research article the study of intelligent vehicle...

Research ArticleThe Study of Intelligent Vehicle Navigation Path Based onBehavior Coordination of Particle Swarm

Gaining Han12 Weiping Fu1 and Wen Wang1

1School of Mechanical and Precision Instrument Engineering Xirsquoan University of Technology Xirsquoan Shaanxi 710048 China2Information Engineering Department Xianyang Normal University Xianyang Shaanxi 712000 China

Correspondence should be addressed to Weiping Fu weipingfxauteducn

Received 26 August 2015 Revised 23 November 2015 Accepted 15 December 2015

Academic Editor Marco A M Armendariz

Copyright copy 2016 Gaining Han et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

In the behavior dynamicsmodel behavior competition leads to the shock problemof the intelligent vehicle navigation path becauseof the simultaneous occurrence of the time-variant target behavior and obstacle avoidance behavior Considering the safety and real-time of intelligent vehicle the particle swarmoptimization (PSO) algorithm is proposed to solve these problems for the optimizationof weight coefficients of the heading angle and the path velocity Firstly according to the behavior dynamics model the fitnessfunction is defined concerning the intelligent vehicle driving characteristics the distance between intelligent vehicle and obstacleand distance of intelligent vehicle and target Secondly behavior coordination parameters that minimize the fitness function areobtained by particle swarm optimization algorithms Finally the simulation results show that the optimization method and itsfitness function can improve the perturbations of the vehicle planning path and real-time and reliability

1 Introduction

In recent years with the increase in the number of carstraffic accidents happen more and more frequently most ofwhich are caused by man-made mistakes and serious threatto peoplersquos live and social stability Thus ldquocar-people-roadrdquomode is converted to ldquocar-computer-roadrdquo mode which canbe liberated from driving environment and is one of the effec-tive ways to reduce traffic accident The unmanned groundvehicle has become a hot issue domestic and internationalresearch [1ndash3] The DARPA (Defense Advanced ResearchProjects Agency) proposed a series of control technologyfor the unmanned vehicle and digital road traffic controlscheme under the urban road environment which is themostimportant for the path planning and controlling [4ndash9]

This paper mainly focused on the intelligent vehiclenavigation problems especially the path planning andguiding the vehicles to the target position At presentthe traditional path planning methods including potentialfield and grid method are prone to local minimum prob-lems [10 11] The bionic path planning algorithms includ-ing genetic algorithm ant colony algorithm and particle

swarm algorithm need to determine the coding scheme andthe optimization of objective function [12ndash17]The intelligentpath planning algorithms including BP neural networkANN and fuzzy neural network learning methods needconstant learning and training to obtain the optimal path thenetwork settings of the intermediate nodes of these methodsare often dependent on experience and constant trial anderror [18ndash21] These are only from the perspective of theplanning However the navigation path of the intelligentvehicle should not only contain trace paths but also containthe time information (namely speed) for real-time trackingcontrol The above methods including field method bionicmethod and intelligent information do not include the timein the path planning

In 1995 behavioral dynamics method is first proposedby Schoner et al [22] it has been applied to a mobile robotsystem complex structured environment navigation controlmachine control arms andmultirobot formations navigationcontrol [23ndash31] Due to its outstanding performance scholarsbegan to pay attention to relevant theorem studying andapplication research [32ndash38] Behavior dynamics method isintroduced into the intelligent vehicle navigation planning

Hindawi Publishing CorporationComputational Intelligence and NeuroscienceVolume 2016 Article ID 6540807 10 pageshttpdxdoiorg10115520166540807

2 Computational Intelligence and Neuroscience

and in-depth research and experiments verify the feasibilityof this method

The aim of the paper is to discuss the intelligent vehiclepath planning problem in dynamic environment In theurban environment the road environment information isacquired through the vehicle sensor which is indispensablefor path planning and obstacles avoiding Behavior dynamicsmethod is used for the path planning to generate competitionbehavior problem while the particle swarm optimizationalgorithm is adopted to improve the behavior coordination

The rest of the paper is as follows Section 2 intro-duces behavior dynamicsmethodThe behavior coordinationof the standard particle swarm optimization algorithm isintroduced in Section 3 Simulation results of illustrationexamples and discussion are presented in Section 4 Finallyin Section 5 the conclusion and further discussion are given

2 Behavior Dynamics

The behavior dynamics were included towards the goal ofbehavior and obstacle avoidance behavior Using the behaviorcharacter of behavior dynamics plan according to the localtarget from a starting point of a path (namely the intelligentvehicle navigation path) and satisfying the vehicle in thedriveway can avoid the pavement and all sorts of obstaclesto the dynamic environmentThe heading angle and the pathvelocity in behavior dynamics model could be represented asthe following form [26]

120595 = 119891

1(120595 119875)

V = 1198912(V 119875)

(1)

where 119875 is position of the intelligent vehicle in the worldcoordinator system and the parameters are the heading angle120595 and the path velocity V of the intelligent vehicle whichgenerally act as the behavioral variables

120595 and V define therate of the heading angle and velocity as a function of theircurrent values

21 Behavior Dynamics Modeling The dynamic approach topath generation of the intelligent vehicles employs 120595 andV as the planning variable [22 23] The path planning iscontinuous in a time course of 120595 and V the target directions120595tar relative to the world coordinate system in which targetlie from the view of the intelligent vehicle and the obstacledirections 120595obs119894 in which obstacles lie from the viewpoint ofthe vehicle relative to the world coordinate system 120595tar and120595obs119894 are represented by attractive and repulsive force-letsacting on heading direction (Figure 1) The velocity of targetVtar and the velocity of obstacle Vobs are constraints that arerepresented by attractive and repulsive force-lets acting onthe path velocity In the dynamics system the variables arethe direction angle and the path velocity lying at obstaclesand target from the current position of the intelligent vehicleshown in Figure 1

According to the behavior variables and the behaviorpattern of the intelligent vehicle the target and obstacleavoidance of dynamics model is built

1

33

2 3

22

v

OY

X y

1

1

x

Obstacle 1

Obstacle 2

Obstacle 3

Vehicle

Veh

Veh

Veh

Target

120595

120595

tar120595obs1

120595obs2

120595obs3

obs

Figure 1 Variables represented in the dynamics

211TheTarget of BehavioralModels Toward target behavioris the behavior of the intelligent vehicle to reach its desti-nation The dynamic behavior of the attractor equation isexpected with negative slope and asymptotic speed convergesto a stable fixed point The attractors of the equation definedthe behavioral variables so that the system is ensured tobe at all times in a stable state The heading angle and thepath velocity of the behavior model are established as thefollowing

