This paper presents a system representation and a control strategy for a multi-rate linear discrete-time plant. No traditional mathematical model such as transfer function or state equation of the plant is employed. The plant dynamics is represented as a set of basis vectors whose elements are multi-rate sampling data of the plant. Then it is shown that, for a reference signal, a dead-beat control input which minimizes a quadratic performance index can readily be computed by means of a convex optimization defined by the data array.
A railway crew scheduling problem is an assignment problem of crews for a given operational schedule under the constraint on labor regulations. In this paper, we discuss an automatic railway crew scheduling method based on numerical optimization for a large-scale and complex railway route with branch lines and multiple bases. The proposed method makes the crew schedule automatically using step by step optimization based on a partial pairing, an elastic embedded set partitioning problem and a graph partitioning.
This study presents an extension of Bayesian learning approach with Gaussian process regression focusing on continuous-time optimal control problem in which stage cost function is unknown. By applying control parametrization method, the optimal control problem can be approximately formulated as a nonlinear programming problem, and the statistics of the cost function estimated by Gaussian process regression is analyzed. To obtain a solution to Bayesian optimization problem, an effective gradient calculation based on variational method is developed. Furthermore, the analysis of optimality in the fashion of bandit problem provides the order of regret bound achieved by the proposed algorithm.
It is important to increase the amount of the trade of farm products. However, a plant quarantine is often problem. The amount of the trade of the peach is the third in Japan, and the 40% peaches are transported to Taiwan. The peach fruit moth is very serious problem in Taiwan. When the peach containing the larva of the peach fruit moth is exported, the trade stops until the countermeasure is approved. This penalty is decided by the terms of the plant quarantine in Taiwan. We are developing the system to scan the inner of the peach using X-ray. The accuracy and the speed of the scan are not only important, but also the consideration of the transportation risk of the peach containing the larva. By the consideration, the necessary function can be enhanced, and the unnecessary function can be removed. Therefore, the development period is decreased and the development cost is saved. This paper describes that the method to decrease the transportation risk of the peach that is transported to Taiwan is considered. The result of the consideration can be adapted to the other fruit, because this result is not limited to the peach. Finally, effectiveness of the part of the methods is verified by using the real system that is installed in a packing house.
In this paper, we investigate optimal management of a data center cooling system in coordination with ICT equipments. In recent data centers, a system manager has become able to access data in servers such as equipment temperatures, power consumption, fan rotation speeds and computational loads. The technological advance would allow one to model the cooling process more accurately and to feedback online data to control algorithms for air conditioning units, which would contribute to enhancing energy efficiency for the cooling system. In this paper, we first model the cooling process based on the above data and then formulate an optimization problem to be solved at a certain time interval to determine the reference set-point of the air conditioning units. The real-time optimization scheme is then demonstrated on a real data center cooling testbed.
In this paper, we address a problem of enhancing damping performance of power systems. The performance enhancement is achieved by controlling the power flow and by supplementarily adding controllers. The power flow control problem is formulated as parameter design in the parameter-dependent power system model. We propose an iterative approach of linear matrix inequalities to solve the parameter design problem. Furthermore, in order to improve damping performance of the power system, controller design is combined with the parameter design.
At the sports and games for elementary and junior high school students, their parents often become personal photographer for the game. They see their child over the frame despite being in the playing field. In addition, problems such as trying to take a good place to compete may occur. We can solve these problems by the conference management side distributing the sports competition videos. So in this paper, we show a method to automatically reconstruct movies using competition movies taken by fixed point cameras and to present high quality movies to viewers.
In this paper, we propose generic object recognition module, which is called as RT component for OpenRTM-aist. Before now, how to design the RT component has not been discussed, clearly. Thus, the component design concept based on object recognition component is also discussed. Moreover, the evaluation experiment for implementations of RT component is also discussed.