CN107672817B

CN107672817B - An intelligent take-off and landing system for a mobile vehicle-mounted UAV

Info

Publication number: CN107672817B
Application number: CN201710861766.2A
Authority: CN
Inventors: 齐咏生; 孙作慧; 李永亭; 刘利强; 孟学斌; 王林
Original assignee: Inner Mongolia University of Technology
Current assignee: Inner Mongolia University of Technology
Priority date: 2017-09-21
Filing date: 2017-09-21
Publication date: 2020-05-22
Anticipated expiration: 2037-09-21
Also published as: CN107672817A

Abstract

The invention discloses an intelligent take-off and landing system of a mobile vehicle-mounted unmanned aerial vehicle. The system comprises a lifting platform device, an unmanned aerial vehicle fixing and automatic charging device. First, four universal balls are installed at the bottom of the landing gear of the drone. When the drone falls on a smooth landing platform, the rotation of the universal balls can automatically slide into the groove of the landing platform. Then, a conductive ring is installed at the fixed position of each foot of the landing gear of the drone, so that the conductive ring of the drone and the charging guide rail in the groove are on the same plane after landing. The feet form an electrode, which is connected to the battery of the drone. Finally, equip the drone with a vision processing module with an embedded chip. The vision processing module is composed of two sensors, the GPS module and the vision camera module, and the precise positioning of the UAV is realized through algorithm control. The invention facilitates the flying and recovery of the unmanned aerial vehicle in the field environment, and can realize autonomous charging.

Description

Intelligent take-off and landing system of mobile vehicle-mounted unmanned aerial vehicle

Technical Field

The invention relates to an intelligent take-off and landing system of a mobile vehicle-mounted unmanned aerial vehicle, and belongs to the technical field of unmanned aerial vehicle control design.

Background

The unmanned aerial vehicle has extremely important functions in the current society by virtue of low cost, strong viability, good maneuvering performance and convenient use, and has wide prospects in the civil field. But will examine the field with the automatic inspection of more extensive application of unmanned aerial vehicle, just need to have the on-vehicle unmanned aerial vehicle intelligence system of taking off and land of a kind of mobility flexibility to come better fly off and retrieve unmanned aerial vehicle to supply unmanned aerial vehicle battery power that can be timely. How steady landing of unmanned aerial vehicle is on vehicle-mounted platform and fixed, the correlation technique that corresponds at present is still rarely reported, and this aspect is mainly still to operate by people as the owner.

For unmanned aerial vehicles, battery endurance and in-flight recharging have been troublesome issues in the industry. Due to the limitation of battery technology, the endurance time of the general unmanned aerial vehicle system is within half an hour, and the standby time is limited due to the limitation of battery capacity. According to the current battery technology level, under the condition of unattended charging, the unmanned aerial vehicle cannot execute some continuous or continuous tasks for a long time, such as grassland ecological monitoring, unmanned aerial vehicle autonomous inspection, natural disaster situation detection and the like. Because the vast population in northwest of China is rare, the environment is complicated, and the unmanned aerial vehicle has to fly for a long time, an automatic charging device is needed, after the unmanned aerial vehicle returns to the vehicle-mounted platform, the unmanned aerial vehicle can be automatically charged, and the unmanned aerial vehicle can continue to execute tasks after being fully charged.

At present, the main problems of the domestic unmanned aerial vehicle-mounted platform are that the unmanned aerial vehicle can not be stably recovered and the endurance of a power supply can not be independently provided for the unmanned aerial vehicle. Therefore, how to let the unmanned aerial vehicle stably land on the vehicle-mounted platform and fix is the first problem that needs to be solved. The invention relates to an intelligent take-off and landing system of a mobile vehicle-mounted unmanned aerial vehicle based on a mechanical transmission technology and an advanced control technology. Mainly solves the following problems:

1. because the take-off and landing system is installed on the intelligent vehicle (or transport vehicle), when the intelligent vehicle (or transport vehicle) moves, the landing of the unmanned aerial vehicle on the platform can be influenced by a plurality of factors. Unmanned aerial vehicle is at the flight in-process, and the fast revolution of rotor can produce very strong air current, influences unmanned aerial vehicle's landing control, and unmanned aerial vehicle landing point is close to ground more moreover, receives the recoil power that comes from ground just bigger, and unmanned aerial vehicle's accurate descending will become difficult more. Therefore, the mesh-shaped vehicle-mounted lifting platform is designed, and multiple times of experimental verification show that when the platform reaches a certain height, the recoil force of the unmanned aerial vehicle from the ground can be greatly reduced according to the aerodynamic theory, so that the problem that the unmanned aerial vehicle stably falls on the vehicle-mounted lifting platform with the mesh can be solved.

2. Because unmanned aerial vehicle descends on intelligent vehicle (or transport vechicle), when the dolly motion, the shake takes place for unmanned aerial vehicle, causes the damage to unmanned aerial vehicle easily, and moreover, unmanned aerial vehicle long-time flight work needs an automatic battery charging outfit to guarantee the continuation of the journey of its power. Consequently, this lift platform design has the smooth circular landing platform of certain radian to there is a recess at the center of landing platform, after unmanned aerial vehicle after the repacking descends to the platform on, can slide in the recess automatically. The groove is internally provided with a charging guide rail which can transversely slide, and the guide rail is provided with an infrared sensor. When the sensor detects that the charging guide rail is effectively contacted with the conducting ring on the unmanned aerial vehicle, the system starts to automatically charge and fix the unmanned aerial vehicle; when unmanned aerial vehicle prepares to leave the platform of taking off and landing, the guide rail that charges separates with unmanned aerial vehicle's conducting ring, and the charging process finishes, realizes flying of unmanned aerial vehicle.

3. Because unmanned aerial vehicle exposes outside for a long time, also can cause the injury to itself under the environment of open-air complicacy. According to the unmanned aerial vehicle recycling device, a Z-shaped lifting structure with an electric push rod is arranged below a platform by utilizing the principle of mechanical transmission, the push rod is automatically stretched by a control mechanism of the electric push rod, the platform is lifted and descended, and the space and resources of an intelligent trolley are fully utilized to solve the problem of recycling of the unmanned aerial vehicle.

