![](bg7.png)
Figure 7: Proposed method when the environment contains
a large number of obstacles.
As expected, the use of a switching strategy yields to
a less smooth trajectory w.r.t. the one obtained by the
traditional APF method. Note that the proposed plan-
ning rules aim at driving the robot to the goal what-
ever is the obstacles configuration with no regards on
other trajectory optimality criteria. It follows that the
obtained robot path could be slower/faster or more
complicated w.r.t. the one obtained with other tech-
niques but, on the other hand, the proposed method
will always provide a solution to the target reaching
with obstacles avoiding problem.
In the third testing configuration, the robot has
been placed in [1,3]
T
with an initial heading of
θ
r
(0) = 0, the goal is in [x
G
,y
G
] = [10,3]
T
and a 16
obstacles, placed as shown in Fig. 7, have been used
to obstruct the robot movements. This testing con-
figuration proposes various local minima situations in
the traditional APF case. For example, if the obstacle
O
16
is equipped with a traditional artificial repulsive
potential, it could prevent the robot from reaching the
goal due to the imposed repulsion when the robot is
close to its target. On the contrary, using the proposed
planning method, the robot is driven to the goal with
no local minima situations and avoiding all the obsta-
cles in the environment, as shown in Fig. 7. In partic-
ular, obstacle O
16
does not affect robot path since it is
placed after the goal, in the robot point of view, and
as a consequence it is out of the tube T (r(t),G,R
m
).
7 CONCLUSIONS
In this work, a novel approach to the artificial poten-
tials method has been proposed to face the path plan-
ning and obstacles avoidance problem for a mobile
robot. The new method has been developed aiming
at overcoming the well known local minima problem
of the traditional APF technique. A novel helicoidal
potential has been proposed to allow for bypassing
an obstacle using only the effect of a single potential
field, with no need for the summation of a repulsive
one and an attractive one, in order to avoid local min-
ima. In this context, the proposed method is based
on the use of a single potential at a time, switching
from attractive to bypassing case depending on a set
of defined switching rules.
Moreover, since only local information is used,
the proposed technique ensures high robustness, in
terms of achieved performance and computational
complexity, w.r.t. the number of obstacles.
The described method has been compared, in a nu-
merical way, with traditional APF technique, and has
shown a more robust behavior w.r.t. it, providing a
feasible path to the robot goal also in the case of a
framework with multiple obstacles to be avoided.
As a future research direction, the proposed tech-
nique will be extended to the case of mobile obstacles.
REFERENCES
Alajlan, M., Kouba, A., Chari, I., Bennaceur, H., and Am-
mar, A. (2013). Global path planning for mobile
robots in large-scale grid environments using genetic
algorithms. In 2013 International Conference on In-
dividual and Collective Behaviors in Robotics (ICBR),
pages 1–8.
Araujo (2006). Prune-able fuzzy art neural architecture for
robot map learning and navigation in dynamic envi-
ronments. IEEE Transactions on Neural Networks,
17(5):1235–1249.
Benzerrouk, A., Adouane, L., and Martinet, P. (2012). Dy-
namic obstacle avoidance strategies using limit cycle
for the navigation of multi-robot system. In IEEE/RSJ
International Conference on Intelligent Robots and
Systems, 4th Workshop on Planning, Perception and
Navigation for Intelligent Vehicles.
Bi, Z., Yimin, Y., and Wei, Y. (2008). Hierarchical path
planning approach for mobile robot navigation under
the dynamic environment. In 2008 6th IEEE Inter-
national Conference on Industrial Informatics, pages
372–376.
Bopardikar, S. D., Englot, B., and Speranzon, A. (2015).
Multiobjective path planning: Localization con-
straints and collision probability. IEEE Transactions
on Robotics, 31(3):562–577.
Chamberland, S., Beaudry, ., Clavien, L., Kabanza, F.,
Michaud, F., and Lauriay, M. (2010). Motion plan-
ning for an omnidirectional robot with steering con-
straints. In 2010 IEEE/RSJ International Conference
on Intelligent Robots and Systems, pages 4305–4310.
Choset, H., Lynch, K. M., Hutchinson, S., Kantor, G. A.,
Burgard, W., Kavraki, L. E., and Thrun, S. (2005).
Principles of Robot Motion: Theory, Algorithms, and
Implementations. MIT Press, Cambridge, MA.
Path Planning and Obstacles Avoidance using Switching Potential Functions
349