, camera is arranged on diaxon The Cloud Terrace, and connects navigational computer; Two motors of diaxon The Cloud Terrace connect a stepper motor driver respectively; Described data processing chip connects navigational computer, connects two stepper motor drivers simultaneously; Described two cover vision processing systems are arranged on the both sides of target area that vehicle lands respectively;

Described airborne control subsystem comprises many rotor flyings platform, flight controller, the second wireless data transmission module; Wherein, described flight controller comprises flight-control computer, and connects the sensor assembly of described flight-control computer; Described sensor assembly comprises ultrasonic height meter, six axle attitude sensor and electronic compasss; Described many rotor flyings platform comprises symmetrical frame and topworks, and the control end of described topworks connects flight-control computer; Described second wireless data transmission module connects flight-control computer, and can connect the first wireless data transmission module by wireless link.

2. the Guidance control method of vehicle-mounted many rotor unmanned helicopters approaching and landing system control system as claimed in claim 1, is characterized in that, comprise the steps:

Step (1), the image transmitting of captured in real-time is constantly given the navigational computer of vehicle-mounted guiding subsystem by two cameras of vehicle-mounted guiding subsystem, navigational computer carries out identifying processing to image, after the many rotor unmanned helicopters being pasted with distinguishing mark image enter camera view, namely navigational computer recognizes many rotor unmanned helicopters;

Step (2), described navigational computer calculates the relative position relation of many rotor unmanned helicopters and each camera based on technique of binocular stereoscopic vision, then constantly adjust the corner of two diaxon The Cloud Terraces, make many rotor unmanned helicopters always be positioned at the center of each camera view;

Step (3), always be positioned at the prerequisite at the center of camera view at many rotor unmanned helicopters under, the number of revolutions count value of each diaxon The Cloud Terrace two motors that navigational computer obtains according to vision processing system, calculates the relative position (x of many rotor unmanned helicopters relative to fiducial cameras ₀, y ₀, z ₀), be specially:

If the camera wherein in a set of vision processing system is initial point O, the camera in another set of vision processing system is an O ₁, with OO ₁for x-axis, straight up for y-axis sets up right hand rectangular coordinate system, if the The Cloud Terrace crab angle of initial point O place camera is α ₁, the angle of pitch is β ₁, some O ₁the The Cloud Terrace crab angle of place's camera is α ₂, the angle of pitch is β ₂, OO ₁=s, then obtain:

$\left\{\begin{array}{c}{x}_0=\frac{s\tan {\mathrm{\α}}_1}{\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2}\\ y_0=\frac{s\tan {\mathrm{\β}}_1}{\cos {\mathrm{\α}}_1(\tan {\mathrm{\α}}_1+{\tan \mathrm{\α}}_2)}\\ z_0=-\frac{s}{\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2}\end{array}\right.$ (formula 1)

According to described positional information (x ₀, y ₀, z ₀) obtain unmanned plane relative vehicle landing point positional information (x, y, z):

$\left\{\begin{array}{c}x=x_0-\frac{s}{2}=\frac{s\tan {\mathrm{\α}}_1}{\tan {\mathrm{\α}}_1+{\tan \mathrm{\α}}_2}-\frac{s}{2}\\ y=y_0=\frac{s\tan {\mathrm{\β}}_1}{\cos {\mathrm{\α}}_1(\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2)}\\ z=z_0=-\frac{s}{\tan {\mathrm{\α}}_1+\tan {\mathrm{\α}}_2}\end{array}\right.$ (formula 2)

Then described positional information (x, y, z) is sent to the airborne control subsystem of unmanned plane by radio transmission apparatus;

Step (4), the airborne control subsystem of many rotor unmanned helicopters receives the positional information (x of described unmanned plane relative vehicle landing point, y, z) after, control many rotor unmanned helicopters by flight-control computer and drop to target area on vehicle, comprise the steps:

(41), the detection data of the sensor assembly Real-time Collection ultrasonic height meter of many rotor unmanned helicopters, six axle attitude sensors and electronic compass, flight-control computer resolves the altitude information, the attitude data that obtain many rotor unmanned helicopters, and according to described positional information (x, y, z) real time data resolve the speed data obtaining many rotor unmanned helicopters relative vehicle;

(42), the speed data of described altitude information, attitude data, station-keeping data and relative vehicle and expected data are compared, obtains error information;

(43), calculate the controlled quentity controlled variable of topworks according to described error information and unmanned aerial vehicle (UAV) control rule, and be sent to the topworks of many rotor unmanned helicopters; Described unmanned aerial vehicle (UAV) control rule comprises attitude control law, height-holding control law, horizontal direction control law; Wherein, described attitude control law comprises angle of pitch control rate, roll angle control rate, crab angle control rate;

Described height-holding control law is:

$\mathrm{\Δ}{\mathrm{\δ}}_c=K_{p_z}\mathrm{\Δz}+K_{p_z}^{\′}\stackrel{\·}{z}+K_{p_z}^{\′\′}\stackrel{\·\·}{z}+K_{i_z}^{\′\′}\∫\stackrel{\·\·}{z}+K_{d_z}^{\′\′}{\stackrel{\·\·}{z}}^{\′}$

Wherein, Δ z is height error data, short transverse speed, short transverse acceleration, Δ δ _cfor altitude channel controlled quentity controlled variable, it is right to represent differentiate, with it is controling parameters;

Many rotor unmanned helicopters horizontal x direction Landing Control is that described angle of pitch control rate is inner looping with the tracing control in horizontal x direction rule for external loop; Wherein, the tracing control rule in described horizontal x direction is:

$\left\{\begin{array}{cc}\mathrm{\Δ}{\mathrm{\δ}}_{\mathrm{\θ}}=K_{p_x}\mathrm{\Δx}+K_{i_x}\∫\mathrm{\Δx}+K_{d_x}\mathrm{\Δ}{x}^{\′}+K_{P_x}^{\′}\stackrel{\·}{x},& \left|\mathrm{\Δx}\right|\≤e\\ {\mathrm{\Δ\δ}}_{\mathrm{\θ}}=K_{p_x}\mathrm{\Δx}+K_{d_x}\mathrm{\Δx}+K_{P_x}^{\′}\stackrel{\·}{x}& \left|\mathrm{\Δx}\right|>e\end{array}\right.$

Many rotor unmanned helicopters horizontal y direction Landing Control is that described roll angle control rate is inner looping with the tracing control in horizontal y direction rule for external loop; Wherein, the tracing control rule in described horizontal y direction is:

$\left\{\begin{array}{cc}{\mathrm{\Δ\δ}}_{\mathrm{\φ}}=K_{p_y}\mathrm{\Δy}+K_{i_y}\∫\mathrm{\Δy}+k_{d_y}{\mathrm{\Δy}}^{\′}+K_{p_y}^{\′}\stackrel{\·}{y},& \left|\mathrm{\Δy}\right|\≤e\\ {\mathrm{\Δ\δ}}_{\mathrm{\φ}}=K_{p_y}\mathrm{\Δy}+K_{d_y}{\mathrm{\Δy}}^{\′}+K_{p_y}^{\′}\stackrel{\·}{y},& \left|\mathrm{\Δy}\right|>e\end{array}\right.$

Wherein, Δ δ _θ, Δ δ _φbe respectively x, y direction controlling amount, be respectively the speed data of x, y direction relative vehicle, the site error that Δ x, Δ x ' are x-axis direction and error change rule, the site error that Δ y, Δ y ' are y-axis direction and error change rule, with the controling parameters in x-axis direction, with be the controling parameters in y-axis direction, e is the fixed error value of setting.