Design of Fully Automatic Drone Parachute System with ...Design of Fully Automatic Drone Parachute System with Temperature Compensation Mechanism for Civilian and Military Applications

Research ArticleDesign of Fully Automatic Drone Parachute System withTemperature Compensation Mechanism for Civilian andMilitary Applications

Basem AL-Madani1 Marius Svirskis2 Gintautas Narvydas2

Rytis Maskeli0nas 3 and Robertas DamaševiIius 3

1Computer Engineering Department College of Computer Science and Engineering King Fahd University of Petroleum andMineralsPO Box 1195 Dhahran 31261 Saudi Arabia2Department of Automation Kaunas University of Technology StudentJ 48 111 51367 Kaunas Lithuania3Department of Multimedia Engineering Kaunas University of Technology Barsausko 59 A338 Kaunas Lithuania

Correspondence should be addressed to Robertas Damasevicius robertasdamaseviciusktult

Received 23 May 2018 Accepted 4 November 2018 Published 15 November 2018

Academic Editor Yair Wiseman

Copyright copy 2018 Basem AL-Madani et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Application of Unmanned Aerial Vehicles (aka drones) is becomingmore popular and their safety is becoming a serious concernDue to high cost of top-end drones and requirements for secure landing development of reliable drone recovery systems is a hottopic now In this paper we describe the development of a parachute system with fall detection based on accelerometer-gyroscopeMPU ndash 6050 and fall detection algorithm based on the Kalman filter to reduce acceleration errors while drone is flying Wedeveloped the compensation algorithm for temperature-related accelerometer errors The parachute system tests were performedfrom a small height on a soft surface Later the system was tested under real-world conditions The system functioned effectivelyresulting in parachute activation times of less than 05s We also discuss the civilian and military applications of the developedrecovery system in harsh (high temperature) environment

1 Introduction

Unmanned Aerial Vehicles (UAVs) popularly known asdrones are autonomously or remotely operated aircraftswhich can fly without on-board human pilot while beingoperated from the ground [1] Currently drones are formultiple purposes such as for entertainment (as a toy [2])for civilian applications (for example monitoring soil erosion[3] and wildlife [4] remote sensing [5] surveying and pho-togrammetry [6] smart agriculture [7] disaster surveyingand management [8] forestry [9] energy harvesting [10]and housing renovation [11]) scientific applications [12] andmilitary applications (surveillance and reconnaissance [13]and attacking the enemy [14 15])

As the evolution of drone technology continues itsbattery capacity and the range of the flight increase thereforealso increasing the range and scope of tasks drones can

be assigned to perform Statistics of accidents during theoperation of an Unmanned Aerial Vehicle (UAV) exhibitsa significantly higher accident rate compared to pilotedaircrafts [16] This poses severe limitations on the possibleadoption of unmanned systems particularly in the civilianair space During the flight the drones may face a varietyof problems both internal (eg an electrical circuit failurebroken connection or mechanical damage) and external(interference of an external obstacle such as birds or hostileforce) Due to these problems drones can become unman-ageable fall down and land on hard surface In such casesnot only the drone and its carrying equipment but alsothe objects below (eg people or valuable property) can bedamaged resulting in large financial losses in addition to lostequipment and information

Hence one important challenge is the implementationof technologies which can deal gracefully with failures and

HindawiJournal of Advanced TransportationVolume 2018 Article ID 2964583 11 pageshttpsdoiorg10115520182964583

2 Journal of Advanced Transportation

ensure safe operation even in the event of engine loss Inultralight aviation emergency parachutes have become animportant tool to avert damages and casualties caused byaircraft failures Parachute recovery is particularly suited totactical fixed wing UAV systems that require a high degreeof mobility by allowing air vehicle recovery onto unpreparedterrain [17] A higher attrition rate for UAVs compared tomanned aircraft can be due to both physical aircraft andhuman operator factors Given that the aircraft does not haveto accommodate a human being it can be made smaller Bymaking the aircraft smaller it is easier to stay undetected inthe sky for longer time compared to a larger aircraftHoweverdue to the size of the UAV it will encounter different envi-ronmental issues that designers and operators may not havehad experience with in manned aircraft thus requiring thedeployment of emergency recovery system such as parachute

Such problems could be avoided by using the drone-mounted parachute system which would expel parachuteduring the fall and ensure safe landing As in the courseof damage or hostile attack such as attempt of seizure theconnection of drone to the control centre may be lostand it may not be possible to activate the parachute bysending a signal from the control centre In addition thedrone on-board sensors could detect a possible damageor loss of equipment during flight much faster and morereliably especially at high altitudes Therefore an automaticparachute system is a desired solution for the protection ofthe UAV [18] It must also be reliable to eliminate possiblemistakes and unintentional firing of the parachute [19]because unnecessary parachuting at low altitude can damageobjects on the ground Avoiding such accidents requiresaccurate identification of the free falling state regardless ofthe axis and angle of the fall [20] but considering varyingairflow the drone is subject to during the flight is a consider-able challenge Analysing drone on-board sensor parameters(such as acceleration [21]) can help to avoid false parachutedeployment while it still ensures the protection of the dronefrom hitting the ground

Operational and exploration requirements for UAVsimpose in particular cases the use of rescue and recoverysystems These systems could possess a large range of con-structive solutions such as recover parachutes [22] airbags[23] safety nets [24] and pneumatic cushions boats [25]The rescue and recovery systems must allow a safe landingin case of emergencies and could become a solution to astandardization regarding landing in extreme weather andgeographical conditions [25]

The aim of this paper is to develop and experimentallyvalidate a drone-mounted parachute system capable of timelyparachuting and safely landing the drone in case of anaccident attack or loss of control

2 State-of-the-Art

Recovery of self-propelled flying drones in a limited space isa rather difficult task [26] During the halt problems suchas hovering before landing precise tracking of the UAVand changing weather conditions are encountered In thissituation there are several possible options for dropdowns

such as parachute network or hook-trapping The descentof these systems can be automated using evolutionary algo-rithms that help prepare the drone for landing braking beforelanding

Recognition of a critical failure (due to internal mal-function or interaction with an external object or force) isa distinctive problem in itself [27 28] Timely detection ofcritical failures can be used to enable activation of dronerecovery system in order to save an aircraft andor sensitivedata from being destroyed or falling into adversary handsSolutions include hardware redundancy to detect faultysensors or equipment [29] and analytical redundancy usingeg fuzzy logic to detect faulty sensor outputs [30] orcapturing an external view of an aircraft and subsequentlyapplying image processing techniques to detect damage to thebody or parts of an aircraft [28] However due to the weightsize and cost limitations it is not always possible to replicateor put additional equipment on a small UAV such as a drone

