You are on page 1of 6

Proceedings of the Fourth International Conference on Machine Learning and Cybernetics, Guangzhou, 18-21 August 2005

FUZZY LOGIC BASED ROBOT PATH PLANNING IN UNKNOWN


ENVIRONMENT
MENG WANG, JAMES N.K. LIU

Department of Computing, The Hong Kong Polytechnic University, Hung Hom, Hong Kong
E-MAIL: {csmwang, csnkliu}@comp.polyu.edu.hk

Abstract:
This paper proposes a new method, minimum risk 2. The path searching behavior
approach, to address the local path planning to escape from
local minimum during goal-oriented robot navigation in The Path-Searching (PS) behavior is one application of
unknown environments. This approach is theoretically proved
our memory grid technique. The memory grid is a new type
to guarantee global convergence even in the long-wall,
unstructured, cluttered, maze-like, and modified environments. of map for modeling the robot environment. Memory grid
The approach adopts a strategy of multi-behavior differs from other maps (e.g. occupancy grid, certainty grid,
coordination, in which a novel path-searching behavior is connectivity graph [8, 9]) chiefly in that it makes use of an
developed to recommend the regional direction with minimum obstacle and trajectory memory dot model. The memory
risk. The paper provides a fuzzy logic framework to grid uses the obstacle memory dots to quantify the fuzzy
implement the behavior design and coordination. It is verified possibility that represents the uncertainty of obstacles
by the simulated and real world tests. detected by sonar sensors, while it uses the trajectory
memory dots to save the trajectory traversed by robot. The
Keywords: longer the robot takes to traverse a grid region, the greater
Fuzzy control; intelligent control systems; mobile robot the number of trajectory memory dots saved in this grid
motion planning; local path planning
(see Fig.6(9)). Therefore, the regional direction with the
minimum number of obstacle and trajectory memory dots
1. Introduction can represent the regional direction with the minimum risk.
This direction is the good choice for the robot in trying to
We call it “blind goal-oriented navigation” that the avoid a previous trajectory, and consequently escape from
robot is required to autonomously reach a desired goal but local minimum and reach the goal. PS behavior does this
it does not have a priori known environmental knowledge. job to recommend the direction that entails minimum risk.
The related methods [1-7] can be categorized as two types:
boundary following, and virtual subgoal. Unfortunately, 2.1. Regional Risk Index
they have difficulties to guarantee the global convergence
in the complex environments since they may get trapped in
Four features are extracted from memory grid data, in
the local minimum (or dead end). This paper proposes a
which iteration risk α and collision risk β are used to
new method, minimum risk approach, to address the local
path planning to handle the local minimum during the blind infer the Risk Index for the fuzzy navigational rules of PS
goal-oriented robot navigation. This approach adopts a behavior, and the trajectory dot intensity κ and obstacle
strategy of multi-behavior coordination, in which a novel dot intensity τ are used in combination with fuzzy logic to
Path-Searching (PS) behavior is developed to recommend calculate the weight of PS behavior. The magnitude of
the regional direction that is offering minimum risk. The iteration risk α is converted into three linguistic fuzzy
final command output is obtained by coordinating three sets {LOW, MEDIUM, HIGH}, with the membership functions
behaviors: PS, Obstacle-Avoidance (OA), and shown in Fig. 1(a). The other three are similar.
Goal-Seeking (GS). The paper provides a fuzzy logic The Fuzzy Rule-Based Risk Index combines the two
framework to implement the behavior design and regional risk parameters into a single indicator of safety of
coordination. Thus errors due to sensor noise and traversal of the region by the mobile robot. The Risk Index
self-localization are effectively handled by our navigation r is represented by three linguistic fuzzy sets {DANGEROUS,
system. UNCERTAIN, SAFE}, with the membership functions shown

0-7803-9091-1/05/$20.00 ©2005 IEEE


813
Proceedings of the Fourth International Conference on Machine Learning and Cybernetics, Guangzhou, 18-21 August 2005

µ (α ) µ (r )
low medium high dangerous uncertain safe
1 1

α r
0 α1 α2 α3 α4 α5 0 r1 r2 r3 r4 r5
(a) (b)
Fig. 1. Membership functions (a) for iteration risk. (b) for Risk Index

in Fig. 1(b). The Risk Index is defined in terms of both the


iteration risk and the collision risk by a set of simple
intuitive fuzzy logic relations as Table I.

