6,788 Hits in 6.8 sec

Beyond imitation: Zero-shot task transfer on robots by learning concepts as cognitive programs [article]

Miguel Lázaro-Gredilla, Dianhuan Lin, J. Swaroop Guntupalli, Dileep George
2018 arXiv   pre-print
Previously learned concepts simplify the learning of subsequent more elaborate concepts, and create a hierarchy of abstractions.  ...  By bringing cognitive science ideas on mental imagery, perceptual symbols, embodied cognition, and deictic mechanisms into the realm of machine learning, our work brings us closer to the goal of building  ...  The progression of concepts learned during multiple E-C iterations demonstrate the advantage of learning to learn (38) in program induction.  ... 
arXiv:1812.02788v1 fatcat:utnlmxi2trdzdeiqnysw2xem5e


D.T. Pham, A.A. Fahmy
2005 IFAC Proceedings Volumes  
First, an inductive learning technique is applied to generate the required modelling rules from input/output measurements recorded in the off-line structure learning phase.  ...  This paper presents a new neuro-fuzzy controller for robot manipulators.  ...  ACKNOWLEDGMENTS The research described in this paper was carried out within the EC project IST-1999-13109 "Supporting Rehabilitation of Disabled Using Industrial Robots for Upper Limb Motion Therapy".  ... 
doi:10.3182/20050703-6-cz-1902.01453 fatcat:uzf2sl4kwjffjayhfykv6v5rca

On the Development of Learning Control for Robotic Manipulators

Dan Zhang, Bin Wei
2017 Robotics  
In this paper, the authors review and discuss the learning control in robotic manipulators and some issues in learning control for robotic manipulators are also illustrated.  ...  This review is able to give a general guideline for future research in learning control for robotic manipulators.  ...  In [35] , the authors designed a new type of feedback controller for robot mechanisms with random communication delays through incorporating the optimal P-type iterative learning controller concept and  ... 
doi:10.3390/robotics6040023 fatcat:k2zex76mz5fk3p6jiuy5mlde54

From explanation to synthesis: Compositional program induction for learning from demonstration [article]

Michael Burke, Svetlin Penkov, Subramanian Ramamoorthy
2019 arXiv   pre-print
This work introduces an approach to learning hybrid systems from demonstrations, with an emphasis on extracting models that are explicitly verifiable and easily interpreted by robot operators.  ...  Hybrid systems are a compact and natural mechanism with which to address problems in robotics.  ...  [23] learn high level concepts (programs) by inducing linear sequences of instructions using example programs, given a known set of atomic instructions for perception and control.  ... 
arXiv:1902.10657v1 fatcat:tqiyn22nbbacjd3gfzkb5atnru

From Explanation to Synthesis: Compositional Program Induction for Learning from Demonstration

Michael Burke, Svetlin Valentinov Penkov, Subramanian Ramamoorthy
2019 Robotics: Science and Systems XV  
This work introduces an approach to learning hybrid systems from demonstrations, with an emphasis on extracting models that are explicitly verifiable and easily interpreted by robot operators.  ...  Hybrid systems are a compact and natural mechanism with which to address problems in robotics.  ...  [23] learn high level concepts (programs) by inducing linear sequences of instructions using example programs, given a known set of atomic instructions for perception and control.  ... 
doi:10.15607/rss.2019.xv.015 dblp:conf/rss/BurkePR19 fatcat:xi5yuy5mgzgklg5m24dpntqzcu

Learning navigation Teleo-Reactive Programs using behavioural cloning

Blanca Vargas, Eduardo F. Morales
2009 2009 IEEE International Conference on Mechatronics  
to achieve simple tasks using an Inductive Logic Programming (ILP) system, and (iii) it learns hierarchical TRPs that express how to achieve goals by following particular sequences of actions using a  ...  In this paper, it is shown how a robot can learn TRPs from human guided traces. A user guides a robot to perform a task and the robot learns how to perform such task in similar dynamic environments.  ...  RELATED WORK In this work we used behavioural cloning to provide examples, and ILP and grammar induction to learn TRPs for mobile robots. We review relevant related work on these areas.  ... 
doi:10.1109/icmech.2009.4957173 fatcat:ca5j6q6e2rcvrlmcaigelvzz3q

An approach to learning mobile robot navigation

Sebastian Thrun
1995 Robotics and Autonomous Systems  
Here EBNN is applied in the context of reinforcement learning, which allows the robot to learn control using dynamic programming.  ...  A mobile robot, equipped with visual, ultrasonic and laser sensors, learns to servo to a designated target object.  ...  This work was supported in part by grant IRI-9313367 from the US National Science Foundation to Tom Mitchell.  ... 
doi:10.1016/0921-8890(95)00022-8 fatcat:mmzhp4xz7nfpbcajdxe4fhtpna

Biologically inspired control and modeling of (bio)robotic systems and some applications of fractional calculus in mechanics

Mihailo Lazarevic
2013 Theoretical and Applied Mechanics  
local positive/negative feedback on control with great amplifying), which allows efficiently realization of control based on iterative natural learning.  ...  Second, the model of (bio)mechanical system may be obtained using another biological concept called distributed positioning (DP), which is based on the inertial properties and actuation of joints of considered  ...  As one of alternatives, the iterative learning control (ILC) method has been developed [25] , where the concept of ILC was originally proposed by Arimoto [26] for accurate tracking of robot trajectories  ... 
doi:10.2298/tam1301163l fatcat:24lt75wsrrghvpcoqaoiibp7w4