In case of critical failure determined the autonomousparachute deployment systems are being the primary dronerecovery method now [31ndash33] Autonomous Emergency sys-tems are responsible for deploying the parachute in case of acritical failure (eg lost radio control signal or electrical orengine failure) [34] The main advantage of the parachute-based drone recovery method is that the UAV may beretrieved at any time anywhere (although an accurate pointlanding is difficult) In addition if a parachute is usedas the primary recovery method it may also be used asthe emergency recovery method saving additional cost andweightThese systems are designed to safely locate aircrafts inemergency situations Usually the UAVflies autonomously toa preselected site and the parachute is deployed ensuring saferecovery Most parachutes are parafoil or cruciform shapedstored in the fuselage In cases where external sensitiveequipment is under the fuselage the parachute is deployedfrom the belly and the vehicle is inverted so that the partsunder the fuselage are protected from the impact A morecomplicated system for protecting the vehicle structure or anysensitive parts is the combination of a parachute (deployednormally on top of the fuselage) and an airbag which isinflated under the fuselage a few seconds before touchdownThe airbag may also be used to keep the UAV afloat whenlanded on water The main problem in parachute recoveryis the transition from horizontal to vertical motion of thevehicle which starts as soon as the parachute is deployedThen although parachute steering is possible the oscillatorymovements of the vehicle present problems in recovery suchas inaccurate point landing or touching down at differentangles which may damage the vehicle structure [35]

Alizadeh et al [23] suggested using airbag systems asa shock absorber for UAVs to assist with rapid parachutelandings An airbag system is activated on touchdown toabsorb the kinetic energy during the impact with the groundBleier et al [34] presented the risk assessment models forgeneration of flight paths which support the automaticdeployment of emergency parachutes for Unmanned AerialVehicles in emergencies due to loss of propulsion Basedon a risk analysis of the area underneath the flight pathsuitable deployment positions are identified whichminimize

Journal of Advanced Transportation 3

the chance of endangering humans on the ground propertydamage and loss of the air vehicle Du [36] introducedsimulation model of a certain kind UAVs for UAV recoverysystem design

Guo et al [37] provided six degrees of freedom (DOF)flight dynamics model of UAV and parachute recoverydynamic model for a quantitative prediction of the motionprocess ofUAVand recovery system and predicted dangeroussituations at recovery stage Kim et al [31] predicted theparachute deployment for landing at the desired point usingthe neural network and the flight conditions such as thedeployment position UAVrsquos velocity and wind velocity asinput data sets and landing points such as the cross rangeand the down range position as output data sets Li et al [32]analysed the motion characteristics of the parachute-UAVcombination in thewhole recycling process and a comparisonof the calculated results to the experimental results to verifythe reliability of the numerical simulation Shao et al [22]evaluated a model of the UAV-parachute system with windfields and a control strategy for recovery

Shyu and Hsiao [33] described the development ofmini-UAV that has the capabilities of bungee launch andparachute recovery Yong and Li [38] studied the optimalparachute decelerating trajectory of mini-UAVs which issolved by sequential quadratic programming method andgenetic algorithm while the results are used to guideresearch of parachute design Zeng and Cai [39] introducedthe parachute recovery system composed of fuselage doorhanger segregator shooting system main parachute drogueparachute unlocking system of doors and binding systemThe parachute system was optimized by wind tunnel testsfor the UAV being in unfavoured states by air-drop experi-ment

Morgan et al [40] explore the build and design of aballistic parachute recovery systemThis system will monitorseveral variables in real-time to determine whether or notthe aircraft is operating in a safe environment The elementsobserved are main battery voltage current GPS coordinatesand current acceleration If the system determines that theaircraft is operating in an unsafe environment the recoverysystem will cut main power and deploy a ballistic parachuteto guide the aircraft safely to the ground

Cristian and Codrea [41] designed parachute recoveryand landing attenuation system for a military reconnaissancedrone The recovery system is able to recover the air vehicleafter the complete mission when the vehicle has landed inrough terrain at altitudes from sea level to 3000 m recoverthe drone during uncontrolled flight conditions and serve assafety device to prevent the air vehicle leaving the boundariesof the zone

Safe landing problems are also relevant in rocket systems[42] Hybrid rockets can be used to control inflatable wingsThe advantage of such a system against a simple parachutesystem is that the landing is controlled and the inflatablewings help to travel to the desired landing location Theinflatable wings occupy a much smaller area than the foldsof similar dimensions However this type of wing system isdifficult to reconcile the wings are not very stable they needan inflated compressed gas control equipment

3 Method

31 Parachute Design Considerations The purpose of aparachute is to decrease speed while maintaining stability ofa falling vehicle or payload The aerodynamic and stabilitycharacteristics of the parachute are determined by its geomet-ric features which influence opening force and movementtrajectory [43] UAVs are subject to increasing demands andare committed to achieving maximum operational reliabil-ity Parachute systems are used to ensure the reliability ofoperation and the safety of the machine It is important toknow the characteristics of the parachute for proper landingParachutes can be of different types and their parameters ofresistance to wind also vary greatly Inflatable parachute issuitable for slowing down at high altitude where there is avery rare atmosphere Parachutes of this type are intended forlowering scientific equipment and measure the direction of ahorizontal moving windTherefore the release of a parachutemust be fast and efficient The effectiveness of the parachutespread depends on its size A simple parachute which doesnot need to be controlled gradually is effective only up to15 m in diameter Parachutes of this type are expandingvery quickly they can be used for discharging the openingcontainer from which the parachute simply falls out Also aspring exhaust mechanism can be used

Larger parachutes must withstand much higher overloaddue to high air resistance during take-off In order to avoidbreaking of the cord due to the occurrence of a shock it isnecessary to tighten the ropes before the parachute is pulledout One of the options is to use a rocket engine that pullsthe parachute forward However such a system takes up quitea lot of space which is difficult and not practical in smallaircraft Another option is to use a small parachute that dragsout a larger parachute All variations must have the appro-priate fall speed during the launch of the parachute Whendesigning a parachute several basic criteria need to be takeninto consideration simplicity of design and constructionresistance to tensile stress and high wind resistance

32 Mathematical Model of Parachute Fall It is difficult tomodel the dynamics of parachute ejection opening andfalling Usually some simplifications are adopted for exam-ple the parachute is considered to be a rigid body which isaffected by aerodynamic drag strains the connection riserand transmits a force to the payload with six degrees offreedom (see Figure 1)

The system of equations presented here is based on theprevious works of [43ndash46] and is defined as follows

1198911 = 1198721198912 = 119883 minus 119898119892 sin 1205791198913 = 119885 + 119898119892 cos 1205791198914 = 120579 minus 120574 minus 120572

(1)

here

119872 = 121205881198812119901119878119901119862119872 + 119865119877 cos 120572 + 119865119877 sin 120572


Fall

velo

city amp

angle

Dist

ance

to im

pact

- Tension Force

- Damping setting

Figure 1 Model of parachute system

119883 = 121205881198812119901119878119901119862119909 minus 119865119877 cos 120572

119885 = 121205881198812119901119878119901119862119911 minus 119865119877 sin 120572

(2)

whereM is the moment X Y and Z are the body forces 119881119901is velocity S is parachute surface area 119865119877 is the riser force(tension) 120572 is attack angle and 120573 is sideslip angle

The deceleration of the parachute is calculated as follows

119889119881119889119905 =

119865119877 minus 119865119863 minus 119898119901119892 sin 120574(119898119901 + 119898119889) (3)

where 119898119901 is the parachute mass 119898119889 is the drone mass 119865119877is riser force 119865119863 is parachute drag force and 120574 is flight pathangle and a is acceleration

The drag force of the parachute is calculated as follows

119865119863 = 121205781205881198812119901119863119878119901 (4)

where119863 is drag coefficient 119878119901 is projected parachute area 119881119901is parachute velocity 120588 is air density and 120578 is an efficiencyfactor

