Learning directly on real world systems such as autonomous robots is a challenging task, especially if the training signal is given only in terms of success or failure (reinforcement learning). However, if successful,...
详细信息
Learning directly on real world systems such as autonomous robots is a challenging task, especially if the training signal is given only in terms of success or failure (reinforcement learning). However, if successful, the controller has the advantage of being tailored exactly to the system it eventually has to control. Here we describe, how a neural network based RL controller learns the challenging task of ball dribbling directly on our middle-size robot. The learned behaviour was actively used throughout the RoboCup world championship tournament 2007 in Atlanta, where we won the first place. This constitutes another important step within our Brainstormers project. The goal of this project is to develop an intelligent control architecture for a soccer playing robot, that is able to learn more and more complex behaviours from scratch.
Power control is an important research topic for ad-hoc Wireless Sensor Networks (WSNs). In today's sophisticated and competitive wireless environment, the control of the energy consumption in a WSN for homecare e...
详细信息
ISBN:
(纸本)9781605580678
Power control is an important research topic for ad-hoc Wireless Sensor Networks (WSNs). In today's sophisticated and competitive wireless environment, the control of the energy consumption in a WSN for homecare e-health makes it possible to guarantee basic levels of system performance, such as connectivity, throughput, delay, QoS and survivability in the presence of both mobility-immobility and a large number of sensor nodes. Recent advances in sensor fabrication technology, low-power digital and analogue electronics, and low-power wireless communication systems have made it possible to develop low-cost, robust and survivable WSNs to support activities such as assisted living and ambient intelligence (Aml). A large variety of approaches for intelligent energy-efficient schemes have been simulated over different performance metrics. In this paper, various decision support schemes are proposed evaluating the selection of different network infrastructures in terms of routing optimization and signal strength selection. Copyright 2008 ACM.
How do you develop a game that is provably educational, that is engaging from pre-school to adult, and that independently targets different language ability and world experience/maturity? Our purpose is to teach stude...
详细信息
ISBN:
(纸本)9781605580678
How do you develop a game that is provably educational, that is engaging from pre-school to adult, and that independently targets different language ability and world experience/maturity? Our purpose is to teach students of German or English as a Second Language in the same way that a baby learns a First Language. Children acquire language automatically without overt teaching, without conjugating verbs, looking up dictionaries or taking complex classes in syntax and morphology. They acquire through being immersed in an environment in which they have to learn to communicate in the language in order to achieve a variety of social and personal goals. In this project we provide a mixed environment with real toys for the learner to play with, and a simulated environment and a simulated teacher/caregiver to interact with. Whilst young children may be happy building towers and bridges out of wooden blocks, older children and adults cannot be expected to find motivating goals from the same tasks as a baby tackles as s/he simultaneously learns about the world and his/her language, culture and society. This paper explores the methodology we are developing to independently control for degree of language knowledge and degree of world experience. Copyright 2008 ACM.
This paper presents a low-cost modular robot called GZ-I which is an improved version of the Y1 modular robot. After a related survey on the previous work, all aspects of the new modular robot are introduced systemati...
详细信息
The paper applies the pseudo-linear algebra to unify the study of irreducibility and reduction for continuous-and discrete-time nonlinear control systems. The necessary and sufficient condition for irreducibility of n...
详细信息
The paper applies the pseudo-linear algebra to unify the study of irreducibility and reduction for continuous-and discrete-time nonlinear control systems. The necessary and sufficient condition for irreducibility of nonlinear input-output equation is presented in terms of the common left factor of two polynomials describing the behaviour of the system. Besides unification, the tools of pseudo-linear algebra allow to extend the results for systems defined by difference, q-shift and q-difference operators.
In 1987, Eitelberg introduced a useful method by adjusting the input weighting factors to recover system response when a direct tuning of the PI(D)'s parameters was prohibited from adjustment. Because of minimum i...
详细信息
In 1987, Eitelberg introduced a useful method by adjusting the input weighting factors to recover system response when a direct tuning of the PI(D)'s parameters was prohibited from adjustment. Because of minimum interruption to an original control system, the Eitelberg's method has been extended to various applications. From previous applications, it was found that the P-element input weighting factor distinctly effects on the system responses. In this paper, the input weighting factor optimization for P-element of PI controllers via the particle swarm optimization (PSO) method is presented. The proposed optimization method is tested against two benchmark-controlled systems. As the simulation results, it was found that the P-element input weighting factor can be appropriated optimized by the PSO-based method and the satisfactory responses can be recovered when the parameter variation occurred.
The structure and coordination of the electrical power system (EPS) are presented in this paper with respect to three concepts: system theory, systems analysis and multi-agent techniques. The above mentioned concepts ...
详细信息
The structure and coordination of the electrical power system (EPS) are presented in this paper with respect to three concepts: system theory, systems analysis and multi-agent techniques. The above mentioned concepts are discussed approaching the EPS behaviour and functioning based on actual problems regarding the globalisation of the energy production and transportation, integration of different renewable sources of energy with the conventional power systems and the safety aspects of connecting the distributed generation to the interconnected large system.
Sensors, actuators or physical components in engineering systems are often subjected to un-permitted or unexpected deviations from usual operating conditions. With a relatively small computing effort and with good res...
详细信息
Sensors, actuators or physical components in engineering systems are often subjected to un-permitted or unexpected deviations from usual operating conditions. With a relatively small computing effort and with good results to unmeasured disturbances, the parity equations build for a specific process are proven to be very useful to provide fault detection residuals in order to identify the fault.
The paper suggests how to compute the order of the minimal realization of the discrete-time nonlinear control system, described by the set of input-output difference equations, or alternatively, by the two polynomial ...
详细信息
The paper suggests how to compute the order of the minimal realization of the discrete-time nonlinear control system, described by the set of input-output difference equations, or alternatively, by the two polynomial matrices with the entries in the non-commutative skew polynomial ring. For that purpose the Ore determinant is specified further by including certain additional restrictions and proved that under these additional restrictions the degree of the dieudonne determinant is equal to the degree of the Ore determinant. Finally, the system order is shown to be equal to the degree of the determinant of one polynomial matrix. Only the special case of two output systems is considered in this paper.
暂无评论