(1) Heading Direction Behavior Dynamics Model In thebehavior dynamics model the heading angle is specified asa time-variant term at any time 119905 Ψ is an angle and is cyclicand defined as 120595 isin [minus1205872 1205872] in the world coordinates andthe 119909-axis is always consistent with the direction of the roadIn [24] the heading direction of behavior dynamics for targetacquisition can be defined by

120595 = 119891tar (120595) = minus120582tar tan (120595 minus 120595tar)

120595tar = arctan(119875tar119910 minus 119875veh119910

119875tar119909 minus 119875veh119909)

(2)

where 120582tar gt 0 sets the attraction strength of the vehiclersquosheading variable 120595tar is the angle between the target and theintelligent vehicle in world coordinate system and 120595tar is theattractor in dynamics system 119891tar(120595) is an attractive force-let and converges to the value so that 120595 goes towards theintended target-state119875tar(119875tar119909 119875tar119910) is target position in theworld coordinator system and is the time-variant target and119875veh(119875veh119909 119875veh119910) is intelligent vehicle position in the worldcoordinator system

(2) The Path Velocity Behavior Dynamics Model Besidesheading direction in order to ensure the formation andmaintenance of the stabilization path velocity must be

Computational Intelligence and Neuroscience 3

adjusted real-time In terms of path velocity considering thatthe security and stability of the vehicle system must meetwith safety distance from maximum contact time 119879max =

(119889tarminus119863119904)V119863119904 is aminimum safe following distance and 119889taris the distance to the target Vtar is an attractor of velocity andV goes towards the required velocity Vtar Literature [26 34 35]used linear velocity equation with certain constraints In thisstudy [25] is used to define the velocity behavior dynamicsthe equation towards the target

V = 119891tarV (V) = minus120582tarV (V minus Vtar) exp[(V minus Vtar)

2

2120590

2

V] (3)

where 120582tarV is strength factor Vtar is the desired vehiclevelocity and 120590V is scope attractors calculated in [36]

212 Obstacle Avoidance Behavior Model Obstacle avoid-ance behavior can be divided into heading direction andvelocity behavior

(1) Heading Direction Behavior Dynamics Model When theintelligent vehicle is driving in the way of the target road itis time to avoid the static or moving obstacles on the roadto reach the destination safely 120595obs119894 is called the repellerwhich is unstable point 120595obs119894 to make the influence of anobstacle go to zero in behavioral dynamics system 120595obs119894defined the behavioral variables Reference [24] sets theheading direction in the obstacle avoiding behavior using thefollowing equation120595obs119894 = 119891obs119894 (120595obs119894) = 120582obs119894 (120595 minus 120595obs119894) exp (minus119862119889obs119894)

sdot exp[minus(120595 minus 120595obs119894)

2

2120590

2

119894

]

(4)

where 120582obs119894 is the repulsion strength 119889obs119894 is the distancebetween the obstacles and intelligent vehicle which cancontinuously be estimated using sensory inputs 119862 is therepulsive force of attenuation coefficient with the increaseof distance 120590

119894is the angular scope of the repeller and 120590V

is velocity scope of the repeller [36] This implies that thestrength and angular range of the repulsion become strongeras the intelligent vehicle gets closer to obstacles

Multiple obstacles heading direction of behavior dynam-ics equation can be written as

120595obs = 119865obs = sum

119894

119891obs119894 (120595obs119894) (5)

(2) The Path Velocity Behavior Dynamics Model The velocityof the intelligent vehicle is related to the distance betweenvehicles and obstacles 119889obs and also to the safety distance119863

119904

Similar to (3) the path velocity could be modeled as

V = 119891obs119894 (V) = minus120582obsV (V minus Vobs119894) exp[(V minus Vobs119894)

2

2120590

2

V] (6)

where 120582obsV is velocity to the repeller of velocity intensityfactor the repeller can be changed by adjusting 120582obsV Vobs119894is the required obstacle velocity and 120590V is scope the repellercalculated in [36]

22 Behavior Dynamics Coordination After the establish-ment of the various dynamics model navigation planningdemands the fusion of several behavior variants in practicalapplications it is necessary to fuse the behavior various andconduct planning for the vehicle Consider a fusion of twoactions

120595 = 120596obs119865obs + 120596tar119891tar (7)

V = 120582obs119891obs119894 (V) + 120582tar119891tar (V) (8)

where 120596obs 120596tar 120582obs 120582tar are the weight coefficients andbehavior of fusion is not considered in literature [36]where four weight coefficients are equal to 1 Thereof whenthe target behavior and obstacle avoidance behavior occursimultaneously two types of behavior lead to a perturbationof the intelligent vehicle Under the absence of behavior offusion the target behavior and obstacle avoidance behavioroccur simultaneously which leads to the perturbation of thevehicle In order to solve this problem behavior coordinationwas studied in depth

(1) Heading Direction Behavior Fusion From (7) in the pro-cess of driving it can be seen that there only exists the targetbehavior with no obstacles in the environment so 120596obs = 0120596tar = 1 When obstacles exist in the process of intelligentvehicle driving obstacle avoidance behavior is stronger thantoward the target behavior so parameters 120596obs = 1 120596tar = 0When the intelligent vehicle passes through obstacles themain obstacle avoidance behavior turned toward the targetbehavior Similarly when the vehicle encounters obstaclesthe vehicle behavior turns from ldquotoward targetrdquo to ldquoobstacleavoidancerdquo along the direction of gradient between the twostates associating with the increasing weight of the obstacleavoidance behavior Therefore the vehicle direction pertur-bations problem arises with the inharmoniousness behaviorswitching because of the rough coordination

(2) The Path Velocity Behavior Fusion Similarly from (8)when there are no obstacles or the obstacles are static thereis only velocity towards the target behavior the parameters120582obs = 0 120582tar = 1 When obstacles are in motion in the pro-cess of intelligent vehicle driving obstacle avoidance behavioris stronger than toward target behavior the parameters120582obs =1 120582tar = 0 in (8) When the vehicle is away from obstaclesthe velocity of behavior is directed towards the target Dueto the change of behavior speed behavior alternates frommoving towards the target to obstacle avoidance so that thetwo behaviors can smooth transition which will depend ontwo behaviors of coordination Otherwise there will be fast orslow phenomenon for the velocity of intelligent vehicle

3 Particle Swarm Optimization Algorithm

According to many research literatures [32 33 35 39]behavior of competitive dynamics method is used for behav-ior coordination Besides the differential form of dynam-ics equation the competition dynamics model is still thedifferential equation and relates to the stability of thesystem making behavior coordination more complicated


