The shortcomings of inertial sensor-based localization for mobile robots are discussed and a method to improve the performance of localization is introduced. Because dead reckoning using wheel speed sensors is insufficient for accurate motion estimation in uneven and slippery environments, many studies have used inertial sensors in order to improve the performance of localization. However, inertial sensor-based localization also has inherent problems caused by an accumulation of sensor errors. To improve the performance of inertial sensor-based localization for mobile robots, a method is suggested that involves the reduction of unexpected noise of inertial sensor using a Multiple Estimation Windows (MEW) filter. The main concept of the MEW filter is to estimate the long-term signal of interest using a single long-term moving window and multiple short-term moving windows. The proposed denoising method can reduce the noise of all frequencies without the loss of information from the signal of interest. The performance improvement of the mobile robot's localization using the MEW filter is confirmed experimentally.

Brain-computer interfaces rely on accurate decoding of cortical activity to understand intended action. Algorithms for neural decoding can be broadly categorized into two groups: direct versus generative methods. Two generative models, the population vector algorithm (PVA) and the Kalman filter (KF), have been widely used for many intracortical BCI studies, where KF generally showed superior decoding to PVA. However, little has been known for which conditions each algorithm works properly and how KF translates the ensemble information. To address these questions, we performed a simulation study and demonstrated that KF and PVA worked congruently for uniformly distributed preferred directions (PDs) whereas KF outperformed PVA for non-uniform PDs. In addition, we showed that KF decoded better than PVA for low signal-to-noise ratio (SNR) or a small ensemble size. The results suggest that KF may decode direction better than PVA with non-uniform PDs or with low SNR and small ensemble size.

This paper addresses an efficient vision based motion estimation method of robot systems for the ego-motion compensation using type-2 fuzzy sets. Every intelligent robot system like walking robots, service robots, automatic vehicles should have the ability to autonomously recognize its surroundings and to make right decisions under unknown environment. To enable a robot system to do this, ego-motion compensation is mandatory. Therefore, we suggest the ego-motion estimation method so that the errors of the environment recognition caused by the egomotion of intelligent robot systems are eliminated. This method uses the disparity map obtained from the stereo-vision and can be divided into three parts - segmentation, feature extraction, estimation. In the segmentation part, a novel type-2 fuzzy sets based segmentation method to extract the objects is used. In the feature extraction module, features are extracted by the wavelet level-set transform. In the estimation part, least square ellipse approximation is used to calculate the displacement for the rotation and translation between image sequences. we can estimate the displacement for the rotation and translation by tracking the least square ellipse and type-2 fuzzy sets based filtering method. From the results of experiments, we can know that the proposed method can be applied to intelligent robot systems effectively.

Intelligent Space is an area to perceive and understand what is happening in it, which provides diverse services to human and robots. We expect that Intelligent Space offers a lot of benefits as a good platform for robots. For example, robots can move by the guidance of Intelligent Space although they don't know about the environment. If a new robot comes in Intelligent Space, the robot wants to receive the guidance service from the space for its navigation. So, the space should perceive and control the robot, since it doesn't know anything about the space. In this paper, we propose a new algorithm for controlling an unknown robot in Intelligent Space. An unknown robot is an unidentified or new coming robot that an intelligent space has not yet known it. Objectives of the algorithm are to estimate the location of robot, to navigate it, and to identify it. The algorithm uses a color-based particle filter to estimate the location of robot and a predefined motion pattern to identify the robot. To evaluate the proposed algorithm, we command that a mobile robot moves according trajectory of robot.

A direct adaptive state-feedback controller is proposed for highly nonlinear systems. We consider uncertain or ill-defined nonaffine nonlinear systems and employ a neural network (NN) with flexible structure, i.e., an online variation of the number of neurons. The NN approximates and adaptively cancels an unknown plant nonlinearity. A control law and adaptive laws for the weights in the hidden layer and output layer of the NN are established so that the whole closed-loop system is stable in the sense of Lyapunov. Moreover, the tracking error is guaranteed to be uniformly asymptotically stable (UAS) rather than uniformly ultimately bounded (UUB) with the aid of an additional robustifying control term. The proposed control algorithm is relatively simple and requires no restrictive conditions on the design constants for the stability. The efficiency of the proposed scheme is shown through the simulation of a simple nonaffine nonlinear system.

