Filters








7 Hits in 7.4 sec

Design of semi-decentralized control laws for distributed-air-jet micromanipulators by reinforcement learning

Laetitia Matignon, Guillaume J. Laurent, Nadine Le Fort-Piat
2009 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems  
The experimental results validate the semi-decentralized reinforcement learning method as a way to design control laws for such distributed systems.  ...  Recently, a great deal of interest has been developed in learning in multi-agent systems to achieve decentralized control.  ...  We propose to use reinforcement learning (RL) control techniques in a semi-decentralized perspective as a way to design control laws for distributed manipulation systems.  ... 
doi:10.1109/iros.2009.5353902 dblp:conf/iros/MatignonLF09 fatcat:goe4nk52ojhm7nyujmtm67l44m

Designing Decentralized Controllers for Distributed-Air-Jet MEMS-Based Micromanipulators by Reinforcement Learning

Laëtitia Matignon, Guillaume J. Laurent, Nadine Le Fort-Piat, Yves-André Chapuis
2010 Journal of Intelligent and Robotic Systems  
The simulation results demonstrate that the reinforcement learning method is a promising way to design control laws for such distributed systems.  ...  The control of such distributed systems is very challenging and usual approaches for contact arrayed system don't produce satisfactory results.  ...  This work was supported by the Smart Surface NRA (French National Research Agency) project (ANR 06 ROBO 0009 03).  ... 
doi:10.1007/s10846-010-9396-9 fatcat:ucqcciu3yfhjvnonar7mpw2bcy

2020 Index IEEE Robotics and Automation Letters Vol. 5

2020 IEEE Robotics and Automation Letters  
Belzile, B., +, LRA Jan. 2020 40-47 Jet engines Modeling, Identification and Control of Model Jet Engines for Jet Powered Robotics.  ...  ., +, LRA July 2020 3846-3853 Modeling, Identification and Control of Model Jet Engines for Jet Powered Robotics.  ... 
doi:10.1109/lra.2020.3032821 fatcat:qrnouccm7jb47ipq6w3erf3cja

2019-2020 Index IEEE Transactions on Industrial Electronics Vol. 67

2020 IEEE transactions on industrial electronics (1982. Print)  
., Novel Topologies of Power Electronics Converter as Active Magnetic Bearing Drive; TIE Feb. 2020 950-959 Jiang, D., see Chen, J., TIE Aug. 2020 6215-6226 Jiang, F., Pourpanah, F., and Hao, Q., Design  ...  , Implementation, and Evaluation of a Neural-Network-Based Quadcopter UAV System; TIE March 2020 2076-2085 Jiang, J., Mohagheghi, A., and Moallem, M., Energy-Efficient Supplemental LED Lighting Control  ...  ., +, TIE June 2020 4957-4968 Optimized Formation Control Using Simplified Reinforcement Learning for a Class of Multiagent Systems With Unknown Dynamics.  ... 
doi:10.1109/tie.2020.3045338 fatcat:gljm7ngg3fakvmnfswcbb5vwiu

Conceptual design and multidisciplinary optimization of in-plane morphing wing structures

Daisaku Inoyama, Brian P. Sanders, James J. Joo, Douglas K. Lindner
2006 Smart Structures and Materials 2006: Modeling, Signal Processing, and Control  
Snap-through buckling is a step further in the direction of using non-linear mechanics for actuator design.  ...  Using the experimental system, 2Hz mono-frequency is used for excitation of ER damper and the damping force is measured by load-cell.  ...  Indeed, each transducer does not need to be addressed by a centralized computer but can simply be used by some decentralized distributed controllers.  ... 
doi:10.1117/12.658686 fatcat:rqnvt2atkjgnvpog5pivwjejzm

German Scientists and the Atom Bomb

William G. Schlecht
1959 Science  
The paths are controlled by factors of topography, the upper-air jet stream, and others, less well understood.  ...  By making this assumption of a “uniform distributionof probability between 0 and 1, Laplace proved his so-called “law of succession.”  ... 
doi:10.1126/science.129.3347.438-c fatcat:lujmi3nn4rdxvfndmaei72o7oe

Motion planning by genetic algorithm for a redundant manipulator using an evaluation function based on criteria of skilled operators

T. Shibata, T. Abe, K. Tanie, M. Nose
Proceedings of 1995 IEEE International Conference on Robotics and Automation  
In order to combine Petri nets and learning methods, we modeled Q-Learning -a variant of Reinforcement Learning -with high-level Petri nets.  ...  The controller is non-adaptive and of sliding mode type, and is designed based on the decentralized form of the dynamics of UVMS.  ... 
doi:10.1109/robot.1995.525631 dblp:conf/icra/ShibataATN95 fatcat:vzk3brbh3rh7xaln5vowvqxnue