In the literature [38] heading direction angle coefficient ofthe mobile robot is optimized by genetic algorithm with thevelocity left constant Reference [39] using particle swarmoptimization algorithm mainly for the interaction betweenmultiple mobile robots for the angle of coordinate did notconsider speed In this paper it introduced the behavior fromthe direction angle and linear velocity at the same time bythe particle swarm optimization algorithm to strengthen theproblem of the intelligent vehicle navigation path

31 Particle Swarm Optimization Algorithm The particleswarm optimization algorithm has many advantages such asbeing easy to describe adjusting less parameters and havingfast converging speed so that it has become a hotspot inintelligent optimization and evolution computing once pro-posed It has been widely used in the function optimizationdynamic environment optimization neural network trainingand fuzzy system control applications [40]

In PSO each potential solution to optimization problemis a bird in the search space called a particle All particles havea function that is optimized by the decision on the fitnessvalue Each particle has a speed that decided they fly out ofdirection and distance and the optimal particles are to followthe current search in the solution space Optimization startsto initialize a set of random particles (random solutions) anddetermines the merits of the solution by the fitness functionEach particle iterative process contains two extremes theindividual extreme and global extreme individual extremecognitive level represents particles themselves and the globalextremes represent the social cognitive level the final optimalsolution is obtained after several iterations [40]

The goal of optimizing function search space for 119898-dimensional (namely the number of optimization variables)is set the number of particles is 119899 the 119894 particle posi-tion is 119883

119894= (119883

1198941 119883

1198942 119883

119894119898)

119879 and speed is 119881119894=

(119881

1198941 119881

1198942 119881

119894119898)

119879The course of the flight in accordancewith(7) and (8) to update their own position and speed of theparticle 119875

119894= (119875

1198941 119875

1198942 119883

119894119898)

119879 for the individual pbest119875

119892= (119875

1198921 119875

1198922 119883

119892119898)

119879 for the global gbest

119881

119894119896(119905 + 1) = 119908119881

119894119896(119905) + 119888

1119903

1(119875

119894119896(119905) minus 119883

119894119896(119905))

+ 119888

2119903

2(119875

119892119896(119905) minus 119883

119894119896(119905))

119883

119894119896(119905 + 1) = 119883

119894119896(119905) + 119881

119894119896(119905 + 1)

(9)

where 119896 = 1 2 119898 119894 = 1 2 119899 119905 is the number ofiterations 119908 is weight coefficient for the particle velocityand the greater the value of 119908 the better global searchingability for PSO 119903

1 1199032are a random number in [0 1] and 119888

1

119888

2are limiting factors for acceleration In iterative process

the particle swarm optimization algorithm has no specificmechanisms to control the velocity of the particles the speedof the particles to limit the particle speed of each dimension iswithin [119881min 119881max] the position of each dimension is limitedwithin [119883min 119883max]

32 PSO Implementation of Behavior Coordination Basedon the driving characteristics of the intelligent vehicle

at a certain velocity when the behavior toward the targetand obstacle avoidance behavior occurs alternately to reachthe destination safely it becomes a major obstacle avoidancebehavior obstacle avoidance behavior becomes a majorbehavior and towards the target of behavior will be weakenedWhen passing through the obstacles toward the targetbehavior becomes dominant Behavior dynamics analysisthrough the front towards the target and obstacle avoidancebehavior including heading direction and the path velocitybehavior the weight coefficient of heading direction and thepath velocity are optimized by particle swarm optimizationalgorithm thereby eliminating perturbations the intelligentvehicle can smoothly move through obstacles to reach thetarget location

In this paper employing themethod of PSO optimizationvariables are expressed by the particle the particles set is119883

120595= (120596obs 120596tar) and 119883V = (120582obs 120582tar) The fitness function

reflects the relationship between the intelligent vehicle andthe environment When the obstacle is very close to theintelligent vehicle the avoidance obstacle behavior plays theleading factor When the intelligent vehicle is away fromobject the target of behavior plays a decisive role the fitnessfunction is designed as follows

119865

120595=

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

119889obs (120595 minus 120595tar)

119889tar (120595 minus 120595obs)

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

(10)

As proposed in [41] (7) and (8) could be solved bythe fourth-order Runge-Kutta method Besides 120596obs 120596tar areconfirmed by the extreme value of fitness function using theparticle swarm optimization algorithm

Similarly the velocity of behavior of coordination fitnessfunction is

119865V =

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

119889obs (V minus Vtar)119889tar (V minus Vobs)

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

(11)

Using the particle swarm optimization algorithm for itsfunction extreme value 120582obs 120582tar can be obtained

4 The Simulation Analysis

In order to verify the behavior dynamics method as well asbehavior coordination algorithm is feasible in the intelligentvehicle path navigation we carried out the experiment underthe condition of the different pavement environment andavoiding obstacle security requirements like lane keepingvehicle following lane changing and overtaking Followthese steps to carry out experiment

(1) Establish the system model with unknown param-eters of the intelligent vehicle the sensors and theroad environment These parameters are all detectedby vehicular sensors and used to parameterize cor-responding term such as road environment startingpoint and target point of the driving initial velocityand heading direction of the vehicle position of theobstacles detecting range and distance of the sensorand so forth


(2) During the process of driving obstacles beyonddetection scope are according to the behavior dynam-ics model towards the target for navigation pathWhen obstacles are detected there is a use of behaviordynamics of the obstacle avoidance behavior andtowards the target behavior of the coordination algo-rithm for navigation path The intelligent vehicle indriving when detecting obstacles outside the safedistance is according to the dynamics of behaviortowards the goal of behavior for navigation pathwhen detecting obstacles within the safe distanceaccording to the behavior of dynamic obstacle avoid-ance behavior and towards the target behavior of thecoordination for navigation path

(3) Nonstationary conditions of the vehicle continuouslydetermine the current vehicle position and the targetposition stop the iteration after reaching the targetotherwise repeat (2)

In this paper MATLAB R2013a simulation experi-ments as shown in Figure 1 the simulation parametersintelligent vehicle initialization parameters for VehiclePos= [21 3 90 36 8 90 1] (initial 119909 119910 coordinates headingspeed distance perception perception range vehicle size)reference vehicles initialization parameters for VehicleRun-Pos1 = [20 15 90 43 120 1] the target position initializationparameters for TargetPos = [VehicleRunPos1 (2) VehicleRun-Pos1 (2)](target 119909 and 119910 coordinates) the obstacle positioninitialization parameters for ObstaclePos = [12 5 90 2 227 30ndash60 2 15] (119909 and 119910 coordinates of the obstacle headingdirection speed size of obstacles) perceive the sector range[minus30∘300∘] scene size of [50 50] Particle swarmoptimizationparameters 1198881 = 1198882 = 2 119908 = 08 1199031 = 1199032 = rand(1) 119905 = 20119898 = 2 119899 = 20 [119881min 119881max] = [0 1] [119883min 119883max] = [0 1]