As a result the following system of equations is con-structed

The falling speed is from the body velocities in three axes

119881 = radic1198812119909 + 1198812119910 + 1198812119911 (5)

The deceleration is calculated as the derivative of the velocityvector

119889119881119889119905 =

119881119909119889119881119909119889119905 + 119881119910119889119881119910119889119905 + 119881119911119889119881119911119889119905119881 (6)

The angle of attack is given by

120572 = tanminus1 (1003816100381610038161003816100381610038161003816100381611988111991111988111990910038161003816100381610038161003816100381610038161003816) (7)

The angle of sideslip is derived by calculating

120573 = sinminus1 (100381610038161003816100381610038161003816100381610038161003816119881119910119881100381610038161003816100381610038161003816100381610038161003816) (8)

33 Calculation of Drone Descent Speed Drone speed iscalculated in several steps In order to knowwhat accelerationthe objectmoves regardless of the accelerometer position thesum of the acceleration values for all axes is expressed as

119866119909119910119911 = radic1198662119909 + 1198662119910 + 1198662119911 (9)

where 119866119909 119866119910 119866119911 are the accelerometer values of theaccelerometer axes (x y z)

Assuming that v0 = 0 m s we can calculate the total fallspeed as follows

119881119896119903 = int119905119896119903

0

100381610038161003816100381610038161003816997888rarr119866100381610038161003816100381610038161003816 119889119905 = Δ119905

119905119896119903sum0

119866119909119910119911 (10)

where 119905119896119903 is the time after which it is checked or the fall speedhas not exceeded the set limit

To eject the parachute emission we apply the conditionthat it must reach a certain speed (15 ms) or more withina specified time interval (025 s) Otherwise the parachutewill not be ejected Under such conditions the drone is notallowed to reach a dangerous fall speed

34 Parameter Control In order to understand the causes ofa drone accident and to identify the distinctive features it isnecessary to analyse previous accidents aiming to find possi-ble solutions to problemsHaving evaluated the real situationsand in order for the algorithm to be able to distinguish thefall data from the movement data the fall parameters mustbe properly described By measuring accelerometer readingsand evaluating parameters with a certain parameter limit itis possible to detect the moment of falling

The system must clearly detect the drop and the riseregardless of the drone orientation in the space [47] Whenknowing the exact moment from when the object began tofall and when it reached the ground it is possible to calculatethe height of the fall Detection is also relevant to protectingelectronics from potential damage during impact Detectinga fall quickly is especially important in cases where theoperation has a very low height This requires an optimizedprogram code and an appropriate microcontroller capable ofprocessing information as quickly as possible

To determine the position and height change it is nec-essary to process the received sensor readings and to rejectincorrect values and calculate speeds at certain time intervalsPosition tracking algorithms with accelerometers can beused in situations where high precision is not required Theintegration algorithm is ideal for low-cost devices due to its


0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

Acce

lera

tion

(g)

Acce

lera

tion

(g)

Temperature (∘C)Temperature (∘C)

Figure 2 Dependence of the accelerometer Z-axis values (1g ndash 981 ms2) upon the temperature during rapid (5∘Cmin) heating (left) andcooling (right)

simplicity Also for all speeds all calculations are performedusing only floating point operations The data is processedusing a low frequency filter Several different sensors can beused to increase the positioning accuracy

35 Calibration of Accelerometer Accurate accelerometervalues are particularly important for drone speed detectionDue to temperature accelerometer chip manufacturing inac-curacies and different stresses occurring during solderingand assembly deviation of accelerometer values occursMini-mal andmaximum values of the scanned values are processedby the Kalman filter [48] (the delay in the submission ofdata is not affected during the calibration) The filtered valuesare compared and recorded at the highest Subsequently thesubroutine is stopped the accelerometer is overturned theaccelerometer axis being measured in the opposite directionand then the scan is continued The same calibration steps arerepeated for each axis

Accelerometers and gyros have high noise and temper-ature variations Therefore it is not enough to take thevalues of the sensor data alone but we also need to knowthe environmental conditions do the sensor values could becompensated for temperature compensation It is necessaryto measure the sensor values at different temperatures inorder to determine possible errors

A device for adjusting the temperature was made forcalibration of the accelerometer The thermoelectric heatingand cooling element was used for this purpose operatingunder the Peltier effect principle Also used are the temper-ature sensor DS18B20 the radiator for excessive heat or coldof the thermoelectric element a fan a 12V power supplya bidirectional motor control TB6612FNG and an Arduinomicrocontroller

By inserting the accelerometer into the temperaturecontrol chamber first of all the chamber is positioned sothat the axis measured by the sensor is perpendicular to theground The fan is mounted on the stand so that it does nottouch the cooling system at all and does not affect the sensorrsquosmeasurement data Then the sensor is cooled to 0∘C thenheated to 50∘C and then cooled again to 0∘C The process is

performed 10 times for each axis of the sensor in a positionthat maximizes the values of that axis

A collection of values has been created in the processingenvironment which reads the values through a serial dataoutput and writes them to a text file The program readssensor values and checks the conditions to avoid false valueschecking whether the value arrives and whether it has morethan 30 ldquocharrdquo elements If conditions are met values arewritten to the text document This cycle is repeated until theprogram is stopped (Figure 2)

Figure 2 shows the deviations of the accelerometer val-ues depending upon the temperature values We can seethat the acceleration values vary greatly We assumed theaccelerometer can not warm up or cool down evenly due torapid heating or cooling (5∘Cmin) Therefore it is necessaryto significantly reduce the temperature variation in thetemperature control chamber By reducing the temperaturechange to 1∘C per minute we obtained the following heatingand cooling charts (Figure 3)

From the graph (Figure 3) using linear regression wehave derived the following linear functions of the changeof acceleration during the cooling and heating process asfollows

119866119911 = minus00047119879 + 11031119866119911 = minus00056119879 + 10967

(11)

where119866119911 is the acceleration measured by the accelerometerrsquosZ-axis

The initial and final values of the graphs are similarTherefore we can assume that during the cooling time theaccelerometer values need to be further elaborated Sinceby heating the values change evenly without any significantdistortion we assume that deviations of the Z-axis of thesensor from the norm can be compensated by using thisfunction The same test is performed with other axes (see theresults presented in Figures 4 and 5)

From Figures 4 and 5 we can note that the temperature-induced variations on the X-axis and the Y-axis are con-siderably smaller than the ones on the Z-axis The Z-axis


0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

Figure 3 Dependence of the accelerometer Z-axis values (1g ndash 981 ms2) upon the temperature during slow (1∘Cmin) heating (left) andcooling (right)

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

Figure 4 Dependence of the accelerometer X-axis values (1g ndash 981 ms2) upon the temperature during slow (1∘Cmin) heating (left) andcooling (right)

temperature compensation is derived from the heating valuefunction and the temperature compensation formula is asfollows

119866119911 119888119900119898119901 = 119866119911 + 00047119879 minus 01031 (12)

where 119866119911 119888119900119898119901 is the compensated acceleration value 119866119911 isthe measured value of the accelerometer Z-axis and T is thetemperature

From the graphs of compensated values (Figure 6) we canconclude that the deviations of the values are compensatedHowever due to differences in heating and cooling (at thesame temperature) the values are compensated to a certainlimit until themaximumdeviation reaches 5Thedeviationof the values before compensation was 14

