![](bg8.png)
(a) Interception started (b) Interception stopped
(c) Hitting (d) Return
Figure 6: The different movements composing a ball hitting
task in the real robot.
5 CONCLUSIONS
Based on attractor dynamics, an autonomous mech-
anism of timed movement generation was imple-
mented for a ball hitting task. The approach was eval-
uated in simulation and successfully brought onto a
real manipulator, enabling the robot to repeatedly in-
tercept and hit a ball rolling on an inclined plane. Sta-
bility and targetted instabilities are inherent properties
of the nonlinear dynamics for trajectory generation
and behavioral organization through which the gen-
erated behavior can be updated online and adjusted
depending on the current situation. This behavioral
flexibility enabled the robot to perform the hitting task
autonomously. The system is capable of responding
quickly and flexibly to changes in the sensed environ-
ment or movement conditions while keeping timing
stable. Limitations in the system’s performance were
primarily due to the small framerate imposed by the
employed camera. The robot’smodest maximal speed
affected the hitting performance on the real robot.
We are currently working on overcoming these
limitations by improving the vision system’s predic-
tion performance through a faster camera and so, an
improved Kalman estimation system. We will also
transfer the implementation to a new robotic platform,
the KUKA lightweight arm, which permits much
faster movements and is, therefore, a more suitable
platform for this application.
REFERENCES
A. J. Ijspeert, J. Nakanishi, S. S. (2002). Movement imi-
tation with nonlinear dynamical systems in humanoid
robots. In IEEE Int Conf on Robotics and Automation.
ICRA.
A. Yilmaz, O. Javed, M. S. (2006). Object tracking: A sur-
vey. In ACM Journal of Computing Surveys. Vol. 38,
No. 4.
B. R. Fajen, W. H. Warren, S. T. L. P. K. (2003). A dynami-
cal model of visually-guided steering, obstacle avoid-
ance and route selection. In International Journal of
Computer Vision. 54(1-2) vol 54 13-34.
E. Bicho, G. S. (1996). The dynamic approach to au-
tonomous robotics demonstrated on a low-level vehi-
cle platform. In International Symposium on Intelli-
gent Robotic Systems. SIRS 96.
Ijspeert, A. J. (2008). Central pattern generators for loco-
motion control in animals and robots. In Neural Net-
works Review. vol. 21, no. 4, pp. 642-653.
Iossifidis, I. and Schoener, G. (2004). Autonomous reach-
ing and obstacle avoidance with the anthropomorphic
arm of a robotic assistant using the attractor dynamics
approach. In IEEE Int Conf on Robotics and Automa-
tion. ICRA.
Jeannerod, M. (1984). The timing of natural prehension
movements. In Journal of Motor Behavior. vol 16,
235-254.
Khatib, O. (1986). Real time obstacle avoidance for manip-
ulators and mobile robots. In International Journal
Robotics Research. vol. 5, pp. 90-98.
M. Tuma, I. Iossifidis, G. S. (2009). Temporal stabiliza-
tion of discrete movement in variable environments
an attractor dynamics approach. In IEEE Int Conf on
Robotics and Automation. ICRA.
Raibert, M. H. (1986). Legged robots that balance. MIT
Press, Massachusetts, 1st edition.
S. Degallier, C. Santos, L. R. A. I. (2006). Movement gen-
eration using dynamical systems: a humanoid robot
performing a drumming task. In International Con-
ference on Humanoid Robots. IEEE-RAS.
S. Schaal, A. Ijspeert, A. B. (2003). Computational ap-
proaches to motor learning by imitation. In Philo-
sophical Transactions of the Royal Society. vol 358
537-547.
Schoener, G. (1990). A dynamic theory of coordination of
discrete movement. In Biological Cybernetics 63. vol
63, 257-270.
Schoener, G. (2002). Timing, clocks and dynamical sys-
tems. In Brain and Cognition. vol 48, pp. 31-51.
Schoener, G. and Dose, M. (1992). A dynamical systems
approach to task-level system integration used to plan
and control autonomous vehicle motion. In Robotics
and Autonomous Systems. vol 10, 253-267.
Schutter, J. D. (1999). Kalman filters: A tutorial. In Journal
of Computing Surveys. vol. 40 (4).
Steinhage, A. and Schoener, G. (1998). Dynamical systems
for the behavioral organization of autonomous robot
navigation. In Sensor Fusion and Decentralized Con-
trol in Robotic Systems. Proc SPIE.
Warren, W. H. (2006). The dynamics of perception and
action. In Psychological Review 113(2). vol 113, 358-
389.
AutonomousTimedMovementbasedonAttractorDynamicsinaBallHittingTask
311