Table I. Rule sets of regional Risk Index r


α β HIGH MEDIUM LOW
HIGH DANGEROUS DANGEROUS DANGEROUS
Table II. Turn rules for path-searching behavior
MEDIUM DANGEROUS DANGEROUS UNCERTAIN
LOW DANGEROUS UNCERTAIN SAFE
Center of Gravity (COG) method is used for defuzzification
[10]. For complexity reduction, we assume that the speed of
2.2. Turn rules and a complement algorithm the robot is only influenced by the OA behavior, not by the
PS or GS behavior. In addition, it is assumed that the robot
The motion control variables of the mobile robot are can only move in the forward direction (i.e., reverse motion
the translational speed and the rotational turn angle. The is not allowed). The region available for robot traversal is
robot speed is represented by three linguistic fuzzy sets divided up into three circular side sectors. The radius of the
{STOP, SLOW, FAST}, with the membership functions shown circular sector is a user-defined regional perception range
in Fig. 2(a). The robot turn angle is represented by five of the robot, i.e., the distance at which we wish the robot to
linguistic fuzzy sets {NB, NS, ZE, PS, PB}, with the react to the regional risk features. The Risk Indices for the
membership functions shown in Fig.2(b), where NB is three regions, rleft , rfront and rright, are inferred by the Risk
negative-big, NS negative-small, ZE zero, PS positive-small, Index rules.
and PB positive-big. Positive and negative mean that the The turn rules for the PS behavior are summarized in
robot turns to the left and right, respectively. Table II. Notice that in Table II, when the robot needs to
µ (θ ) turn, but the left and right sectors have the same Risk
µ (v ) fast Indices, then the recommended turn angle is GOAL, where
stop slow NB NS ZE PS PB
1 GOAL implies that the recommended turn angle should be
toward the direction close to the goal location. Another
v θ
0 v1 v2 v3 −θ3 −θ 2 −θ1 0 θ1 θ2 θ3 Algorithm: MINIMUM COLLISION & ITERATION (MCI):
(a) (b) {
Fig. 2. Membership functions (a) for speed. (b) for turn angle. Step 1: IF the turn direction recommended by the turn
rules has a lower iteration risk than the threshold α1 , THEN
maintain this original turn direction and return; OTHERWISE
go to Step 2.
Step 2: To check whether or not there are regional
directions with a lower collision risk than the threshold β1 . IF
not, THEN maintain the original turn direction and return;
OTHERWISE go to Step 3.
Step 3: IF there exists only one regional direction that has
Fig.3. Turn angle determination of PS behavior by both Turn Rules and a lower collision risk, THEN this regional direction is
MCI Algorithm recommended as the turn direction and return; OTHERWISE
go to Step 4.
Step 4: IF there exist multiple regional directions that
The Risk Index is used to develop simple fuzzy rules have a lower collision risk, THEN a regional direction that has
for determination of the robot turn angle as shown in Fig.3. a minimum iteration risk value is recommended as the turn
direction and return.
FLC implies the fuzzy logic controller. We use a Mamdani }
model as a fuzzy inference engine, and the Centroid or

