In this paper, the levitation control method of a MIMO Magnetic Levitation (Maglev) transportation system with 3-DOF is presented. Fluctuations of magnetic poles cause the Maglev system to become critically unstable. We propose a design method of the MIMO Maglev controller based on SISO Maglev control technology to correct the suspension and compensate for the effect of rotational motions. In addition, a force loop controller is designed for placement in front of each sub-controller of an electromagnet for stability improvement. The proposed control method is evaluated using simulations and real experiments using the developed Maglev transportation system.
This paper clarifies the relations in properties and structures between fuzzy inference methods based on α-cuts and the generalized mean. The group of the inference methods is named the α-GEM (α-cut and generalized-mean-based inference) family. A unified platform is proposed for the inference methods in the α-GEM family by the effective use of the above-mentioned relations. For the unified platform, a criterion is made clear to uniquely determine the value of a parameter in fuzzy-constraint propagation control for facts given by singletons. Moreover, conditions are derived to make the inference methods in the α-GEM family equivalent to singleton-consequent-type fuzzy inference which has been successfully applied to a wide variety of fields. Thereby, the unified platform can contribute to the construction of an inference engine for both the methods in the α-GEM family and singleton-consequent-type fuzzy inference. Such scheme of the inference engine provides an effective way to make these inference methods transformed into each other in learning for selecting the inference methods as well as for optimizing fuzzy rules.
We have been constructing a swimming ability improvement support system. One of the issues to be addressed is the automatic classification of swimming styles (backstroke, breaststroke, butterfly, and front crawl). The mainstream swimming style classification technique of conventional researches is based on non-ensemble learning; in their classification, breaststroke and butterfly are mixed up with each other. To improve its generalization performance, we need to use better classifiers and more adaptive feature values than previously considered. Therefore, this research has introduced (1) random forest technique, one of ensemble learning techniques, and (2) feature values specific to breaststroke and butterfly to construct a four-swimming-style classifier that has resolved this issue. From subjects with 7 to 20 years history of swimming races, we have obtained their sensor data during swimming and have divided the data into learning data and test data. We have also converted them into feature values that represent their body motions. We have selected from those body-motion-representing feature values the important data to classify four swimming styles and feature values specific to breaststroke and butterfly. We have used the learning data to construct a swimming style classifier, and the test data to evaluate its classification accuracy. The evaluation results show that (1’) the introduction of ensemble learning has improved the mean value of F-measure for breaststroke and butterfly by 0.053, and (2’) the introduction of feature values specific to breaststroke and butterfly has improved the mean value of F-measure for breaststroke and butterfly by 0.121 as compared with (1’). The proposed swimming style classifier has performed a mean F-measure of 0.981 for the four swimming styles as well as good classification accuracies for front crawl and backstroke. Therefore, we have concluded that the swimming style classifier we have constructed has resolved the problem of mixing up breaststroke and butterfly, as well as can properly classify all different swimming styles.
In this paper, we propose a method called Convolutional Neural Network-Markov Random Field (CNN-MRF) to estimate the crowd count in a still image. We first divide the dense crowd visible image into overlapping patches and then use a deep convolutional neural network to extract features from each patch image, followed by a fully connected neural network to regress the local patch crowd count. Since the local patches have overlapping portions, the crowd count of the adjacent patches has a high correlation. We use this correlation and the Markov random field to smooth the counting results of the local patches. Experiments show that our approach significantly outperforms the state-of-the-art methods on UCF and Shanghaitech crowd counting datasets.
This study presents an aerial robotic application of deep reinforcement learning that imparts an asynchronous learning framework and trust region policy optimization to a simulated quad-rotor helicopter (quadcopter) environment. In particular, we optimized a control policy asynchronously through interaction with concurrent instances of the environment. The control system was benchmarked and extended with examples to tackle continuous state-action tasks for the quadcoptor: hovering control and balancing an inverted pole. Performing these maneuvers required continuous actions for sensitive control of small acceleration changes of the quadcoptor, thereby maximizing the scalar reward of the defined tasks. The simulation results demonstrated an enhancement of the learning speed and reliability for the tasks.
Prediction of rising stars has become a core issue in data mining and social networks. Prediction of rising venues could unveil rapidly emerging research venues in citation network. The aim of this research is to predict the rising venues. First, we presented five effective prediction features along with their mathematical formulations for extracting rising venues. The underlying features are composed by incorporating the citation count, publications, cited to and cited by information at venue level. For prediction purpose, we employ four machine learning algorithms including Bayesian Network, Support Vector Machine, Multilayer Perceptron and Random Forest. Experimental results demonstrate that proposed features set are effective for rising venues prediction. Our empirical analysis spotlights the rising venues that demonstrate the continuous improvement over time and finally become the leading scientific venues.
Six years have passed since the publication of our Special Issue on Human-Robot Interaction Systems in 2011. Since then, artificial intelligence and robotics have developed rapidly, and the opportunities for human beings and robots to work together have increased. The objective of this special issue’s twelve articles is to activate and expand high-quality research.
