Showing papers presented at "Robot and Human Interactive Communication in 2000"

Effect of sensor fusion for recognition of emotional states using voice, face image and thermal image of face

Abstract: A new integrated method is presented to recognize the emotional expressions of human using both voices and facial expressions. For voices, we use such prosodic parameters as pitch signals, energy, and their derivatives, which are trained by hidden Markov model for recognition. For facial expressions, we use feature parameters from thermal images in addition to visible images, which are trained by neural networks for recognition. The thermal images are observed by infrared ray which is not influenced by lighting conditions. The total recognition rates show better performance than that obtained from each single experiment. The results are compared with the recognition by human questionnaire.

Fuzzy logic implementation for photovoltaic maximum power tracking

Abstract: The paper presents a simulation study and an experimental implementation of a fuzzy logic controller (FLC) for Cuk converter in a stand alone photovoltaic (PV) energy scheme. DC-DC converters are used to convert the unregulated DC input into a regulated DC output at a desired voltage level. A FL algorithm is selected and used to control the PV nonlinear system. The control objective for Cuk converter is to move the operating point of the PV system to its peak power point. A system that consists of a PV generator, converter, AC PWM inverter and load models is simulated, analyzed, and experimentally implemented. The simulated system with FLC is investigated at different solar insolation levels. The fixed structure FLC shows a robust performance when applied on a wide operating range of the proposed converter and AC PWM inverter.

A tactile display using surface acoustic wave

Abstract: We propose a novel method to provide human tactile sensation using surface acoustic wave (SAW). A pulse modulated driving voltage excites temporal distribution of shear force or friction shift on the surface of SAW substrates. The force/friction distribution can be perceived as tactile sensation at mechanoreceptors in the finger skin. A first prototype could express roughness successfully.

User centered design for intelligent service robots

Abstract: This paper describes the development of a fetch-and-carry robot to assist physically impaired people in an office environment. Different methods involving users are employed in the project, including the task analysis, Hi-Fi simulation trials and focus group sessions. Through an iterative design process, a prototype robot system has been developed, with an enhanced robot platform including a graphical user interface and natural language interface. The users' need for continuous feedback from the robot has led to the development of an animated character (CERO), which relates the two interface components and indicates the robot's current state by using simple gestures.

Robotic analyses of output force distribution developed by human limbs

Abstract: We propose an introduction of a muscle coordinate system consisting of the bi-articular muscles as well as the mono-articular muscles instead of a conventional joint coordinate system consisting of only the mono-articular muscles without the bi-articular muscles to discuss an interface between man and machine. The output force distribution developed at the wrist joint of the human arm under isometric and maximal effort conditions demonstrates a hexagonal shape. Since complex muscle arrangements of the human arm could be simplified into one antagonistic pair of the bi-articular muscles and two antagonistic pairs of the mono-articular muscles (functionally effective muscular strengths), a robot arm provided with three pairs of muscles was built. The robot arm operated with the muscle coordinate system demonstrated hexagonal shape output force distributions. Whereas, the robot arm operated with the joint coordinate system showed quadrangular output force distributions. A characteristic of the hexagonal force distribution made it possible to evaluate individual functionally effective muscular strengths from the human arm output force distribution. The results obtained here will lead a new virtual human model to be more realistic than the conventional model operated with the joint coordinate system. The idea will also be useful for a human-friendly robot.

Human-robot interface using agents communicating in an XML-based markup language

Abstract: The paper concerns an agent-based human-robot interface via the Internet. A user client and an embedded software are viewed as agents with limited computational and communication resources. To facilitate the communication between the real-time embedded agents and the user interface agents via the communication channel of uncertain quality the proxy agent is proposed as a mediator. The functions assigned to the proxy agent target reduction of inter-agent communication load and minimization of computational resources taken by the embedded agents and user interface agents for communication-related tasks. An XML-based language, RoboML, is designed to serve as a common language for robot programming, agent communication and knowledge representation. The human-robot interface software prototype is developed for an autonomous guided vehicle to evaluate the proposed techniques.

An epigenetic approach to human-robot communication

Abstract: This paper proposes a developmental approach to social intelligence, especially communication ability, for robots and other artificial systems. Any social being has to have two essential features: naturalistic embodiment, i.e., having a body similar to others; and socio-cultural situatedness, i.e., being able to communicate with others and to participate in the social activity. However, we still have an open question: how does the body become situated in the social environment? Our answer is epigenesis, where (1) we create a humanoid with minimum innate abilities, namely a primordial form of joint attention and indirect experience, then (2) through the attentional and imitative interaction with human caregivers, the humanoid autonomously explores how to interact socially with people. As an epigenetic embodiment, the authors are building an upper-torso humanoid, Infanoid, which is to acquire social situatedness in the human community.

Assessment of mental fatigue during VDT task using event-related potential (P300)

Abstract: The study was designed to measure and evaluate mental fatigue induced during VDT tasks using the P300 component of the event-related potential. The experimental task induced mentally and physically fatigued states to each subject from the viewpoints of performance decrement and subjective feeling of fatigue. In accordance with this result, the P300 amplitude tended to decrease after the experimental task. The P300 latency increased after the experimental task. The results indicated that the appearance of fatigue was reflected in the two P300 components, that is, the P300 amplitude and the P300 latency. The prolonged P300 latency that indicated the delayed cognitive information processing and the decreased P300 amplitude that was related to the decreased activity of cognitive information processing were found to be effective measures of mental fatigue.

