This study proposes an automatically customizable micro process planning system that reflects the change of properties of an actual machine tool to generate optimal machining information. The system consists of an updatable machining database and a database-oriented micro process planning algorithm. A machining database is updated based on analyzing NC data that are adaptively generated for an actual machine tool by skilled process planners. From the updated machining database, a database-oriented micro process planning algorithm is generated by a decision tree and a regression tree. Machining strategy and cutting tools are determined using IF-THEN rules that are generated from the database by the decision tree method. Cutting conditions are determined from a feasible regression equation. Regression equations and selection rules of these equations are generated from the database by the regression tree method. An example of micro process planning using the generated algorithm is also shown.
For detecting potential problems of a cutter path, cutting force simulation in the NC milling process is necessary prior to actual machining. A milling operation is geometrically equivalent to a Boolean subtraction of the swept volume of a cutter moving along a path from a solid model representing the stock shape. In order to precisely estimate the cutting force, the subtraction operation must be executed for every small motion of the cutter. By using GPU, the required time for the subtraction can be drastically reduced. In this paper, the computation speed of two known GPU accelerated geometric milling simulation methods, which are the depth buffer based method and the parallel processing based method with CUDA language, are compared. Computational experiments show that the implementation with CUDA is several times faster than the depth buffer based method when the cutter motion in the simulation process is sufficiently small.
In recent years, a large-scale logistic center plays an important role in mail-order business with Internet. In the logistic center, the efficient managing is required to deliver products to customers as soon as possible. Researches to efficiently control the logistic center have been done in the various approaches. This study proposed a new method for the order-picking problem considering worker’s jamming at the same shelf in the logistic center. In the proposed method, we formulate worker’s scheduling in the logistic center as Job-shop Scheduling Problem and optimize this problem. Numerical experiments show the proposed method improve worker’s scheduling compared with rule-based scheduling.
Investigations of human motor control using functional magnetic resonance imaging (fMRI) are increasingly receiving attention, with applications in fields such as motor learning and rehabilitation. In these neuroscience studies, force and position sensors are used to control haptic devices and safely interact with the human motion in an MR environment. However, conventional force sensors such as strain gauges are known to cause electromagnetic interference originating from electrical cables, transducers, and electronics. Light transmission through optical fibers is one alternative that avoids these problems. Since optical fibers do not produce electromagnetic noise, they can be used in an MR environment without electromagnetic interference. In this paper, we propose a novel design of an MRI-compatible grasping force sensor based on these principles. The sensor structure was designed to fit into an MRI scanner with its inclined double parallel mechanism, and was specifically adapted to precision grip tasks. This paper presents the sensor design and preliminary characterization in a non-MR environment.