[in Japanese]

View full abstract

[in Japanese]

View full abstract

In mechatronics the field of oscillation damping control for flexible structures is one of the key areas resulting in interesting new functionalities. Reasons are given by the necessity of exact adjustment of systematic evaluation of dynamic properties, adequate sensor and actuator systems, efficient and capable hardware and software implementation. It can be clearly seen that this combination is an ideally reflection of the three main areas of mechatronics: mechanical engineering, electrical engineering and computer science. In the talk at four examples of oscillation damping control respectively control of mechanical structures the idea of mechatronic design from a control perspective is given. First example is the oscillation damping control for a fire turntable ladder, second example is an automated slewing crane, third example is the oscillation damping control for a reconnaissance and disarming robot system of the German army, and the fourth example is dealing with a mechatronic approach for the control of a deformable, flexible mirror in adaptive optics.

View full abstract

A 6DOF parallel-link-type active vibration reducer is installed on a cart to damp liquid sloshing in a cylindrical container. The liquid sloshing in the transfer is damped with the vibration reducer using an optimal regulator. The regulator's gains are determined by using a genetic algorithm. In addition, the regulator's gains are calculated through gain scheduling when the static liquid level is changed. The effectiveness of the present method is clarified through simulation and experimental results.

View full abstract

When the liquid container is rapidly transferred, the sloshing (liquid vibration) is caused. Thus, transfer motion for conventional speed control by making speed slow spends a lot of transfer time and makes the tact time delay. Therefore, in this research, optimum sloshing suppression control of liquid with high-speed transfer is proposed. In order to express the sloshing, mathematical model comprised of the continuous equation and the pressure equation is used, and the sloshing phenomena are analyzed by using Boundary Element Method (BEM). Further, BEM model is transferred into the state-variable model. The state-variable model has nice characteristics such that control design is easier. In this research, a rectangular container is used. Then, liquid surface is expressed by the curve line when a container is transferred in a single direction. This curve line is divided into a number of evaluation points by using BEM. The effectiveness of proposed liquid surface model is shown by comparing with CFD (computational fluid dynamics) simulation and experiments with liquid container transfer. The proposed model can estimate not only first-order mode sloshing but also higher-order mode sloshing, and predict the future behavior of liquid level more exactly. In order to suppress the sloshing, Generalized Predictive Control (GPC) is used. GPC is one the most popular MPC method in industry. Firstly, the behavior of liquid surface is predicted about specified predictive region. Then, the acceleration input of container transfer is calculated such that the predicted value approaches zero. Therefore, a liquid is transferred without sloshing. As experiment results by using the liquid transferring robot, sloshing is completely suppressed, and the liquid transfer with proposed control method is reached to a goal earlier than that by the previous control method. Therefore, the transfer time is shortened and the proposed liquid transfer control is effective to improve casting productivity.

View full abstract

[in Japanese]

View full abstract

High speed operation of industrial transfer robot arm generates residual vibration, which in many cases may limit productivity of the factory. In this paper, we use input shaping approach to reduce the residual vibration. The approach needs value of natural frequency and damping ratio to be identified in advance. We present an integrated tool of parameter identification and vibration control for higher modes vibration systems, as an easy and effective tool to help industrial people to analyze and solve the vibration problem. A semiconductor wafer transfer robot arm is used as an application case. Experiment has been carried out, and the results showed significant reduction in settling time as well as total movement time of the robot arm.

View full abstract

[in Japanese]

View full abstract

This paper clarifies the dynamic behaviors of a mass measurement system with conveyor belt (a checkweigher). The purpose of this paper is to construct a mathematical model of the mass measurement system. The checkweigher with electromagnetic force balance system can be approximated by a spring-mass-damper system as the physical model, and the equation of motion is derived. Finally, the validity of the proposed model can be confirmed by comparison of the simulation results with the realistic responses.

