311
1 INTRODUCTION
Amongmanytypesofrobots,somemoveinwater,in
the air, or on land. The latter group can be divided
into two basic types‐wheeled and walking‐each
with its advantages and disadvantages. One of the
undisputedadvantagesofwheeledrobotsisthatthey
arerelatively
simpletoconstructandcontrolsinceno
complex programmes are responsible for their
movement. However, their locomotion is severely
limited‐while moving on flat terrain is not a
problem, larger, uneven surfaces can easily reduce
their ability to move, even if we consider tracked
robots. In such case, solution to the
problem is to
replacethewheelswithlegs.Withthisimprovement,
the robot can easily cross the obstacle or climb on
them. Yet, as the capabilities increase, so do the
hardwareandsoftwarerequirements.Inorderforthe
robottomove,itmust be equipped withsufficiently
powerful actuators, and
each movement must be
programmed so that it does not lose balance and
movesinthespecifieddirection.
Despite the difficulties associated with building
walkingrobots,manypeoplearedecidetobuildand
furtherimprovethem.Forexample,theauthorshave
builtafour‐leggedwalkingrobotmodelledoninsects
[1,2]
or mammals [3,4]. A general review of the
development of such machines has been done for
four‐leggedrobots[5,6].
Otherauthorshaveconstructedsix‐leggedrobots,
calledhexapodsorʹspidersʺ,tooptimizetheirmotion
[7,8], track body position [9], or study physical
properties [10]. One of the most
interesting issuesis
presented in an article describing the Ball‐on‐Plate
(BoP) problem [11], in which the authors propose a
new method of motion designed for robots with six
legsinstalledparalleltoeachother.
The Hexapod Modernization Process and Its Impact on
the Locomotion Efficiency of the Unit
N.Popowniak
GdyniaMaritimeUniversity,Gdynia,Poland
ABSTRACT:This paper presentsthe processof retrofittinga six‐point‐o
f
‐freedomwalking unit‐ ahexapod
remotelycontrolledbyBluetoothcommunicationbasedonaproprietaryprogramme.Acomparisonismadeto
showhowdifferencesinthedesignofthewalkingunit,andinparticularthenumberoffreedompointsperleg,
canaffecttheefficiencyoflocomotion.Inthefirstsection,
theworkofotherauthorsrelatedtowalkingrobotsis
presented. In the next section of the article, the individual components of the robot and the software are
analysedandcomparedwiththeoriginalunittoshowthedifferencesandadvantagesobtainedbytheupgrade.
Thelastpartofthearticle
summarisestheworkdoneandshowsfurtherpossibilitiesforthedevelopmentofthe
project.
http://www.transnav.eu
the International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 18
Number 2
June 2024
DOI:10.12716/1001.18.02.06
312
2 MODERNIZATION
The robot presented in this paper is the second
versionofawalkingrobotdesignedbytheauthor.In
the original version, the main problems were low
toleranceforunevennessofthesurfaceonwhichthe
robot was moving, slow tempo of operation, which
led to the impression
that the design worked
sluggishly. To increase the robotʹs mobility, it was
decidedtoadda servoto each leg,giving theentire
design18actuators.Tomeetthenewrequirements,it
wasdecidedtoreplacetheservoswiththosecapable
of lifting the structure, which led to necessity
of
improving the power supply to ensure the stable
operation of the robot. To increase the dynamics, it
wasdecidedtoaddcommandstothenewservosand
tomake therobotʹsmovement speeddepend onthe
inclinationofthejoystick.
In new version of the robot, for each leg,
two
servosareresponsibleforverticalmovementandone
forhorizontalmovement,asshowninthefigure1.:
Figure1.Servomovement
Aswiththeoriginalversion,controloftherobotis
based on wireless communication enabled by
Bluetoothmodules.ThemasterunitusesanArduino
Nano, while the slave unit uses an Arduino Mega.
Therobotʹslimbsaredividedintotwosectionsthree
legs each, which work together to execute most
commands. They are chosen to give the robot the
greatestpossiblestaticstabilitywhenwalking(when
theotherlimbstakeoff)byformingatrianglewiththe
longestpossiblesides.Sucharrangementincludesthe
twooppositelegsononesideandthemiddlelegon
the other side. The servos
included in such groups
performthepositionchangesimultaneously[Fig.2.].
Figure2.Servogroupsandsamplemotion
To take full advantage of the new servosʹ
capabilities, the master unitʹs control algorithm was
modernised. First, it was a matter of making the
robotʹs speed dependent on the tilt of the joystick.
Then,parts ofthe programmewereaddedtoenable
theoperation ofthebuttonsresponsible for
the new
commands.
3 RESULTS
Thefirststepintheretrofitwastochangetheentire
structure of the robot. Adding new servos required
obtainingnew parts,which were printedusinga3D
printer. Another important step was to modify the
software to take full advantage of the new
capabilities.
Theadditionofonepointoffreedomper
leg meant that the robotʹs body could move
horizontallyaswellasvertically[Fig.3.].
Figure3.Frontalview
Whentherobotencountersanobstacleinitspath,
it can not only avoid it but at the same time move
over or under it. In addition, a new command has
beenprogrammedthatusestheabilitytochangethe
heightofthetorso,allowingtherobottosteppingon
obstacles
upto4cm.
