Masayoshi Tomizuka

Bio: Masayoshi Tomizuka is an academic researcher from University of California, Berkeley. The author has contributed to research in topics: Control theory & Control system. The author has an hindex of 80, co-authored 1111 publications receiving 30069 citations. Previous affiliations of Masayoshi Tomizuka include University of California & Western Digital.

Robust motion controller design for high-accuracy positioning systems

Abstract: This paper presents a controller structure for robust high speed and accuracy motion control systems. The overall control system consists of four elements: a friction compensator; a disturbance observer for the velocity loop; a position loop feedback controller; and a feedforward controller acting on the desired output. A parameter estimation technique coupled with friction compensation is used as the first step in the design process. The friction compensator is based on the experimental friction model and it compensates for unmodeled nonlinear friction. Stability of the closed-loop is provided by the feedback controller. The robust feedback controller based on the disturbance observer compensates for external disturbances and plant uncertainties. Precise tracking is achieved by the zero phase error tracking controller. Experimental results are presented to demonstrate performance improvement obtained by each element in the proposed robust control structure.

Control issues in automated highway systems

Abstract: This article describes vehicle control issues that must be faced in designing a fully automated highway system (AHS). In particular, requirements for a control system architecture as well as issues of lateral and longitudinal "platoon" control are addressed. Interest in AHS is clearly expanding at a rapid pace due to the ever-increasing problems of freeway congestion and the potential for a technological solution. The approach described is based on five years of research as part of the California PATH program. >

Sparse R-CNN: End-to-End Object Detection with Learnable Proposals

Abstract: We present Sparse R-CNN, a purely sparse method for object detection in images. Existing works on object detection heavily rely on dense object candidates, such as $k$ anchor boxes pre-defined on all grids of image feature map of size $H\times W$. In our method, however, a fixed sparse set of learned object proposals, total length of $N$, are provided to object recognition head to perform classification and location. By eliminating $HWk$ (up to hundreds of thousands) hand-designed object candidates to $N$ (e.g. 100) learnable proposals, Sparse R-CNN completely avoids all efforts related to object candidates design and many-to-one label assignment. More importantly, final predictions are directly output without non-maximum suppression post-procedure. Sparse R-CNN demonstrates accuracy, run-time and training convergence performance on par with the well-established detector baselines on the challenging COCO dataset, e.g., achieving 45.0 AP in standard $3\times$ training schedule and running at 22 fps using ResNet-50 FPN model. We hope our work could inspire re-thinking the convention of dense prior in object detectors. The code is available at: https://github.com/PeizeSun/SparseR-CNN.

Control of Rotary Series Elastic Actuator for Ideal Force-Mode Actuation in Human–Robot Interaction Applications

Abstract: To realize ideal force control of robots that interact with a human, a very precise actuating system with zero impedance is desired. For such applications, a rotary series elastic actuator (RSEA) has been introduced recently. This paper presents the design of RSEA and the associated control algorithms. To generate joint torque as desired, a torsional spring is installed between a motor and a human joint, and the motor is controlled to produce a proper spring deflection for torque generation. When the desired torque is zero, the motor must follow the human joint motion, which requires that the friction and the inertia of the motor be compensated. The human joint and the body part impose the load on the RSEA. They interact with uncertain environments and their physical properties vary with time. In this paper, the disturbance observer (DOB) method is applied to make the RSEA precisely generate the desired torque under such time-varying conditions. Based on the nominal model preserved by the DOB, feedback and feedforward controllers are optimally designed for the desired performance, i.e., the RSEA: (1) exhibits very low impedance and (2) generates the desired torque precisely while interacting with a human. The effectiveness of the proposed design is verified by experiments.

I and i

Abstract: 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 …

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

Abstract: 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.

Tabu Search

Abstract: 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.

Information flow and cooperative control of vehicle formations

Abstract: 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.