In the first article, Y. Tamura, T. Akashi, and H. Osumi propose a computational model of robot’s gaze. In the second article, S. Hoshino and K. Uchida propose an interactive motion planner for robot navigation in dynamic environments. In the third article, T. Iio, Y. Yoshikawa, and H. Ishiguro develop a conversational robotic system based on human response. In the fourth article, K. Sakai, F. Dalla Libera, Y. Yoshikawa, and H. Ishiguro propose a method for generating bystander robots’ actions that is based on an analysis of the relative probabilities of human responses to robot actions. In the fifth article, T. Matsumaru and M. Narita present a newly developed support system for learning calligraphy strokes. In the sixth article, E. Tamura, Y. Yamashita, T. Yamashita, E. Sato-Shimokawara, and T. Yamaguchi present a method of driving a car simply by gesturing. In the seventh article, A. Kurosu and T. Hashimoto develop an eye robot with two degrees of freedom. It is intended for use as a communication robot. In the eighth article, T. Hashimoto, Y. Munakata, R. Yamanaka, and A. Kurosu report on a method for retrieving episodic memories. In the ninth article, Y. Nishikawa, Y. Kagawa, and A. Okazaki develop a spiral movement robot for inpatients. In the tenth article, Y. Umesawa, K. Doi, and H. Fujimoto develop an interface device that creates kinaesthetic illusions by inducing vibrations in muscle tendons, vibrations that coordinate with dual-joint movements. In the eleventh article, R. Horio, N. Uchiyama, and S. Sano propose a human-operated biped robot for transporting objects over rough terrain or up steps. In the closing contribution, T. Sakuraba, N. Uchiyama, S. Sano, and T. Sakaguchi present the design of a spring-based regenerative brake, and they verify its effectiveness by driving a system that uses it.
We thank the referees for their comprehensive reviews and the staff members of Fuji Technology Press, Ltd. for their encouragement and advice.
For a robot to smoothly interact with humans, it has to possess the capability to manipulate human attention to a certain degree. In this study, we start with a hypothesis that humans cannot correctly perceive what a robot is looking at. To examine the hypothesis, an experiment, which focuses on the relationship between a robot’s geometrical gaze point and the gaze point perceived by a human, was conducted. The results of the experiment supported the hypothesis. Based on the results, we propose a computational model that calculates where robots are to look in order to guide human’s attention to the desired area. The validity of the proposed model was demonstrated by cross validation.
In dynamic environments, taking static and moving obstacles into consideration in motion planning for mobile robot navigation is a technical issue. In this paper, we use a single mobile robot, for which humans are moving obstacles. Since moving humans sometimes get in the way of the robot, it must avoid collisions with them. Furthermore, if a part of the environment is crowded with humans, it is better for the robot to detour around the congested area. For this navigational challenge, we focus on the interaction between humans and the robot, so this paper proposes a motion planner for successfully getting through the human-robot interaction. The interactive motion planner is based on the hybrid use of global and local path planners. Furthermore, the local path planner is executed repetitively during the navigation. Through the human-robot interaction, the robot is enabled not only to avoid the collisions with humans but also to detour around congested areas. The emergence of this movement is the main contribution of this paper. We discuss the simulation results in terms of the effectiveness of the proposed motion planner for robot navigation in dynamic environments that include humans.
In human-robot conversation in a real environment, low speech recognition and unnatural response generation are critical issues. Most autonomous conversational robotic systems avoid these issues by restricting user input and robot responses. However, such restrictions often render the interaction boring because the conversation becomes predictable. In this study, we propose the use of multiple robots as a solution for this problem. To explore the effect of multiple robots on a conversation, we developed an autonomous conversational robotic system and conducted a field trial in a real event. Our system adopted a button interface, which restricted user input within positive or negative intention, and maintained a conversation by choosing the most suitable of the prepared static scenarios. Through the field trial, we found that visitors who conversed with multiple robots continued their conversation for a more prolonged period, and their experience improved their impression on the conversation, in contrast to the visitors who conversed with a single robot.
This paper describes a method of rule extraction for generation of appropriate actions by a robot in a multiparty conversation based on the relative probability of human actions in a similar situation. The proposed method was applied to a dataset collected from multiparty interactions between two robots and one human subject who took on the role of supporting one robot. By computing the relative occurrence probabilities of human actions after the execution of the robots’ actions, twenty rules describing human behavior in such a role were identified. To evaluate the rules, the human role was filled by a new bystander robot and other subjects were asked to report their impressions of video clips in which the bystander robot acted or did not act in accordance with the rules. The reported impressions and a quantitative analysis of the rules suggest that the behavior of listening and the supporting role that the subjects play can be reproduced by a bystander robot acting in accordance with the rules identified by the proposed method.
