3,108 Hits in 5.1 sec

Policy-contingent abstraction for robust robot control [article]

Joelle Pineau, Geoffrey Gordon, Sebastian Thrun
2012 arXiv   pre-print
To the best of our knowledge, this work is a unique instance of applying POMDPs to high-level robotic control problems.  ...  This paper presents a scalable control algorithm that enables a deployed mobile robot system to make high-level decisions under full consideration of its probabilistic belief.  ...  This paper describes a new probabilistic planning algo rithm called PolCA (for Policy-Contingent Abstraction).  ... 
arXiv:1212.2495v1 fatcat:vtpbosxqdnb6rbof3loqtzezhm

Learning to Coordinate Controllers - Reinforcement Learning on a Control Basis

Manfred Huber, Roderic A. Grupen
1997 International Joint Conference on Artificial Intelligence  
The use of an abstract system model in the automatically derived supervisor reduces the complexity of the learning problem.  ...  To achieve this, the hybrid control architecture presented in this paper uses reinforcement learning on top of a Discrete Event Dynamic System (DEDS) framework to learn to supervise a set of basis controllers  ...  To achieve more autonomous behavior of a robot system, however, the system has to be able to adapt to novel situations and task contingencies without the need for outside supervision.  ... 
dblp:conf/ijcai/HuberG97 fatcat:ihplmktg2nbr5aofs3pgptatpy

FC^3: Feasibility-Based Control Chain Coordination [article]

Jason Harris, Danny Driess, Marc Toussaint
2022 arXiv   pre-print
Hierarchical coordination of controllers often uses symbolic state representations that fully abstract their underlying low-level controllers, treating them as "black boxes" to the symbolic action abstraction  ...  For a given task, FC^3 first automatically constructs a library of potential controller chains using a symbolic action tree, which is then used to coordinate controllers in a chain, evaluate task feasibility  ...  Designing such a policy goes beyond generating a single plan, but requires a form of reactive planning or contingency planning. Paxton et al.  ... 
arXiv:2205.04362v1 fatcat:avhtshf7uzcafpyalcd342dss4

Robotic self-representation improves manipulation skills and transfer learning [article]

Phuong D.H. Nguyen, Manfred Eppe, Stefan Wermter
2020 arXiv   pre-print
Cognitive science suggests that the self-representation is critical for learning and problem-solving.  ...  of robotic manipulation skills.  ...  ACKNOWLEDGMENT We thank Nicolas Frick for the help on the NICOL design and part of the NICOL simulation used in this paper.  ... 
arXiv:2011.06985v1 fatcat:5kyven7c75fevfwpyux7lv5lfq

Learning Generalizable Control Programs

Stephen Hart, Roderic Grupen
2011 IEEE Transactions on Autonomous Mental Development  
The framework encourages behavioral re-use, the acquisition of domain general strategies for interacting with the world, and the efficient generalization of control policies to different contexts.  ...  In this paper, we present a framework for guiding autonomous learning in robots.  ...  The authors would like to thank Shiraj Sen and Arjan Gijsberts for their useful feedback.  ... 
doi:10.1109/tamd.2010.2103311 fatcat:lq55qv7bkfalvbhwmkbyckv7bu

Meta-Modeling of Assembly Contingencies and Planning for Repair [article]

Priyam Parashar, Aayush Naik, Jiaming Hu, Henrik I. Christensen
2021 arXiv   pre-print
The World Robotics Challenge (2018 & 2020) was designed to challenge teams to design systems that are easy to adapt to new tasks and to ensure robust operation in a semi-structured environment.  ...  We propose a model for characterizing failures using this model and discuss repairs. Simple failures are by far the most common in our WRC system and we also present how we repaired them.  ...  This contingency requires some additional knowledge to either replan and get back to π or plan for the next most optimal policy, sayπ, from contingent state to the goal-state.  ... 
arXiv:2103.07544v1 fatcat:rxij7zh72rflvprj2sdj2szl3i

Software Mode Changes for Continuous Motion Tracking [chapter]

Deepak Karuppiah, Patrick Deegan, Elizeth Araujo, Yunlei Yang, Gary Holness, Zhigang Zhu, Barbara Lerner, Roderic Grupen, Edward Riseman
2000 Lecture Notes in Computer Science  
Moreover, in multi-robot systems, there are typically many ways in which to compensate for inadequate performance.  ...  Robot control in nonlinear and nonstationary run-time environments presents challenges to traditional software methodologies.  ...  Over time, robust plans for interacting with specific problem domains are compiled these policies into rich, comprehensive reactive policies.  ... 
doi:10.1007/3-540-44584-6_13 fatcat:fn5ylww5zzehrcqrvlpoappdlm

A game-theoretic procedure for learning hierarchically structured strategies

Benjamin Rosman, Subramanian Ramamoorthy
2010 2010 IEEE International Conference on Robotics and Automation  
We argue that this type of factored task specification and learning is a necessary ingredient for robust autonomous behaviour in a "large-world" setting.  ...  This paper addresses the problem of acquiring a hierarchically structured robotic skill in a nonstationary environment.  ...  However, oftentimes, discussions of robustness are focussed on variations within the robot system.  ... 
doi:10.1109/robot.2010.5509632 dblp:conf/icra/RosmanR10 fatcat:6el5kt5yx5bqpdoavabogobhia

