Part I of this paper presents a study of an autonomous trajectory generation system (ATGS) which is useful for industrial robots carrying out mechanical operations. It is a very simple job with such an ATGS to generat...
详细信息
Part I of this paper presents a study of an autonomous trajectory generation system (ATGS) which is useful for industrial robots carrying out mechanical operations. It is a very simple job with such an ATGS to generate planar curved trajectories f(x, y) = 0, an attractive factor in such applications. Two alternative variations of the ATGS are presented which possess particular characteristics. Part II deals with the trajectory control problem and uses the ATGS as a first stage of a model reference adaptive control loop of a robot. The theoretical results are illustrated by simulation examples worked out using the software packages developed for this purpose.
A very important subject for the consolidation of neural networks is the study of their capabilities. In this paper, the relationships between network size, training set size and generalization capabilities are examin...
详细信息
A very important subject for the consolidation of neural networks is the study of their capabilities. In this paper, the relationships between network size, training set size and generalization capabilities are examined. The phenomenon of overtraining in backpropagation networks is discussed and an extension to an existing algorithm is described. The extended algorithm provides a new energy function and its advantages, such as improved plasticity and performance along with its dynamic properties, are explained. The algorithm is applied to some common problems (XOR, numeric character recognition and function approximation) and simulation results are presented and discussed.
This paper presents an Expert Data Rejection (EDR) system, developed to accompany a real-time data acquisition tool for industrial plants. The EDR system is an integrated software tool responsible for the first stage ...
详细信息
This paper presents an Expert Data Rejection (EDR) system, developed to accompany a real-time data acquisition tool for industrial plants. The EDR system is an integrated software tool responsible for the first stage process of multiple raw measurements coming from the plant. It provides different algorithms, which operate in real time, recognizing faulty patterns according to the user requirements. For each specific measurement the user may define a different way of treatment. The system checks the signal values and filters the incoming data. Various adaptive techniques have been adopted, such as digital filtering and variations of the recursive least-squares algorithm, as well as the neural network based algorithms. The system was applied and tested in a subprocess of a refinery plant in Athens. The EDR system removes the measurement noise and provides reliable data for further processing. It also gives information about the existence of faulty instruments monitoring the operation of the industrial process.
Proportional plus integral plus derivative (PID) control is a well established technique for industrial control and possesses a high flexibility for use in many applications. However the tuning procedure is in many si...
详细信息
Proportional plus integral plus derivative (PID) control is a well established technique for industrial control and possesses a high flexibility for use in many applications. However the tuning procedure is in many situations nontrivial, which implies that it is difficult to find the values of the parameters that guarantee a desired closed-loop performance. The purpose of this paper is to present a PID-tuning technique which is based on a combination of two popular methods;namely the pole placement and the parameter optimization methods. The resulting PID self-tuning controller was applied to several simulated examples. This study showed that the controller has an improved performance over the simple pole placement or the simple parameter optimization self-tuning PID controller.
This paper presents and investigates a neural network structure which can perform general fuzzy inference. This system consists of a number of parallel neural network units which are called ''flexible inferenc...
详细信息
This paper presents and investigates a neural network structure which can perform general fuzzy inference. This system consists of a number of parallel neural network units which are called ''flexible inference cells'' (FICs). Each FIC implements a single-input/single-output (SISO) IF-THEN rule of a fuzzy knowledge base. The assumption of SISO fuzzy rules allows the implementation of any complex fuzzy inference algorithm (for control or other decision making purposes), since any MIMO (multi-input/multi-output) rule can be decomposed into an equivalent set of MISO (multi-input/single-output) rules, and a MISO rule can be decomposed to a set of SISO rules. The paper discusses the assumptions and requirements for the proposed neurofuzzy structure, and classifies the FICs into four categories. Some results derived by simulation using 3125 exemplar patterns produced computationally are provided.
Neural networks or connectionist models are massively parallel interconnections of simple neurons that work as a collective system, can emulate human performance and provide high computation rates. On the other hand, ...
详细信息
Neural networks or connectionist models are massively parallel interconnections of simple neurons that work as a collective system, can emulate human performance and provide high computation rates. On the other hand, fuzzy systems are capable to model uncertain or ambiguous situations that are so often encountered in real life. One way for implementing fuzzy systems is through utilizations of the expert system architecture. Recently, many attempts have been made to ''fuse'' fuzzy systems and neural nets in order to achieve better performance in reasoning and decision making processes. The systems that result from such a fusion are called neuro-fuzzy inference systems and possess combined features. The purpose of the present paper is to propose such a neuro-fuzzy system by extending and improving the system of Keller et al. (1992). The present system makes use of Hamacher's fuzzy intersection function and Sugeno's complement function. After a brief outline of the operation of the system its features are established with the aid of four theorems which are fully proved. The capabilities of the system are shown by a set of simulation results derived for the case of trapezoidal fuzzy sets. These results are shown to be better than the ones obtained with the original neuro-fuzzy system of Keller et al.