In figure blue filled circles are an obstacle red filled circleis detected in the movement of obstacles black thick lineand blue dotted line are for the lane green rectangle is forintelligent vehicle blue sector is for intelligent vehicle sensingarea black rectangles are for the target vehicle and red signsare for intelligent vehicle running track

The simulation results are shown in Figure 2 In thesame scene two kinds of behavior coordination algorithmoptimize the simulation results shown in Figure 2 so thevelocity of obstacles behavior weight coefficient is 0 onlyweight coefficient of heading direction behavior coordina-tion Figure 2(a) shows running track for PSO behaviorcoordination in the lane for safe obstacle avoidance and thedestination Figure 2(b) shows running track to Competitivebehavior coordination for the lane safe obstacle avoidanceand the destination Figures 2(c) and 2(d) show the intelligentvehicle heading direction variation for PSO behavior coordi-nation and competition behavior coordination Figures 2(e)and 2(f) show the weight coefficient variation in headingdirection for PSO behavior coordination and competitionbehavior coordination From the simulation data and dia-gram it can be seen as in PSO behavior to coordinate thenavigation path and heading direction change closer to theactual vehicle characteristics of the vehicle while the anglechange is too large for competitive behavior coordinated

For Figures 2(e) and 2(f) the heading direction of obstacleavoidance and towards the target weight coefficient optimiza-tion and the particle swarm optimization method of twoparameters optimization more conform to the behavior ofthe competition law the particle swarmoptimizationmethodis easy to realize in the design less parameter adjustmentfast convergence and so on However using the method ofthe differential equation of competition more parameters areadjusted making the coordination more complicated

Figure 3 is the optimization effect of the PSO headingdirection and velocity behavior coordination in dynamicobstacles Figure 3(a) shows PSO behavior coordination tothe navigation path of intelligent vehicle Figures 3(b) and3(c) show PSO behavior coordination in the heading direc-tion and velocity changes and Figures 3(d) and 3(e) showthe heading direction and velocity of weight coefficient ofvariation to PSObehavior coordination Figures 3(a) and 3(b)can be seen as the simulation trajectory and angle change isconsistent with the actual vehicle Figures 3(d) and 3(e) canbe seen as heading direction and velocity weight coefficientoptimization point of view PSO weight coefficient is moreconsistent with the competition behavior change law and cancontrol the intelligent vehicle safe avoidance obstacle towardsthe target

Figure 4 shows lane changing and overtaking usingbehavior dynamics method and PSO behavior coordinationfrom running track of the intelligent vehicle heading direc-tion change and behavioral weight factors variation law canbe seen behavior dynamics can be well for path planning ofthe intelligent vehicle

Simulation results show that behavior dynamics methodand PSO behavior coordination can effectively navigate pathof the intelligent vehicle for the vehicle follow lane keepinglane changing and overtaking behavior PSO algorithmsolves the obstacle avoidance behavior toward the targetbehavior coordination due to behavior competition arisingperturbations In the process of driving at the same timethe change rule of heading direction and driving direction areconsistent and the vehiclersquos velocity is adjusted in the processof obstacle avoidance to remain consistent with the actualtraffic characteristics

5 Conclusion

(1) Based on the characteristics of intelligent vehicle drivingcombined with behavioral dynamics method we establishedmethod of the navigation path of the intelligent vehicle Thetarget is the time-variant

(2) Using PSO algorithm and competition behavior algo-rithm toward the target and avoidance weight coefficientsto optimize shows that PSO algorithm in behavior weightcoefficient optimization has advantages fast convergence andless parameter setting using behavioral dynamics methodto realize the control of the intelligent vehicle for the lateralheading control and the longitudinal velocity control so thatthe intelligent vehicle along a desired direction path can passsafely through the obstacles to target position


0 10 20 30 40 50

Lane keeping behavior running track of PSO

y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50

(a) Running track for PSO behavior coordination

Lane keeping behavior running trace of competition

0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50

(b) Running track to competitive behavior coordination

Hea

ding

dire

ctio

n an

gle (

∘ )

minus10

0

10

20

5 10 15 20 25 300Time (s)

(c) Heading direction of PSO behavior coordination

Hea

ding

dire

ctio

n an

gle (

∘ )

minus10

minus5

0

5

5 10 15 20 25 30 350Time (s)

(d) Heading direction of competitive behavior coordination

0

02

04

06

08

1

5 10 15 20 25 300Time (s)

120596ta

r120596ob

s

120596obs

120596tar

(e) PSO coordination factor of heading direction

0

02

04

06

08

1

5 10 15 20 25 30 350Time (s)

120596ta

r120596ob

s

120596obs

120596tar

(f) Competitive coordination factors of heading direction

Figure 2 Straight to heading direction coordinate

In addition using the behavior dynamics to the naviga-tion path can provide the relevant parameters for the nextmotion control Lateral control is concernedwith steering thevehicle automatically to follow the reference path The lon-gitudinal control is concerned with accelerating the vehicleautomatically The variable behavior dynamics is the relationbetween

120595 = 119891tar and the yaw the relation of V = 119891tarV(V)

longitudinal accelerationThe intelligent vehicle is controlledby using actuators such as the brakes the accelerator and thesteering wheel so that it adheres to the reference path [42]

(3)The experimental results show that behavioral dynam-ics method has the real-time performance and reliability andthe PSO algorithm can be a good alternative to the behaviorcoordination problems


Lane keeping behavior running track

0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50

(a) PSO behavior coordination of running track

minus5

0

5

10

15

Hea

ding

dire

ctio

n an

gle (

∘ )

5 10 15 20 25 300Time (s)

(b) PSO behavior coordination of heading direction

0

5

10

15

Path

velo

city

(km

h)

5 10 15 20 25 300Time (s)

tarobs

(c) PSO behavior coordination of velocity

5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar

(d) PSO coordination factor of heading direction

5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120582ta

r120582ob

s

120582obs

120582tar

(e) PSO coordination factors of velocity

Figure 3 Straight to heading direction and velocity of PSO behavior coordination


0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50Lane changing behavior running track

(a) The planning path of PSO lane change behavior coordination

0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50Overtaking behavior running track

(b) The planning path of PSO overtaking behavior coordination

Hea

ding

dire

ctio

n an

gle (

∘ )

0

10

20

30

40

50

5 10 15 20 25 30 350Time (s)

(c) PSO lane changing behavior coordination heading

Hea

ding

dire

ctio

n an

gle (

∘ )

minus50

0

50

100

5 10 15 20 25 30 35 40 450Time (s)