An unmanned aerial vehicle, abbreviated as "unmanned aerial vehicle" ("UAV"), is an unmanned aerial vehicle that is operated using a radio remote control device and a self-contained program control device. Compared with manned aircraft, it has advantages of small volume, low cost, convenient use, etc., and has wide application in industries such as police, city management, agriculture, geology, meteorology, electric power, emergency rescue and relief, video shooting, etc. Wherein, patrol the open-air inspection of line unmanned aerial vehicle mainly used and detect, then pass back through the picture and supply the professional to look over, very big manpower of having saved, the security is high moreover. But at present, the line patrol unmanned aerial vehicle faces the biggest problem of how to fly off and retrieve in the field, how to accomplish the autonomic continuation of journey of battery, and these two problems determine the time and the stability of line patrol unmanned aerial vehicle flight, and the indirect efficiency and the quality of line patrol unmanned aerial vehicle operation of determining.

To the problem of stopping and continuing the journey of unmanned aerial vehicle, there have been some technologies that are applied to unmanned aerial vehicle and retrieve and automatic charging at present. But its recovery on fixed platform, rather than the recovery on-vehicle moving platform, the problem that is difficult to control when can not solving unmanned aerial vehicle and retrieving also does not possess the function that can go up and down. Moreover, how to solve the problem of endurance of the field line patrol unmanned aerial vehicle is very few reports in China at present. Through the literature retrieval, patent publication number is 104503459, and patent number is 2014106823229, the name is many rotor unmanned aerial vehicle recovery system. The active recovery of the invention adopts an adsorption device and an airborne passive recovery adsorption device, and the mechanism of the active recovery device is completely different from that of the recovery device. Due to the uncertainty of control, the invention is difficult to realize the accurate landing of the unmanned aerial vehicle; the invention only refers to the recovery of the unmanned aerial vehicle, but the system does not contain an automatic lifting function and does not consider the protection of the unmanned aerial vehicle; although the invention refers to the use of a wireless communication device to complete the landing positioning of the unmanned aerial vehicle, no control algorithm (implementation algorithm) is mentioned, and obviously, the invention only protects the designed mechanical structure. In the invention, the lifting platform is designed into a smooth circular landing platform with a certain radian, and the platform is provided with meshes, so that the recoil force of landing of the unmanned aerial vehicle can be overcome, and the accurate control is realized; in order to enable the unmanned aerial vehicle to land on a preset position accurately, the invention uses a multi-sensor fusion algorithm based on vision to realize the accurate landing of the unmanned aerial vehicle; the invention also installs a Z-shaped lifting structure with an electric push rod under the platform, and the push rod automatically extends and retracts through a control mechanism of the electric push rod, thereby realizing the lifting and the falling of the platform.

To sum up, the prior art does not mention how to solve on-vehicle unmanned aerial vehicle system of taking off and land and the continuation of journey problem of intelligent charging. According to the working environment and working condition of the unmanned aerial vehicle and the intelligent vehicle, the unmanned aerial vehicle and the intelligent vehicle adopt new design ideas from mechanical design to control algorithm so as to achieve the purposes of reasonably controlling the coordinated motion of the unmanned aerial vehicle and the intelligent vehicle and realizing the recovery, the release and the automatic charging of the unmanned aerial vehicle.

Firstly, the existing invention mentioned above has no lifting platform device, and in practical situations, the unmanned aerial vehicle falls on the vehicle-mounted platform, occupies a large space, has a large shaking amplitude, and cannot play a role in protection. Therefore, the invention develops a new way, adopts a Z-shaped lifting structure with an electric push rod, and drives the push rod to stretch through a control mechanism of the electric push rod so as to realize the lifting and the falling of the platform; secondly, aiming at the problem of how to stably land the unmanned aerial vehicle on the platform, the mesh-shaped lifting platform is designed, and when the platform is lifted to a certain height through the lifting platform device, the recoil force of the unmanned aerial vehicle from the ground can be greatly reduced, so that the stable landing and the accurate positioning of the unmanned aerial vehicle are realized; then, to how fixed problem with automatic charging of unmanned aerial vehicle, this lift platform design has the smooth circular descending platform of certain radian to there is a recess at the center of descending platform, has installed four rotor unmanned aerial vehicle of universal ball (or universal wheel) additional on the descending support in can the automatic slip recess. Two semicircular charging guide rails are arranged in the groove, each section of charging guide rail is connected with the push rod motor, and an infrared sensor is arranged on the guide rail, so that effective contact between the charging guide rail and the unmanned aerial vehicle conducting ring can be detected. When the unmanned aerial vehicle is stabilized in the groove, the system controls the push rod motor to drive the two charging guide rails to transversely advance so as to be in contact with the conducting rings on the unmanned aerial vehicle, so that the unmanned aerial vehicle starts to be charged and is fixed; when the unmanned aerial vehicle is ready to leave the platform, the system controls the charging guide rail to transversely retreat, so that the charging guide rail is separated from the conducting ring of the unmanned aerial vehicle, the charging process is finished, and the unmanned aerial vehicle can fly. Finally, the invention uses a vision-based multi-sensor fusion algorithm to automatically land the unmanned aerial vehicle, thereby realizing that the unmanned aerial vehicle accurately lands to a preset position.

Although the existing unmanned aerial vehicle flying and recovering technology is applied to a certain extent in the corresponding application field, and a certain effect is achieved, some problems still exist if the existing unmanned aerial vehicle flying and recovering technology is directly applied to a vehicle-mounted platform.

