Robot Action Diagnosis and Experience Correction by Falsifying Parameterised Execution Models

When faced with an execution failure, an intelligent robot should be able to identify the likely reasons for the failure and adapt its execution policy accordingly. This paper addresses the question of how to utilise knowledge about the execution process, expressed in terms of learned constraints, in order to direct the diagnosis and experience acquisition process. In particular, we present two methods for creating a synergy between failure diagnosis and execution model learning. We first propose a method for diagnosing execution failures of parameterised action execution models, which searches for action parameters that violate a learned precondition model. We then develop a strategy that uses the results of the diagnosis process for generating synthetic data that are more likely to lead to successful execution, thereby increasing the set of available experiences to learn from. The diagnosis and experience correction methods are evaluated for the problem of handle grasping, such that we experimentally demonstrate the effectiveness of the diagnosis algorithm and show that corrected failed experiences can contribute towards improving the execution success of a robot.

[1]  Richard E. Korf,et al.  Depth-First Iterative-Deepening: An Optimal Admissible Tree Search , 1985, Artif. Intell..

[2]  Judea Pearl,et al.  The algorithmization of counterfactuals , 2011, Annals of Mathematics and Artificial Intelligence.

[3]  Javier García,et al.  A comprehensive survey on safe reinforcement learning , 2015, J. Mach. Learn. Res..

[4]  George Konidaris,et al.  Learning Symbolic Representations for Planning with Parameterized Skills , 2018, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[5]  Gerhard Lakemeyer,et al.  Representation and Experience-Based Learning of Explainable Models for Robot Action Execution , 2020, 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[6]  Franz Wotawa,et al.  Model-Based Diagnosis or Reasoning from First Principles , 2003, IEEE Intell. Syst..

[7]  Louise Travé-Massuyès,et al.  Preferential Discrete Model-based Diagnosis for Intermittent and Permanent Faults , 2018, DX@Safeprocess.

[8]  Patrick Rodler,et al.  Evaluating Active Learning Heuristics for Sequential Diagnosis , 2018, ArXiv.

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

[10]  Fuminori Saito,et al.  Development of Human Support Robot as the research platform of a domestic mobile manipulator , 2019 .

[11]  Oliver Kroemer,et al.  Learning Robust Manipulation Strategies with Multimodal State Transition Models and Recovery Heuristics , 2019, 2019 International Conference on Robotics and Automation (ICRA).

[12]  Meir Kalech,et al.  On Fault Detection and Diagnosis in Robotic Systems , 2018, ACM Comput. Surv..

[13]  Lynne E. Parker,et al.  Adaptive Causal Models for Fault Diagnosis and Recovery in Multi-Robot Teams , 2006, 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[14]  Brian C. Williams,et al.  Diagnosing Multiple Faults , 1987, Artif. Intell..

[15]  Tom Obry,et al.  Computer-aided Diagnosis via Hierarchical Density Based Clustering , 2018, DX@Safeprocess.

[16]  Wolfram Burgard,et al.  Adaptive Curriculum Generation from Demonstrations for Sim-to-Real Visuomotor Control , 2020, 2020 IEEE International Conference on Robotics and Automation (ICRA).

[17]  Michael Beetz,et al.  Learning action failure models from interactive physics-based simulations , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[18]  Peter Struss,et al.  CASE STUDIES IN MODEL-BASED DIAGNOSIS AND FAULT ANALYSIS OF CAR-SUBSYSTEMS , 1998 .

[19]  Bradley Hayes,et al.  Robust Robot Learning from Demonstration and Skill Repair Using Conceptual Constraints , 2018, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[20]  Ciarán M Lee,et al.  Improving the accuracy of medical diagnosis with causal machine learning , 2020, Nature Communications.

[21]  Aude Billard,et al.  Robot Learning from Failed Demonstrations , 2012, Int. J. Soc. Robotics.