Advertisement

Adaptive Optimal Stochastic Trajectory Planning

Numerics and Real-time Application
  • Andreas Aurnhammer
  • Kurt Marti
Chapter

Abstract

In Optimal Stochastic Trajectory Planning of industrial or service robots the problem can be modelled by a variational problem under stochastic disturbances that compared to ordinary deterministic engineering techniques also accounts for stochastic model parameters. Using stochastic optimisation theory, this variational problem is transformed into a nonlinear mathematical program, that can be solved by means of standard optimisation routines like SQP. However, these methods are not applicable in the on-line control process of robots, since they are not capable of solving mathematical programs in real-time. Hence, Neural Networks are trained based on solutions obtained from a standard optimisation algorithm.

Keywords

Path Planning Configuration Space Trajectory Planning Chance Constraint Path Parameter 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    J. A. Anderson, E. Rosenfeld: Neurocomputing: Foundations of Research, MIT Press, 1988.Google Scholar
  2. 2.
    G. A. Bekey, K. Y. Goldberg: Neural Networks in Robotics, Kluwer Academic Puplishers, 1993.MATHCrossRefGoogle Scholar
  3. 3.
    J. E. Bobrow, S. Dubowsky, J. S. Gibson: Time-optimal Control of Robotic Manipulators, Int. J. Robot. Res. 4(3), 1985, pp. 3–17.CrossRefGoogle Scholar
  4. 4.
    C. de Boor: A Practical Guide to Splines, Springer-Verlag, New York, 1978.MATHCrossRefGoogle Scholar
  5. 5.
    P. E. Gill, W. Murray, M. A. Saunders: User’s Guide for SNOPT5.3: A FORTRAN Package for Large-Scale Nonlinear Programming, University California/Stanford University, March, 1998.Google Scholar
  6. 6.
    R. Hecht-Nielsen: Neurocomputing, Addison-Wesley, Reading, MA, 1990.Google Scholar
  7. 7.
    R. Hettich, P. Zencke: Numerische Methoden der Approximation und semi-infiniten Optimierung, B.G.Teubner, Stuttgart, 1982.MATHGoogle Scholar
  8. 8.
    R. Isermann: Identifikation dynamischer Systeme, Springer Verlag, Berlin, 1992.Google Scholar
  9. 9.
    R. Johanni: Optimale Bahnplanung bei Industrierobotern, VDI Fortschrittberichte, Reihe 18, Nr. 51, Düsseldorf, 1988.Google Scholar
  10. 10.
    P. Kall, St. Wallace: Stochastic Programming, Wiley, Chichester, 1995.MATHGoogle Scholar
  11. 11.
    U.M. Loeve: Probability Theory I, Springer-Verlag, New York-Heidelberg-Berlin 1977.MATHCrossRefGoogle Scholar
  12. 12.
    K. Marti: Approximation stochastischer Optimierungsprobleme Hain, Königstein/Ts., 1979.Google Scholar
  13. 13.
    K. Marti: Path Planning for Robots under Stochastic Uncertainty, Optimization, 45, 1999, pp. 163–195.MathSciNetMATHCrossRefGoogle Scholar
  14. 14.
    K. Marti: Robust Adaptive Control of Robots by Stochastic Optimization Methods, Proceedings 2nd Int. Conf. on Engineering Computational Technology, 6–8 September 2000, Leu wen, Belgium, Civil-Comp Press 2000.Google Scholar
  15. 15.
    K. Marti: Adaptive Optimal Stochastic Trajectory Planning and Control (AOSTPC) for Robots, Preprint 00–21, DFG-priority program ‘Real-time Optimization of Large Scale Systems’, available for download1, October 2000.Google Scholar
  16. 16.
    K. Marti, A. Aurnhammer. Adaptive Stochastic Path Planning and Control (ASPPC). In: G. I. Schuèller, P. Kafka (eds.), Safety and Reliability, A. A. Balkeema, Rotterdam 1999, Volume 2, pp. 1609–1614.Google Scholar
  17. 17.
    K. Marti, S. Qu: Optimal Trajectory Planning for Robot Considering Stochastic Parameters and Disturbances. In: K. Marti, P. Kail (eds.), Stochastic Programming: Numerical Techniques and Engineering Application, LNEMS, Vol. 423, Springer-Verlag, New York-Berlin-Heidelberg 1995, pp.268–288.Google Scholar
  18. 18.
    K. Marti, S. Qu: Optimal Trajectory Planning for Robot Considering Stochastic Parameters and Disturbances — Computation of an Efficient Open-Loop Strategy, J. Intelligent and Robotic Systems 15, 1996, pp. 19–23.CrossRefGoogle Scholar
  19. 19.
    K. Marti, S. Qu: Path Planning for Robots by Stochastic Optimization Methods, J. Intelligent and Robotic Systems 22,1998, pp. 117–127.MATHCrossRefGoogle Scholar
  20. 20.
    F. Pfeiffer, R. Johanni: A Concept for Manipulator Trajectory Planning, IEEE J.Robot. Automat. RA-3(3), 1987, pp. 115–123.CrossRefGoogle Scholar
  21. 21.
    F Pfeiffer, E. Reithmeier: Roboterdynamik, B.G.Teubner, Stuttgart, 1987.MATHGoogle Scholar
  22. 22.
    M. Prüfer: Reibungsanalyse und Identifikation von Dynamikparametern bei direk-tangetriebenen und getriebebehafteten Robotern, Fortschritte in der Robotik 1, Shaker, Aachen, 1996.Google Scholar
  23. 23.
    S. Qu: Optimale Bahnplanung für Roboter unter Berücksichtigung stochastischer Parameterschwankungen, VDI Verlag, Düsseldorf, 1995.Google Scholar
  24. 24.
    S. Qu: Stochastic Trajectory planning and its application for Robot Manutec r3, In: K. Marti, P. Kail (eds.), Stochastic Programming Methods and Technical Applications, LNEMS, Vol. 458, Springer-Verlag, New York-Heidelberg-Berlin 1998, pp. 382–393.CrossRefGoogle Scholar
  25. 25.
    Z. Schiller, S. Dubowsky: Robot Path Planning with Obstacles, Actuator, Gripper and Payload Constraints, Int. J. Robot. Res. 8(6), 1989, pp. 3–18.CrossRefGoogle Scholar
  26. 26.
    K. G. Shin, N. D. McKay: Minimum-time Control of Robotic Manipulator with Geometric Path Constraints, IEEE Trans. Automat. Control AC-30(6), 1985, pp. 370–375.Google Scholar
  27. 27.
    L. L. Schumaker: Spline Functions: Basic Theory, John Wiley & Sons, New York, 1981.MATHGoogle Scholar
  28. 28.
    M. Turk: Zur Modellierung der Dynamik von Robotern mit rotatorischen Gelenken, VDI-Verlag, Düsseldorf, 1990.Google Scholar
  29. 29.
    A. Zell: Simulation neuronaler Netze, Addison-Wesley, Bonn, 1996.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2001

Authors and Affiliations

  • Andreas Aurnhammer
    • 1
  • Kurt Marti
    • 1
  1. 1.Institut für Mathematik und InformatikUniversität der Bundeswehr MünchenGermany

Personalised recommendations