Cognitive robots learning failure contexts through real-world experimentation

Learning is essential for cognitive robots as humans to gain experience and to adapt to the real world. We propose an experiential learning method for robots to build their experience online and to transfer knowledge among appropriate contexts. Experience gained through learning is used as a guide to future decisions of the robot for both efficiency and robustness. We use Inductive Logic Programming (ILP) learning paradigm to frame hypotheses represented in first-order logic that are useful for further reasoning and planning processes. Furthermore, incorporation of background knowledge is also possible to generalize the framed hypotheses. Partially specified world states can also be easily represented by these hypotheses. All these advantages of ILP make this approach superior to the other supervised learning methods. We have analyzed the performance of the learning method on our autonomous mobile robot and on our robot arm both building their experience on action executions online. It has been observed in both domains that our experience-based learning and learning-based guidance methods frame sound hypotheses that are useful for constraining and guiding the future tasks of the robots. This learning paradigm is promising especially for the contexts where abstraction is useful for efficient transfer of knowledge.

[1]  William W. Cohen.  Learning Approximate Control Rules of High Utility , 1990, ML.

[2]  Patrick Doherty,et al.  A Temporal Logic-Based Planning and Execution Monitoring System , 2008, ICAPS.

[3]  D. Kolb.  Experiential Learning: Experience as the Source of Learning and Development , 1983 .

[4]  Richard Dearden,et al.  Manipulation planning using learned symbolic state abstractions , 2014, Robotics Auton. Syst..

[5]  Saurav Agarwal,et al.  Periodic-Node Graph-Based Framework for Stochastic Control of Small Aerial Vehicles , 2015 .

[6]  Moritz Tenorth,et al.  RoboEarth - A World Wide Web for Robots , 2011, ICRA 2011.

[7]  Rocı́o G. Durán.  Integrating Macro-Operators and Control-Rules Learning , 2006 .

[8]  Moritz Tenorth,et al.  KnowRob: A knowledge processing infrastructure for cognition-enabled robots , 2013, Int. J. Robotics Res..

[9]  Francesco Mondada,et al.  Integration of Online Learning into HTN Planning for Robotic Tasks , 2012, AAAI Spring Symposium: Designing Intelligent Robots.

[10]  Sanem Sariel,et al.  Failure Handling In a Planning Framework , 2012, AAAI.

[11]  Tara A. Estlin,et al.  Learning to Improve both Efficiency and Quality of Planning , 1997, IJCAI.

[12]  S. Kambhampati,et al.  Learning Explanation-Based Search Control Rules for Partial Order Planning , 1994, AAAI.

[13]  Pedro Isasi Viñuela,et al.  Using genetic programming to learn and improve control knowledge , 2002, Artif. Intell..

[14]  Jonathan P. How,et al.  Threat-aware path planning in uncertain urban environments , 2010, 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[15]  Jussi Rintanen,et al.  Incorporation of Temporal Logic Control into Plan Operators , 2000, ECAI.

[16]  Peter Øhrstrøm,et al.  Temporal Logic , 1994, Lecture Notes in Computer Science.

[17]  Sanem Sariel,et al.  Cognitive Robots Learning Failure Contexts Through Experimentation , 2015, AAMAS.

[18]  L. P. Kaelbling,et al.  Learning Symbolic Models of Stochastic Domains , 2007, J. Artif. Intell. Res..

[19]  Tucker Hermans,et al.  Representing and learning affordance-based behaviors , 2014 .

[20]  Enric Plaza,et al.  Case-Based Learning of Strategic Knowledge , 1991, EWSL.

[21]  Michel Barbeau,et al.  Planning Control Rules for Reactive Agents , 1997, Artif. Intell..

[22]  Jonathan Schaeffer,et al.  Macro-FF: Improving AI Planning with Automatically Learned Macro-Operators , 2005, J. Artif. Intell. Res..

[23]  Seyedshams Feyzabadi,et al.  Risk-aware path planning using hirerachical constrained Markov Decision Processes , 2014, 2014 IEEE International Conference on Automation Science and Engineering (CASE).

[24]  Sanem Sariel,et al.  Dynamic Temporal Planning for Multirobot Systems , 2011, Automated Action Planning for Autonomous Mobile Robots.