1) Steady descending during unable solution recovery unmanned aerial vehicle

The existing unmanned aerial vehicle recovery technology is only suitable for landing on a fixed large platform, and due to the requirement of field inspection, when the unmanned aerial vehicle is about to land on a mobile vehicle-mounted small platform or an intelligent trolley, the existing technology hardly meets the landing requirement of the unmanned aerial vehicle. Because the fast rotation of rotor can produce very big impact air current, influences unmanned aerial vehicle's accurate landing control, and unmanned aerial vehicle landing point is close to ground more moreover, receives the recoil power that comes from ground just bigger, and unmanned aerial vehicle's control will become more difficult.

2) Fixing and automatic charging of unmanned aerial vehicle on take-off and landing platform cannot be achieved

Because unmanned aerial vehicle's stability and wind resistance are relatively weak, so unmanned aerial vehicle must fix on moving platform, and can realize automatic continuation of the journey that charges. However, the unmanned aerial vehicle has limited precision when landing on the intelligent vehicle, namely, the unmanned aerial vehicle can only be guaranteed to land within a certain range, but not at an accurate point. Consequently, unmanned aerial vehicle's the fixed problem and the automatic guide rail problem degree of difficulty that carries on charging are very big, and this has also greatly restricted unmanned aerial vehicle in the development of on-vehicle automatic aspect of receiving and releasing. At present, a device for fixing an unmanned aerial vehicle is few, and the fixing problem of a vehicle-mounted platform cannot be solved. The invention has the important innovation that the problems of fixation and automatic charging of the unmanned aerial vehicle are solved, and the cooperative cooperation of the field inspection robot is further ensured. The application of the unmanned aerial vehicle to the automobile platform also has the problems, so that the unmanned aerial vehicle charging system can be applied to the vehicle-mounted platform of a manned automobile, and the automatic fixing and charging continuation of the journey of the unmanned aerial vehicle are realized.

3) How to solve the problem of the lifting of the take-off and landing system of the airplane

At present, the landing platform of the existing unmanned aerial vehicle mostly has no lifting device. The influence of the recoil air flow of the ground or the platform when the unmanned aerial vehicle lands is considered, so that the landing platform of the unmanned aerial vehicle needs to occupy a large area and is not beneficial to the advance of the trolley. And the four rotors have very large shaking amplitude under the influence of airflow, and the platform cannot well play a role in protection. Therefore, the platform is designed into a lifting system with meshes, and the influence of the recoil airflow of the ground and the lifting platform on the accurate positioning and landing of the unmanned aerial vehicle can be effectively reduced. How to design a suitable lifting system is also an important issue to be solved by the present invention.

4) How to realize accurate landing of unmanned aerial vehicle to preset position

At present, the conventional unmanned aerial vehicle autonomous landing is mainly realized through GPS navigation. However, the fact proves that the independent GPS navigation cannot provide accurate position information for the unmanned aerial vehicle, and accurate landing is difficult to realize in a complex environment in the field. In order to make the unmanned aerial vehicle land on a preset position accurately, a plurality of sensors (including a vision sensor) are matched and a corresponding algorithm is needed to realize the landing.

Disclosure of Invention

Aiming at the problems, the invention provides an intelligent take-off and landing system of a mobile vehicle-mounted unmanned aerial vehicle, which is convenient for the flying and recovery of the unmanned aerial vehicle in a field environment and can realize autonomous charging; according to the particularity of the take-off and landing platform of the vehicle-mounted unmanned aerial vehicle, the take-off and landing platform can fundamentally complete the flying and recovery of the unmanned aerial vehicle and can automatically charge the unmanned aerial vehicle through mechanical design and control means.

To the problem that can't solve steady landing when retrieving unmanned aerial vehicle: the existing related inventions and technologies are all suitable for flying and recovering the unmanned aerial vehicle on a large fixed platform, and for a movable vehicle-mounted lifting platform, the range of a landing point is greatly reduced, the stability degree in the landing process is also greatly reduced, and the difficulty of automatic recovery of the vehicle-mounted unmanned aerial vehicle is increased.

In order to solve the problem, the invention designs a mesh-shaped vehicle-mounted lifting platform, and obtains a relatively ideal lifting height through calculation and experiments according to aerodynamics.

To solving the fixed and automatic problem of charging of unmanned aerial vehicle on the platform that takes off and land: because unmanned aerial vehicle falls on the intelligent vehicle, when the dolly motion, unmanned aerial vehicle must be fixed, otherwise can take place the shake, cause unmanned aerial vehicle's damage easily. Moreover, the problem of endurance of the unmanned aerial vehicle is also a big problem that the unmanned aerial vehicle cannot carry out long-distance inspection, and an automatic charging device is needed to guarantee the endurance of the power supply of the unmanned aerial vehicle. The conventional device for fixing the unmanned aerial vehicle is few, cannot solve the problem of fixing the vehicle-mounted platform, and does not have the capability of autonomous charging. The invention designs the lifting platform into a smooth circular landing platform with a certain radian, so that the radian of the platform cannot be too large or too small, the unmanned aerial vehicle with too large radian is easy to turn over, and the unmanned aerial vehicle with too small radian is easy to block.

Through a plurality of simulation experiments, the radian of the platform is designed to be 15 degrees. The center of landing platform has a recess, and through visual algorithm location, unmanned aerial vehicle can be the landing of basic accuracy on the platform. Because install universal ball on the unmanned aerial vehicle descending support, even there is slight deviation in the descending process, unmanned aerial vehicle also can slide automatically in the recess that descends the platform. The guide rails that charge are respectively installed on the two sides of the groove and connected with the push rod motor, and the push rod motor is fixed on the support platform. Every section all is equipped with infrared sensor on the guide rail that charges, can detect the effective contact of guide rail and unmanned aerial vehicle conducting ring that charges. When the unmanned aerial vehicle needs to be charged autonomously, the push rod motor is controlled to drive the two charging guide rails to move forwards transversely to be in contact with the conducting ring on the unmanned aerial vehicle, and then autonomous charging is started and the unmanned aerial vehicle is fixed; when unmanned aerial vehicle prepares to leave the platform, through control push rod motor, drive two guide rails that charge and transversely retreat, make it and unmanned aerial vehicle's conducting ring separation, the charging process finishes to realize flying of unmanned aerial vehicle.