(d) PSO overtaking behavior coordination heading

5 10 15 20 25 35300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar

(e) PSO lane changing behavior coordination factor

5 10 15 20 25 3530 45400Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar

(f) PSO overtaking behavior coordination factors

Figure 4 Lane changing and overtaking heading direction coordination behavior

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

This work is supported by Project (10872160) of the NationalNatural Science Foundation of China Shaanxi Province


Important Disciplines of the Vehicle Engineering Con-struction of China Scientific Research Projects of ShaanxiProvince Education Office Key Laboratory (13JS070) andScientific Research Project in Shaanxi Province Departmentof Education (14jk1796)

References

[1] M Hada and T Tsuji ldquoA layout design method of human-vehicle systems based on equivalent inertia indicesrdquo Transac-tions of the Society of Instrument amp Control Engineers vol 43no 5 pp 400ndash407 2007

[2] G K D H Z C C Minglan ldquoProgress of the human-vehicle closed-loop system manoeuvrailityrsquos evaluation andoptimizationrdquo Chinese Journal of Mechanical Engineering vol39 pp 27ndash35 2003

[3] Z Q Liu F Lian J J Mao and L V Xue ldquoResearch on vehiclestability in driver-vehicle-road closed-loop systemrdquo Vehicle ampPower Technology no 2 pp 18ndash34 2012

[4] M Montemerlo S Thrun H Dahlkamp D Stavens and SStrohband ldquoWinning the DARPA grand challenge with anAI robotrdquo in Proceedings of the 21st National Conference onArtificial Intelligence (AAAI rsquo06) vol 1 Boston Mass USA July2006

[5] D Ferguson and A Stentz ldquoField Dlowast an interpolation-basedpath planner and replannerrdquo in Robotics Research vol 28 ofSpringer Tracts in Advanced Robotics pp 239ndash253 SpringerBerlin Germany 2005

[6] M Buehler K Iagnemma and S Singh ldquoThe 2005 DARPAgrand challengerdquo in Proceedings of 11th International Sympo-sium on Computational Intelligence in Robotics amp Automation(ICARCV rsquo10) pp 661ndash692 Singapore December 2010

[7] C Urmson J Anhalt D Bagnell et al ldquoAutonomous driving inurban environments boss and the urban challengerdquo Journal ofField Robotics vol 25 no 8 pp 425ndash466 2008

[8] M Montemerlo J Becker S Shat et al ldquoJunior the Stanfordentry in the urban challengerdquo Journal of Field Robotics vol 25no 9 pp 569ndash597 2008

[9] X-J Jing ldquoBehavior dynamics based motion planning ofmobile robots in uncertain dynamic environmentsrdquo Robotics ampAutonomous Systems vol 53 no 2 pp 99ndash123 2005

[10] Y Kuwata J Teo G Fiore S Karaman E Frazzoli andJ P How ldquoReal-time motion planning with applications toautonomous urban drivingrdquo IEEE Transactions on ControlSystems Technology vol 17 no 5 pp 1105ndash1118 2009

[11] A T Rashid A A Ali M Frasca and L Fortuna ldquoPathplanning with obstacle avoidance based on visibility binary treealgorithmrdquo Robotics amp Autonomous Systems vol 61 no 12 pp1440ndash1449 2013

[12] H Chen and Z Xu ldquoPath planning based on a new geneticalgorithmrdquo in Proceedings of the International Conference onNeural Networks and Brain (ICNNampB rsquo05) pp 788ndash792 BeijingChina October 2005

[13] M Gemeinder and M Gerke An Active Search AlgorithmExtending GA Based Path Planning for Mobile Robot SystemsSpringer London UK 2002

[14] M A P Garcia O Montiel O Castillo R Sepulveda and PMelin ldquoPath planning for autonomous mobile robot navigationwith ant colony optimization and fuzzy cost function evalua-tionrdquoApplied Soft Computing Journal vol 9 no 3 pp 1102ndash11102009

[15] D Blum and U Dortmund ldquoAnt colony optimization (ACO)rdquoJournal of Interdisciplinary Mathematics vol 8 pp 157ndash1682013

[16] H Mo and L Xu ldquoResearch of biogeography particle swarmoptimization for robot path planningrdquo Neurocomputing vol148 pp 91ndash99 2015

[17] L Lu and D Gong ldquoRobot path planning in unknown envi-ronments using particle swarm optimizationrdquo in Proceedings ofthe 4th International Conference onNatural Computation (ICNCrsquo08) pp 422ndash426 IEEE Jinan China October 2008

[18] J E Naranjo C Gonzalez R Garcıa and T De Pedro ldquoLane-change fuzzy control in autonomous vehicles for the overtakingmaneuverrdquo IEEE Transactions on Intelligent TransportationSystems vol 9 no 3 pp 438ndash450 2008

[19] R Schubert K Schulze and GWanielik ldquoSituation assessmentfor automatic lane-change maneuversrdquo IEEE Transactions onIntelligent Transportation Systems vol 11 no 3 pp 607ndash6162010

[20] Y Wang T Wei and X Qu ldquoStudy of multi-objective fuzzyoptimization for path planningrdquoChinese Journal of Aeronauticsvol 25 no 1 pp 51ndash56 2012

[21] M L Cummings J JMarquez andNRoy ldquoHuman-automatedpath planning optimization anddecision supportrdquo InternationalJournal of Human Computer Studies vol 70 no 2 pp 116ndash1282012

[22] G Schoner M Dose and C Engels ldquoDynamics of behaviortheory and applications for autonomous robot architecturesrdquoRobotics amp Autonomous Systems vol 16 no 2-4 pp 213ndash2451995

[23] A Steinhage and G Schoner ldquoThe dynamic approach toautonomous robot navigationrdquo in Proceedings of the IEEEInternational Symposiumon Industrial Electronics (ISIE rsquo97) vol11 pp S7ndashS12 Guimaraes Portugal July 1997

[24] G Schoner and M Dose ldquoA dynamical systems approachto task-level system integration used to plan and controlautonomous vehicle motionrdquo Robotics amp Autonomous Systemsvol 10 no 4 pp 253ndash267 1992

[25] E Bicho P Mallet and G Schoner ldquoUsing attractor dynamicsto control autonomous vehicle motionrdquo in Proceedings of the24thAnnual Conference of the IEEE Industrial Electronics Society(IECON rsquo98) vol 2 pp 1176ndash1181 IEEE Aachen GermanyAugust-September 1998

[26] S Monteiro and E Bicho ldquoAttractor dynamics approach toformation control theory and applicationrdquoAutonomous Robotsvol 29 no 3-4 pp 331ndash355 2010