4 Drone Design and Testing

41 Settings Drone testing was performed under real condi-tions (see Figure 7) The drone frame consisted of a balsamtree reinforced with a glass fibre This type of frame isextremely lightweight and has low production and repaircosts The electric motor bearing structures consisted ofaluminium hollow bars The Emax company controller wasselected for drone engine management The drone is con-trolled using the 9-channel receiver using the S-Bus type

protocol between the receiver and the master controllerThe main controller has several sensors an accelerometer amagnetometer a barometer and a gyroscope In order forthe drone to fly stably it was necessary for this controller toadjust the PID parameters The settings have been adjustedmanually

42 Electronics A system for unlocking the parachute ejec-tion mechanism was created (Figure 8) which deploys thespring holding rope This system is particularly lightweightand has a small additional space since it no longer requiresa heavy servo motor The Nichrome wire is used to burnthe cord A MOS type transistor is used to prevent the cordfrom slipping off when the current is started Transistorbandwidth is controlled by a PWM type signal The systemis managed by the Arduino microcontroller which processesthe sensor data and controls the outputs A battery of 110mAh lithium polymer has been selected for power supplyIts charging system has an integrated battery charger withbattery protection from excessive discharge The battery ischarged through the micro USB connector The RGB LEDis used for indicating the operation The LED at the start ofthe system in colour indicates the level of the battery voltageand informs about the operation of the system and the falldetection


0 10 20 30 40 50

12

115

11

105

1

095

09

085

080 10 20 30 40 50

12

115

11

105

1

095

09

085

08

Figure 5 Dependence of the accelerometer Y-axis values (1g ndash 981 ms2) upon the temperature during slow (1∘Cmin) heating (left) andcooling (right)

0 10 20 30 40 50

12

115

11

105

1

095

09

085

0 10 20 30 40 50

12

115

11

105

1

095

09

085

Figure 6 Dependence of the compensated accelerometer Z-axis values (1g ndash 981 ms2) upon the temperature during slow (1∘Cmin) heating(left) and cooling (right)

Figure 7 Ready-to-fly drone

43 Control Algorithm For fall detection we have adoptedan approach that is based on the detection of the falling stateof the drone The approach in itself in not new [49] and weselected it due to its simplicity however we have improvedthe approach by introducing compensation for temperaturecaused variability The parachute system program (Figure 9)measures the battery voltage then the accelerometer andgyroscope data The Z-axis data on the accelerometer iscompensated by the temperature-dependent deviation The

1

2 4

3

56

7

8

Figure 8 Parachute system electronic circuit diagram 1 ndash chargingunit 2 - switch 3 ndash Lithiumpolymer battery 4 -microcontroller 5 ndashaccelerometer - gyroscope 6 ndash LED diode 7 ndashMOS type transistor8 ndash nichrome wire

axis values are filtered using a Kalman filter Using the filtereddata the combined acceleration of all axles is calculatedThen the condition is checked whether the speed limit isfixed If the limit is fixed - the time from the moment ofcapture is measured After a limited time the fall accelerationis checked again If the condition fits the parachute is ejectedAlso one of the main conditions for detecting the fall state orthe unstable operation of a drone is the measurement of the


START

Is voltagegt38V

Initiate sensor readout

Calculate acceleration

on all axis

Is accel Value equal to fall

value

Initiate angle readout

Is angle gt90o

Deadlock on

Drop the parachute

Read accel data

End of programme

First variable is marked as 0

Is this an initial accel Readout

Start calculating time set first variable to 1

Read angledata

No

Yes

YesNo

No

No

Yes

Yes

No

Yes

Figure 9 Parachute system operation algorithm

angle of inclinationWhen the limit value (at 90∘ ) is exceededthe parachute is ejected If no condition is satisfied the cycleis repeated until the condition is satisfied or the shutdownbutton is depressed

44 Parachute Exhaust System The parachute was madefrom a rugged nylon material Parachute dimensions arechosen such that a 2 kg drone does not fall faster than 5 msUnder the selected dimensions a 16-segment dome-shapedparachute with overlock stitches was made (Figure 10) Thediameter of the parachute is 145 m

The parachute exhaust housings (Figure 11) were madeof PVC pipe glass fibre board and a bottom-mountedmounting piece made of three-dimensional printing

45 Testing To determine the efficiency of the parachutesystem we attempted to detect a fall without the drone itself

(the system was thrown onto a soft surface) First of allthe simple fall detection algorithm was tested when theaverage of the total values of the acceleration was 15 andthe threshold was applied to it When crossing the limitthe parachute system was triggered The system workedreliably However when placed on the drone no fall wasrecorded due to excessive vibrations Using the Kalman filterfor this algorithm the system worked efficiently The fallwas recorded in 05 s In order to prevent the unintentionalejection the fall acceleration was checked for a second timeafter a specified period of time Only in accordance withthe second measurement condition the parachute is to beopened During critical conditions when vibrations haveextremely high amplitude and frequency both Kalman andthe compliant filter failed to properly filter the errors As aresult the fall was sometimes erroneously detected Howeverin real conditions we have not detected any false falls The


Figure 10 The 16-segment dome-shaped parachute

Figure 11 Parachute system body

operating parachute system during opening stage is depictedin Figure 12

During the tests an attempt was made to accelerate theparachute opening and the parachute was covered with Talcpowder from the bonding of thematerial A special parachutefolding technique is used when the parachute is folded insuch a way as to spill as soon as possible As a result foldedparachute can open much faster However the minimumheight required for the parachute to landwas about 20m fromthe ground

5 Evaluation

We evaluate the developed UAV parachute system accordingto the requirements for UAV recovery systems formulated byWyllie [17] as follows

(i) Safety a mandatory requirement during landing(ii) Protection protect drone and sensitive on board

equipment from damage during landing(iii) Accuracy ensure high accuracy when landing to a

planned landing point to minimize landing damage(iv) Automation reduce the number of manual opera-

tions by the drone operator(v) Reliability ensure predictability of operation(vi) Repeatability ensure that recovery may be repeated

many times during lifetime of the drone

Since the areaswhere drones are used are expandingwith newapplications proposed every year it is important that dronescontinue to be used both safely and efficiently in multitude

Figure 12 System testing under real conditions

of different environments while ensuring the avoidance ofobstacles [50] For example the meteorology drones collectdata on air temperature humidity pressure wind force radi-ation etc and should be able to withstand the adverse forcesof nature such as strong wind gusts If flying at high altitudesalso the temperature factor should be taken into account [51]There are significant challenges related to predicting tem-perature variations when using drones in Arctic [52] wheretemperature may fall below -50∘C or in deserts [53] wheretemperature may exceed +50∘C but also for more commoncivilian applications executed during nigh time [54] or forforest firemonitoring [55] when temperatures may fall belowor well exceed the range of temperatures most consumerdrones are designed for which is usually 0∘C ndash 40∘C Inthis paper we have addressed the challenge of designing areliable parachute-based recovery system for a drone whichuses the temperature-compensated accelerometer data forparachute ejection Note that still many other temperature-related challenges for drone designers exist most notablythe impact of temperature on the battery performance andcharacteristics

