the first part of this paper presents a logico-algebraic reconstruction of conceptual graph fundamentals using an appropriately extended binary relation algebra. the algebraisation comprises axioms which in a straight...
详细信息
A simulator of autonomous robots in a non-structured environment is presented. this simulator is used to develop alternative programming techniques for robot control. these techniques consist basically of using geneti...
详细信息
A simulator of autonomous robots in a non-structured environment is presented. this simulator is used to develop alternative programming techniques for robot control. these techniques consist basically of using genetic algorithms to train neural networks that are used to control the autonomous robots. the robots' autonomous control is presented and the computational aspects are discussed.
the categorization process defines sensor and action categories from elementary sensor readings and basic actions so that the necessary elements for solving a task are correctly perceived and manipulated. In reinforce...
详细信息
the categorization process defines sensor and action categories from elementary sensor readings and basic actions so that the necessary elements for solving a task are correctly perceived and manipulated. In reinforcement learning, a previous categorization process is needed to define sensor and action categories with special requirements that we analyze and that, in general, are difficult to achieve, especially in complex tasks such as those that arise when working with autonomous robots. We show how these special requirements should be relaxed and we sketch a reinforcement learning algorithm that uses a less restrictive form of sensory categorization than existing algorithms. Additionally, we show how a given sensory categorization can be improved so that it better fits the demands of the previous algorithm.
System modeling is highly relevant in the automation and simulation processes. Until now, there have been two main ways to deal withthe problem. the first is to collect the equations, normally differential, which dir...
详细信息
System modeling is highly relevant in the automation and simulation processes. Until now, there have been two main ways to deal withthe problem. the first is to collect the equations, normally differential, which direct the dynamics of the system and to solve them mainly by the S transform. the other way is to collect enough data from the process and, based on a predefined structure of the model, use a method for the parameter adjustment such as the least mean squares technique. In this paper an alternative method is presented. Based on the technique like genetic programming, a particular application of genetic algorithms where the structures under adaptation are "computer programs", a tray for the induction of models in the block diagram representation using simple discretized systems is made. the genetic program needs a way of performing parameter adjustment. For this purpose, a genetic algorithm has been applied with highly convincing results.
A systematic procedure is presented for generating Boolean logic equations that represent the desired control action to be carried out by a discrete logic controller. A key component of the controller design is the sp...
详细信息
A systematic procedure is presented for generating Boolean logic equations that represent the desired control action to be carried out by a discrete logic controller. A key component of the controller design is the specification of set and reset operations that define the next value of each state variable in the Boolean logic equations for the controller. the logic equations can be implemented on a PLC or PC in order to realize the desired control action. Two examples are given to illustrate the procedure.
Autonomous navigation applications demand sensors with a low sample time to be able to increase speed. We have developed a stereo vision algorithm, capable to deliver dense disparity maps for single, high-resolution s...
详细信息
Autonomous navigation applications demand sensors with a low sample time to be able to increase speed. We have developed a stereo vision algorithm, capable to deliver dense disparity maps for single, high-resolution scanlines at high speed (40 ms/line), even for wide disparity ranges. We have tested the algorithm with synthetic and real images. Our algorithm is based on a dynamic programming schema with a cost function based on a weighted sum of squared intensify errors. Weight factors are based on gradient values. the algorithm includes explicitly detection of occlusion. Occlusion cost changes dynamically depending on gradient values of matched points.
A productive scheme can be represented with an oriented network in which nodes are the productive elements connected through different input/output arcs, which reflect the material's flow between pairs of nodes. S...
详细信息
A productive scheme can be represented with an oriented network in which nodes are the productive elements connected through different input/output arcs, which reflect the material's flow between pairs of nodes. Such a scheme can be modelled by means of linear programming models that use a constant coefficient for input/output connections. We have proposed a software system to automatically generate such models so that a user only needs to write the scheme based on table rules. these tables support the dates and the scheme structure. the method permits the development of productive models in a short time, making it easier to introduce changes in the model and facilitating sensitivity analysis to gain confidence in the process of reaching a final model.
this paper considers the minimum cycle time (MCT) computation in deterministically timed weighted marked graphs (WMGs) under infinite server semantics. A transformation from a live and bounded WMG system into a live a...
详细信息
this paper considers the minimum cycle time (MCT) computation in deterministically timed weighted marked graphs (WMGs) under infinite server semantics. A transformation from a live and bounded WMG system into a live and bounded MG system is proposed in which the untimed behavior (represented by the Petri net language) and the minimum cycle time are preserved. By applying the well known LPP to the transformed MG system, the MCT can be computed exactly. the computational complexity of our approach is polynomial with respect to |X|, the size of the elementary T-semiflow of the original WMG.
A reactive with tags classifier system (RTCS) is a special classifier system. this system combines the execution capabilities of symbolic systems and the learning capabilities of genetic algorithms. A RTCS is able to ...
详细信息
A reactive with tags classifier system (RTCS) is a special classifier system. this system combines the execution capabilities of symbolic systems and the learning capabilities of genetic algorithms. A RTCS is able to learn symbolic rules that allow to generate sequence of actions, chaining rules among different time instants, and react to new environmental situations, considering the last environmental situation to take a decision. the capacity of RTCS to learn good rules has been prove in robotics navigation problem. Results show the suitability of this approximation to the navigation problem and the coherence of extracted rules.
An opportunistic maintenance strategy is proposed to maintain a production line consisting of k nonidentical processors and without intermediary stocks. Operational characteristics of processors are degraded with usag...
详细信息
An opportunistic maintenance strategy is proposed to maintain a production line consisting of k nonidentical processors and without intermediary stocks. Operational characteristics of processors are degraded with usage. In this strategy, preventive maintenance activities are combined with corrective ones as soon as a certain technical and economical conditions are satisfied. this might reduce the total number of maintenance activities and consequently the total maintenance cost incurred. Performance evaluation of a such a strategy is carried out using a simulation model.
暂无评论