Process Discovery for Structured Program Synthesis [article]

Dell Zhang, Alexander Kuhnle, Julian Richardson, Murat Sensoy
2020 arXiv   pre-print
The proposed algorithm works by iteratively applying a few graph rewriting rules to the directly-follows-graph of activities.  ...  In comparison with the popular top-down recursive inductive miner, our proposed agglomerative miner enjoys the similar theoretical guarantee to produce sound process models (without deadlocks and other  ...  Most existing process discovery algorithms seem to overcome this obstacle to learn from positive examples (observed traces) only by imposing a strong inductive bias against duplicate activities in the  ... 
arXiv:2008.05804v1 fatcat:i2neriv3lrg3lowwgm2vvsry6i

Educational Technologies for Precollege Engineering Education

M. Riojas, S. Lysecky, J. Rozenblit
2012 IEEE Transactions on Learning Technologies  
In addition, we outline how these technologies align with deductive and inductive teaching methods that emphasize direct-instruction, inquiry-, problem-, and project-based methods, as studies have shown  ...  Studies have shown these students are at a critical age where exposure to engineering and other related fields such as science, mathematics, and technology greatly impact their career goals.  ...  The learning experiences that can be afforded with these products include deductive and inductive methods.  ... 
doi:10.1109/tlt.2011.16 fatcat:rkdewhkhorgjxdgiwfyggqetcm

Iterative Learning Control: Brief Survey and Categorization

Hyo-Sung Ahn, YangQuan Chen, Kevin L. Moore
2007 IEEE Transactions on Systems Man and Cybernetics Part C (Applications and Reviews)  
In this paper the iterative learning control (ILC) literature published between 1998 and 2004 is categorized and discussed, extending earlier reviews presented by two of the authors.  ...  Historically, the first novel idea related to a multi-pass control strategy can be traced back to [116] , published in 1974, though the stability analysis was restricted to classical control concepts and  ...  For example, in [391] , an adaptive learning (A-L) control scheme was developed for robot manipulator tracking; in [272] , Miyasato proposed a hybrid adaptive control scheme (enhanced by ILC), and in  ... 
doi:10.1109/tsmcc.2007.905759 fatcat:6vmsbnf3a5axno3mbv2nljgjhu

The Classification of the Applicable Machine Learning Methods in Robot Manipulators

Hadi Hormozi, Elham Hormozi, Hamed Rahimi Nohooji
2012 International Journal of Machine Learning and Computing  
Index Terms-Machine learning, adaptive control, repetitive control, robot manipulators. Hadi Hormozi is currently serving as a senior Lecturer at  ...  This paper describes various supervised machine learning classification techniques used in robotic manipulators.  ...  Learning to learn learns its own inductive bias based on previous experience for robots [1] . Supervised algorithms are more basic than other classifications in manipulating of robots.  ... 
doi:10.7763/ijmlc.2012.v2.189 fatcat:3fdzyqsvnvdzlfs656op43qy4a

Neuro-fuzzy controller for control and robotics applications

D.H. Rao, M.M. Gupta
1994 Engineering applications of artificial intelligence  
The proposed control scheme is implemented for controlling a class of unknown nonlinear dynamic systems, and for computing the in verse kinematic transformations of a two-linked robot.  ...  The purpose of this paper is to develop a neuro-fuzzy controller (NFC) for adaptive tracking in unknown nonlinear dynamic systems, and for on-line computation of inverse kinematic transformations of robot  ...  Example 4. 3 : 3 In this example a nonlinear plant represented by equation (28) was considered, where the relation between y(k+ 1) and the past values of the control input u(k-j) was assumed to be linear  ... 
doi:10.1016/0952-1976(94)90027-2 fatcat:ndryj6a44ndebfpwdmwe327zfy

Guest Editorial

Lus Gomes, Seta Bogosyan
2009 IEEE transactions on industrial electronics (1982. Print)  
Pereira, allows the integration of mixed-reality experiments with virtual learning environments, introducing the concept of interchangeable components, with emphasis on education in control and automation  ...  Her research interests are nonlinear control and estimation techniques for electromechanical systems with applications in direct-drive systems, sensorless control of induction motors, hybrid and autonomous  ... 
doi:10.1109/tie.2009.2033630 fatcat:a2hl5irs6beh7elfiwov5acvg4

Few-Shot Induction of Generalized Logical Concepts via Human Guidance

Mayukh Das, Nandini Ramanan, Janardhan Rao Doppa, Sriraam Natarajan
2020 Frontiers in Robotics and AI  
We consider the problem of learning generalized first-order representations of concepts from a small number of examples. We augment an inductive logic programming learner with 2 novel contributions.  ...  First, we define a distance measure between candidate concept representations that improves the efficiency of search for target concept and generalization.  ...  BACKGROUND AND RELATED WORK Our approach to Concept Learning is closely related to Stroulia and Goel (1994) 's work, which learns logical problem-solving concepts by reflection.  ... 
doi:10.3389/frobt.2020.00122 pmid:33501288 pmcid:PMC7805948 fatcat:khw63epsjbbdlhmq35necqfjaq
« Previous Showing results 1 — 15 out of 6,788 results