[27] E Bicho and S Monteiro ldquoFormation control for multiplemobile robots a non-linear attractor dynamics approachrdquoin Proceedings of the IEEERSJ International Conference onIntelligent Robots and Systems (IROS rsquo03) vol 2 pp 2016ndash2022October 2003

[28] D Tian J Zhou Y Wang G Zhang and H Xia ldquoAn adaptivevehicular epidemic routingmethod based on attractor selectionmodelrdquo Ad Hoc Networks vol 36 part 2 pp 465ndash481 2016

[29] A C Hernandes H B Guerrero M Becker J-S Jokeit and GSchoner ldquoA comparison between reactive potential fields andAttractor Dynamicsrdquo in Proceedings of the 5th IEEE ColombianWorkshop on Circuits and Systems (CWCAS rsquo14) pp 1ndash5 IEEEBogota Colombia October 2014

[30] V Faria J Silva M Martins and C Santos ldquoDynamicalsystem approach for obstacle avoidance in a Smart Walkerdevicerdquo in Proceedings of the IEEE International Conference on


Autonomous Robot Systems and Competitions (ICARSC rsquo14) pp261ndash266 IEEE Espinho Portugal May 2014

[31] I Rano and I Iossifidis ldquoModelling human arm motionthrough the attractor dynamics approachrdquo in Proceedings ofthe IEEE International Conference on Robotics and Biomimetics(ROBIO rsquo13) pp 2088ndash2093 Shenzhen China December 2013

[32] F U Wei-Ping P F Zhang and S Q Yang ldquoBehavioraldynamics of mobile robot and rolling windows algorithm topath planningrdquo Computer Engineering and Applications vol 45pp 212ndash214 2009

[33] Y Shiqiang F Weiping and L Dexin ldquoApplication of dynamicsystem theory to mobile robot navigationrdquo Mechanical Scienceand Technology for Aerospace Engineering vol 29 pp 100ndash1042010

[34] F W H Dapeng Y Shiqiang and W Wen ldquoStudy on thenavigation method of behavior dynamics in mobile robotrdquoMechanical Science and Technology for Aerospace Engineeringvol 32 no 10 pp 1488ndash1491 2013

[35] Y Lei Q Zhu and Z Feng ldquoDynamic path planning usingvelocity obstacles and behavior dynamicsrdquo Journal of HuazhongUniversity of Science and Technology vol 39 no 4 pp 15ndash192011

[36] F W Han Gaining H Dapeng and W Wang ldquoStudy on themotion planning method of intelligent vehicle based on thebehavior dynamicsrdquo Mechanical Science and Technology forAerospace Engineering vol 34 no 2 pp 301ndash306 2015

[37] P Althaus and H I Christensen ldquoBehavior coordination instructured environmentsrdquo Advanced Robotics vol 17 no 7 pp657ndash674 2003

[38] S Yang W Fu D Li andWWang ldquoResearch on application ofgenetic algorithm for intelligent mobile robot navigation basedon dynamic approachrdquo in Proceedings of the IEEE InternationalConference on Automation and Logistics (ICAL rsquo07) pp 898ndash902 IEEE Jinan China August 2007

[39] L YanminResearch on dynamic path planningmethod formulti-robot systems [PhD thesis] Harbin Engineering UniversityHarbin China 2011

[40] R Coban ldquoA fuzzy controller design for nuclear researchreactors using the particle swarm optimization algorithmrdquoNuclear Engineering amp Design vol 241 no 5 pp 1899ndash19082011

[41] L Qingyang Numerical Analysis Tsinghua University Press2008

[42] G Tagne R Talj and A Charara ldquoDesign and validation ofa robust immersion and invariance controller for the lateraldynamics of intelligent vehiclesrdquo Control Engineering Practicevol 40 pp 81ndash92 2015

Submit your manuscripts athttpwwwhindawicom

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Distributed Sensor Networks


Advances in

FuzzySystems

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014


ReconfigurableComputing

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014


Applied Computational Intelligence and Soft Computing

thinspAdvancesthinspinthinsp

Artificial Intelligence

HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014

Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Electrical and Computer Engineering

Journal of

Journal of

Computer Networks and Communications


Hindawi Publishing Corporation

httpwwwhindawicom Volume 2014

Advances in

Multimedia


Biomedical Imaging


ArtificialNeural Systems

Advances in


RoboticsJournal of



Computational Intelligence and Neuroscience

Industrial EngineeringJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in



and in-depth research and experiments verify the feasibilityof this method

The aim of the paper is to discuss the intelligent vehiclepath planning problem in dynamic environment In theurban environment the road environment information isacquired through the vehicle sensor which is indispensablefor path planning and obstacles avoiding Behavior dynamicsmethod is used for the path planning to generate competitionbehavior problem while the particle swarm optimizationalgorithm is adopted to improve the behavior coordination

The rest of the paper is as follows Section 2 intro-duces behavior dynamicsmethodThe behavior coordinationof the standard particle swarm optimization algorithm isintroduced in Section 3 Simulation results of illustrationexamples and discussion are presented in Section 4 Finallyin Section 5 the conclusion and further discussion are given

2 Behavior Dynamics

The behavior dynamics were included towards the goal ofbehavior and obstacle avoidance behavior Using the behaviorcharacter of behavior dynamics plan according to the localtarget from a starting point of a path (namely the intelligentvehicle navigation path) and satisfying the vehicle in thedriveway can avoid the pavement and all sorts of obstaclesto the dynamic environmentThe heading angle and the pathvelocity in behavior dynamics model could be represented asthe following form [26]

120595 = 119891

1(120595 119875)

V = 1198912(V 119875)

(1)

where 119875 is position of the intelligent vehicle in the worldcoordinator system and the parameters are the heading angle120595 and the path velocity V of the intelligent vehicle whichgenerally act as the behavioral variables

120595 and V define therate of the heading angle and velocity as a function of theircurrent values

21 Behavior Dynamics Modeling The dynamic approach topath generation of the intelligent vehicles employs 120595 andV as the planning variable [22 23] The path planning iscontinuous in a time course of 120595 and V the target directions120595tar relative to the world coordinate system in which targetlie from the view of the intelligent vehicle and the obstacledirections 120595obs119894 in which obstacles lie from the viewpoint ofthe vehicle relative to the world coordinate system 120595tar and120595obs119894 are represented by attractive and repulsive force-letsacting on heading direction (Figure 1) The velocity of targetVtar and the velocity of obstacle Vobs are constraints that arerepresented by attractive and repulsive force-lets acting onthe path velocity In the dynamics system the variables arethe direction angle and the path velocity lying at obstaclesand target from the current position of the intelligent vehicleshown in Figure 1