For solving the problem of lifting of the lifting system: existing unmanned aerial vehicle's recovery platform mostly does not have elevating gear, and unmanned aerial vehicle can occupy great space on falling the platform, and ground and elevating platform recoil air current make unmanned aerial vehicle range of rocking great, can not play the effect of protection. Therefore, the ordinary unmanned aerial vehicle recovery platform cannot be suitable for the task of field inspection. The invention adopts a Z-shaped lifting structure with an electric push rod through a mechanical transmission principle, and is respectively connected with a bracket base fixed in a vehicle body and a bracket platform supporting a landing platform, and a control mechanism of the electric push rod drives a motor to work, thereby realizing the lifting and the falling of the platform.

The problem of how to realize accurate positioning of the unmanned aerial vehicle is as follows: traditional unmanned aerial vehicle independently lands mostly mainly through GPS navigation realization, but solitary GPS navigation can not provide accurate positional information for unmanned aerial vehicle, under the environment of open-air complicacy, hardly realizes accurate landing moreover. In order to enable the unmanned aerial vehicle to land on a preset position accurately, the invention adopts a vision-based multi-sensor fusion algorithm to realize the automatic landing of the unmanned aerial vehicle.

In order to meet the requirements of the take-off and landing system of the vehicle-mounted unmanned aerial vehicle on the flying recovery and automatic charging of the unmanned aerial vehicle, the technical scheme adopted by the invention is an intelligent take-off and landing system of the mobile vehicle-mounted unmanned aerial vehicle, and the system comprises a lifting platform device and an unmanned aerial vehicle fixing and automatic charging device.

The lifting platform device consists of three parts, namely a bracket platform (10), a Z-shaped lifting structure and a bracket base (18) from top to bottom. The lower part of the bracket platform (10) is connected with a Z-shaped lifting structure.

The Z-shaped lifting structure comprises eight same lifting rods (13), three bracket beams (17), a lifting rod push rod (14), an electric push rod (15) and a plurality of screws. The lifting rods (13) are eight in total and divided into two groups, each group is four, each group is respectively placed on two sides of the support base (18), and the corresponding lifting rods (13) of each group are connected together through support beams (17) and lifting rod push rods (14). The four lifting rods (13) on each side are divided into an upper group and a lower group, the centers of the lifting rods (13) are fixed together by screws, and the bracket beam (17) is connected to the lifting rods (13) to form a lifting structure; the electric push rod (15) is connected with the lifting rod (13) through the lifting rod push rod (14), and the lifting rod (13) is lifted or lowered through the electric push rod (15). The electric push rod (15) is connected with the lifting rod push rod (14), the electric push rod (15) is installed on the control mechanism (16), and the control mechanism (16) and the electric push rod (15) are fixed in a central groove (19) of the support base (18) together.

The support platform (10) is a mesh-shaped rectangular platform and is a support structure of the unmanned aerial vehicle. Two lifting rod fixing seats (11) and two fixed sliding rails (12) are respectively arranged at the bottom of the support platform (10). The lifting rod fixing seat (11) is placed on one side of the support platform (10) and plays a role in fixing a fulcrum of the lifting rod (13); the fixed slide rail (12) is arranged on the other side of the support platform (10) to facilitate the synchronous ascending and descending of the lifting rod (13).

The support base (18) is the base of the entire system. The central groove (19) is positioned in the middle of the bracket base (18).

The unmanned aerial vehicle fixing and automatic charging device consists of a vision positioning unit, a landing platform (2) and an automatic charging and fixing unit. The vision positioning unit comprises universal balls (3), a conducting ring (4) and a vision processing module provided with an embedded chip and arranged on the unmanned aerial vehicle, and the four universal balls (3) and the conducting ring (4) are additionally arranged at the bottom of the landing gear of the unmanned aerial vehicle. The automatic charging and fixing unit comprises a charging guide rail (6), a push rod motor (9) and an infrared sensor (7). The vision processing module is composed of a GPS module (1) and a vision camera module (5), and the unmanned aerial vehicle is accurately positioned through algorithm control.

The landing platform (2) is a mesh-shaped circular platform with a certain radian and smoothness, and is fixed on the support platform (10). A groove (8) is formed in the center of the landing platform, and the charging guide rail (6) is installed in the groove (8). Push rod motor (9) are fixed on support platform (10), and push rod motor (9) link to each other with guide rail (6) that charges to the lateral shifting of guide rail (6) charges is driven, and infrared sensor (7) are installed on guide rail (6) charges. The upper part of the bracket platform (10) is connected with a landing platform (2).

This system realizes the steady stop of design platform and independently charges on four rotor unmanned aerial vehicle bases. Firstly, install four universal ball additional in unmanned aerial vehicle undercarriage bottom, when unmanned aerial vehicle fell on the smooth descending platform, in the recess that can the automatic sliding landing platform through universal ball's rotation. Then, all install the conducting ring additional in the fixed position department of every horn of unmanned aerial vehicle undercarriage, make descending back unmanned aerial vehicle's conducting ring and the guide rail that charges in the recess on same plane, an electrode is constituteed to every two adjacent horns, is connected with unmanned aerial vehicle's battery. And finally, a visual processing module with an embedded chip is equipped on the unmanned aerial vehicle. The vision processing module is composed of a GPS module and a vision camera module, and the unmanned aerial vehicle is accurately positioned through algorithm control.

The radian of the landing platform is 15 degrees. The center of descending platform has a recess, and through vision landing algorithm, unmanned aerial vehicle can descend to the recess of platform accurately in, even unmanned aerial vehicle descends to have the deviation, installs the unmanned aerial vehicle of universal ball and also can the automatic slip recess.

The infrared sensor is used for detecting effective contact of the charging guide rail and the unmanned aerial vehicle conducting ring. When unmanned aerial vehicle will independently charge, through control push rod motor, drive two guide rails that charge and transversely advance, take place the contact with the conducting ring on the unmanned aerial vehicle, begin to charge and fix unmanned aerial vehicle. When unmanned aerial vehicle prepares to leave the platform, the control guide rail that charges transversely retreats, and with unmanned aerial vehicle's conducting ring separation, the charging process finishes, realizes flying of unmanned aerial vehicle.