814
Proceedings of the Fourth International Conference on Machine Learning and Cybernetics, Guangzhou, 18-21 August 2005

important note, a turn maneuver is not initiated when the turn, but the left and right sectors have the same obstacle
three sectors have the same dangerous risk indices as distance, then the recommended turn angle is GOAL, where
shown in the (1,1) element of the top layer in Table II. The GOAL implies that the recommended turn angle should be
turn rule does not at this stage force the robot to arbitrarily toward the direction close to the goal location. This is
choose between left and right, but maintains the turn angle similar to the turn rules for PS behavior. One last important
at zero. The final selection will be made by a complement note: when the three sectors have the same VERYNEAR
algorithm, Minimum Collision & Iteration (MCI). The MCI obstacle distance as shown in the (3,3) element of the top
algorithm is exact, not fuzzy. layer in Table IV(a), a large left turn (PB) angle is
recommended. This turn rule enables the robot to escape
2.3. Weight rules from its current embarrassed situation.
d right
The weighting factor wps represents the strength by far near verynear
d front
d left far
which the PS behavior recommendation is taken into GOAL PS PB
verynear
account to calculate the final motion command. The weight near
NS GOAL PS
of PS behavior is represented by three linguistic fuzzy sets verynear
NB NS PB
{SMALL, MEDIUM, LARGE}, and is derived directly from d right
both the trajectory dot intensity κ and obstacle dot
far near verynear
dleft far d front
GOAL PS PS
intensity τ of a square region, using the rule sets in Table near
near
NS ZE PS
III. verynear
Table III. Weight rules of the PS behavior NS NS ZE

κ τ HIGH MEDIUM LOW far


d right
near verynear
HIGH LARGE LARGE LARGE d left far d front
ZE ZE PS
far
MEDIUM LARGE LARGE MEDIUM near
ZE ZE PS
LOW LARGE MEDIUM SMALL verynear
NS NS ZE

(a)
d right
3. The obstacle avoidance behavior far near verynear
dleft far d front
large large large
The local Obstacle-Avoidance (OA) behavior is near
verynear
large large large
actually a sensor-based behavior which implements a verynear
large large large
control strategy based on external sensing. OA behavior is d right
effective if obstacles are close. We adopt a fuzzy logic far near verynear
d front
controller for the implementation. The sonar readings of the dleft far
medium large large
near
robot are grouped into three sectors (left, front, right). For near
large large large
example, our robot has a ring of eight forward ultrasonic verynear
large large large
sonars that produce a set of obstacle distances {d0, d1, d2, d right
d3, d4, d5, d6, d7}. We obtain three groups of obstacle far near verynear
dleft far d front
distances by the following equations: dleft = min(d0, d1); small small large
far
dfront = min(d2, d3, d4, d5); dright = min(d6, d7). The near
small medium large
obstacle distance of each sector is represented by three verynear
large large large
linguistic fuzzy sets {VERYNEAR, NEAR, FAR}. The obstacle
(b)
avoidance navigation rules are discussed below. Table IV. (a) Turn rules for OA behavior. (b) Weight rules for OA behavior
The turn rules for the OA behavior are summarized in
Table IV(a). Observe that the rules exhibit such a behavior The robot speed v is based on the close obstacle
characteristic: if the obstacle distance in any sector is distance in the three sectors. The speed rules are as follows.
VERYNEAR, the robot should turn away to find a safer 1) IF dfront is VERYNEAR, THEN v is STOP.
direction. For instance, the (1,3) element of the bottom 2) IF dfront is NEAR, THEN v is SLOW.
layer in Table IV(a) can be written out as the rule: 3) IF dfront is FAR AND dleft is FAR AND dright is FAR, THEN v is
IF dfront is FAR AND dleft is FAR AND dright is VERYNEAR, THEN FAST.
4) IF dfront is FAR AND (dleft is VERYNEAR OR dleft is NEAR OR
θ oa is PS. dright is VERYNEAR OR dright is NEAR), THEN v is SLOW.
In addition, in Table IV(a), when the robot needs to The last rule enables the robot to decrease its speed