The problem of optimizing the operation of an autonomous manufacturing/supply chain is considered. The chain consists of an upper level manufacturer and a hierarchy of intermediate suppliers/subsuppliers that provide ...
详细信息
The problem of optimizing the operation of an autonomous manufacturing/supply chain is considered. The chain consists of an upper level manufacturer and a hierarchy of intermediate suppliers/subsuppliers that provide the required materials. Each subsystem involved in the chain incurs some costs that have to be handled effectively. This paper considers the case of ordering and holding costs. It is assumed that each level in the chain may add some delay in the materials' procurement due to the extra processing taking place or to the time required to supply the corresponding materials. Three possible scenaria are examined. In scenario I, the manufacturer optimizes his own costs and imposes the resulting policy to the (sub)suppliers without any question. In scenario II, all levels of the chain are assumed to cooperate in the sense that their common objective is to minimize the overall operational cost. Finally, in scenario III, each level is optimizing a local cost function in a decentralized way, which may be in conflict with that of the other levels. A partial cooperation is assumed, in the sense that each part minimizes its own costs, plus appropriately weighted deviations of its order profile from a set of calculated reference variables that depend upon the dynamics of the immediate downstream level. A complete numerical example illustrates the above cases.
The problem of scheduling n dependent tasks, with arbitrary processing times, on m identical machines so as to minimize the makespan criterion is considered. Since this problem is NP-hard in the strong sense, it can b...
详细信息
The problem of scheduling n dependent tasks, with arbitrary processing times, on m identical machines so as to minimize the makespan criterion is considered. Since this problem is NP-hard in the strong sense, it can be solved only suboptimally using heuristic approaches. Two new heuristic algorithms (dispatching rules), namely MVT/MISF acid DMVT/MISF algorithms, for this problem are proposed. These algorithms are then used, together with the existing ones CP/MISF and DHLF/MISF, as a dispatching rule base of a new adaptively weighted combinatorial dispatching (AWCD) rule. This combinatorial dispatching rule has a superior behaviour compared to simple dispatching rules. Extended experimentation with these algorithms supports this argument. Here a representative robotic dynamics computation example is included. In addition, some empirical rules are derived and proposed for the selection of a simple dispatching rule (heuristic) if such a selection is required, for each particular input data set. These methods, as well as the existing optimal algorithms for special solvable cases of the considered problem, have been integrated in a decision support system (DSS).
Flexibility and versatility are two basic requirements for industrial robots used in manufacturing areas. In the present paper, a kinematic transformation for redundant robots between Cartesian and joint coordinates i...
详细信息
Flexibility and versatility are two basic requirements for industrial robots used in manufacturing areas. In the present paper, a kinematic transformation for redundant robots between Cartesian and joint coordinates is achieved in the trajectory planner. This problem is treated by taking into account both the dynamics of the system expressed by its Lagrangian and the kinematic constraints of the robot. Given the joint trajectories as above, a control law for joint trajectory tracking is generated here using the Model Based Predictive control (MBPC) approach. The MBPC strategy is based on an explicit robot model to predict the process output over a long-range time period. The postulated control law is constructed such that to fit a desired command acceleration for the servo motors of the robotic system. The actual control law is calculated by minimizing a suitable objective function. A numerical example concerning a robot with one degree of redundancy illustrates the method and shows the excellent performance of the proposed MBPC scheme. The drifting problem that emerges in the resolved motion control of redundant robots is successfully faced.
Expert system technology is receiving increasing popularityand acceptance in the engineering community. This is due tothe fact that there actually exists a close match betweenthe capabilities of the current generat...
详细信息
ISBN:
(数字)9783642840487
ISBN:
(纸本)9783642840500
Expert system technology is receiving increasing popularity
and acceptance in the engineering community. This is due to
the fact that there actually exists a close match between
the capabilities of the current generation expert systems
and the requirements of engineering practice. Prepared by a
distinguished team of experts, this book provides a balanced
state-of-the-art presentation of the design principles of
engineering expert systems, and a representative picture of
their capabilities to assist efficiently the design,
diagnosis and operation of complex industrial plants.
Among the application areas covered are the following:
hardware synthesis, industrial plant layout design, fault
diagnosis, process control, image analysis, computer
communication, electric power systems, intelligentcontrol,
robotics, and manufacturing systems. The book is appropriate
for the researcher and the professional. The researcher can
save considerable time in searching the scattered technical
information on engineering expert systems. The professional
can have readily available a rich set of guidelines and
techniques that are applicable to a wide class of
engineering domains.
暂无评论