Electric putter drives zigzag elevation structure and risees the platform to 80cm department, and unmanned aerial vehicle docks on descending platform through the multisensor fusion algorithm based on vision, and the unmanned aerial vehicle of having installed universal ball freely slides on the platform, finally docks in the recess of descending platform. The guide rail that charges in the recess passes through the push rod motor and transversely gos forward, and when infrared sensor detected the guide rail that charges and took place effective contact with the unmanned aerial vehicle conducting ring, unmanned aerial vehicle began independently to charge and is fixed. The Z-shaped lifting structure drives the platform to descend, and the whole system descends into the vehicle body. When unmanned aerial vehicle prepares to leave the platform, zigzag elevation structure drives entire system and rises, and the control guide rail that charges separates with unmanned aerial vehicle's conducting ring, and the charging process finishes, and unmanned aerial vehicle obtains the instruction and takes off from the platform is automatic.

Drawings

FIG. 1.1 is the wind pattern for a platform height of 80 cm.

FIG. 1.2 is the wind direction diagram at a platform height of 30cm

Fig. 2 is an overall schematic diagram of the intelligent take-off and landing system of the mobile vehicle-mounted unmanned aerial vehicle, (a) a schematic diagram of the modified unmanned aerial vehicle, (b) a schematic diagram of the unmanned aerial vehicle landing on the platform, and (c) a front view of the intelligent take-off and landing system of the mobile vehicle-mounted unmanned aerial vehicle.

Fig. 3 is an overall configuration diagram of the system of the present invention.

Fig. 4.1 is a flow chart of the unmanned aerial vehicle recovery system.

Fig. 4.2 is a flow chart of the flying drone system.

FIG. 5 visual landing identifier map.

In the figure: 1. the GPS module, 2, descending platform, 3, universal ball, 4, conducting ring, 5, vision camera module, 6, charging guide rail, 7, infrared sensor, 8, groove, 9, push rod motor, 10, support platform, 11, lifter fixing seat, 12, fixed slide rail, 13, lifter, 14, lifter push rod, 15, electric putter, 16, control mechanism, 17, support beam, 18, support base, 19, central groove.

Detailed Description

Let P_∞Is static pressure, p_∞Is the density of the air flow, V_∞Is the air velocity, then the dynamic air pressure

At this time, the air lift

Wherein C is_lFor the lift coefficient of the drone, S represents the wing area.

The wind patterns produced by the airfoils at platform heights of 80cm and 30cm, respectively, are shown in FIG. 1. It can be seen that when the height of platform apart from ground is about 80cm, unmanned aerial vehicle comes from the recoil force on ground and will reach a smaller value (continue to increase apart from the recoil force and tend to the constant value, elevating platform cost increases), reaches the compromise point of manufacturing cost (degree of difficulty) and distance this moment, also shows through the experiment that unmanned aerial vehicle can steadily descend on the elevating platform basically this moment in addition.

The system comprises the following implementation steps:

step 1: the stroke motor with the lift platform device designs for the biggest extension degree and is 80cm, when unmanned aerial vehicle prepares to descend to the platform of taking off and landing on, electric putter's control mechanism begins the extension of control push rod, and mechanical structure slowly lifts up whole system. When the platform reaches the maximum height, the system automatically stops moving, at which point the platform stabilizes at 80 cm.

Step 2: unmanned aerial vehicle is fixed and independently charges. Unmanned aerial vehicle docks on the platform of taking off and landing through the multisensor fusion algorithm based on vision, because the universal ball is equipped with on the unmanned aerial vehicle undercarriage, and the landing platform is smooth and has certain radian circular platform moreover, and unmanned aerial vehicle is through the skew of the free slip at the landing platform in the time of overcoming the landing to a certain extent of the drop point. At the moment, the landing gear of the unmanned aerial vehicle is supported by the support platform, and the conducting ring on the undercarriage and the charging guide rail in the groove are in the same plane. When unmanned aerial vehicle is stable in the recess, control push rod motor transversely gos forward, drives the guide rail that charges and takes place the contact with unmanned aerial vehicle conducting ring. When infrared sensor on the guide rail that charges detected the guide rail that charges and the effective contact of unmanned aerial vehicle conducting ring, start unmanned aerial vehicle charge controller and begin work, and unmanned aerial vehicle has been fixed by the guide rail this moment.

And step 3: and (6) landing the platform. In order to guarantee the stability and the security of unmanned aerial vehicle on the system, descend whole operating system to the automobile body in. Electric putter's control mechanism begins control push rod to shorten, and the platform that takes off and land descends, finally makes entire system contain in the automobile body, has both saved the space, has also protected unmanned aerial vehicle stability and safety in the marcing.

And 4, step 4: and (5) flying the unmanned aerial vehicle. When unmanned aerial vehicle prepares to leave the platform, drive entire system through control lift platform device again and rise, when the platform rises to the height that can fly away, electric putter automatic stop motion. At this moment, the push rod motor is controlled to transversely retreat, and the charging guide rail is driven to be separated from the unmanned aerial vehicle conducting ring. The charging process is finished, and the unmanned aerial vehicle realizes independent takeoff.

The overall structure of the system of the invention is shown in FIG. 3; general block diagram of system architecture of the invention a flow chart of the control process of the invention is shown in fig. 4.1-4.2

(3) Multi-sensor fusion algorithm based on vision

In the invention, because the landing control of the unmanned aerial vehicle is involved and the control precision has higher requirements, the traditional unmanned aerial vehicle autonomous landing is mostly realized by GPS navigation. But the independent GPS navigation cannot provide accurate position information for the unmanned aerial vehicle, and cannot position in the indoor and some other specific scenes, so that it is difficult to realize accurate landing of the unmanned aerial vehicle on the landing platform, and the expected accuracy requirement cannot be met. Therefore, the invention adopts a new control strategy, and the GPS navigation and the vision camera are matched to work through an algorithm, namely, the multi-sensor fusion algorithm based on vision is adopted to realize the accurate landing of the unmanned aerial vehicle.