According to the behavior variables and the behaviorpattern of the intelligent vehicle the target and obstacleavoidance of dynamics model is built

1

33

2 3

22

v

OY

X y

1

1

x

Obstacle 1

Obstacle 2

Obstacle 3

Vehicle

Veh

Veh

Veh

Target

120595

120595

tar120595obs1

120595obs2

120595obs3

obs

Figure 1 Variables represented in the dynamics

211TheTarget of BehavioralModels Toward target behavioris the behavior of the intelligent vehicle to reach its desti-nation The dynamic behavior of the attractor equation isexpected with negative slope and asymptotic speed convergesto a stable fixed point The attractors of the equation definedthe behavioral variables so that the system is ensured tobe at all times in a stable state The heading angle and thepath velocity of the behavior model are established as thefollowing

(1) Heading Direction Behavior Dynamics Model In thebehavior dynamics model the heading angle is specified asa time-variant term at any time 119905 Ψ is an angle and is cyclicand defined as 120595 isin [minus1205872 1205872] in the world coordinates andthe 119909-axis is always consistent with the direction of the roadIn [24] the heading direction of behavior dynamics for targetacquisition can be defined by

120595 = 119891tar (120595) = minus120582tar tan (120595 minus 120595tar)

120595tar = arctan(119875tar119910 minus 119875veh119910

119875tar119909 minus 119875veh119909)

(2)

where 120582tar gt 0 sets the attraction strength of the vehiclersquosheading variable 120595tar is the angle between the target and theintelligent vehicle in world coordinate system and 120595tar is theattractor in dynamics system 119891tar(120595) is an attractive force-let and converges to the value so that 120595 goes towards theintended target-state119875tar(119875tar119909 119875tar119910) is target position in theworld coordinator system and is the time-variant target and119875veh(119875veh119909 119875veh119910) is intelligent vehicle position in the worldcoordinator system

(2) The Path Velocity Behavior Dynamics Model Besidesheading direction in order to ensure the formation andmaintenance of the stabilization path velocity must be





2

2120590

2

V] (3)





2

2120590

2

119894

]

(4)





120595obs = 119865obs = sum

119894

119891obs119894 (120595obs119894) (5)


119904



2

2120590

2

V] (6)



120595 = 120596obs119865obs + 120596tar119891tar (7)












119894= (119883

1198941 119883

1198942 119883

119894119898)

119879 and speed is 119881119894=

(119881

1198941 119881

1198942 119881

119894119898)


119894= (119875

1198941 119875

1198942 119883

119894119898)


119892= (119875

1198921 119875

1198922 119883

119892119898)


119881

119894119896(119905 + 1) = 119908119881

119894119896(119905) + 119888

1119903

1(119875

119894119896(119905) minus 119883

119894119896(119905))

+ 119888

2119903

2(119875

119892119896(119905) minus 119883

119894119896(119905))

119883

119894119896(119905 + 1) = 119883

119894119896(119905) + 119881

119894119896(119905 + 1)

(9)



1

119888








119865

120595=

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

119889obs (120595 minus 120595tar)

119889tar (120595 minus 120595obs)

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

(10)



119865V =

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816


1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

(11)















5 Conclusion




0 10 20 30 40 50


y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50



0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50


Hea

ding

dire

ctio

n an

gle (

∘ )

minus10

0

10

20

5 10 15 20 25 300Time (s)


Hea

ding

dire

ctio

n an

gle (

∘ )

minus10

minus5

0

5

5 10 15 20 25 30 350Time (s)


0

02

04

06

08

1

5 10 15 20 25 300Time (s)

120596ta

r120596ob

s

120596obs

120596tar


0

02

04

06

08

1

5 10 15 20 25 30 350Time (s)

120596ta

r120596ob

s

120596obs

120596tar









0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50


minus5

0

5

10

15

Hea

ding

dire

ctio

n an

gle (

∘ )

5 10 15 20 25 300Time (s)


0

5

10

15

Path

velo

city

(km

h)

5 10 15 20 25 300Time (s)

tarobs


5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120582ta

r120582ob

s

120582obs

120582tar




0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



Hea

ding

dire

ctio

n an

gle (

∘ )

0

10

20

30

40

50

5 10 15 20 25 30 350Time (s)


Hea

ding

dire

ctio

n an

gle (

∘ )

minus50

0

50

100

5 10 15 20 25 30 35 40 450Time (s)


5 10 15 20 25 35300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 3530 45400Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar





Acknowledgments




References




















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in







2

2120590

2

V] (3)





2

2120590

2

119894

]

(4)





120595obs = 119865obs = sum

119894

119891obs119894 (120595obs119894) (5)


119904



2

2120590

2

V] (6)



120595 = 120596obs119865obs + 120596tar119891tar (7)












119894= (119883

1198941 119883

1198942 119883

119894119898)

119879 and speed is 119881119894=

(119881

1198941 119881

1198942 119881

119894119898)


119894= (119875

1198941 119875

1198942 119883

119894119898)


119892= (119875

1198921 119875

1198922 119883

119892119898)


119881

119894119896(119905 + 1) = 119908119881

119894119896(119905) + 119888

1119903

1(119875

119894119896(119905) minus 119883

119894119896(119905))

+ 119888

2119903

2(119875

119892119896(119905) minus 119883

119894119896(119905))

119883

119894119896(119905 + 1) = 119883

119894119896(119905) + 119881

119894119896(119905 + 1)

(9)



1

119888








119865

120595=

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

119889obs (120595 minus 120595tar)

119889tar (120595 minus 120595obs)

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

(10)



119865V =

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816


1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

(11)















5 Conclusion




0 10 20 30 40 50


y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50



0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50


Hea

ding

dire

ctio

n an

gle (

∘ )

minus10

0

10

20

5 10 15 20 25 300Time (s)


Hea

ding

dire

ctio

n an

gle (

∘ )

minus10

minus5

0

5

5 10 15 20 25 30 350Time (s)


0

02

04

06

08

1

5 10 15 20 25 300Time (s)

120596ta

r120596ob

s

120596obs

120596tar


0

02

04

06

08

1

5 10 15 20 25 30 350Time (s)

120596ta

r120596ob

s

120596obs

120596tar









0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50


minus5

0

5

10

15

Hea

ding

dire

ctio

n an

gle (

∘ )

5 10 15 20 25 300Time (s)


0

5

10

15

Path

velo

city

(km

h)

5 10 15 20 25 300Time (s)

tarobs


5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120582ta

r120582ob

s

120582obs

120582tar




0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



Hea

ding

dire

ctio

n an

gle (

∘ )

0

10

20

30

40

50

5 10 15 20 25 30 350Time (s)