We introduce a hybrid architecture that dwells on the ideas of fuzzy rule-based computing and an approximation scheme (SOPNN). The hybrid system is combined to get a novel heuristic approximation method. This composite structure overcomes the shortcomings of the individual methods especially it solves drawbacks of SOPNN while maintaining their desirable features. The combined method is efficient and much more accurate than either of the two individual schemes as well as other modeling methods. A three-input nonlinear static function is demonstrated for the utility of the proposed approach.

A direct adaptive state-feedback controller for highly nonlinear systems is proposed. This paper considers uncertain or ill-defined nonaffine nonlinear systems and employ a static fuzzy logic system (FLS) with flexible structure, i.e., on-line variation of the number of fuzzy rules. The FLS approximates and adaptively cancels an unknown plant nonlinearity. A control law and adaptive laws for unknown fuzzy parameters and bounding constant are established so that the whole closed-loop system is stable in the sense of Lyapunov. The tracking error is guaranteed to be uniformly asymptotically stable rather than uniformly ultimately bounded with the aid of an additional robustifying control term. No a priori knowledge of an upper bound on a lumped uncertainty is required.

This paper proposes an autonomous motion controller for mobile robots. It combines a biologically inspired motion planner and a nonlinear feedback controller in order to make a simple and practical motion controller in natural unknown environments. The motion planner part is designed in the frame of behavior-based scheme and adopts the biological concept "homeostasis". It plays the role of determining desired temporal target posture and modifying controller's gain parameters. The other nonlinear controller part drives the robot to the given target posture from the planner. As a result, the stable nonlinear feedback controller in ideal obstacle free space becomes practical one in natural environments constrained by obstacles. The proposed controller is simple mainly due to its behavior-based characteristics and stable due to the feedback controller considering robot's physical constraints. Simulated results and some experimental feature for autonomous navigation of the differential drive mobile robot, named "MARI", are shown for the present.

In this paper, a flexible microassembly system based on hybrid manipulation scheme is proposed and applied to the assembly of a photonic component. In order to achieve both high precision and dexterity in microassembly, we propose a hybrid microassembly system with sensory feedbacks of vision and force. This system consists of the distributed 6-DOF micromanipulation units, the stereo microscope, and haptic interface for the force feedback-based microassembly. A hybrid assembly method, which combines the vision-based microassembly and the scaled teleoperated microassembly with force feedback, is proposed. The feasibility of the proposed method is investigated via experimental studies for assembly micro opto-electrical components. Experimental results show that the hybrid microassembly system is feasible for applications to the assembly of photonic components in the commercial market with better flexibility and efficiency.

This paper presents a robust adaptive flux observer for induction motor control. The proposed flux observer guarantees the robustness against parametric uncertainties, and needs no estimation schemes with employing additional robustifying signal to cope with parametric uncertainties. The proposed adaptive scheme is determined so that the observer dynamics are stable in the sense of Lyapunov. The estimated rotor speed is then used to generate feedback control signal for speed sensorless vector control system. The simulation results are presented to show the validity and efficiency of the proposed system.

We discuss a new hybrid architecture that dwells on the ideas of combining neural networks (NNs) with self-organizing approximators (self-organizing polynomial neural networks: SOPNN). The hybrid system is combined to get a novel heuristic approximation method which can find a workable synergistic environment for nonlinear system modeling. NNs have been used successfully in order to describe the static and dynamics of nonlinear process system. SOPNN is an analysis technique for identifying nonlinear relationships between inputs and outputs of the system and builds hierarchical polynomial regressions of required complexity. To harmonize NN with SOPNN and find a workable synergistic environment, systematic approach, neural networks based SOPNN, has been investigated and contained its comparative study in this paper. Identification results of nonlinear system with two inputs will be demonstrated to show the performance of the proposed approach. As a result, it is shown that the proposed method is efficient and much more accurate than either of the two individual schemes as well as other modeling methods.