1. GPS navigation initial adjustment

The Global Positioning System (GPS) is a space intersection fixed-point navigation System capable of timing and ranging, has navigation, Positioning and timing functions of omnipotence (land, sea, air and space), globality, all weather, continuity and real-time, and can provide high-precision three-dimensional coordinates, three-dimensional speed and time information for various users. The basic principle of GPS navigation is that the instantaneous position of a satellite moving at high speed is used as known calculation data, and the position of a point to be measured is determined by adopting a space distance rear intersection method.

This design has added the GPS module at the unmanned aerial vehicle unit, and when unmanned aerial vehicle was far away from on-vehicle take off and land system, the coarse adjustment of unmanned aerial vehicle descending was realized to the position of utilizing the navigation of GPS to roughly fix a position on-vehicle take off and land system. At this moment, the vision camera module is in the off-state, can save unmanned aerial vehicle's electric quantity from this, guarantees unmanned aerial vehicle's continuation of the journey. When the unmanned aerial vehicle flies to the position close to the preset position, the vision camera is started, the fine adjustment stage of unmanned aerial vehicle landing is started, the vision landing identification on the landing platform is searched, the GPS navigation is closed at the moment, and the unmanned aerial vehicle accurate landing based on the vision algorithm is started.

2. Visual algorithm tweaking

The algorithm designs a visual landing identifier, the visual landing identifier can be rapidly detected in real time through a self-adaptive threshold value method and a relatively simple landing identifier identification method in the algorithm, and the position parameters required by the unmanned aerial vehicle can be calculated from the landing identifier. And converting the position information in the image into real position information through geometric operation, and transmitting the real position information to the aircraft controller for navigation.

Firstly, designing a visual landing identifier on a landing platform of an intelligent take-off and landing system: the outer large circular frame and the inner small circular frame are shown in figure 5. The outer large circular frame is used for calculating the position parameters of the unmanned aerial vehicle relative to the landing target, and the inner small circular frame is used for estimating the landing position of the unmanned aerial vehicle. Visual landing indication is that the external graphic and the internal graphic have strong contrast. Firstly, the unmanned aerial vehicle estimates the position information of the unmanned aerial vehicle by using a large outer circular frame; and after the unmanned plane flies over the external large circular frame mark, continuously estimating the position information of the unmanned plane according to the internal small circular frame.

The camera is installed in the positive center of unmanned aerial vehicle along the direction of vertical decurrent, consequently replaces the position of unmanned aerial vehicle central point with the position of camera central point, namely the light center.

The self-adaptive threshold method is adopted to calculate the image h (i, j) of the visual landing marker map acquired by the camera

Wherein 0 is black and 1 is white; and scanning the image line by line, and then calculating the average value of the pixels of the first S pixels of the current scanning point. And when the point pixel is smaller than (1-T) times of the mean value of the S pixels before the point, or the point pixel is smaller than (1+ T) times of the mean value of the pixels which are judged to be black pixels in the S pixels before the point, judging that the value of the point pixel is 0. f (i, j) is the pixel value of the current pixel point; ds (i, j) represents the pixel average value of S pixel points before the current point; t is a number from 0 to 1; the values of S and T are S-width/8, wherein the width represents the width of an image, and the best effect is achieved when T is 0.15; r_b(i, j) represents the average value of pixels judged as black pixels in the first S pixels of the point, and b represents the number. And the binary judgment of the image is more refined, so that a high-quality binary image is generated, and a foundation is laid for the detection of the landing identifier. And obtaining pixel offset x and y after the contour judgment is successful. The offset refers to the relative position of the central point of the unmanned aerial vehicle and the central point of the landing identifier, and the control parameters of the flight controller to the aircraft are based on a real coordinate system, so that the solved pixel offset is converted into real offsets X and Y. And calculating the difference value between the coordinate of the landing identification center point in the image and the coordinate of the image center point through the pixel offset. The true offset is obtained by the following formula using the pixel offset. The concrete formula is as follows:

where X, Y denote pixel offsets, X, Y denote true offsets, f_x,f_yIs shown in the x directionAnd focal length in y-direction, c_x,c_yAnd H represents the height of a moment drop point of the central point of the camera.

The method adopts a vision-based multi-sensor fusion technology, namely, the GPS navigation and the vision camera are matched to work, so that the endurance of the unmanned aerial vehicle battery is ensured, the time spent in positioning the unmanned aerial vehicle is saved, and the accurate landing of the unmanned aerial vehicle on the vehicle-mounted mobile platform is finally realized.

The overall schematic diagram of the intelligent take-off and landing system of the mobile vehicle-mounted unmanned aerial vehicle is shown in fig. 2.

The originality of the invention is mainly embodied in a mesh-shaped lifting platform, a lifting structure of a vehicle-mounted platform, an unmanned aerial vehicle fixing and autonomous charging structure and a vision-based multi-sensor fusion algorithm.

1) Mesh-shaped lifting platform: the invention designs a circular mesh-shaped liftable platform with a certain radian and smoothness, through a plurality of simulation experiments, the radian of the selected platform is designed to be about 15 degrees, and according to the theoretical basis of aerodynamics, a more ideal lifting height is obtained through calculation and experiments, so that the accurate landing of an unmanned aerial vehicle can be realized.

2) Lifting structure of vehicle platform: the invention adopts a Z-shaped lifting structure with an electric push rod through a mechanical transmission principle, and is respectively connected with a bracket base fixed in a vehicle body and a bracket platform supporting a landing platform, and a control mechanism of the electric push rod drives a motor to work, thereby realizing the lifting and the falling of the vehicle-mounted platform.