6 Conclusion

Wehave proposed a parachute-based system for drone recov-ery with the accelerator sensor temperature compensationmechanism The parachute system has been successfullytested in real-life conditions The fall is detected almostimmediately (within 05 s) However it takes muchmore timefor the parachute to unfold so at less than 20 m height theparachute still may not be able to unfold in time The systemalso includes a mechanism for prevention of unintendedparachute deployment In calculating the rate of fall theresulting acceleration errors even after compensation werehighThe calculated speed data could not be relied upon evenat small intervals assuming that an initial speed of falling isequal to 0 ms The acceleration deviations can be offset byapplying the Kalman filter

The accelerometer values depend significantly upon theenvironment temperature so the compensation mechanismbased on linear regression was introduced to allow stabledetection of drone fall state during harsh environment con-ditions such as in a desert environment during a hot sum-mer


Data Availability

The data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

The authors declare that they have no conflicts of interest

Acknowledgments

The authors acknowledge the support by King Fahd Univer-sity of Petroleum amp Minerals (KFUPM) in publishing thiswork

References

[1] E N Barmpounakis E I Vlahogianni and J C Golias ldquoUn-manned Aerial Aircraft Systems for transportation engineeringCurrent practice and future challengesrdquo International Journal ofTransportation Science and Technology vol 5 no 3 pp 111ndash1222016

[2] C-H Poh andC-K Poh ldquoRadioControlled ldquo3DAerobatic Air-planesrdquo as Basis for Fixed-Wing UAVs with VTOL CapabilityrdquoOpen Journal of Applied Sciences vol 04 no 12 pp 515ndash5212014

[3] S DrsquoOleire-Oltmanns I Marzolff K D Peter and J B RiesldquoUnmanned aerial vehicle (UAV) for monitoring soil erosionin Moroccordquo Remote Sensing vol 4 no 11 pp 3390ndash3416 2012

[4] M E Goebel W L Perryman J T Hinke et al ldquoA smallunmanned aerial system for estimating abundance and size ofAntarctic predatorsrdquo Polar Biology vol 38 no 5 pp 619ndash6302015

[5] K Whitehead and C H Hugenholtz ldquoRemote sensing of theenvironment with small unmanned aircraft systems (UASs)part 1 a review of progress and challengesrdquo Journal of Un-manned Vehicle Systems vol 02 p 69 2014

[6] F Cuesta F Lopez-Rodriguez and A Esteban ldquoA New BlondinSystem for Surveying andPhotogrammetryrdquo Sensors vol 13 no12 pp 16894ndash16914 2013

[7] E Honkavaara H Saari J Kaivosoja et al ldquoProcessing andassessment of spectrometric stereoscopic imagery collectedusing a lightweight UAV spectral camera for precision agricul-turerdquo Remote Sensing vol 5 no 10 pp 5006ndash5039 2013

[8] Z Xu J Yang C Peng et al ldquoDevelopment of an UAS forpost-earthquake disaster surveying and its application inMs70Lushan Earthquake Sichuan ChinardquoComputers ampGeosciencesvol 68 p 22 2014

[9] L Tang and G Shao ldquoDrone remote sensing for forestryresearch and practicesrdquo Journal of Forestry Research vol 26 no4 pp 791ndash797 2015

[10] Z Zhao X Pu C Du et al ldquoFreestanding Flag-Type Triboelec-tric Nanogenerator for Harvesting High-AltitudeWind Energyfrom Arbitrary Directionsrdquo ACS Nano vol 10 no 2 pp 1780ndash1787 2016

[11] A AMustaffaM FHasmori A S Sarif N F Ahmad andN YZainun ldquoTheUse ofUAV inHousingRenovation IdentificationA Case Study at Taman Manis 2rdquo IOP Conference Series Earthand Environmental Science vol 140 p 012003 2018

[12] K Whitehead C H Hugenholtz S Myshak et al ldquoRemotesensing of the environment with small unmanned aircraft

systems (UASs) part 2 scientific and commercial applicationsrdquoJournal of UnmannedVehicle Systems vol 02 no 03 pp 86ndash1022014

[13] C Paucar L Morales K Pinto et al ldquoUse of drones for surveil-lance and reconnaissance of military areasrdquo in Developmentsand Advances in Defense and Security MICRADS 2018 SmartInnovation Systems and Technologies A Rocha and T GuardaEds vol 94 pp 119ndash132 Springer Cham 2018

[14] J Drew ldquoDrone strike Long-range attack UAVs being devel-oped from aerial targetsrdquo Aviation Week amp Space Technologyvol 179 no 4 pp 42-43 2017

[15] D Hambling ldquoMilitary tackles the problem and the potential ofdronesrdquo Aviation Week amp Space Technology vol 180 no 8 pp44ndash47 2018

[16] GWild J Murray and G Baxter ldquoExploring Civil Drone Acci-dents and Incidents to Help Prevent Potential Air DisastersrdquoAerospace vol 3 no 3 p 22 2016

[17] T Wyllie ldquoParachute recovery for UAV systemsrdquo AircraftEngineering and Aerospace Technology vol 73 no 6 pp 542ndash551 2001

[18] K Graham and S Cartwright Feasibility of Parachute RecoverySystems for Small UAVs University of New South Wales NewSouth Wales 2008

[19] D Lim C Park N H Kim S-H Kim and Y S Yu ldquoFall-Detection Algorithm Using 3-Axis Acceleration Combinationwith Simple Threshold and Hidden Markov Modelrdquo Journal ofAppliedMathematics vol 2014 Article ID 896030 8 pages 2014

[20] J Luo B Zhong and D Lv ldquoFall Monitoring Device for OldPeople based on Tri-Axial Accelerometerrdquo International Journalof Advanced Computer Science and Applications vol 6 2015

[21] K Seifert ldquoOscar Camacho Implementing Positioning Algo-rithms Using AccelerometersrdquoApplication Note AN3397 Rev 0Freescale Semiconductor 2007

[22] P Shao C Wu and S Ma ldquoResearch on key problems inassigned-point recovery of UAV using parachuterdquo in Proceed-ings of the 2013 IEEE International Conference of IEEE Region 10(TENCON 2013) pp 1ndash4 2013

[23] M Alizadeh A Sedaghat and E Kargar ldquoShape and orificeoptimization of airbag systems for UAV parachute landingrdquoInternational Journal of Aeronautical and Space Sciences vol 15no 3 pp 335ndash343 2014

[24] S Ramasamy R Sabatini and A Gardi ldquoA Unified AnalyticalFramework for Aircraft Separation Assurance and UAS Sense-and-Avoidrdquo Journal of Intelligent amp Robotic Systems vol 91 no3-4 pp 735ndash754 2018

[25] V Prisacariu ldquoRecovery System Of The Multi-Helicopter UavrdquoReview of the Air Force Academy vol 1 no 31 pp 91ndash98 2016

[26] S Khantsis Control system design using evolutionary algorithmsfor autonomous shipboard recovery of unmanned aerial vehiclesDoctor of Philosophy (PhD) Aerospace Mechanical andManu-facturing Engineering RMIT University 2006

[27] C H Lo E H Fung and Y K Wong ldquoIntelligent automaticfault detection for actuator failures in aircraftrdquo IEEE Transac-tions on Industrial Informatics vol 5 no 1 pp 50ndash55 2009