Interactive classifier system for real robot learning

Abstract: We describe a fast learning method for a mobile robot which acquires autonomous behaviors from interaction between a human and a robot. We develop a behavior learning method ICS (interactive classifier system) using evolutionary computation and a mobile robot is able to quickly learn rules so that a human operator can directly teach a physical robot. Also the ICS is a novel evolutionary robotics approach, using an adaptive classifier system, to environmental changes. The ICS has two major characteristics for evolutionary robotics. For one thing, it can speedup learning by means of generating initial individuals from human-robot interaction. For another, it is a kind of incremental learning method which adds new acquired rules to priori knowledge by teaching from human-robot interaction at any time.

Proto-conversations with an anthropomorphic robot

Abstract: We present proto-dialog skills of an anthropomorphic robot, Kismet. The implementation is modeled after the proto-conversational skills of human infants. We have supplemented Kismet's vocal turn-taking skills with other para-linguistic social cues. We have found that naive subjects intuitively read and use these cues to regulate the exchange, making it smoother over time.

Genetic symbiosis algorithm for multiobjective optimization problem

Abstract: Evolutionary algorithms are often well-suited for optimization problems. Since the mid-1980's, interest in multiobjective problems has been expanding rapidly. Various evolutionary algorithms have been developed which are capable of searching for multiple solutions concurrently in a single run. In this paper, we proposed a genetic symbiosis algorithm (GSA) for multi-object optimization problems (MOP) based on the symbiotic concept found widely in ecosystem. In the proposed GSA for MOP, a set of symbiotic parameters are introduced to modify the fitness of individuals used for reproduction so as to obtain a variety of Pareto solutions corresponding to user's demands. The symbiotic parameters are trained by minimizing a user defined criterion function. Several numerical simulations are carried out to demonstrate the effectiveness of proposed GSA.

Learning performance of neuron model based on quantum superposition

Abstract: Concerns the use of quantum computer methods to develop a distributed and strongly connectionist system that achieves parallel and fast information processing. We have proposed a qubit-like neuron model based on quantum mechanics and constructed the quantum backpropagation learning rule (QBP). In this paper, we show our improved QBP neural network model and discuss its performance on solving the 4 bit parity check problem, the function and the gray-scale pattern identification problem. Then, we find our model is more excellent than the conventional one in information processing efficiency.

Dynamical analysis and suppression of human hunting in the excavator operation

Abstract: The paper deals with human hunting phenomena, which is one of the coupled vibration phenomena between the vehicle and the operator on board, appearing at the operation of the small-sized excavator and degrades operativity. This phenomena is related not only to the vehicle dynamical characteristic but also to human-operator's dynamical characteristic. To analyze this phenomenon, it is necessary to consider the vehicle-operator dynamic model. We first derived a mathematical model, which has six degrees of freedom to analyze this phenomenon. Up and down motion of the excavator-boom is simulated, and it agrees with the phenomenon well. The linearized mathematical model shows that some control lever parameters are related to a stability criterion. By simulation, it is confirmed that the proper selected control parameters can suppress the human hunting phenomena.

A wearable robotic arm with high force-reflection capability

Abstract: Many exoskeletal robotic arms have been developed for teleoperation having force reflection. They can measure an operator's arm motion and apply reactive force to the operator as well. Previous research has emphasized control performance of motion tracking and force reflection mainly but ignored how comfortable an operator feels, that is, issues of human wearability. Most robotic arms are heavy and give an operator excess fatigue during the teleoperation. We propose a robotic arm that satisfies high wearability and high force-reflection simultaneously. In order to accomplish these two objectives, the proposed arm has a parallel mechanism, one for a forearm and the other for a brachium. The parallel mechanism has an analogy to human muscular structure in which many extensors and flexors interact with each other and generate torque. The prismatic joints in the parallel mechanism act as virtual human muscles and determine joint torque by contraction and relaxation. The configuration of the prismatic joints enhances human wearability because all the joints are placed on the circumference of cylinders. Consequently, this circular configuration distributes internal force over an arm and thus contributes to reduce human fatigue during operation. In addition, the robotic arm does not make any singularity in its kinematics. The kinematic parameters in the robotic arm are selected to maximize the operator's workspace while considering performance of force reflection.

Physical interaction and multi-aspect representation for information intensive environments

Abstract: Embedded technology and networking allow for the integration of physical objects with information infrastructure and augmentation of the physical environment with information and computing functions. The concept of intelligent artifacts associated with robots will be restructured to many different forms of intelligent and interactive artifacts and environments that range from simple objects with a tagging chip to complex physical systems with local intelligence and actuators connected to a large-scale information infrastructure. This opens a new problem space of interactive systems design: how to control remotely connected objects, how to interact with intelligent environments, and how to combine physical and media entities. This paper explores the issues of interactivity in physical space and media space, models of physical interaction, system architecture and the design methodology from the viewpoint of human-centered system design.