3) Unmanned aerial vehicle is fixed and independently charge structure: according to the invention, the charging guide rails are respectively arranged at two sides of the groove and connected with the push rod motor, and the push rod motor is fixed on the support platform. Every section all is equipped with infrared sensor on the guide rail that charges, can detect the effective contact of guide rail and unmanned aerial vehicle conducting ring that charges. When the unmanned aerial vehicle needs to be charged autonomously, the push rod motor is controlled to drive the two charging guide rails to move forwards transversely to be in contact with the conducting ring on the unmanned aerial vehicle, and then autonomous charging is started and the unmanned aerial vehicle is fixed; when unmanned aerial vehicle prepares to leave the platform, through control push rod motor, drive two guide rails that charge and transversely retreat, make it and unmanned aerial vehicle's conducting ring separation, the charging process finishes to realize flying of unmanned aerial vehicle.

4) The vision-based multi-sensor fusion algorithm: traditional unmanned aerial vehicle independently lands mostly mainly through GPS navigation realization, but solitary GPS navigation can not provide accurate positional information for unmanned aerial vehicle, under the environment of open-air complicacy, hardly realizes accurate landing moreover. In order to enable the unmanned aerial vehicle to land on a preset position accurately, the invention adopts a vision-based multi-sensor fusion algorithm to realize the automatic landing of the unmanned aerial vehicle.

Claims

1. an intelligent take-off and landing system of a mobile vehicle-mounted unmanned aerial vehicle, is characterized in that: this intelligent take-off and landing system comprises lifting platform device, unmanned aerial vehicle fixed and automatic charging device;

The lifting platform device is composed of three parts from top to bottom, a bracket platform (10), a Z-shaped lifting structure and a bracket base (18); the lower part of the bracket platform (10) is connected to the Z-shaped lifting structure;

The Z-shaped lifting structure includes eight identical lifting rods (13), three support beams (17), a lifting rod push rod (14), an electric push rod (15) and a number of screws; the lifting rod (13) has a total of eight and divided into two groups, each group of four, each group is placed on both sides of the bracket base (18) respectively, and the corresponding lifting rods (13) of each group are connected to each other through the bracket beam (17) and the lifting rod push rod (14) together; the four lifting rods (13) on each side are divided into upper and lower groups. Lifting structure; the electric push rod (15) is connected with the lifting rod (13) through the lifting rod push rod (14), and the lifting or lowering of the lifting rod (13) is realized by the electric push rod (15); the electric push rod (15) is connected with The lift rod push rod (14) is connected, the electric push rod (15) is installed on the control mechanism (16), and the control mechanism (16) and the electric push rod (15) are fixed together in the center groove ( 19);

The support platform (10) is a mesh-shaped rectangular platform, which is the support structure of the UAV; two lifting rod fixing seats (11) and two fixed sliding rails (12) are respectively installed at the bottom of the support platform (10) ; The lifting rod fixing seat (11) is placed on one side of the support platform (10) to play the role of fixing the fulcrum of the lifting rod (13); the fixed slide rail (12) is placed on the other side of the support platform (10), which is convenient for The lift rod (13) rises and falls synchronously;

The support base (18) is the base of the entire intelligent take-off and landing system; the central groove (19) is located in the middle of the support base (18);

The UAV fixing and automatic charging device consists of three parts: a visual positioning unit, a landing platform (2), and an automatic charging and fixing unit; the visual positioning unit includes a universal ball (3), a conductive ring (4), and a A vision processing module with an embedded chip, four universal balls (3) and conductive rings (4) are installed at the bottom of the landing gear of the drone; the automatic charging and fixing unit includes a charging guide rail (6), a push rod motor ( 9) and infrared sensor (7); the vision processing module is composed of two sensors, the GPS module (1) and the visual camera module (5), and the precise positioning of the UAV is realized through algorithm control;

The landing platform (2) is a mesh-shaped circular platform with a certain arc and smooth, and the landing platform is fixed on the support platform (10); a groove (8) is arranged in the center of the landing platform, and the charging guide rail (6) is installed in the concave. in the slot (8); the push rod motor (9) is fixed on the support platform (10), and the push rod motor (9) is connected with the charging guide rail (6) to drive the lateral movement of the charging guide rail (6), and the infrared sensor (7) ) is installed on the charging guide rail (6); the upper part of the support platform (10) is connected with a landing platform (2).

2. The intelligent take-off and landing system of a mobile vehicle-mounted unmanned aerial vehicle according to claim 1, is characterized in that: at first, four universal balls are installed at the bottom of the landing gear of the unmanned aerial vehicle, when the unmanned aerial vehicle falls on the When it is on a smooth landing platform, it can automatically slide into the groove of the landing platform through the rotation of the universal ball; The conductive ring of the drone is on the same plane as the charging rail in the groove, and every two adjacent machine feet form an electrode, which is connected to the battery of the drone; The vision processing module of the type chip; the vision processing module is composed of two sensors, the GPS module and the vision camera module, and the precise positioning of the drone is realized through algorithm control.

3. The intelligent take-off and landing system of a mobile vehicle-mounted unmanned aerial vehicle according to claim 1, is characterized in that: the radian of the landing platform is 15°; the center of the landing platform has a groove, and through the visual landing algorithm, no one The drone can accurately land in the groove of the platform, and even if the drone deviates when landing, the drone equipped with the universal ball can automatically slide into the groove.

4. The intelligent take-off and landing system of a mobile vehicle-mounted unmanned aerial vehicle according to claim 1, wherein the infrared sensor is used to detect the effective contact between the charging guide rail and the conducting ring of the unmanned aerial vehicle; when the unmanned aerial vehicle is about to be autonomous When charging, by controlling the push rod motor, the two charging rails are driven to move forward horizontally, contact with the conductive ring on the drone, and start charging and fix the drone; when the drone is ready to leave the platform, control the charging rail to move back horizontally. , separated from the conductive ring of the drone, the charging process is over, and the drone is released.