By using stronger servos and increasing the
numberofservos,therobotcannotonlyturnbutalso
movedirectlytotheleftorrightbychangingitscentre
ofgravityandsupportingitselfwithitsfeet.Thishas
madeit possible toreducethetime
ittakestoavoid
obstaclesbyusingasmallernumberofcommands.
However, this system also has its drawbacks. In
theoriginalversion,therobotʹsfootwasata90°angle
mostofthetime,sotheworktheservoshadtodoto
stayinplacewaszero,which
couldbeseenwhenthe
power wasturned off‐the robotstopped. After the
upgrade,theservoshavetoholdthearmatanangle
allthetime,resultinginconstantpowerconsumption,
the higher the angle the greater the power usage.
Considering this and the increased weight and
strength
of the servos, the power requirements
increased many times over, determinating a power
supplychange.
Asummaryoftheupgradeisshownintable1.
The servosused inthe secondversion had metal
gears, which added to the weight of the robot.
However,thisallowedtherobottomovefaster.
Using
morecommands(from4to7)helpsreducethetimeit
takestoavoidanobstacleandallowsittodriveinto
or overlarger obstacles. To increase theprecision of
313
the movement, three different speeds are
programmed, depending on the inclination of the
joystick.Thecontrolunitprogrammeitselfconsistsof
amainfile,asourcefileandalibrary.
Table1.Robotversioncomparison
________________________________________________
VersionI VersionII
________________________________________________
Weight[kg]~1.2 ~3
Obstacleavoidance[s]26 9
Distance30cm[s]40 20
Speedcontrol‐Yes(3speeds)
Numberofprogrammed
commands4 7
Obstacleclimbing[cm]<=0.5 <=4
Suspensionheight[cm]4 0‐9
________________________________________________
The appearance of the robot after the upgrade is
showninthefigure4.
Figure4.Robot
4 CONCLUSIONS
Thearticleisa preludeto futureplannedwork. The
addition of another degree of freedom has greatly
improved the robotʹs locomotion capabilities. It can
now climb over obstacles, change the position of its
torso,andshiftitscentreofgravity.Forthenextwork,
it is planned
to add the possibility of autonomous
movementoftherobotbasedonthevaluesmeasured
by the sensors, so that the robot can independently
reachhard‐to‐reachplaces. Also underconsideration
is the possibility of modifyingthe control system so
that the robotʹs steps are controlled by artificial
intelligencealgorithmsinthefuture.
REFERENCES
[1]P.M. James, Amal Prakash, Virbhadrappa Kalburgi,
PramodSreedharan“Design,analysis,manufacturingof
four‐leggedwalkingrobotwithinsecttypeleg”,Volume
46,Part10,2021,Pages4647‐4652
[2]Virbhadrappa Kalburgi, P.M. James, Pramod
Sreedharan, “Control system design for four‐legged
walkingrobotwithinsecttypelegusingROS”,Volume
46,Part10,2021,Pages5092‐5097
[3]Jun He, Feng Gao, “Type Synthesis for Bionic
Quadruped Walking Robots” Journal of Bionic
Engineering,Volume12,Issue4,2015,Pages527‐538
[4]Anurag Narayan Sarmah, Abhijit Boruah, Daisy Kalita,
Darshana Neog, Saumay Paul, “A Bio‐Inspired
Implementation of Walking and Stair
Climbing on a
Quadruped Robot”, Procedia Computer Science,
Volume143,2018,Pages671‐677
[5]PriyaranjanBiswal,PrasesK.Mohanty, ”Developmentof
quadruped walking robots: A review, Ain Shams
Engineering Journal”, Volume 12, Issue 2, 2021, Pages
2017‐2031.
[6]Zhijun Chen, Jimu Liu, Feng Gao, “Real‐time gait
planning method for
six‐legged robots to optimize the
performancesofterrainadaptabilityandwalkingspeed”
MechanismandMachineTheory,Volume168,2022
[7]Haitao Yu, Haibo Gao, Zongquan Deng, “Enhancing
adaptability with local reactive behaviors for hexapod
walking robot via sensory feedback integrated central
pattern generator” RoboticsandAutonomous Systems,
Volume124,
2020
[8]DominikBelter,PiotrSkrzypczyński,“IntegratedMotion
Planning for a Hexapod Robot Walking on Rough
Terrain,“IFACProceedingsVolumes,Volume44,Issue
1,2011,Pages6918‐6923
[9]Gang Chen, Bo Jin, Ying Chen, “Accurate and robust
body position trajectory trackingof six‐legged walking
robots with nonsingular terminal
sliding mode control
method,“AppliedMathematicalModelling,Volume77,
Part2,2020,Pages1348‐1372
[10]S.Yu. Misyurin, G.V. Kreinin, N.Yu. Nosova, A.P.
Nelubin, “Six‐Legged Walking Robot (Hexabot),
Kinematics, Dynamics and Motion Optimization”,
Procedia Computer Science, Volume 190, 2021, Pages
604‐610
[11]ZhijunChen,FengGao,QiaoSun,
YuanTian,JimuLiu,
Yinan Zhao, “Ball‐on‐plate motion planning for six‐
parallel‐legged robots walking on irregular terrains
using pure haptic information”, Mechanism and
MachineTheory,Volume141,2019,Pages136‐150