Empirical Analysis of Probabilistic Methods for Failure Isolation in Robots

Finding the root causes of failures is crucial for robots to prevent them reoccur. For this purpose, execution must be continually monitored and failure isolation procedures must be applied to discover underlying reasons and recover from them effectively. A failure may be rooted from multiple reasons and these may not be directly related to the action in execution. In such cases, temporal probabilistic methods are more suitable for efficient failure isolation. In this paper, we analyse two probabilistic methods for failure isolation problem: Hierarchical Hidden Markov Models (HHMMs) and Particle Filters (PFs). Experimental evaluation is performed on realworld data taken by an autonomous robot during an object manipulation scenario. The preliminary results indicate that HHMMs and PFs provide promising solutions to the failure isolation problem.

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

[2]  Sanem Sariel,et al.  Cognitive robots learning failure contexts through real-world experimentation , 2015, Autonomous Robots.

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

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

[5]  R. Morales-Menendez,et al.  Towards a new fault diagnosis system for electric machines based on dynamic probabilistic models , 2005, Proceedings of the 2005, American Control Conference, 2005..

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

[7]  Alessandro Saffiotti,et al.  Handling uncertainty in semantic-knowledge based execution monitoring , 2007, 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[8]  Muhammad Tahir Khan,et al.  Fault detection in mobile robots using sensor fusion , 2015, 2015 10th International Conference on Computer Science & Education (ICCSE).

[10]  Christian Fritz,et al.  Execution Monitoring – A Survey – Fall 2005 , 2011 .

[11]  Roberto Micalizio,et al.  ACTION FAILURE RECOVERY VIA MODEL‐BASED DIAGNOSIS AND CONFORMANT PLANNING , 2013, Comput. Intell..

[12]  Franz Wotawa,et al.  Robust Plan Execution Using Model-Based Reasoning , 2009, Adv. Robotics.

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

[14]  Gerald Steinbauer,et al.  A dependable perception-decision-execution cycle for autonomous robots , 2012, 2012 IEEE International Conference on Robotics and Automation.

[15]  N. Ramdani,et al.  Robust Fault Detection and Isolation applied to Indoor Localization , 2015 .

[16]  Alessandro Saffiotti,et al.  Monitoring the execution of robot plans using semantic knowledge , 2008, Robotics Auton. Syst..

[17]  Bernhard Nebel,et al.  Coming up With Good Excuses: What to do When no Plan Can be Found , 2010, Cognitive Robotics.

[18]  Sanem Sariel,et al.  Probabilistic Failure Isolation for Cognitive Robots , 2014, FLAIRS.

[19]  Yoram Singer,et al.  The Hierarchical Hidden Markov Model: Analysis and Applications , 1998, Machine Learning.

[20]  Rowan McAllister,et al.  Autonomous Reconfiguration of a Multi-Modal Mobile Robot , 2011 .

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

[22]  Reid Simmons,et al.  Probabilistic models for monitoring and fault diagnosis , 2002 .

[23]  Robin R. Murphy,et al.  Follow-up analysis of mobile robot failures , 2004, IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004.

[24]  Van Nostrand,et al.  Error Bounds for Convolutional Codes and an Asymptotically Optimum Decoding Algorithm , 1967 .

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

[26]  Manuela M. Veloso,et al.  Plan execution monitoring through detection of unmet expectations about action outcomes , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[27]  Bo Zhao,et al.  Particle Filter for Fault Diagnosis and Robust Navigation of Underwater Robot , 2014, IEEE Transactions on Control Systems Technology.

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

[29]  Sebastian Thrun,et al.  Real-time fault diagnosis [robot fault diagnosis] , 2004, IEEE Robotics & Automation Magazine.

[30]  Alexandre M. Amory,et al.  Distributed fault diagnosis for multiple mobile robots using an agent programming language , 2015, 2015 International Conference on Advanced Robotics (ICAR).

[31]  Ola Pettersson,et al.  Execution monitoring in robotics: A survey , 2005, Robotics Auton. Syst..

[32]  L. Baum,et al.  Statistical Inference for Probabilistic Functions of Finite State Markov Chains , 1966 .