5. the intelligent take-off and landing system of a mobile vehicle-mounted unmanned aerial vehicle according to claim 1, is characterized in that: the electric push rod drives the zigzag lifting structure to raise the platform to 80cm place, and the unmanned aerial vehicle passes the vision-based The multi-sensor fusion algorithm is docked on the landing platform, and the drone installed with the universal ball slides freely on the platform, and finally stops in the groove of the landing platform; the charging rail in the groove advances laterally through the push rod motor, and when the infrared When the sensor detects the effective contact between the charging rail and the conducting ring of the drone, the drone starts to charge itself and is fixed; the Z-shaped lifting structure drives the platform down, and the entire intelligent take-off and landing system descends into the vehicle body; when the drone is ready When leaving the platform, the Z-shaped lifting structure drives the entire intelligent take-off and landing system to rise, and controls the charging guide rail to separate from the conductive ring of the drone. After the charging process is over, the drone is instructed to take off automatically from the platform.

6. the intelligent take-off and landing system of a kind of mobile vehicle-mounted unmanned aerial vehicle according to claim 1, is characterized in that:

Let P _∞ be the static pressure, ρ _∞ the airflow density, and V _∞ the airflow velocity, then the dynamic pressure

At this time, the air lift

Among them, C _l is the lift coefficient of the UAV, and S is the wing area;

The platform heights are 80cm and 30cm respectively; when the height of the platform from the ground is 80cm, the recoil force of the drone from the ground will reach a small value;

The implementation steps of the intelligent take-off and landing system are as follows:

Step 1: The stroke motor of the lifting platform device is designed to have a maximum extension of 80cm. When the drone is ready to land on the landing platform, the control mechanism of the electric push rod starts to control the extension of the push rod, and the mechanical structure will lift the entire intelligent The landing system is lifted slowly; when the platform reaches the maximum height, the intelligent take-off and landing system automatically stops moving, at this time, the platform is stable at 80cm;

Step 2: The drone is fixed and autonomously charged; the drone is docked on the take-off and landing platform through a vision-based multi-sensor fusion algorithm. Since the landing gear of the drone is equipped with universal balls, and the landing platform is smooth and has a certain arc Circular platform, the UAV overcomes the deviation of the landing point to a certain extent through the free sliding of the landing platform; at this time, the landing gear of the UAV is supported by the bracket platform, and the conductive ring and groove on the tripod The inner charging rails are in the same plane; when the drone is stabilized in the groove, the push rod motor is controlled to advance laterally, driving the charging rail to contact the conductive ring of the drone; when the infrared sensor on the charging rail detects the charging rail When it is in effective contact with the UAV conductive ring, start the UAV charging controller and start working, and the UAV has been fixed by the guide rail at this time;

Step 3: Landing the platform; in order to ensure the stability and safety of the drone on the intelligent take-off and landing system, the entire intelligent take-off and landing system is lowered into the vehicle body; the control mechanism of the electric push rod begins to control the shortening of the push rod, and the take-off and landing platform descends , and finally the entire intelligent take-off and landing system is included in the vehicle body, which not only saves space, but also protects the stability and safety of the drone during travel;

Step 4: The drone is released; when the drone is ready to leave the platform, the entire intelligent take-off and landing system is driven up again by controlling the lifting platform device. When the platform reaches the height that can be released, the electric push rod automatically stops moving; at this time , control the push rod motor to retreat laterally, and drive the charging guide rail to separate from the UAV conductive ring; after the charging process is over, the UAV achieves autonomous take-off.

7. The intelligent take-off and landing system of a mobile vehicle-mounted unmanned aerial vehicle according to claim 6, characterized in that: in step 2, a visual landing sign has been designed, through the adaptive threshold method in the algorithm and the relatively simple The landing mark recognition method can quickly detect the visual landing mark in real time, and can calculate the position parameters required by the UAV from the landing mark; and then convert the position information in the image into real position information through geometric operations and transmit it to the aircraft controller. to navigate;

First, a visual landing sign is designed on the landing platform of the intelligent take-off and landing system: a large outer circular frame and an inner small circular frame; the outer large circular frame is used to calculate the position parameters of the drone relative to the landing target, The inner small circular box is used to estimate the landing position of the drone; the visual landing mark is that the outer graphics and the inner graphics have a strong contrast; first, the drone uses the outer large circular box to estimate the position information of the drone ; When no one flies over the large outer circular frame mark, continue to estimate the position information of the UAV according to the inner small circular frame;

The camera is installed in the center of the drone in the vertical downward direction, so the position of the center point of the drone is replaced by the position of the center point of the camera, that is, the position of the optical center;

The adaptive threshold method is adopted to calculate the image h(i,j) of the visual landing sign map collected by the camera.

Among them, 0 is black and 1 is white; scan the image line by line, and then calculate the average value of the first S pixels of the current scanning point; when the pixel of this point is less than the average value of the first S pixels of the point (1-T ) times, or the pixel of this point is less than (1+T) times the average value of the pixels of the black pixels in the first S pixels of the point, then the pixel value of this point is determined to be 0; f(i, j) is the current pixel The pixel value of the point; Ds(i,j) represents the pixel average value of the S pixels before the current point; T is a value from 0 to 1; the values of S and T are S=width/8, and width represents the width of the image , when T=0.15, the effect is the best; R _b (i, j) represents the average value of pixels that are judged to be black pixels among the S pixels before the point, and b represents the number; The refined processing generates a high-quality binary image, which lays the foundation for the detection of the landing mark; after the contour determination is successful, the pixel offset x, y is obtained; the offset refers to the center point of the drone and the center point of the landing mark Since the control parameters of the aircraft controller to the aircraft are based on the real coordinate system, the calculated pixel offset should be converted into the real offset X, Y; the pixel offset is calculated by calculating the The difference between the coordinates of the center point of the landing mark and the coordinates of the center point of the image is obtained; the real offset is obtained by using the pixel offset by the following formula; the specific formula is:

Where x, y represent the pixel offset, X, Y represent the real offset, f _x , f _y represent the focal length in the x and y directions, c _x , c _y represent the camera optical center and image center point offset , H represents the height of the camera center point from the landing point.