815
Proceedings of the Fourth International Conference on Machine Learning and Cybernetics, Guangzhou, 18-21 August 2005

when an obstacle is closing. Because in the environmental goal-oriented navigation task, this solution path must have
maps of the tests the exit location is very narrow and often the minimum obstacle and trajectory memory dots, i.e. the
situated in the lateral wall, a fast speed may cause the robot minimum risk. And the detected regions of the current robot
to miss the exit. position must have a regional direction that has a minimum
Like the weight of PS behavior, the weight of OA risk. The minimum risk approach can guarantee to finally
behavior woa is represented by three linguistic fuzzy sets recommend the region direction with minimum risk. Thus
{SMALL, MEDIUM, LARGE}, and is derived directly from the solution path must be found and the robot finally
obstacle distances in the three sectors. The weight rules for reaches the goal. That’s the global convergence. ■
the OA behavior are summarized in Table IV(b). Corollary 1: Even in the modified environment, the
minimum risk approach can guarantee global convergence
4. The goal seeking behavior if a solution path exists.

Unlike other approaches [11], we first assume that the


Goal-Seeking (GS) behavior does not influence the speed
of the robot, and contributes only to the rotational turn
angle. Second, we use a very simple analytical model rather
than a set of fuzzy logic navigation rules. We set the turn (a) (b)
angle recommended by the GS behavior as the heading Fig.5. “trial-and-return” behavior phenomenon.
error between the current robot heading and goal direction.
Thus, the value domain of this turn angle is (-180 ° , 180 ° ]. The minimum risk approach produces the
The weight of GS behavior wgs is based on the weights “trial-and-return” behavior phenomenon as shown in
of both OA and PS behaviors. Fig. 4 shows the weight Fig.5(a) and (b). Obviously, if there is not obstacle to block
determination of three behaviors. The simple weight rules the nearest exit, the “trial-and-return” behavior
are as follows: phenomenon must enable the robot to find it and escape
1) IF wps is LARGE OR woa is LARGE OR wps is MEDIUM OR woa from local minimum. This is the Theorem 2. It is verified
is MEDIUM, THEN wgs is SMALL. by the experiments in Section 6.
2) IF wps is SMALL AND woa is SMALL, THEN wgs is LARGE. Theorem 2: If there is no obstacle blocking the nearest
Importantly, the weight of GS behavior is suppressed exit, the minimum risk approach is guaranteed to find a
and small when any one weight of the OA and PS behaviors nearest exit to escape from the local minimum and
is not SMALL. When the weights of both OA and PS are ultimately reach the goal. ■
SMALL, the GS behavior makes a dominant contribution to
the final control command. Although the GS behavior is 6. Experimental results
often suppressed, the GOAL factor is reflected in the turn
rules of both OA and PS behaviors (see Tables II and IV). It Our approach is tested on a mobile robot (vehicle).
is an important factor for the minimum risk approach to The robot is equipped with a ring of eight forward
guarantee global convergence. ultrasonic sonars that can detect obstacles within three
meters. We verify the performance of our minimum risk
approach that is applied to an unknown long-wall
environment with local minimum as shown in Fig.6.
When the robot starts to move at a normal (maximum)
speed, the number of the memory dots (including the
obstacle and trajectory memory dots) saved in the memory
Fig.4. Weight determination of OA, PS, GS behaviors grid is SMALL, so that the trajectory and obstacle dot
intensity are SMALL (see Section 2). The weight of the PS
5. Performance analysis in theory behavior is thus small. At this time the weight of the OA
behavior is small as well because the front obstacle is
Theorem 1: If there exists a solution path for a blind distant. Consequently, at this time the GS (line c) behavior
goal-oriented navigation task, the minimum risk approach makes the dominant contribution to the final motion output
can guarantee the global convergence. (Fig.6 (1)). When in response to a nearby obstacle the robot
Proof: Limited by the page numbers, we briefly gave decreases its speed, the number of memory dots increases
the proof analysis. If there exists a solution path for a blind and the memory dot intensity becomes MEDIUM or LARGE.

