Periodic motions are often desired in engineering applications. The traditional method of trajectory planning followed by feedback tracking may work well for maintaining the planned motion. However, it is desired for certain applications to achieve the motion as autonomous rather than forced oscillations so as to adapt the motion profile in response to disturbance and changing environment. This motivates us to consider feedback control design to embed a limit cycle oscillation in the closed-loop dynamics with the notion of orbital stability. Here we provide a tutorial review of basic concepts and analysis tools for limit cycles arising from autonomous nonlinear dynamics. The Floquet theory and contraction analysis are reviewed, simple planar oscillators are illustrated, and synchronization and coordination of coupled oscillators are discussed.
Rendezvous orbital maneuvers are planned operations which intend to make two spacecraft meet, while avoiding collisions. Aspects such as trajectory safety and robustness, as well as obstacle avoidance, are fundamental for the mission success. The aim of this paper is to provide an overview of some recent algorithms for guidance and control systems, and also their combined exploitation, which can provide significant enhancements. This overview focuses on those algorithms that are characterized by key selected features: the discussed schemes guarantee low computational effort, low fuel consumption, and high safety and robustness. Different experimental setups, which can be used to evaluate the performance of the considered algorithms, are also taken into account and presented.
In the last decade, diversification of communication systems is increasing with the information technology development, and our human relationships are getting more and more complex. Sometimes, we need helps to overcome a difficult situation that we must face in such complicated relationships. Psychodrama is one of the effective counseling methods which are used for finding hints to solve human relationship problems. Despite many advantages of psychodrama have been reported, some weak points also have been pointed out. The first one is that, before conducting a psychodrama event, the participants have to adjust their schedules. The second one is that, some participants may hesitate to talk their problems of human relationships in a public space. These factors make it difficult to conduct psychodrama events easily and comfortably. In this paper, a self psychological therapy tool is proposed as an approach for the problems. The proposed tool aims to provide similar effects with creating social network diagram, which is a basic technique of psychodrama. Experiments have been conducted in order to evaluate the proposed tool. The results show that the proposed tool can provide some hints for the users who had troubles in current human relationships and want to change the situations.
To increase the accuracy of real-time sleep stage estimation when only a small number of sleep data can be obtained such as after going to bed, the information for the subject's own sleep and other's past sleep can be used. However, the types of other sleep data such as the subject's own sleep data or other's sleep data, which can contribute to the increase of the sleep stage estimation accuracy, are unknown. Therefore, this paper focuses on these two types of sleep data (i.e., the subject's own sleep data and other's sleep data) and aims to investigate the usefulness of these data for increasing the sleep stage estimation accuracy. Using human subject experiments, the following conclusions have been revealed: (1) the accuracy of the sleep stage estimation is improved using the similarity of either the subject's own past sleep or others' sleep and (2) the accuracy of the estimation method using the sleep data of six other people is higher than those obtained using the data for none, one, or two people.
This paper reports our experimental results on analyzing a human's goal finding process in continuous learning. The objective of this research is to make clear the mechanism of continuous learning. To fill in the missing piece of reinforcement learning framework for a learning robot, we focus on two human mental learning processes, awareness as pre-learning process and reflection as post-learning process. To observe mental learning processes of a human, we propose a new method for visualizing them by the reflection subtask for human to be aware of the goal finding process in continuous learning with invisible mazes. Two-layered task is introduced. The first layer is the main task of continuous learning designing the environmental mastery task to accomplish the goal for any environment. The second layer is the reflection subtask to make clear the goal finding process in continuous learning. The reflection cost is evaluated to analyze it.
The popularization of video-viewing systems enables both adults and children to continuously watch countless video clips. However, such extended video viewing might cause health problems, particularly for children. Rule-making tendencies are weaker for video-viewing systems than they are for television programming. Children struggle with self-motivated video-viewing stops because of attractive, rich clips. In this study, we propose a video recommendation system that arranges video clips based on a pre-set time, to support self-motivated video-viewing stops. Our proposed system enables parents to limit video-viewing time in advance and provides video clips that are arranged to finish exactly at a pre-set time. We conducted two experiments. The first experiment targeted adults to confirm the effectiveness of our approach. Our results from the first experiment suggest that our proposed system creates a good video-viewing ending time and could support self-motivated video-viewing stops. Our second experiment targeted children and aimed to obtain children's reactions to the use of our developed system, as well as feedback from their parents. We observed the children's self-motivated video-viewing stops. In addition, we obtained good impressions of the system from parents.
In this paper, we develop a text chat system to support graphical interaction for on-line group discussion and investigate whether our graphical text chat system increases a speaker's feeling of conveying and a listener's feeling of understanding in on-line group discussion. For on-line communication, in particular at group discussion, text chat is a very popular and widely used communication tool. Not only text messages but also graphical ones are available in recent chat systems, and many graphical text chat systems exist. However, whether and how such graphical text chat systems contribute to both feelings of conveying and understanding between users remains unknown. In this study, we developed a graphical text chat system that enables users to exchange interpretations of the concepts contained in each other's text messages as images. Our experimental results demonstrate that the developed system increased both conveying and understanding feelings between users from both the speaker and listener viewpoints.
This paper addresses the simultaneous design of observer-based robust scaled-H∞ controllers and scaling matrices for discrete-time linear time-invariant polytopic systems. Based on the so-called dilated linear matrix inequality approach, we propose a design method for our problem with a structural constraint for the decision matrix which is introduced in “dilation” procedure. The structural constraint inevitably introduces conservatism; however, numerical examples illustrate the effectiveness of the simultaneous design of controller matrices, i.e. observer gain, state-feedback gain, the state transition matrix of the observer, and the input or the output matrix of the observer, as well as the scaling matrix.
This paper provides a component-based modeling of a small wheeled mobile robot with indispensable nonlinearities. We obtain a model of each fundamental component for a wheeled mobile robot, and then we provide the whole robot model by integrating the resultant component models. First, we model the DC motor considering the brush voltage drop and sink current effects. Second, we model the torque loss of the gearbox as a function of the angular velocity. Third, the rolling resistance and cornering resistance models are provided. Particularly, in measuring the cornering resistance, we use the computer vision with image processing technique. Then, we derive the cornering resistance model by numerical optimization. Moreover, we develop a differential wheeled robot as an experimental test-bed, and demonstrate the validity of each component model and the whole robot model by comparing with the experimental behaviors of the test-bed.
This study proposes a new approach for parameter estimation of random-phase varying multiple periodic signals. A phase-varying multiple signal is typically observed in digital communication signals wherein the phase of a signal is discretely varied due to modulation. Additionally, the study investigates the performance of the proposed method with respect to signals in which the phase varies continuously in addition to discrete phase varying. The proposed method enables blind and joint estimation of the period, the number of sources (NOS), and the fundamental waveform for each of the multiple signals. The performance of the method is verified by using a computer simulation.