View full abstract

For sampled-data underactuated three degree-of-freedom ships, the design of discrete-time surge and yaw state feedback laws that make a ship track a desired straight-line path while maintaining a desired nonzero constant forward speed is considered. First a line-of-sight guidance algorithm is introduced and by using the Euler approximate model of tracking error dynamics, discrete-time surge and yaw state feedback laws are designed. Then applying the nonlinear sampled-data control theory and the stability theory of parameterized discrete-time cascade interconnected systems, the designed surge and yaw state feedback laws achieve sampled-data straight-line path following control. Experimental results are also given to show the efficiency of the design method.

View full abstract

[in Japanese]

View full abstract

In this paper, we adopt reference governor control to satisfy constraints such as input saturation. When the number of facets of maximal output addmissible sets is large, then we can not apply maximal output addmissible sets in control scheme using reference governeor becasue of the real time computing time limitation. In this respect, we propose to use a inner approximation of the maximal output addmissible set. We approximate curved surfaces decided by many neighboring tangents by hyperplanes. To demonstrate the usefuleness of the proposing method simulation results are shown.

View full abstract

[in Japanese]

View full abstract

This paper describes tracking control method for constrained system. We propose a control method combining idea of 2 degree of freedom servo systems for linear systems and reference shaping method. Moreover, we take advantage that Feedforward reference signal and reference signal for feedback loops can be chosen independently. We examine this by simulation.

View full abstract

This paper proposes a novel friction compensation method for achieving higher system stability and responsiveness by actively utilizing the friction force of feed drive systems driven by a linear motor. The proposed method is applied to general feedback position control system combined with the proposed compensator. The proposed compensator replaces the real friction characteristics of feed drive mechanism by the virtual friction characteristics that provide higher damping characteristics. In the compensator design, first, the precise nonlinear friction characteristics of linear roller guides are mathematically modeled for use in the proposed compensator. Then, To evaluate the proposed friction model, step responses under various conditions were measured and simulated. As the results, it was confirmed that the proposed nonlinear friction model can express the actual behaviors. Next, the virtual friction model was designed by using the proposed friction model that provides the feed drive system with higher damping characteristic as compared with the real one. To evaluate the proposed compensation method, the positioning performances of a linear motor servo system using the proposed compensator were simulated and measured. The results confirmed that the proposed compensator corrects for the decrease in stability of the control system and enables higher servo gains, leading to high responsiveness.

View full abstract

Machining accuracy as well as consumed energy saving is an important issue in machining by multi-axis feed drive systems. Contour error, which is defined as the error component orthogonal to the desired contour curve, is a better indication of machining precision. This paper presents a novel sliding mode contouring controller with nonlinear sliding surface to improve the machining accuracy for biaxial feed drive systems and reduce the consumed energy. Unlike the conventional sliding mode control, the proposed nonlinear sliding surface depends on the output so that the damping ratio of the system changes from its initial low value to final high value as the output changes from its initial value to the set point. Hence, the proposed algorithm allows a closed-loop system to simultaneously achieve low overshoot and short settling time, resulting in small error and small energy consumption. Because contour error is more important than the tracking error with respect to each feed drive, the contour error component is included in the proposed sliding surface. Simulation result for a typical biaxial feed drive system with bounded disturbance has shown an improved performance of the proposed method in terms of a contouring error and robustness to disturbance.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

In this study, we develop a feedback control scheme based on RL for partially unknown nonlinear systems. In order to obtain a value function, we employ the stabilizing technique of polynomial systems. Stabilizing controller of our scheme is constructed by using Lyapunov function as a value function and switched to LQ controller in the estimated DOA. We apply the proposed controller to the underactuated inverted pendulum to demonstrate the usefulness of this control scheme.

View full abstract

A new method of measuring mass under weightless has been proposed. It utilizes the fixed point of a mass-spring system with a dynamic vibration absorber so that mass is estimated independently of the damping in the absorber. It is characterized by using the phase-looked loop to achieve the tuning condition. The principal of measurement is described based on a mathematical model. A measuring apparatus is designed and fabricated. Several basic characteristics are studied experimentally. The effects of damping in the primary system are studied both analytically and experimentally.

View full abstract

Since the Galileo probe discovered the binary asteroid, i.e., "Ida" and "Dactyl" in 1993, much attention has been paid on the space mission design and analysis of the spacecraft approaching to binary astroids. In this study, we will explore dynamics of a binary asteroid to derive relative equilibria in the context of the cotangent bundle reduction. Then, we will show the stability regions of the relative equilibria by using the energy-momentum method.

