Most information systems always require us to represent our information in a formal and strict way. This lack of flexibility reduces the usefulness of the information systems. In this paper, we introduce two weakly structured representation techniques for easily constructing and retrieving ill-organized information. The first technique is a word-list-query which is a free ordered sequence of words used to represent user's information needs. The second is an associative graph which is a kind of semantic network consisting of concept nodes, IS-A links and unnamed horizontal links. To parse a word-list-query, we use a concept reduction method which is recursive rewriting of words according to information represented in associative graph. This integration of parsing and information retrieval disambiguates word-list-queries. We implemented an experimental system using these techniques, and show that this system is easy for unskilled users and its accuracy is reasonable.
Many adaptive control schemes have been proposed for the systems whose parameters and time delays are unknown. Especially, from the simplicity of the control structure, self-tuning control (STC) schemes have been widely studied. In this paper, a STC scheme based on the minimum variance control (MVC) method is proposed for the nonminimum phase and/or unknown time delay systems, considering an augmented system equipped with a feedforward compensator on a parallel to the system. Since the poles of the closed-loop system constructed by the proposed method can be assigned arbitrarily, the good performance can be obtained relatively easily. Moreover, it is shown that the closed-loop system can be interpreted as a system constructed by the generalized minimum variance control (GMVC) method, or as a system equipped with a local feedback compensator like a Smith predictor. Finally, a numerical example and an example applied to an experimental plant are shown to illustrate the effectiveness of the proposed method.
Recently a new approach is required for a scheduling system which has flexibility against change of condition. This paper deals with re-scheduling for change of the due date, addition of a product, trouble of a machine and increase of processing time, respectively, in an open-shop manufacturing process. For solving this problem, the genetic algorithm (GA) is applied. First a method for constructing GA is proposed for an open-shop problem without change of conditions. By comparing GA with the random search method and the local search method, it is confirmed that the accuracy of GA is better than those of other methods. Secondly, when we solve the re-scheduling problem for change of several conditions, the initial population is revised in such a way that a number of individuals made randomly are added to the population. The effectiveness of the revised algorithm is confirmed through numerical results as compared with the algorithm without revision of the initial population.
In a multi-commodity game, we consider e-cores, least cores and nucleoli, which are originally defined in a conventional n-person cooperative game in characteristic function form. First we review the definition of a multi-commodity game and the concept of the core, and then introduce concepts of e-cores and least cores. Employing excess functions in the multi-commodity game, the cores, the ε-cores and the least cores are redefined, and nucleoli are defined. To compute solutions, we classify excess functions into two categories and give two instances for each category, and briefly mention the difference among them. We develop computational methods for obtaining a point in each of the least cores under some assumptions and illustrate one of the methods with a numerical example. Finally we refer to methods for obtaining a point in each of the nucleoli.
This paper proposes a method to measure the 3D position and attitude of a vehicle by use of corner cubes and laser fan beams. The method uses three sets of laser transceivers mounted on the vehicle and retro-reflecting targets (cornercubes) set along a side of the course. Each laser transceiver consists of a laser beam transmitter that emits a fan shaped laser and photo sensor for detecting the laser reflected by the retro-reflecting targets. The three laser beams are set with different angles and offset distances from each other, so that the time data of detecting the returned laser depend on the position and attitude of the vehicle. Then the on-board computer calculates the position and attitude from the time data. This paper reports the measurement principle, simulation and experimental results.