We discuss a novel design methodology of self-organizing approximator technique (self-organizing polynomial neural networks) (SOPNN) in the framework of genetic algorithm (GA). SOPNN dwells on the ideas of group method of data handling (GMDH). Its each node exhibits a high level of flexibility and realizes a polynomial type of mapping between input and output variables. But the performances of SOPNN depend strongly on a few factors. They are number of input variables available to the model, number of input variable and type (order) of the polynomials to each node. In most cases, these factors are determined by the trial and error method. Moreover, SOPNN algorithm is a heuristic method so it does not guarantee that the obtained SOPNN is the best one for nonlinear system modeling. Therefore, more attention must be paid to solve the drawbacks. We alleviate these problems by using GA. Comparisons with other modeling methods and conventional SOPNN show that the proposed design method has better performance.

We propose and analyze a robust adaptive fuzzy controller for uncertain nonlinear systems without information on the input gain sign. The proposed scheme completely overcomes the singularity problem which occurs in the indirect adaptive feedback linearizing control. No projection in the estimated parameters and no switching in the control input are needed. The stability of the closed-loop system is guaranteed in the Lyapunov viewpoint.

This paper presents micropositioning system using feedback of multiple magnification images from CCD cameras mounted on an optical microscope. In order to guarantee of the accurate micropositioning to micro assembly system, high precise sensors are required. But many kinds of sensors, which are used in macro world, have insufficient resolution and size. The visual data from an optical microscope have vast information, however, the single field-of-view of optical microscope essentially limits the workspace of the micromanipulator and the low depth-of-field makes it difficult to handle micro parts. To overcome these problems and increase and precision of micropositioning system, we use multiple magnification images that have same viewpoint. The use of multiple magnification images reduces the time consuming for handling of micro parts and increases precision of micropositioning. We propose this method for micropositioning system.

A resource sharing architecture (RSA) is a robot architecture for a mobile robot in a networked and intelligent space. A robot based on a RSA performs its tasks by sharing many external resources around it and thus has a simple, flexible and low priced structure. However, RSA has a problem as it must guarantee a real-time condition, because it is designed to use the network protocol JINI. For an example, a time delay occurs in transferring data from a vision service to a robot control service. In this paper, we propose a data accessing service (DAS) to solve this problem. It is a service of a network protocol JINI, which transfers data between two services. Its basic function is to find a data source service that a data destination service needs, and after doing this, the DAS transfers data between two services using not JINI, but another network protocol such as TCP/IP or UDP. As a result, we can reduce the time delay to access data. Some experimental results show the validity of our proposition.

This paper describes the design of a robust adaptive fuzzy observer for uncertain nonlinear dynamical system. The Lyapunov synthesis approach is used to guarantee a uniform ultimate boundedness property of the state estimation error, as well as of all other signals in the closed-loop system. Especially, we have focused on the realization of minimal dynamic order of the observer. For the purpose, we propose a new method in which no strictly proper (SPR) condition is needed and combine dynamic rule insertion scheme with on-line estimation of fuzzy parameters. No a priori knowledge of upper bounds on the optimal parameters and modeling errors is required. The theoretical results are illustrated through a simulation example.

We present a control method for a general nonlinear systems using Takagi-Sugeno fuzzy models and develop an adaptation law to adjust the parameters of the fuzzy system. It is proved that the closed-loop system is stable in the sense of Lyapunov and all signals including tracking error, fuzzy parameters and estimate of upper bound for approximation error are bounded. Finally, simulation results verify the effectiveness of the proposed control method.

In this paper, by using network protocol Jini, we propose an advanced resource sharing architecture (RSA) for a mobile robot. We made several services with all resources around the robot by using Jini. There is a task managing service for a certain task, and the service uses other service to perform a task as well as make a new service of other services. To verify our architecture we built a simple experimental space. In this space we performed the experiments such as a detection of desired objects, path generation and motion control for a robot, obstacle avoidance, and so on. Experimental results show that the mobile robot with our advanced RSA performs various tasks successfully.

In a robot soccer game, a minimal trajectory generating method and accurate motion trajectory control are very important. We suggest a trajectory generating method which is very simple and effective. This method uses only arcs and lines. We also propose a new motion tracking control method. In most motion tracking control the main control parameter is a steering angle. But we use a steering radius instead of the steering angle as a control parameter. In this method we don't have to consider a dynamic equation of the robot. As a result we can control robots simply and accurately. These proposed methods are applied to the robot soccer system, named 'Crimson', which has participated in Robot WorldCup'98 in Paris. Simulation results show that these methods are simple and effective compared with the conventional methods.