![](bg4.png)
In the bottom panel, we can note that the direction
of the robot,
ˆ
φ
h
, (greencontinuousline) changes when
obstacles in the trajectory are detected (shadow ar-
eas). These changes are required for obstacle circum-
navigation. However, as soon as the obstacles were
circumnavigated, the heading direction converges to
the goal direction
ˆ
ψ
tar
. This experiment demonstrated
0 2 4 6 8 10 12 14 16 18
0
2
4
6
8
20
Dx,Dx
^
(m)
0 2 4 6 8 10 12 14 16 18
-0.5
-0.25
0
0.25
0.5
Dy,Dy
^
(m)
20
0 2 4 6 8 10 12 14 16 18
−0.5
0
0.5
1
1.5
20
Time (s)
^
φ ,
h
ψ
tar
^
(rad)
Figure 5: Top panel (Middle): Measured D
x
(D
y
) by the
robot’s visual system (blue continuous line), and estimated
ˆ
D
x
(
ˆ
D
y
) by the EKF (continuous green line). Bottom: Esti-
mated robot’s heading direction
ˆ
φ
h
(green continuous line),
and the angle
ˆ
ψ
tar
(red dashed line) that the robot has to
follow.
that the EKF is able to filter and to estimate sensorial
information, in this case visual information about the
goal, and simultaneously, to ensure that the temporal
nature of the nonlinear dynamical system approach is
not degraded, even when the robot has to deal with
dynamical obstacles in a cluttered environment.
6 CONCLUSIONS
The purpose of this paper was to address the problem
of generating timed trajectories for an autonomous
mobile robot equipped with noisy and low-level sen-
sory information, while simultaneously has to esti-
mate the goal location using an EKF.
We have successfully demonstrated that the inte-
gration of a standard EKF and a nonlinear dynamical
system to robotics in the same approach without de-
grading the temporal nature of the proposed architec-
ture. Moreover, the inclusion of the EKF has allowed
to reduce the error between the final position of the
robot and the position of the goal.
For future work, we intend to address other more
complex and cluttered environments, as well as to op-
erate in a real hospital environment.
ACKNOWLEDGEMENTS
Work supported by the Portuguese Science Foun-
dation (grant PTDC/EEA-CRO/100655/2008),
and by project FCT PEst-OE/EEI/LA0009/2011.
Jorge B. Silva is supported by PhD Grant
SFRH/BD/68805/2010, granted by the Portuguese
Science Foundation.
REFERENCES
Cui, X., Zhu, Y., Zang, X., Tang, S., and Zhao, J. (2010).
Cpg based locomotion control of pitch-yaw connect-
ing modular self-reconfigurable robots. IEEE Inter-
national Conference on Information and Automation
(ICIA), pages 1410 – 1415.
Ijspeert, A., Nakanishi, J., and Schaal, S. (2002). Learning
attractor landscapes for learning motor primitives. In
Advances in Neural Information Processing Systems
15, pages 1547–1554. MIT Press.
Ronsse, R., Vitiello, N., Lenzi, T., van den Kieboom, J.,
Carrozza, M., and Ijspeert, A. (2010). Human-robot
synchrony:flexible assistance using adaptive oscilla-
tors. IEEE Transactions on Biomedical Enginnering,
58:1001 – 1012.
Santos, C. (2004). Generating timed trajectories for an au-
tonomous vehicle: a non-linear dynamical systems ap-
proach. in Proc. of the IEEE Int. Conf. on Robotics
and Automation (ICRA).
Santos, C. and Matos, V. (2011). Gait transition and mod-
ulation in a quadruped robot: A brainstem-like modu-
lation approach. Robot. Auton. Syst., 59:620 – 634.
Sch¨oner, G. (1994). Dynamic theory of action - perception
patterns: The time before-contact-paradigm. Human
Mov. Science, 3:415 – 439.
Sch¨oner, G. and Santos, C. (2001). Control of movement
time and sequential action through attractor dynamics:
A simulation study demonstrating object interception
and coordination. in Proc. of the 9th Int. Symposium
on Intelligent Robotic Systems (SIRS).
Silva, J., Santos, C., and Matos, V. (2010a). Generating tra-
jectories with temporal constraints for an autonomous
robot. In 8th IEEE International Workshop on Safety,
Security & Rescue, Bremen, Germany, July 26-30.
Silva, J., Santos, C., and Matos, V. (2010b). Timed trajec-
tory generation for a toy-like wheeled robot. In 36th
Annual Conference of the IEEE Industrial Electronics
Society, Glendale, USA, November 07-10, pages 1645
– 1650.
Tuma, M., Iossifidis, I., and Sch¨oner, G. (2009). Temporal
stabilization of discrete movement in variable environ-
ments: an attractor dynamics approach. in IEEE In-
ternational Conference on Robotics and Automation,
Kobe International Conference Center, Kobe, Japan,
pages 863 – 868.
ICINCO2012-9thInternationalConferenceonInformaticsinControl,AutomationandRobotics
434