816
Proceedings of the Fourth International Conference on Machine Learning and Cybernetics, Guangzhou, 18-21 August 2005

(1) (2) (3) (4) (5)

(6) (7) (8)


100
OA
80 B PS
GS
60
E Final
) F
e 40
er
g
e 20
d( S A
e
gl 0
n T
a
nr -20
ut
at -40
l
e
D
-60

-80
D Time
C
-100
0 50 100 150 200 250 300 350 400 450 500

(9) (10)
Fig. 6. Minimum risk approach to the long-wall environment with local minimum. (1-8) S is the start, T the goal target. OA is the red line “a”,
PS is the blue line “b”, GS is the black line “c”. (9) Underlying memory grid is shown by the spaced horizontal and vertical lines; The obstacle
memory dots are drawn as the black squares of different size. The larger is the size of the square, the higher the possibility of the obstacle is; The
trajectory memory dots are drawn as the red circles of different size. The larger is the circle, the more the number of trajectory dots is. (10) Turn
angles recommended by different behaviors. For the display, the GS turn angle is a half.

Consequently, the weight of PS behavior increases, and PS environment as shown in Fig.8(a), because the robot
behavior (line b) is effective in these cases (see encounters another local minimum at the location “b” and
(2)(3)(4)(5)(6) in Fig.6); When the front of the robot is “c” when it is working under the influence of previous
closing upon the obstacles, the weight of OA behavior (line virtual subgoal. Krishna and Kalra’s method [3] has a good
a) becomes large (see (2)(4)(6) in Fig.6); When the OA or result as shown in Fig.8(b). But it highly depends on the
PS behaviors are dominant, the GS behavior is suppressed landmark recognition and exact coordination localization.
and its weight is small (see (2)(3)(4)(5)(6) in Fig.6). When In addition, it is difficult to choose a correct direction to
the robot is far from the obstacles and is approaching the follow the wall boundary as seen in Fig.8(e). Maaref and
goal T at a normal speed, the weights of both OA and PS Barret’s method [4] fails to reach the goal in such large
behavior are small and only GS behavior is dominant (Fig.6 concave environment, because it detects the local minimum
(7)). Fig. 6(9) shows the underlying memory grid as well as using a restricted criterion that all sensors must give the
the control interface. small obstacle distances at the same time. Fig.8(c) shows
Observe that in Fig. 6 (8) and (10), the labels A, B, C, the result of our minimum risk approach. The robot exhibits
D, E, F, S, and T represent the robot locations. The turn the typical “trial-and-return” behavior phenomenon. This
angles recommended by OA and PS behavior are consistent property can help the robot to find the nearest exit to escape
during most of the entire task period. At the locations B, C, from the local minimum. It’s further verified in the Fig.8(f).
and D, the goal T is switched from the left of the robot to Fig.8(d) shows the result of Huang and Lee’s method [1].
the right, or from the right to the left. This is why the robot This method has a conservative leaving criterion that makes
leaves the wall at location D and turn toward location E. the robot traverse a longer path compared with other
Fig.7(a) and (b) show the results in the maze-like, methods that adopt the boundary-following strategy. More
unstructured and cluttered environments respectively. importantly, it is still difficult to choose the correct
We compare our minimum risk approach with other boundary-following direction. Similar problems occur on
related methods. The virtual target method [7] fails to reach the Distbug method [2] and Virtual-target-side method [5].
the goal in such concave and recursive U-shape Now we exhibit a real world test in a modified

