Abstract
This paper explores the idea that robots can learn safe behaviors without prior knowledge about its environment nor the task at hand, using intrinsic motivation to reverse actions. Our general idea is that if the robot learns to reverse its actions, all the behaviors that emerge from this principle are intrinsically safe. We validate this idea with experiments to benchmark the performance of obstacle avoidance behavior. We compare our algorithm based on an abstract intrinsic reward with a Q-learning algorithm for obstacle avoidance based on external reward signal. Finally, we demonstrate that safety of learning can be increased further by first training the robot in the simulator using the intrinsic reward and then running the test with the real robot in the real environment.
The experimental results show that the performance of the proposed algorithm is on average only 5-10% lower than of the Q-Learning algorithm. A physical robot, using the knowledge obtained in simulation, in real world performs 10% worse than in simulation. However, its performance reaches the same success rate with the physically trained robot after a short learning period. We interpret this as the evidence confirming the hypothesis that our learning algorithm can be used to teach safe behaviors to a robot.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Ryan, R.M., Deci, E.L.: Intrinsic and extrinsic motivations: classic definitions and new directions. Contemporary Educational Psychology 25(1), 54–67 (2000)
Prescott, T.J., Montes Gonzalez, F.M., Gurney, K., Humphries, M.D., Redgrav, P.: A robot model of the basal ganglia: Behavior and intrinsic processing. Neural Networks 19(1), 31–61 (2006)
Schmidhuber, J.: Exploring the predictable. In: Ghosh, A., Tsutsui, S. (eds.) Advances in Evolutionary Computing, pp. 579–612. Springer, Heidelberg (2003)
Schmidhuber, J.: Self-Motivated Development Through Rewards for Predictor Errors / Improvements. In: 2005 AAAI Spring Symposium on Developmental Robotics, pp. 1994–1996 (2005)
Barto, A.G., Singh, S., Chentanez, N.: Intrinsically Motivated Learning of Hierarchical Collections of Skills. In: ICDL 2004, pp. 112–119 (2004)
Stout, A., Konidaris, G.D., Barto, A.G.: Intrinsically Motivated Reinforcement Learning-A Promising Framework For Developmental Robot Learning. In: The AAAI Spring Symposium on Developmental Robotics (2005)
Kaplan, F., Oudeyer, P.Y.: Motivational principles for visual know-how development. In: 3rd International Workshop on Epigenetic Robotics, pp. 73–80 (2003)
Oudeyer, P.Y., Kaplan, F.: Intrinsic Motivation Systems for Autonomous Mental Development. IEEE Trans. Evol. Comput. 11, 265–286 (2007)
Oudeyer, P.Y., Kaplan, F.: What is intrinsic motivation? A topology of computational approaches. In: Front. Neurorobotics, vol. 1 (2007)
Breazeal, C.: Designing Sociable Robots. Bradford Books/MIT Press, Cambridge (2002)
Kruusmaa, M., Gavshin, Y., Eppendahl, A.: Don’t Do Things You Can’t Undo: Reversibility Models for Generating Safe Behaviours. In: ICRA 2007, pp. 1134–1139 (2007)
Gavshin, Y., Kruusmaa, M.: Comparative experiments on the emergence of safe behaviours. In: TAROS 2008, pp. 65–70 (2008)
Gerkey, B., Vaughan, R., Howard, A.: The player/stage project: Tools for multi-robot and distributed sensor systems. In: ICAR 2003, pp. 317–323 (2003)
Lin, M., Zhu, J., Sun, Z.: Learning Obstacle Avoidance Behavior Using Multi-agent Learning with Fuzzy States. In: Bussler, C.J., Fensel, D. (eds.) AIMSA 2004. LNCS (LNAI), vol. 3192, pp. 389–398. Springer, Heidelberg (2004)
Gutnisky, D.A., Zanutto, B.S.: Learning Obstacle Avoidance with an Operant Behavior Model. Artificial Life 10(1), 65–81 (2004)
Macek, K., Petrovic, I., Peric, N.: A Reinforcement Learning Approach to Obstacle Avoidance of Mobile Robot. In: IEEE AMC 2002, pp. 462–466 (2002)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Gavshin, Y., Kruusmaa, M. (2011). Emergence of Safe Behaviours with an Intrinsic Reward. In: Bouchachia, A. (eds) Adaptive and Intelligent Systems. ICAIS 2011. Lecture Notes in Computer Science(), vol 6943. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-23857-4_20
Download citation
DOI: https://doi.org/10.1007/978-3-642-23857-4_20
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-23856-7
Online ISBN: 978-3-642-23857-4
eBook Packages: Computer ScienceComputer Science (R0)