Hirokazu Ihara

FUZZY CONTROL FOR AUTOMATIC TRAIN OPERATION SYSTEM

A fuzzy control that uses rules based on a skilled human operators experience is applied to automatic train operations. Essentially, train operation is a nonlinear control function. Up to now, this system has been developed by linearlized control using a target pattern. However, it was difficult to control the train in a manner similar to a human operator. In this paper, a fuzzy control which uses linguistical control rules is proposed. Furthermore, an automatic train operation system which evaluates safety, comfort, traceability of target velocity, and the accuracy of a stop gap has been developed. According to the result of the simulation, our newly developed automatic control system having the fuzzy control realizes a comfortable and accurate stop control, and its control rules are easy to understand and are connected to the performance indices. Therefore, system performance can be directly adjusted as desired.

Dependable onboard computer systems with a new method-stepwise negotiating voting

An algorithm for software voting, called stepwise negotiating voting, which can tolerate the faults in up to N-1 subsystems is introduced. The voter behaves as if it were a majority voter if the number of remaining subsystems is sufficient for majority voting, and standby redundancy is realized if the number of remaining subsystems becomes insufficient. With this voting method, the system can survive if more than one subsystem remains. The authors introduce a method for evaluating the dependability of systems. It is based on the viewpoint that not only the hardware reliability but also the reliability of data processing is important. It is assumed that only transient faults take place in the software behavior. The authors concept can be applied to computers in critical application fields, such as space development or engine control.<<ETX>>

Autonomous Controllability of Decentralized System Aiming at Fault-Tolerance

Abstract To design a decentralized control system which is fault-tolerant against failures of any parts, the system is partitioned into a number of autonomous subsystems. Each autonomous subsystem should prevent the propagation of fault-caused errors. Hence autonomous controllability is defined as controllability for the combination of the functioning subsystems except the failed subsystems. The necessary and sufficient conditions for autonomous controllability are obtained, and it is shown that these results can be used to design a fault-tolerant control system either by integrating subsystems or dividing a system.

A track correlation algorithm for multi-sensor integration

A track correlation algorithm in a multi-sensor integration system for a surveillance mission is proposed. The performance of such track correlation processing is strongly dependent not only on the target state estimation error distribution but also on the target spatial density distribution. Therefore, a track correlation problem is formulated as the likelihood ratio test problem which can take both target state estimation error distribution and target state spatial density distribution into consideration. From this formulation, the correlation algorithm for on-line processing is derived by modifications and approximations. Through analytical evaluations and simulation studies, it is shown that the proposed algorithm is superior to the conventional nearest neighbor algorithm.

Autonomous decentralized control and its application to the rapid transit system

Abstract A large scale control system has three intrinsic requirements of fault tolerance, on-line extension and maintenance for non-stop operation. For attaining these requirements, the new design philosophy for autonomous decentralized control system, called HAL (Harmonous, Autonomous and Localities) concept, has been proposed based on the biological analogy. The viewpoint of this concept is that it is quite normal for a system to have failed parts. Sufficient conditions for an autonomous decentralized control system are given that every subsystem is uniform and equal in function and structure, and utilizes only local information for control and coordination. As one of the applications to the system components of this new concept, ADL-Net (Autonomous Decentralized Loop Network) has been developed, where fault tolerance of the system is attained by giving autonomy for fault detection, recovery and transmission to every network control processor. The HAL concept has also been applied to the train traffic control system for subways with three intrinsic requirements for supporting their non-stop operations.

Fuzzy Control for Automatic Train Operation System

Abstract A fuzzy control that uses rules based on a skilled human operators experience is applied to automatic train operations. Essentially, train operation is a nonlinear control function. Up to now, this system has been developed by linearlized control usina a target pattern. However, it was difficult to control the train in a manner similar to a human operator. In this paper, a fuzzy control which uses linguistical control rules is proposed. Furthermore, an automatic train operation system which evaluates safety, comfort, traceability of target velocity, and the accuracy of a stop gap has been developed. According to the result of the simulation, our newly developed automatic control system having the fuzzy control realizes a comfortable and accurate stop control, and its control rules are easy to understand and are connected to the performance indices. Therefore, system performance can be directly adjusted as desired.

Fuzzy Logic for Control Systems

Abstract Ever since fuzzy set theory was first proposed by Professor Zadeh, People in the control field have examined very carefully its applicability to the real world. Very few results of practical application were reported namely due to the difficulty of proving validity. One application to subway control in Japan awoke and encouraged people who had been expecting fuzzy logic. Control experts now cannot help considering its presence. Following the results shown from subway train control and crane control, the possibility of application to critical control, such as aerospace, is surveyed and discussed.

Autonomous decentralization concept and its application to railway control systems

An autonomous decentralization concept based on a biological analogy has been developed. It yields a higher degree of reliability and flexibility by designing system structure, control and expansion on the assumptions that even if any subsystem fails, the other subsystems will continue working without shutting down. A subway transit system based on this concept has been realized. The functions of power supply control, train traffic control, train maintenance and dispatching are decentralized so as to prevent local trouble from affecting the whole system and to maintain transportation service. The autonomous decentralized concept has actually been applied as follows. Station control systems connected to an autonomous decentralized loop network system (ADL), are autonomous subsystems equivalent with others, having timetable information for each station. They can regulate train traffic based on only local information about how trains have been delayed or will be delayed.

Fault-tolerant computer system with stepwise negotiating voting

In the conventional fault-tolerant system, it is difficult to satisfy both requirements of long system life and safety. This paper proposes a method in which the output data from subsystems is selected based on the reliabilities of the data processings in the redundant subsystems, which are estimated by the self- and mutual checks of each subsystem. By the proposed method, a new fault-tolerant system can be realized. This system contains both the long life, as a feature of the stand-by redundant method, and a highly reliable data processing, as a feature of the majority redundant method. It is expected that the proposed method will be applied to the satellite computer. In the proposed method, the trade-off between the life and the reliability of the data processing can easily be achieved according to the purpose of application, by adequately defining the required level for the reliability. The method can cope with any multiplicity, making it applicable to various problems.

Microcomputer Based on Board Real-Time Control System for Automated Guideway Rapid Transit System

Abstract For the purpose of realizing a more reliable and flexible system, this on-board system utilizes a modular structure. In the hardware structure, the on-board system consists of two microcomputer modules. They are connected by communication lines for exchanging control information and for synchronizing each other, and are coupled loosely to prevent adverse effects in the case of one modules failure. In the software structure, structured programing technology is applied to increase the flexibility of the system. The programs are implemented in the form of a functional cluster structure in accordance with the functional structure of the automatic train operation. In this system, the Real-time Monitor plays) an important role in maintaining the consistency of the control information and the control timing of the two processors. Autonomous synchronization is applied to maintain the fault tolerance of the system. This on-board system has been applied to the automatic train operation of the subways and medium sized automated guideway rapid transit system.