817
Proceedings of the Fourth International Conference on Machine Learning and Cybernetics, Guangzhou, 18-21 August 2005

environment. Fig.9(a) shows a picture of robot movement. 7. Conclusions


One person first blocks the exit “A” (Fig. 9(b)), which
forces the robot to turn around in order to look for another The minimum risk approach is theoretically proved to
exit. Then the person moves to location “B” where he is guarantee global convergence even in long-wall, large
closing the robot but leaves the exit “A” clear. The robot concave, recursive U-shape, unstructured, cluttered,
finds the exit “A” while avoiding the moving obstacle (i.e. maze-like, and modified environments. The fuzzy logic
the person). Fig.9(b) shows the actual trajectory and the framework outperforms the analytical methods of detecting
obstacle memory dots. Fig.9 (c) shows the memory grid and leaving criteria adopted by other approaches. The test
that saves the trajectory and obstacle memory dots. The results can be checked on website [12].
result verifies that there is little influence by odometry drift
problem if our minimum risk approach is applied in a small Acknowledgements
scale of space where the local minimum occurs and even it
is dynamic. For the local minimum problem of a large scale The authors would like to acknowledge the partial
of space, we have to choose other localization technique to support of the Hong Kong Polytechnic University via RGC
compensate for the drift error. grant B-Q515 and departmental grant H-Z87.

References

[1] H.P. Huang, P.C. Lee, “A real-time algorithm for obstacle


avoidance of autonomous mobile robots”, Robotica, Vol.10,
217-227, 1992
[2] I.Kamon, E.Rivlin, “Sensory-based motion planning with global
proofs”, IEEE Transactions on Robotics and Automation, Vol.13,
Iss.6, 814-822, 1997
[3] K.M. Krishna, P.K. Kalra, “Perception and remembrance of the
environment during real-time navigation of a mobile robot”,
Robotics and Autonomous Systems, vol.37, 25-51, 2001
(a) (b) [4] H.Maaref, C.Barret, “Sensor-based navigation of a mobile robot in
Fig.7. Minimum risk approach to address the local minimum problem (a) in an indoor environment”, Robotics and Autonomous Systems,
maze-like environment. (b) in unstructured and cluttered environment. Vol.38, 1-18, 2002
[5] R.Chatterjee, F.Matsuno, “Use of single side reflex for autonomous
navigation of mobile robots in unknown environments”, Robotics
and Autonomous Systems, Vol.35, 77-96, 2001
[6] F.G. Pin, S.R. Bender, “Adding memory processing behavior to the
fuzzy behaviorist approach: Resolving limit cycle problems in
mobile robot navigation”, Intelligent Automation and Soft
Computing, Vol.5, Iss.1, 31-41, 1999
[7] W.L. Xu, S.K. Tso, “Sensor-based fuzzy reactive navigation for a
mobile robot through local target switching”, IEEE Transactions on
Systems, Man and Cybernetics, Part C, Vol.29, No.3, 451-459, 1999
[8] G.Oriolo, G.Ulivi, M.Vendittelli, “Real-time map building and
(a) (b) (c) navigation for autonomous robots in unknown environments”, IEEE
Trans. on SMC, Part B, Vol.28, No.3, 316-333, June 1998
[9] A.C.Victorino, P.Rives, J.J.Borrelly, “Safe navigation for indoor
mobile robots. Part II: exploration, self-localization, and map
building”, The International Journal of Robotics Research, Vol.22,
1019-1039, Dec. 2003
[10] Meng Wang, James N.K. Liu, "Autonomous Robot Navigation
using Fuzzy Logic Controller", IEEE Conference on Machine
Learning and Cybernetics, Shanghai, China, 691-696, August 2004
(d) (e) (f)
[11] H.Seraji, A.Howard, “Behavior-based robot navigation on
Fig.8. (a) virtual target method. (b) Krishna and Kalra’s method. (c) our
challenging terrain: A fuzzy logic approach”, IEEE Transactions on
minimum risk method. (d) Huang and Lee’s method. (e) Krishna and Kalra’s
Robotics and Automation, Vol.18, Iss.3, 308-321, Jun 2002
method. (f) our minimum risk method.
[12] http://www4.comp.polyu.edu.hk/~csnkliu/polyuibot/

(a) (b) (c)


Fig.9. Performance in dynamic real world with local minimum

818

You might also like