[28] Y Wiseman ldquoDevice for detection of fuselage defective partsrdquoInternational Information Institute (Tokyo) Information vol 17no 9 pp 4189ndash4194 2014

[29] D Rehage U B Carl and A Vahl ldquoRedundancy manage-ment of fault tolerant aircraft system architectures - Reliabilitysynthesis and analysis of degraded system statesrdquo Journal ofAerospace Science and Technology vol 9 no 4 pp 337ndash3472005


[30] W A Kwong K M Passino E G Laukonen and S YurkovichldquoExpert supervision of fuzzy learning systems for fault toleranceaircraft controlrdquo Proceedings of the IEEE vol 83 no 3 pp 466ndash482

[31] I Kim S Park W Park and C-K Ryoo ldquoPrediction of theparachute deploy for landing at the desired pointrdquo in Proceed-ings of the 2012 Proceedings of SICE Annual Conference SICE2012 pp 1629ndash1634 IEEE Akita Japan August 2012

[32] M Li H Zhang and J Chen ldquoDynamic analysis for parachuterecovery system of unmanned aerial vehiclerdquo Journal of SystemSimulation vol 24 no 6 pp 1303ndash1307 2012

[33] L-S Shyu and Y-C Hsiao ldquoMini UAV design andmanufacturewith bungee launched Parachute recoveryrdquoApplied Mechanicsand Materials vol 610 pp 97ndash100 2014

[34] M Bleier F SetteleM Krauss A Knoll and K Schilling ldquoRiskassessment of flight paths for automatic emergency parachutedeployment in UAVsrdquo IFAC-PapersOnLine vol 28 no 9 pp180ndash185 2015

[35] W J Crowther and K Prassas ldquoPost Stall Landing for FieldRetrieval of UAVsrdquo in Proceedings of the 14th Bristol Interna-tional unmanned air vehicle systems conference 1999

[36] KMDu ldquoSimulation analysis on the problem of openingmainparachute used in UAV recovery systemrdquo Applied Mechanicsand Materials vol 390 pp 57ndash61 2013

[37] L Guo H Zhang and M Tong ldquoDynamics analysis onparachute recovery of unmanned aerial vehiclerdquo Journal ofNanjing University of Aeronautics and Astronautics vol 44 no1 pp 14ndash19 2012

[38] Z Yong and L Li ldquoParachute Decelerate Trajectory Optimiza-tion Design Based on Genetic Algorithmrdquo in Proceedings of the2009 International Workshop on Intelligent Systems and Appli-cations ISA 2009 IEEE Wuhan China 2009

[39] L Zeng and J Q Cai ldquoScheme design of a parachute recoverysystem for UAVrdquo in Proceedings of the 64th International Astro-nautical Congress 2013 IAC 2013 pp 8404ndash8412 China 2013

[40] A Morgan and R Chapman Ballistic Parachute Recovery Sys-tem for Unmanned Aerial Vehicles (UAVs) Make Magazine2016

[41] A-I Cristian and S Codrea ldquoParachute deployment systemsfor drones applied on uav-sact projectrdquo in 15th InternationalConference The Knowledge-Based Organization Applied Techni-cal Sciences and AdvancedMilitary Technologies vol 6 pp 260ndash266 2009

[42] C Sudduch Design of a Hybrid RocketInflatable Wing UAVOklahoma State University USA 2012

[43] G Guglieri ldquoParachute-Payload System Flight Dynamics andTrajectory Simulationrdquo International Journal of Aerospace Engi-neering vol 2012 Article ID 182907 17 pages 2012

[44] K-F Doherr and H Schilling ldquoNine-degree-of-freedom simu-lation of rotating parachute systemsrdquo Journal of Aircraft vol 29no 5 pp 774ndash781 1992

[45] V N Dobrokhodov O A Yakimenko and C J Junge ldquoSix-degree-of-freedom model of a controlled circular parachuterdquoJournal of Aircraft vol 40 no 3 pp 482ndash493 2003

[46] X Gao Q Zhang and Q Tang ldquoParachute dynamics andperturbation analysis of precision airdrop systemrdquo ChineseJournal of Aeronautics vol 29 no 3 pp 596ndash607 2016

[47] M Clifford ldquoDetecting Freefall with Low-G AccelerometersrdquoSensor and Analog Products Division Tempe AZ AN3151 Rev 02006

[48] E A Wan and R Van Der Merwe ldquoThe unscented Kalmanfilter for nonlinear estimationrdquo in Proceedings of the IEEEAdap-tive Systems for Signal Processing Communications and Con-trol Symposium (AS-SPCC 2000) pp 153ndash158 IEEE AlbertaCanada October 2000

[49] K L Crandall and M A Minor ldquoUAV fall detection from adynamic perch using Instantaneous Centers of Rotation andinertial sensingrdquo in Proceedings of the 2015 IEEE InternationalConference on Robotics and Automation ICRA 2015 pp 4675ndash4679 USA 2015

[50] A Ivanovas A Ostreika R Maskeliunas R DamaseviciusD Połap and M Wozniak ldquoBlock Matching Based ObstacleAvoidance for Unmanned Aerial Vehiclerdquo in Artificial Intelli-gence and Soft Computing pp 58ndash69 Springer InternationalPublishing 2018

[51] E H Teets Jr C J Donohue K Underwood and J E BauerldquoAtmospheric considerations for uninhabited aerial vehicle(UAV) flight test planningrdquo NASATM-1998-206541 1998

[52] M Ader and D AxelssonDrones in Arctic Environments KTHStockholm Sweden 2017

[53] D Gallacher ldquoApplications of micro-UAVs (drones) for desertmonitoring current capabilities and requirementsrdquo in Proceed-ings of the Kubuqi International Desert Forum 2015 pp 249ndash2592015

[54] S R Herwitz K Allmendinger R Slye et al ldquoNighttime UAVVineyard Mission Challenges of See-and-Avoid in the NASrdquoin Proceedings of the AIAA 3rd ldquoUnmanned Unlimitedrdquo Techni-cal Conference Workshop and Exhibit Chicago Illinois USASeptember 2004

[55] P Skorput S Mandzuka and H Vojvodic ldquoThe use of Un-manned Aerial Vehicles for forest fire monitoringrdquo in Proceed-ings of the 58th International Symposium ELMAR ELMAR 2016pp 93ndash96 Croatia September 2016

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018


Active and Passive Electronic Components

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of



Journal ofEngineeringVolume 2018

SensorsJournal of



RotatingMachinery


Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation


Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom


ensure safe operation even in the event of engine loss Inultralight aviation emergency parachutes have become animportant tool to avert damages and casualties caused byaircraft failures Parachute recovery is particularly suited totactical fixed wing UAV systems that require a high degreeof mobility by allowing air vehicle recovery onto unpreparedterrain [17] A higher attrition rate for UAVs compared tomanned aircraft can be due to both physical aircraft andhuman operator factors Given that the aircraft does not haveto accommodate a human being it can be made smaller Bymaking the aircraft smaller it is easier to stay undetected inthe sky for longer time compared to a larger aircraftHoweverdue to the size of the UAV it will encounter different envi-ronmental issues that designers and operators may not havehad experience with in manned aircraft thus requiring thedeployment of emergency recovery system such as parachute