View full abstract

Much concern has been focused upon the development of fish robots and underwater vehicles from the viewpoint of ocean probe and environmental study. However, in order to understand the propulsion mechanism of fish swimming, it is inevitable to fully utilize geometric notions such as Lie groups and principal bundles, which have not been familiar in engineering. In this paper, we will show a mechanical model for fish swimming, which consists of three rigid bodies in a two-dimensional perfect fluid and then will show how the fish can swim by making use of the action of special Euclidean groups and with Lagrangian reduction theory. Finally, we will illustrate some numerical simulations on fish propulsion.

View full abstract

This document aims to generate the joint trajectory to hit a table tennis ball with backspin by a robot manipulator. First, the maximum joint velocity at the hit is analyzed with the limitation of the velocity. Second, the joint trajectories are generated in the joint local space between the initial, hitting, stopping and final(initial) states. The limitations of the angle and velocity are checked in the cases where the ball is hit to the 9 separated areas of the opponent's court. It is confirmed that appropriate initial state is important to satisfy the joint velocity limitation.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

In this study, we propose a trajectory generation method for a personal mobility. The proposed method takes into account ride quality of the personal mobility and velocity constraints, which are imposed around hazardous regions to reduce the risk of the autonomous locomotion. The numerical simulations and the experiments show the validity of the proposed method.

View full abstract

The mobile experience of infant stage is important, in order to encourage the development of the cognitive faculty. We are developing some vehicles, which has various functions for children with special needs. We have improved the operability and the safety, through experiments by handicapped children.

View full abstract

[in Japanese]

View full abstract

In the medical treatment welfare facilities, sometimes bed is used for patient's movement because a large physical load is necessary for patient's movement to the wheelchair from the bed. In author's laboratory, to decrease nursing load, omni-directional mobile bed with power assist system is developed. However, by reason that there is a risk of collision when in a narrow space, attendant always necessary to keep in mind the width of the bed and distance to obstacle. Such navigation can produce stress in the attendant. In this paper, to reduce the stress of nursing, collision avoidance system in power-assist system in developed.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

A personal mobilities have been focused on an interest as the new future vehicle. The paper design of seating wheeled inverted pendulum type personal mobility, and describes the effectiveness of the seating posture and the seating type is examined. The results of the numerical simulation was shown the possibility of the realization for the attitude stabilization control of the proposed personal mobility.

View full abstract

In order to achieve the logistics system in sparsely populated areas, we defined the model of the areas, and set the specifications of the transportation system. Main problems of the system are achieving high reliable system control, operation conditions under hazardous weather and robot control system including human behavior. Solving these problems, co-operated control system was developed and the control system estimates human behavior in real time. Robots and humans have same traffic rules as the Road Traffic Law of Japan and they are controlled by autonomous control. The proposed the distributed system is autonomous system that can be managed by human control. We designed the mobile robot mechanism based on the specifications of the system and tested the basic robot function.

View full abstract

Most of existing omnidirectional wheels have complex mechanism, and are not practical. Generally, they have low self positioning accuracy using internal sensor. Because, it is hard to measure a movement of free roller. On the other hand, active-caster wheels feature simple mechanism and high positioning accuracy, and they will enable the robot to improve self positioning accuracy. In this study, the control of self positioning with active-caster wheeled robot are implemented. The effectiveness of the control is evaluated through experiments.

View full abstract

Vehicles with active-caster wheels which are a kind of omnidirectional mechanism using normal pneumatic tires can move arbitrary directions immediately and can move freely from noise and vibrations. The control of vehicles with active casters, however, is not enough studied. This study aims to establish a control method of personal vehicles with active casters with respect to high safety, using independent drive and steering of active caster vehicles and dynamic model of vehicle.

View full abstract

Under today's complicated and unstable society, it is becoming of special importance to make a rational decision from comprehensive point of view. Accordingly, many interests have been paid recently on the studies on multi-objective optimization. In this paper, we have first proposed a new method for multi-objective analysis termed elite-induced MOEA that combines complementally conventional method and recent multi-objective evolutionary algorithm (MOEA). Through this simple combination, the proposed method can derive precise Pareto front and manipulate its distribution well. These prospects are verified in numerical experiments.