[25]  Manuela M. Veloso,et al.  Lazy Incremental Learning of Control Knowledge for Efficiently Obtaining Quality Plans , 1997, Artificial Intelligence Review.

[26]  Alfred Horn,et al.  On sentences which are true of direct unions of algebras , 1951, Journal of Symbolic Logic.

[27]  Sanem Sariel,et al.  A Robust Planning Framework for Cognitive Robots , 2012, CogRob@AAAI.

[28]  Arvin Agah,et al.  A robot decision making framework using constraint programming , 2011, Artificial Intelligence Review.

[29]  Petek Yildiz,et al.  Learning Guided Planning for Robust Task Execution in Cognitive Robotics , 2013, AAAI 2013.

[30]  Manuela Veloso,et al.  Incremental Learning of Control Knowledge for Improvement of Planning Efficiency and Plan Quality , 1994 .

[31]  Petek Yildiz,et al.  Bilişsel Robotlarda Deneyimsel Öğrenme Experimental Learning in Cognitive Robots , 2013 .

[32]  Roni Khardon,et al.  Learning Action Strategies for Planning Domains , 1999, Artif. Intell..

[33]  Fahiem Bacchus,et al.  Using temporal logics to express search control knowledge for planning , 2000, Artif. Intell..

[34]  Xuemei Wang,et al.  Learning Planning Operators by Observation and Practice , 1994, AIPS.

[35]  Brian M. Sadler,et al.  Trading Safety Versus Performance: Rapid Deployment of Robotic Swarms with Robust Performance Constraints , 2015, ArXiv.

[36]  Hulya Yalcin,et al.  Scene Interpretation for Self-Aware Cognitive Robots , 2014, AAAI Spring Symposia.

[37]  Alexander Ferrein,et al.  Belief Management for High-Level Robot Programs , 2011, IJCAI.

[38]  Ingrid Zukerman,et al.  Inductive Learning of Search Control Rules for Planning , 1998, Artif. Intell..

[39]  Stephen Muggleton,et al.  Inverse entailment and progol , 1995, New Generation Computing.

[40]  Fahiem Bacchus,et al.  Planning with Resources and Concurrency: A Forward Chaining Approach , 2001, IJCAI.

[41]  Sanem Sariel,et al.  Action monitoring in cognitive robots , 2014, 2014 22nd Signal Processing and Communications Applications Conference (SIU).

[42]  Henrik I. Christensen,et al.  Efficient Organized Point Cloud Segmentation with Connected Components , 2013 .

[43]  Marc Toussaint,et al.  Planning with Noisy Probabilistic Relational Rules , 2010, J. Artif. Intell. Res..

[44]  Koichi Furukawa,et al.  Special issue on inductive logic programming , 2009, New Generation Computing.

[45]  Hulya Yalcin,et al.  Scene Interpretation for Lifelong Robot Learning , 2014 .

[46]  Malik Ghallab,et al.  Learning how to combine sensory-motor functions into a robust behavior , 2008, Artif. Intell..

[47]  Sanem Sariel,et al.  Robots That Create Alternative Plans against Failures , 2012, SyRoCo.

[48]  Sanem Sariel,et al.  Robust task execution through experience-based guidance for cognitive robots , 2015, 2015 International Conference on Advanced Robotics (ICAR).

[49]  Morgan Quigley,et al.  ROS: an open-source Robot Operating System , 2009, ICRA 2009.

[50]  Hulya Yalcin,et al.  Extracting Spatial Relations Among Objects for Failure Detection , 2013, KIK@KI.

[51]  Karen Zita Haigh,et al.  Learning situation-dependent costs: improving planning from probabilistic robot execution , 1998, AGENTS '98.

[52]  Vincent Lepetit,et al.  Gradient Response Maps for Real-Time Detection of Textureless Objects , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[53]  Ricardo Aler,et al.  Using Previous Experience for Learning Planning Control Knowledge , 2004, FLAIRS.

[54]  Hema Swetha Koppula,et al.  RoboBrain: Large-Scale Knowledge Engine for Robots , 2014, ArXiv.