Such problems could be avoided by using the drone-mounted parachute system which would expel parachuteduring the fall and ensure safe landing As in the courseof damage or hostile attack such as attempt of seizure theconnection of drone to the control centre may be lostand it may not be possible to activate the parachute bysending a signal from the control centre In addition thedrone on-board sensors could detect a possible damageor loss of equipment during flight much faster and morereliably especially at high altitudes Therefore an automaticparachute system is a desired solution for the protection ofthe UAV [18] It must also be reliable to eliminate possiblemistakes and unintentional firing of the parachute [19]because unnecessary parachuting at low altitude can damageobjects on the ground Avoiding such accidents requiresaccurate identification of the free falling state regardless ofthe axis and angle of the fall [20] but considering varyingairflow the drone is subject to during the flight is a consider-able challenge Analysing drone on-board sensor parameters(such as acceleration [21]) can help to avoid false parachutedeployment while it still ensures the protection of the dronefrom hitting the ground

Operational and exploration requirements for UAVsimpose in particular cases the use of rescue and recoverysystems These systems could possess a large range of con-structive solutions such as recover parachutes [22] airbags[23] safety nets [24] and pneumatic cushions boats [25]The rescue and recovery systems must allow a safe landingin case of emergencies and could become a solution to astandardization regarding landing in extreme weather andgeographical conditions [25]

The aim of this paper is to develop and experimentallyvalidate a drone-mounted parachute system capable of timelyparachuting and safely landing the drone in case of anaccident attack or loss of control

2 State-of-the-Art

Recovery of self-propelled flying drones in a limited space isa rather difficult task [26] During the halt problems suchas hovering before landing precise tracking of the UAVand changing weather conditions are encountered In thissituation there are several possible options for dropdowns

such as parachute network or hook-trapping The descentof these systems can be automated using evolutionary algo-rithms that help prepare the drone for landing braking beforelanding

Recognition of a critical failure (due to internal mal-function or interaction with an external object or force) isa distinctive problem in itself [27 28] Timely detection ofcritical failures can be used to enable activation of dronerecovery system in order to save an aircraft andor sensitivedata from being destroyed or falling into adversary handsSolutions include hardware redundancy to detect faultysensors or equipment [29] and analytical redundancy usingeg fuzzy logic to detect faulty sensor outputs [30] orcapturing an external view of an aircraft and subsequentlyapplying image processing techniques to detect damage to thebody or parts of an aircraft [28] However due to the weightsize and cost limitations it is not always possible to replicateor put additional equipment on a small UAV such as a drone

In case of critical failure determined the autonomousparachute deployment systems are being the primary dronerecovery method now [31ndash33] Autonomous Emergency sys-tems are responsible for deploying the parachute in case of acritical failure (eg lost radio control signal or electrical orengine failure) [34] The main advantage of the parachute-based drone recovery method is that the UAV may beretrieved at any time anywhere (although an accurate pointlanding is difficult) In addition if a parachute is usedas the primary recovery method it may also be used asthe emergency recovery method saving additional cost andweightThese systems are designed to safely locate aircrafts inemergency situations Usually the UAVflies autonomously toa preselected site and the parachute is deployed ensuring saferecovery Most parachutes are parafoil or cruciform shapedstored in the fuselage In cases where external sensitiveequipment is under the fuselage the parachute is deployedfrom the belly and the vehicle is inverted so that the partsunder the fuselage are protected from the impact A morecomplicated system for protecting the vehicle structure or anysensitive parts is the combination of a parachute (deployednormally on top of the fuselage) and an airbag which isinflated under the fuselage a few seconds before touchdownThe airbag may also be used to keep the UAV afloat whenlanded on water The main problem in parachute recoveryis the transition from horizontal to vertical motion of thevehicle which starts as soon as the parachute is deployedThen although parachute steering is possible the oscillatorymovements of the vehicle present problems in recovery suchas inaccurate point landing or touching down at differentangles which may damage the vehicle structure [35]

Alizadeh et al [23] suggested using airbag systems asa shock absorber for UAVs to assist with rapid parachutelandings An airbag system is activated on touchdown toabsorb the kinetic energy during the impact with the groundBleier et al [34] presented the risk assessment models forgeneration of flight paths which support the automaticdeployment of emergency parachutes for Unmanned AerialVehicles in emergencies due to loss of propulsion Basedon a risk analysis of the area underneath the flight pathsuitable deployment positions are identified whichminimize








3 Method






(1)

here

119872 = 121205881198812119901119878119901119862119872 + 119865119877 cos 120572 + 119865119877 sin 120572


Fall

velo

city amp

angle

Dist

ance

to im

pact

- Tension Force

- Damping setting


119883 = 121205881198812119901119878119901119862119909 minus 119865119877 cos 120572

119885 = 121205881198812119901119878119901119862119911 minus 119865119877 sin 120572

(2)



119889119881119889119905 =

119865119877 minus 119865119863 minus 119898119901119892 sin 120574(119898119901 + 119898119889) (3)



119865119863 = 121205781205881198812119901119863119878119901 (4)




119881 = radic1198812119909 + 1198812119910 + 1198812119911 (5)


119889119881119889119905 =

119881119909119889119881119909119889119905 + 119881119910119889119881119910119889119905 + 119881119911119889119881119911119889119905119881 (6)


120572 = tanminus1 (1003816100381610038161003816100381610038161003816100381611988111991111988111990910038161003816100381610038161003816100381610038161003816) (7)


120573 = sinminus1 (100381610038161003816100381610038161003816100381610038161003816119881119910119881100381610038161003816100381610038161003816100381610038161003816) (8)


119866119909119910119911 = radic1198662119909 + 1198662119910 + 1198662119911 (9)



119881119896119903 = int119905119896119903

0

100381610038161003816100381610038161003816997888rarr119866100381610038161003816100381610038161003816 119889119905 = Δ119905

119905119896119903sum0

119866119909119910119911 (10)







0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

Acce

lera

tion

(g)

Acce

lera

tion

(g)












119866119911 = minus00047119879 + 11031119866119911 = minus00056119879 + 10967

(11)





0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08


0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08



119866119911 119888119900119898119901 = 119866119911 + 00047119879 minus 01031 (12)








0 10 20 30 40 50

12

115

11

105

1

095

09

085

080 10 20 30 40 50

12

115

11

105

1

095

09

085

08


0 10 20 30 40 50

12

115

11

105

1

095

09

085

0 10 20 30 40 50

12

115

11

105

1

095

09

085




1

2 4

3

56

7

8




START

Is voltagegt38V



on all axis


value


Is angle gt90o

Deadlock on

Drop the parachute

Read accel data

End of programme




Read angledata

No

Yes

YesNo

No

No

Yes

Yes

No

Yes












5 Evaluation










6 Conclusion




Data Availability




Acknowledgments


References




























































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia









3 Method






(1)

here

119872 = 121205881198812119901119878119901119862119872 + 119865119877 cos 120572 + 119865119877 sin 120572


Fall

velo

city amp

angle

Dist

ance

to im

pact

- Tension Force

- Damping setting


119883 = 121205881198812119901119878119901119862119909 minus 119865119877 cos 120572

119885 = 121205881198812119901119878119901119862119911 minus 119865119877 sin 120572

(2)



119889119881119889119905 =

119865119877 minus 119865119863 minus 119898119901119892 sin 120574(119898119901 + 119898119889) (3)