Active Inference and Behavior Trees for Reactive Action Planning and Execution in Robotics [article]

Corrado Pezzato, Carlos Hernandez, Stefan Bonhof, Martijn Wisse
2021 arXiv   pre-print
We propose a hybrid combination of active inference and behavior trees (BTs) for reactive action planning and execution in dynamic environments, showing how robotic tasks can be formulated as a free-energy  ...  The proposed approach allows to handle partially observable initial states and improves the robustness of classical BTs against unexpected contingencies while at the same time reducing the number of nodes  ...  for low level adaptive control [12] , [13] .  ... 
arXiv:2011.09756v3 fatcat:5u3rbxil5jcyjnl2nkniy72v7m

Design Abstraction and Processes in Robotics: From Code-Driven to Model-Driven Engineering [chapter]

Christian Schlegel, Andreas Steck, Davide Brugali, Alois Knoll
2010 Lecture Notes in Computer Science  
We present a novel overall vision of a model-driven engineering approach for robotics that fuses strategies for robustness by design and robustness by adaptation.  ...  Advanced software engineering is the key factor in the design of future complex cognitive robots. It will decide about their robustness, (run-time) adaptivity, cost-effectiveness and usability.  ...  We think that both robustness by design and robustness by adaptation can be achieved by raising the level of abstraction at which both, the system engineer and the robot reason about the problem and the  ... 
doi:10.1007/978-3-642-17319-6_31 fatcat:gu7g4xpjirdxjn3s2b4eur66ei

Autonomous rovers for Mars exploration

R. Washington, K. Golden, J. Bresina, D.E. Smith, C. Anderson, T. Smith
1999 1999 IEEE Aerospace Conference. Proceedings (Cat. No.99TH8403)  
The Pathfinder mission demonstrated the potential for robotic Mars exploration but at the same time indicated the need for more robust rover autonomy.  ...  Future planned missions call for long traverses over unknown terrain, robust navigation and instrument placement, and reliable operations for extended periods of time.  ...  Hans Thomas and the rest of the Intelligent Mechanisms Group at NASA Ames Research Center for helping us integrate our architecture with the Marsokhod user interface and control software.  ... 
doi:10.1109/aero.1999.794236 fatcat:yx44somhyrgzhed55bahxefcki

Learning and Multiagent Reasoning for Autonomous Agents

Peter Stone
2007 International Joint Conference on Artificial Intelligence  
One goal of Artificial Intelligence is to enable the creation of robust, fully autonomous agents that can coexist with us in the real world.  ...  This paper presents current research directions in machine learning, multiagent reasoning, and robotics, and advocates their unification within concrete application domains.  ...  Various machine learning techniques have proven to be useful in finding control policies for a wide variety of IJCAI-07 robots including helicopters [Bagnell and Schneider, 2001; Ng et al., 2004] , biped  ... 
dblp:conf/ijcai/Stone07 fatcat:2br3vltrvfefffmgztnr2ha75m

Survivable Robotic Control through Guided Bayesian Policy Search with Deep Reinforcement Learning [article]

Sayyed Jaffar Ali Raza, Apan Dastider, Mingjie Lin
2021 arXiv   pre-print
Ideally, like biological creatures, a robotic agent should be able to reconfigure its control policy by adapting to dynamic adversaries.  ...  Many robot manipulation skills can be represented with deterministic characteristics and there exist efficient techniques for learning parameterized motor plans for those skills.  ...  For example, in [12] , a 4-leg robot has been demonstrated to recover from a disabled leg autonomously without pre-stored contingent policies by continuous self-modeling.  ... 
arXiv:2106.15653v1 fatcat:4bij365gj5awfpq2cgafabadca

Autonomous task sequencing in a robot swarm

Lorenzo Garattoni, Mauro Birattari
2018 Science Robotics  
on planning in robotics.  ...  Robot swarms mimic natural systems in which collective abilities emerge from the interaction of individuals.  ...  Mark I 3 and Mark I 4 In Mark I 3 , all robots execute the same control software but autonomously assume different roles depending on the contingencies that they encounter.  ... 
doi:10.1126/scirobotics.aat0430 pmid:33141726 fatcat:o62imwbj25gu3felg5a73ioc24

Combining Task and Motion Planning: Challenges and Guidelines

Masoumeh Mansouri, Federico Pecora, Peter Schüller
2021 Frontiers in Robotics and AI  
By doing so, this article aims to provide a guideline for designing combined TAMP solutions that are adequate and effective in the target scenario.  ...  All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.  ...  Many pioneers in using automated task planning for deriving robot behaviors use the term plan-based robot control to distinguish planning for robots from planning for other systems.  ... 
doi:10.3389/frobt.2021.637888 pmid:34095239 pmcid:PMC8170405 fatcat:4dmykcbhezbktfjzafv7t7wjfa
« Previous Showing results 1 — 15 out of 3,108 results