View full abstract

In this paper, we discuss on the efficiency improvement of reinforcement learning by introducing fixed mode states. For a long term learning problem such as the waist trajectory generation of biped robots, the learning efficiency is significantly degenerated due to the rare opportunity to get rewards because of the randomness of the action. To cope with this problem, we showed that the efficiency was improved by changing a state into a fixed mode state, if the state had learned enough. But the evaluation method of the transition strategy has not been established yet. Therefore, in this paper, we propose a practical evaluation method based on the multi start strategy, and evaluate the learning parameters to determine the optimal values.

View full abstract

[in Japanese]

View full abstract

In multiobjective optimization, a final solution is usually selected among Pareto solutions on the basis of the value judgment of decision maker. In this regard, robust design in multiobjective optimization is important for considering not only the robustness for each objective function but also that of Pareto optimality. In this report, we suggest a method for robust design in multiobjective optimization using Satificing Trade-off Method.

View full abstract

Particle Swarm Optimization was used for the facility location problem. As an example of the retail store, it was assumed that the guest gathered by the distance, simulation the customer more than a rival store was able to be acquired, and the place in which a lot of customers were gathered was able to be searched out. Moreover, simulation condition was added a river, bridge and must cross this bridge no matter what, and the place in which the customer was similarly gathered was able to be searched out.

View full abstract

Under the global and agile manufacturing, in this study, to optimize large scale logistics network design efficiently, a parallel computing algorithm has been investigated through a binary PSO developed one of the authors. In terms of the hierarchical hybrid method developed previously, its parallel implementation has been developed, and effect of topology is analyzed in particular. Finally, numerical experiments revealed some properties associated with the configuration and the efficiency.

View full abstract

[in Japanese]

View full abstract

In this paper, we are concerned with optimization problems with a symmetric-matrix-valued variable. Firstly, we introduce the eigenvalues-eigenvectors representation of the symmetric matrix which converts the problem to the one with a pair of orthogonal and diagonal matrices as the set of variables. Secondly, we obtain a representation by a multi-dimensional polar coordinates system of the orthogonal matrix in order to remove the constraints of the orthogonality and the normality. Thirdly, we show a connection rule of the angular parameters by which we can find a point, with the same value of the function, in the domain from any point outside the domain. A numerical example is provided for better understanding of the result of this paper.

View full abstract

[in Japanese]

View full abstract

Distributed Genetic Algorithm (DGA) has been proposed to solve complex optimization problems with keeping diversity among gene population. This study proposes to introduce complex network topology with sociality instead of lattice network topology into neighborhood model with aiming to take balance between exploitation and exploration. The feasibility of the proposal is confirmed by computer simulation results.

View full abstract

This note proposes a method to measure body weight continuously for health monitoring. For elderly care, the precise continuous monitoring of their body weight is crucial to diagnose their health. We install four sensors at the bottoms of bed's legs and acquire he total weight (human body weight + bed weight) continuously. The note discusses a data processing method to retrieve the momentarily change of the body weight.

View full abstract

This note proposes a method to detect falling detection of elderlies by vibration sensors and sound sensors. The sensors are installed at the bottoms of bed legs, which detect vibration or sound caused by falling accidents. The propagation delay between sound and vibration is proportion to the distance from the place of the accidents. The magnitude of the sensors' signal is compensated by the distance. The system detects falling accidents based on the compensated signals.

View full abstract

A biosignal recorder has been developed for assessment of daily lifestyle. The target biosignals are forehead electroencephalogram, electrocardiogram and body acceleration during daily lives, because these signals enable us to estimate one's sleep depth and physical activities in ordinary days, not in extraordinary circumstances as in polysomnography. The size of the device (45mm x 25mm x 65mm, 62.5g) is appropriate for ambulatory recoding. An experimental 4-day measurement shows the availability of the developed device for activity estimation.

View full abstract

[in Japanese]

View full abstract