Advertisement

Deep Coverage: Motion Synthesis in the Data-Driven Era

  • David A. Surovik
  • Kostas E. BekrisEmail author
Conference paper
Part of the Springer Proceedings in Advanced Robotics book series (SPAR, volume 10)

Abstract

Effective robotic systems must be able to produce desired motion in a sufficiently broad variety of robot states and environmental contexts. Classic control and planning methods achieve such coverage through the synthesis of model-based components. New applications and platforms, such as soft robots, present novel challenges, ranging from richer dynamical behaviors to increasingly unstructured environments. In these setups, derived models frequently fail to express important real-world subtleties. An increasingly popular approach to deal with this issue is the use of end-to-end machine learning architectures, which adapt to such complexities through a data-driven process. Unfortunately, however, data are not always available for all regions of the operational space, which complicates the extensibility of these solutions. In light of these issues, this paper proposes a reconciliation of classic motion synthesis with modern data-driven tools towards the objective of “deep coverage”. This notion utilizes the concept of composability, a feature of traditional control and planning methods, over data-derived “motion elements”, towards generalizable and scalable solutions that adapt to real-world experience.

References

  1. 1.
    Agha-mohammadi, A.a., Chakravorty, S., Amato, N.M.: FIRM: Sampling-based feedback motion-planning under uncertainty and imperfect measurements. IJRR 33(2), 268–304 (2014)Google Scholar
  2. 2.
    Akametalu, A.K., Fisac, J.F., Gillula, J.H., Kaynama, S., Zeilinger, M.N., Tomlin, C.J.: Reachability-based safe learning with Gaussian processes. In: CDC, pp. 1424–1431 (2014)Google Scholar
  3. 3.
    Alessio, A., Bemporad, A.: A Survey on Explicit Model Predictive Control (2009)Google Scholar
  4. 4.
    Bai, H., Hsu, D., Lee, W.S.: Integrated perception and planning in the continuous space: a POMDP approach. IJRR 33(9), 1288–1302 (2014)Google Scholar
  5. 5.
    Berkenkamp, F., Moriconi, R., Schoellig, A.P., Krause, A.: Safe learning of regions of attraction for uncertain systems with Gaussian processes. In: CDC, pp. 4661–4666 (2016)Google Scholar
  6. 6.
    Burridge, R.R., Rizzi, A.A., Koditschek, D.E.: Sequential composition of dynamically dexterous robot behaviors. IJRR 18(6), 534–555 (1999)Google Scholar
  7. 7.
    Clune, J., Mouret, J.B., Lipson, H.: The evolutionary origins of modularity. Proc. R. Soc. B 280(1755), 2012–2863 (2013)CrossRefGoogle Scholar
  8. 8.
    Deimel, R., Brock, O.: A novel type of compliant and underactuated robotic hand for dexterous grasping. IJRR 35(1–3), 161–185 (2016)Google Scholar
  9. 9.
    Dollar, A.M., Howe, R.D.: The highly adaptive SDM hand: design and performance evaluation. IJRR 29(5), 585–597 (2010)Google Scholar
  10. 10.
    Fu, J., Levine, S., Abbeel, P.: one-shot learning of manipulation skills with online dynamics adaptation and neural network priors. In: IROS (2016)Google Scholar
  11. 11.
    Guldner, J., Utkin, V.I.: Sliding mode control for gradient tracking and robot navigation using artificial potential fields. IEEE TRA 11(2), 247–254 (1995)Google Scholar
  12. 12.
    Hou, Z.S., Wang, Z.: From model-based control to data-driven control: survey, classification and perspective. Inf. Sci. 235, 3–35 (2013)MathSciNetCrossRefGoogle Scholar
  13. 13.
    Ijspeert, A.J., Nakanishi, J., Hoffmann, H., Pastor, P., Schaal, S.: Dynamical movement primitives: learning attractor models for motor behaviors. Neural Comput. 25(2), 328–373 (2013)MathSciNetCrossRefGoogle Scholar
  14. 14.
    Kavraki, L.E., Svestka, P., Latombe, J.C., Overmars, M.H.: Probabilistic roadmaps for path planning in high-dimensional configuration spaces. IEEE TRA 12(4), 566–580 (1996)Google Scholar
  15. 15.
    Lee, C.S., Elgammal, A.: Human motion synthesis by motion manifold learning and motion primitive segmentation. In: Articulated Motion and Deformable Objects, pp. 464–473 (2006)Google Scholar
  16. 16.
    Levine, S., Finn, C., Darrell, T., Abbeel, P.: End-to-end training of deep visuomotor policies. JMLR 17(39), 1–40 (2016)MathSciNetzbMATHGoogle Scholar
  17. 17.
    Li, Y., Littlefield, Z., Bekris, K.E.: Asymptotically optimal sampling-based kinodynamic planning. IJRR 35(5), 528–564 (2016)Google Scholar
  18. 18.
    Lin, M., Manocha, D.: Efficient contact determination in dynamic environments. Int. J. Comput. Geom. Appl. 07(01n02), 123–151 (1997)Google Scholar
  19. 19.
    Onal, C.D., Rus, D.: Autonomous undulatory serpentine locomotion utilizing body dynamics of a fluidic soft robot. Bioinspiration Biomimetics 8(2), 3–26 (2013)CrossRefGoogle Scholar
  20. 20.
    Pollack, J.B., Lipson, H., Ficici, S., Funes, P., Hornby, G., Watson, R.A.: Evolutionary techniques in physical robotics. In: Evolvable Systems: From Biology to Hardware (2000)Google Scholar
  21. 21.
    Reist, P., Preiswerk, P., Tedrake, R.: Feedback-motion-planning with simulation-based LQR-trees. IJRR 35(11), 1393–1416 (2016)Google Scholar
  22. 22.
    Rennie, C., Bekris, K.E.: Discovering a library of rhythmic gaits for spherical tensegrity locomotion. In: IEEE ICRA (2018)Google Scholar
  23. 23.
    Sabelhaus, A.P., Bruce, J., Caluwaerts, K., Manovi, P., Firoozi, R.F., Dobi, S., Agogino, A.M., SunSpiral, V.: System design and locomotion of SUPERball, an untethered tensegrity robot. In: IEEE ICRA, pp. 2867–2873 (2015)Google Scholar
  24. 24.
    Tai, L., Liu, M.: Deep-learning in mobile robotics - from perception to control systems: a survey on why and why not (2016). arXiv:1612.07139 [cs]
  25. 25.
    Tedrake, R., Manchester, I.R., Tobenkin, M., Roberts, J.W.: LQR-trees: feedback motion planning via sums-of-squares verification. IJRR 29(8), 1038–1052 (2010)Google Scholar
  26. 26.
    Walsh, G., Tilbury, D., Sastry, S., Murray, R., Laumond, J.P.: Stabilization of trajectories for systems with nonholonomic constraints. IEEE Trans. Autom. Control 39(1), 216–222 (1994)MathSciNetCrossRefGoogle Scholar
  27. 27.
    Yang, Y., Brock, O.: Elastic roadmaps–motion generation for autonomous mobile manipulation. Auton. Robot. 28(1), 113 (2010)Google Scholar
  28. 28.
    Zhou, X., Bi, S.: A survey of bio-inspired compliant legged robot designs. Bioinspiration Biomimetics 7(4), (2012)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Computer Science Department of RutgersThe State University of New JerseyNew BrunswickUSA

Personalised recommendations