There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

While the Transformer architecture has become the de-facto standard for natural language processing tasks, its applications to computer vision remain limited. In vision, attention is either applied in conjunction with convolutional networks, or used to replace certain components of convolutional networks while keeping their overall structure in place. We show that this reliance on CNNs is not necessary and a pure transformer applied directly to sequences of image patches can perform very well on image classification tasks. When pre-trained on large amounts of data and transferred to multiple mid-sized or small image recognition benchmarks (ImageNet, CIFAR-100, VTAB, etc.), Vision Transformer (ViT) attains excellent results compared to state-of-the-art convolutional networks while requiring substantially fewer computational resources to train.

An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

From the Publisher:
This book explores the meta-heuristics approach called tabu search, which is dramatically changing our ability to solve a hostof problems that stretch over the realms of resource planning,telecommunications, VLSI design, financial analysis, scheduling, spaceplanning, energy distribution, molecular engineering, logistics,pattern classification, flexible manufacturing, waste management,mineral exploration, biomedical analysis, environmental conservationand scores of other problems. The major ideas of tabu search arepresented with examples that show their relevance to multipleapplications. Numerous illustrations and diagrams are used to clarifyprinciples that deserve emphasis, and that have not always been wellunderstood or applied. The book's goal is to provide ''hands-on' knowledge and insight alike, rather than to focus exclusively eitheron computational recipes or on abstract themes. This book is designedto be useful and accessible to researchers and practitioners inmanagement science, industrial engineering, economics, and computerscience. It can appropriately be used as a textbook in a masterscourse or in a doctoral seminar. Because of its emphasis on presentingideas through illustrations and diagrams, and on identifyingassociated practical applications, it can also be used as asupplementary text in upper division undergraduate courses. 

Finally, there are many more applications of tabu search than canpossibly be covered in a single book, and new ones are emerging everyday. The book's goal is to provide a grounding in the essential ideasof tabu search that will allow readers to create successfulapplications of their own. Along with the essentialideas,understanding of advanced issues is provided, enabling researchers togo beyond today's developments and create the methods of tomorrow.

Tabu Search

We consider the problem of cooperation among a collection of vehicles performing a shared task using intervehicle communication to coordinate their actions. Tools from algebraic graph theory prove useful in modeling the communication network and relating its topology to formation stability. We prove a Nyquist criterion that uses the eigenvalues of the graph Laplacian matrix to determine the effect of the communication topology on formation stability. We also propose a method for decentralized information exchange between vehicles. This approach realizes a dynamical system that supplies each vehicle with a common reference to be used for cooperative motion. We prove a separation principle that decomposes formation stability into two components: Stability of this is achieved information flow for the given graph and stability of an individual vehicle for the given controller. The information flow can thus be rendered highly robust to changes in the graph, enabling tight formation control despite limitations in intervehicle communication capability.

/pdf/information-flow-and-cooperative-control-of-vehicle-4ws61wioze.pdf

Information flow and cooperative control of vehicle formations

This paper presents a novel control method for torque generation in four-stroke spark ignition (SI) engines. A model-based approach is employed to control engine torque output by adjusting throttle air intake with considerations of robust stability and performance. Discrete event engine model (DEM) is adopted with an addition of torque generation dynamics. Disturbance observer (DOB) techniques are utilized to achieve robust stability and performance by regarding the discrepancy between the actual plant and the nominal plant with desired plant characteristics as an equivalent disturbance input, which is estimated and cancelled. The desired plant behavior is stably realized by the DOB up to a bandwidth which is sufficient for a torque control application discussed in authors’ previous work. A switching scheme for desired nominal plant is proposed to further enhance robust performance and stability. Numerical results show the effectiveness of the proposed schem.Copyright © 2007 by ASME

Engine Torque Control Based on Discrete Event Model and Disturbance Observer

Joint Synthesis of Controller and Fault Detector and Its Application to Motor Drive Control System

In this paper, by studying the robot dynamics and error characteristics, we propose a feedforward input generation scheme using radial basis function NN [8‐10] to learn and predict the model following error. Learning/training is performed for the NNs prior to the prediction for real-time control stage. The prediction and control problem is properly decoupled into subproblems for each individual joint to reduce the algorithm complexity and computation requirements. The performance of the proposed approach is experimentally evaluated and compared with the sensor based learning control, which requires additional learning and sensing for each new trajectory.

Feedforward Input Generation Based on Neural Network Prediction in Multi-Joint Robots

The paper discusses state-space generalized predictive control and the preview control algorithms. The optimization procedure used in the derivation of predictive control algorithms is considered. The performance index associated with the generalized predictive controller (GPC) is examined and compared with the linear quadratic (LQ) optimal control formulation used in preview control. A new performance index and consequently a new algorithm is proposed dynamic performance predictive controller (DPPC) that combines the features of both GPC and preview controller. This algorithm minimizes the performance index through a dynamic optimization. A simple example illustrates the features of the three algorithms and prompts a discussion on what is actually minimized in predictive control. The DPPC algorithm, derived in this paper, provides for a minimum of the predictive performance index. The differences and similarities between the preview control and the predictive control have been discussed and optimization approach of predictive control has been explained.

State-space dynamic performance preview-predictive controller

The idea of "sliding mode control" (SMC) as a robust control method is utilized to control systems whose dynamics can be described by x(t)=(A+/spl Delta/A)x(t)+(B+/spl Delta/B)u(t)+d(t) where /spl Delta/A, /spl Delta/B and d(t) characterize unknown plant parameters and unexpected disturbances respectively. The design is described in the discrete form using the Euler operator, which approaches the Laplace operator as the sampling interval approaches zero. Uncertainty transformation from continuous to discrete time domain, which has not been adequately addressed in the control literature, is included throughout the analysis. Selection of the sliding surface, s=0, is not arbitrary but is dependent on the nominal parameters of the system and on the chosen sampling time. The control is constructed based on the estimation of system perturbations using time delay control. This methodology, towards SMC design, offers a robust control action which is continuous with possibly lower gains than the conventional "bang-bang" input, thus avoiding undesired chattering. Such issues are addressed and the asymptotic stability of the controlled system is analyzed. Finally the proposed controller is studied through simulations and experiment on a direct drive NSK robot arm. The results indicate that the controller is robust to parameter variations and unexpected disturbances. >

Masayoshi Tomizuka

Papers

Engine Torque Control Based on Discrete Event Model and Disturbance Observer

Joint Synthesis of Controller and Fault Detector and Its Application to Motor Drive Control System

Feedforward Input Generation Based on Neural Network Prediction in Multi-Joint Robots

State-space dynamic performance preview-predictive controller

Sliding control of discretized continuous systems via the Euler operator