Hea

ding

dire

ctio

n an

gle (

∘ )

minus50

0

50

100

5 10 15 20 25 30 35 40 450Time (s)


5 10 15 20 25 35300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 3530 45400Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar





Acknowledgments




References




















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in








119894= (119883

1198941 119883

1198942 119883

119894119898)

119879 and speed is 119881119894=

(119881

1198941 119881

1198942 119881

119894119898)


119894= (119875

1198941 119875

1198942 119883

119894119898)


119892= (119875

1198921 119875

1198922 119883

119892119898)


119881

119894119896(119905 + 1) = 119908119881

119894119896(119905) + 119888

1119903

1(119875

119894119896(119905) minus 119883

119894119896(119905))

+ 119888

2119903

2(119875

119892119896(119905) minus 119883

119894119896(119905))

119883

119894119896(119905 + 1) = 119883

119894119896(119905) + 119881

119894119896(119905 + 1)

(9)



1

119888








119865

120595=

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

119889obs (120595 minus 120595tar)

119889tar (120595 minus 120595obs)

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

(10)



119865V =

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816


1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

1003816

(11)















5 Conclusion




0 10 20 30 40 50


y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50



0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50


Hea

ding

dire

ctio

n an

gle (

∘ )

minus10

0

10

20

5 10 15 20 25 300Time (s)


Hea

ding

dire

ctio

n an

gle (

∘ )

minus10

minus5

0

5

5 10 15 20 25 30 350Time (s)


0

02

04

06

08

1

5 10 15 20 25 300Time (s)

120596ta

r120596ob

s

120596obs

120596tar


0

02

04

06

08

1

5 10 15 20 25 30 350Time (s)

120596ta

r120596ob

s

120596obs

120596tar









0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50


minus5

0

5

10

15

Hea

ding

dire

ctio

n an

gle (

∘ )

5 10 15 20 25 300Time (s)


0

5

10

15

Path

velo

city

(km

h)

5 10 15 20 25 300Time (s)

tarobs


5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120582ta

r120582ob

s

120582obs

120582tar




0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



Hea

ding

dire

ctio

n an

gle (

∘ )

0

10

20

30

40

50

5 10 15 20 25 30 350Time (s)


Hea

ding

dire

ctio

n an

gle (

∘ )

minus50

0

50

100

5 10 15 20 25 30 35 40 450Time (s)


5 10 15 20 25 35300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 3530 45400Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar





Acknowledgments




References




















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in













5 Conclusion




0 10 20 30 40 50


y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50



0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50


Hea

ding

dire

ctio

n an

gle (

∘ )

minus10

0

10

20

5 10 15 20 25 300Time (s)


Hea

ding

dire

ctio

n an

gle (

∘ )

minus10

minus5

0

5

5 10 15 20 25 30 350Time (s)


0

02

04

06

08

1

5 10 15 20 25 300Time (s)

120596ta

r120596ob

s

120596obs

120596tar


0

02

04

06

08

1

5 10 15 20 25 30 350Time (s)

120596ta

r120596ob

s

120596obs

120596tar









0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50


minus5

0

5

10

15

Hea

ding

dire

ctio

n an

gle (

∘ )

5 10 15 20 25 300Time (s)


0

5

10

15

Path

velo

city

(km

h)

5 10 15 20 25 300Time (s)

tarobs


5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120582ta

r120582ob

s

120582obs

120582tar




0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



Hea

ding

dire

ctio

n an

gle (

∘ )

0

10

20

30

40

50

5 10 15 20 25 30 350Time (s)


Hea

ding

dire

ctio

n an

gle (

∘ )

minus50

0

50

100

5 10 15 20 25 30 35 40 450Time (s)


5 10 15 20 25 35300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 3530 45400Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar





Acknowledgments




References




















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in




0 10 20 30 40 50


y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50



0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50


Hea

ding

dire

ctio

n an

gle (

∘ )

minus10

0

10

20

5 10 15 20 25 300Time (s)


Hea

ding

dire

ctio

n an

gle (

∘ )

minus10

minus5

0

5

5 10 15 20 25 30 350Time (s)


0

02

04

06

08

1

5 10 15 20 25 300Time (s)

120596ta

r120596ob

s

120596obs

120596tar


0

02

04

06

08

1

5 10 15 20 25 30 350Time (s)

120596ta

r120596ob

s

120596obs

120596tar









0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50


minus5

0

5

10

15

Hea

ding

dire

ctio

n an

gle (

∘ )

5 10 15 20 25 300Time (s)


0

5

10

15

Path

velo

city

(km

h)

5 10 15 20 25 300Time (s)

tarobs


5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120582ta

r120582ob

s

120582obs

120582tar




0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



Hea

ding

dire

ctio

n an

gle (

∘ )

0

10

20

30

40

50

5 10 15 20 25 30 350Time (s)


Hea

ding

dire

ctio

n an

gle (

∘ )

minus50

0

50

100

5 10 15 20 25 30 35 40 450Time (s)


5 10 15 20 25 35300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 3530 45400Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar





Acknowledgments




References




















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in





0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45

50


minus5

0

5

10

15

Hea

ding

dire

ctio

n an

gle (

∘ )

5 10 15 20 25 300Time (s)


0

5

10

15

Path

velo

city

(km

h)

5 10 15 20 25 300Time (s)

tarobs


5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 300Time (s)

0

02

04

06

08

1

120582ta

r120582ob

s

120582obs

120582tar




0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



Hea

ding

dire

ctio

n an

gle (

∘ )

0

10

20

30

40

50

5 10 15 20 25 30 350Time (s)


Hea

ding

dire

ctio

n an

gle (

∘ )

minus50

0

50

100

5 10 15 20 25 30 35 40 450Time (s)


5 10 15 20 25 35300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 3530 45400Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar





Acknowledgments




References




















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in




0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



0 10 20 30 40 50y (m)

0

5

10

15

x (m

)

20

25

30

35

40

45



Hea

ding

dire

ctio

n an

gle (

∘ )

0

10

20

30

40

50

5 10 15 20 25 30 350Time (s)


Hea

ding

dire

ctio

n an

gle (

∘ )

minus50

0

50

100

5 10 15 20 25 30 35 40 450Time (s)


5 10 15 20 25 35300Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar


5 10 15 20 25 3530 45400Time (s)

0

02

04

06

08

1

120596ta

r120596ob

s

120596obs

120596tar





Acknowledgments




References




















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in





References




















































Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in
























Advances in

FuzzySystems


Volume 2014












Journal of

Journal of





Advances in

Multimedia


Biomedical Imaging



Advances in


RoboticsJournal of










Advances in



research article the study of intelligent vehicle...

Documents