This paper presents a newly developed calligraphy-stroke learning support system. The system incorporates the following functions: a) Displaying brushwork, trajectory, and handwriting; b) recording and playback of an expert’s calligraphy-stroke; and c) teaching a learner a calligraphy-stroke. The following features of the system demonstrate the contributions of our study. (1) The system, which consists of a sensor and projector, is simple and compact, so as to be easily introduced to the usual educational fields and practical leaning situations. (2) Three-dimensional calligraphy-stroke is instructed by presenting two-dimensional visual information. (3) A trajectory region is generated in the form of continuous squares, calculated using a brush model based on the brush position information measured by a sensor. (4) Handwriting is expressed by mapping a handwriting texture image according to ink concentration and the brush handling state. The results of the trial experiment suggest the effectiveness of the learning support function in terms of letter form and calligraphy-stroke.
Finger pointing is an intuitive method for people to direct a robot to move to a certain location. We propose a system that enables the movement operation of a mobility robot by using finger-pointing gestures for an automatic and intuitive driving experience. We employ a method to recognize gestures by using video images from a USB camera mounted on a wearable device. Our method does not require the use of infrared sensors. Three movement commands for forward motion, turning, and stopping are chosen based on gesture recognition, face orientation detection, and an intelligent safety system. We experimentally demonstrate the usefulness of the system using a scooter-type mobility robot.
The crucial role of nonverbal communication by gaze in social interactions has been highlighted. In this study, an eye robot having two degrees-of-freedom was developed as a communication robot, and the motion of the sight line of this robot in a standby state with an absence of communication with people was investigated. We compared and evaluated the impression provided by a robot and a person, who imitated the action of the robot, to observers and demonstrated that both the robot and person provided improved impression if the line of sight is stationary.
Developments in robotics have advanced the development of robots that can communicate with humans. However, a few robots are only capable of stereotypic responses, and this leads to barriers in smooth communication between humans and robots. In this study, in order to represent mood congruence effects, an episodic memory retrieval method is proposed based on a robot’s emotional values that represent its internal state. In the study, impression evaluation experiments were conducted to prove the effectiveness of the proposed method.
It is necessary to reduce the stress level among patients hospitalized in pediatric wards since elevated stress levels can lead to treatment rejection or unnecessary nurse calls, which can increase the nursing workload. In particular, the cribs used by pediatric patients are surrounded by a tall rail to prevent falls, which can increase the patient’s sense of isolation or anxiety. In this study, we developed a robot designed to operate in these cribs, with the objective of reducing the sense of isolation or anxiety experienced by hospitalized children. The developed robot is a spiral-movement robot, capable of moving up and down the rail pole as well as moving from pole to pole. Experiments were carried out, and they confirmed the robot’s mobile capabilities.
If kinesthetic sensation can be generated using artificial means, we can experience dynamic sensations in the virtual reality space. Subsequently, it can be used as an instruction tool for rehabilitation. By means of kinesthetic illusion, it is possible to create kinesthetic sensation. In this study, we developed an interface device that creates kinesthetic illusions by inducing vibrations in muscle tendons that coordinate dual joint movements. First, we produced a vibrating device using four vibrators. The rotation of motors moving eccentric weights generated the vibrations. Each motor was independently controlled using specially developed software. Second, we produced vibrator fixation structures, which firmly attached the vibrators to the muscle tendons. Using these structures, the vibrators were maintained in position and allowed to transmit forces to the muscle tendons. Furthermore, we conducted an experiment to evaluate the performance of the kinesthetic illusion device. Accordingly, we created the kinesthetic illusion of drawing figures on a horizontal surface by inducing vibrations in muscle tendons that contribute to dual joint movements. The results demonstrated that, by using this device, it was possible to induce kinesthetic illusions of dual joint movements.
The recognition of a robot operator’s intention/command is important in human-robot collaboration systems. This paper presents a novel approach to estimating the human operator’s force applied to a robotic system. In our previous study, we proposed a human-operated biped robot for transporting objects on rough terrain, steps or stairs. In this paper, we consider a new control system for the proposed robot, which enables the estimation of the support force applied by a human operator. The dynamics of the proposed robot are modeled by assuming that a support force applied by an operator is considered as a disturbance to each joint. The observer was designed to estimate the disturbance based on a high-gain observer; it was proven that the observer could estimate the disturbance with sufficient accuracy. Simulation results show that the observer successfully estimated the support force as a disturbance even though the disturbance property was completely unknown. In this study, the proposed biped robot system with the observer was expected to provide support to human operators for the cooperative transportation of objects up the stairs.
In human-operated mechanical systems such as automobiles, electric bicycles, and electric wheelchairs, energy saving is an important criterion. Hybrid systems consisting of combustion engines and electric motors have found successful applications in automobiles. However, it is difficult to apply this type of hybrid system to personal mobilities and industrial machines in a factory, as there is a requirement to reduce their energy consumption owing to recent environmental and energy resource problems. Therefore, a previous study has focused on the use of a mechanical spring as a regenerative brake in a hybrid bicycle. This study, however, presents a new type of hybrid system that includes the use of a mechanical spring. An experimental wheeled mobile system is designed, and its effectiveness is confirmed through comparative experiments in which a reduction of more than 30% in the consumed energy is achieved in acceleration periods as compared to a conventional system.