119865119863 = 121205781205881198812119901119863119878119901 (4)




119881 = radic1198812119909 + 1198812119910 + 1198812119911 (5)


119889119881119889119905 =

119881119909119889119881119909119889119905 + 119881119910119889119881119910119889119905 + 119881119911119889119881119911119889119905119881 (6)


120572 = tanminus1 (1003816100381610038161003816100381610038161003816100381611988111991111988111990910038161003816100381610038161003816100381610038161003816) (7)


120573 = sinminus1 (100381610038161003816100381610038161003816100381610038161003816119881119910119881100381610038161003816100381610038161003816100381610038161003816) (8)


119866119909119910119911 = radic1198662119909 + 1198662119910 + 1198662119911 (9)



119881119896119903 = int119905119896119903

0

100381610038161003816100381610038161003816997888rarr119866100381610038161003816100381610038161003816 119889119905 = Δ119905

119905119896119903sum0

119866119909119910119911 (10)







0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

Acce

lera

tion

(g)

Acce

lera

tion

(g)












119866119911 = minus00047119879 + 11031119866119911 = minus00056119879 + 10967

(11)





0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08


0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08



119866119911 119888119900119898119901 = 119866119911 + 00047119879 minus 01031 (12)








0 10 20 30 40 50

12

115

11

105

1

095

09

085

080 10 20 30 40 50

12

115

11

105

1

095

09

085

08


0 10 20 30 40 50

12

115

11

105

1

095

09

085

0 10 20 30 40 50

12

115

11

105

1

095

09

085




1

2 4

3

56

7

8




START

Is voltagegt38V



on all axis


value


Is angle gt90o

Deadlock on

Drop the parachute

Read accel data

End of programme




Read angledata

No

Yes

YesNo

No

No

Yes

Yes

No

Yes












5 Evaluation










6 Conclusion




Data Availability




Acknowledgments


References




























































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia



Fall

velo

city amp

angle

Dist

ance

to im

pact

- Tension Force

- Damping setting


119883 = 121205881198812119901119878119901119862119909 minus 119865119877 cos 120572

119885 = 121205881198812119901119878119901119862119911 minus 119865119877 sin 120572

(2)



119889119881119889119905 =

119865119877 minus 119865119863 minus 119898119901119892 sin 120574(119898119901 + 119898119889) (3)



119865119863 = 121205781205881198812119901119863119878119901 (4)




119881 = radic1198812119909 + 1198812119910 + 1198812119911 (5)


119889119881119889119905 =

119881119909119889119881119909119889119905 + 119881119910119889119881119910119889119905 + 119881119911119889119881119911119889119905119881 (6)


120572 = tanminus1 (1003816100381610038161003816100381610038161003816100381611988111991111988111990910038161003816100381610038161003816100381610038161003816) (7)


120573 = sinminus1 (100381610038161003816100381610038161003816100381610038161003816119881119910119881100381610038161003816100381610038161003816100381610038161003816) (8)


119866119909119910119911 = radic1198662119909 + 1198662119910 + 1198662119911 (9)



119881119896119903 = int119905119896119903

0

100381610038161003816100381610038161003816997888rarr119866100381610038161003816100381610038161003816 119889119905 = Δ119905

119905119896119903sum0

119866119909119910119911 (10)







0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

Acce

lera

tion

(g)

Acce

lera

tion

(g)












119866119911 = minus00047119879 + 11031119866119911 = minus00056119879 + 10967

(11)





0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08


0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08



119866119911 119888119900119898119901 = 119866119911 + 00047119879 minus 01031 (12)








0 10 20 30 40 50

12

115

11

105

1

095

09

085

080 10 20 30 40 50

12

115

11

105

1

095

09

085

08


0 10 20 30 40 50

12

115

11

105

1

095

09

085

0 10 20 30 40 50

12

115

11

105

1

095

09

085




1

2 4

3

56

7

8




START

Is voltagegt38V



on all axis


value


Is angle gt90o

Deadlock on

Drop the parachute

Read accel data

End of programme




Read angledata

No

Yes

YesNo

No

No

Yes

Yes

No

Yes












5 Evaluation










6 Conclusion




Data Availability




Acknowledgments


References




























































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia



0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

Acce

lera

tion

(g)

Acce

lera

tion

(g)












119866119911 = minus00047119879 + 11031119866119911 = minus00056119879 + 10967

(11)





0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08


0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08



119866119911 119888119900119898119901 = 119866119911 + 00047119879 minus 01031 (12)








0 10 20 30 40 50

12

115

11

105

1

095

09

085

080 10 20 30 40 50

12

115

11

105

1

095

09

085

08


0 10 20 30 40 50

12

115

11

105

1

095

09

085

0 10 20 30 40 50

12

115

11

105

1

095

09

085




1

2 4

3

56

7

8




START

Is voltagegt38V



on all axis


value


Is angle gt90o

Deadlock on

Drop the parachute

Read accel data

End of programme




Read angledata

No

Yes

YesNo

No

No

Yes

Yes

No

Yes












5 Evaluation










6 Conclusion




Data Availability




Acknowledgments


References




























































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia



0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08


0 10 20 30 40 50

12

115

11

105

1

095

09

085

08

0 10 20 30 40 50

12

115

11

105

1

095

09

085

08



119866119911 119888119900119898119901 = 119866119911 + 00047119879 minus 01031 (12)








0 10 20 30 40 50

12

115

11

105

1

095

09

085

080 10 20 30 40 50

12

115

11

105

1

095

09

085

08


0 10 20 30 40 50

12

115

11

105

1

095

09

085

0 10 20 30 40 50

12

115

11

105

1

095

09

085




1

2 4

3

56

7

8




START

Is voltagegt38V



on all axis


value


Is angle gt90o

Deadlock on

Drop the parachute

Read accel data

End of programme




Read angledata

No

Yes

YesNo

No

No

Yes

Yes

No

Yes












5 Evaluation










6 Conclusion




Data Availability




Acknowledgments


References




























































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia



0 10 20 30 40 50

12

115

11

105

1

095

09

085

080 10 20 30 40 50

12

115

11

105

1

095

09

085

08


0 10 20 30 40 50

12

115

11

105

1

095

09

085

0 10 20 30 40 50

12

115

11

105

1

095

09

085




1

2 4

3

56

7

8




START

Is voltagegt38V



on all axis


value


Is angle gt90o

Deadlock on

Drop the parachute

Read accel data

End of programme




Read angledata

No

Yes

YesNo

No

No

Yes

Yes

No

Yes












5 Evaluation










6 Conclusion




Data Availability




Acknowledgments


References




























































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia



START

Is voltagegt38V



on all axis


value


Is angle gt90o

Deadlock on

Drop the parachute

Read accel data

End of programme




Read angledata

No

Yes

YesNo

No

No

Yes

Yes

No

Yes












5 Evaluation










6 Conclusion




Data Availability




Acknowledgments


References




























































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia







5 Evaluation










6 Conclusion




Data Availability




Acknowledgments


References




























































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia



Data Availability




Acknowledgments


References




























































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia




RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia


Documents

Design of Fully Automatic Drone Parachute System with ...Design of Fully Automatic Drone Parachute System with Temperature Compensation Mechanism for Civilian and Military Applications