InterRobot: a speech driven embodied interaction robot

Abstract: The concept of a speech driven embodied interaction robots system is introduced for supporting human interaction by generating the expressive actions or motions of robots coherently related to speech input. Then, the prototype of the system which consists of 2 interaction robots called InterRobots with both functions of speaker and listener is developed in which two models are incorporated: one is a listener's action model in which nodding, blinking and the motions of head, arms and body are estimated by the hierarchical moving average (MA) model of the burst-pause (ON-OFF) of speech to the nodding; the other is a speaker's action model in which the motions of head, arms and body are estimated by its own MA model of the ON-OFF of speech to the head motion. By the sensory evaluation and behavioral analysis in human-robot interaction, the effectiveness of InterRobot's interaction is demonstrated. The effectiveness of the embodied interaction robots system is also confirmed in remote communication.

Control properties of a two-joint link mechanism equipped with mono- and bi-articular actuators

Abstract: When a two-joint link mechanism was installed with an antagonistic pair of bi-articular actuators in addition to antagonistic pairs of mono-articular actuators the two-joint link mechanism could demonstrate perfect human-like control properties at the endpoint. First, only a single command signal informing the target point could control any desired output force direction at the endpoint via a simple spinal level neural network. Second the existence of three antagonistic pairs of actuators made it possible to control three parameters of a viscoelastic ellipse exerted at the endpoint independently, and consequently to induce coincidence of trajectory and output force exerted at the endpoint. Thus, the mechanism could lead smooth, rapid and precise movements without use of positional feedback signal from the endpoint, i.e. open loop control.

Integration model of learning mechanism and dialogue strategy based on stochastic experience representation using Bayesian network

Abstract: We propose a method for personal robots to acquire autonomous behaviors gradually based on interaction between human and robots. In this method, behavior decision models and dialogue control models are integrated using Bayesian networks. This model can treat interaction experiences using statistical processes, and sureness of decision making is represented by certainty factors using stochastic reasoning. The robots not only decide behavior, but also make suggestions to and ask questions of the user using the certainty factors. Consequently, the certainty factors enable the behavior acquisition to be more effective. We investigate the feasibility of this method with obstacle avoidance tasks for mobile robots. Through experiments on a real mobile robot, we have confirmed that the mobile robot acquires robust behavior decision models against changes of environment and uncertainties of sensors, with only a few teaching and learning sessions.

Task constraints on robot realism: the case of talking heads

Abstract: Sophisticated "talking head" systems have been developed and applied to a broad range of tasks. These tasks generally depend on human interaction, which may be influenced by the type and degree of realism portrayed by the robotic system. The thesis of this paper is that achieving ideal robot realism is task dependent and cannot be equated with maximal realism of a single kind. We discuss this issue in terms of our experience designing talking head animation systems for investigation of the production and perception of multimodal (auditory and visual) speech.

EventScope: amplifying human knowledge and experience via intelligent robotic systems and information interaction

Abstract: The EventScope program develops publicly accessible "reality browsers" that display both archived and updating representations of remote environments derived from on-site robotic sensors. The interface encourages collaborative work within a community of users. Public exploration of real remote sites presents a variety of interface issues addressed by EventScope, including time delay, public exploration via a single robot and communication between geographically separate users from diverse backgrounds. Merging public interface with educational and contextual information extends the notion of "interface" to "remote reality library." EventScope is a NASA and private foundation funded project based at Carnegie Mellon University.

Information filtering for emergency management

Abstract: Information filtering is a function to select useful or interesting information for the user among a large amount of information. This function has become very important as network technology develops rapidly. This paper focuses on information filtering for emergency management. We believe that special purpose "information filtering system" should be developed to cope with the problem of information flood in emergency management. In this paper, we first investigate the features of information in emergency situations. Then we propose an information filtering system for emergency management, and evaluate it by computer simulation.

Des fictions créatrices : les identités nationales

Abstract: L'avenement des nations en Europe, a partir de la fin du XVIIIe siecle, correspond a des mutations politiques et economiques. Mais la formation des nations a ete preparee par la creation culturelle des identites nationales, references construites pour devenir support d'une conscience collective. Ecrivains, artistes, erudits ont elabore les patrimoines culturels et symboliques nationaux, selon un modele transnational. C'est a travers des echanges constants d'idees et de savoir-faire qu'ont ete forges les principaux elements des identites nationales (langue, historiographie, paysage, monuments historiques, folklore). the constant exchange of ideas and know-how that the main elements of national identity lan guage historiography landscape historic monuments and folklore were forged

Ergonomics and cognitive engineering for robot-human cooperation

Abstract: Describes the importance of ergonomics and cognitive engineering for robot-human cooperation. First, the concepts of ergonomics and cognitive engineering are briefly explained. Together, the advantages and disadvantages of humans and machines (robots) are listed to compare features between humans and machines and discriminate humans from machines (robots). The transition of robot-human cooperation from the coexistence type to the cooperation type with advanced automation technology and the elaboration of intelligent machines or robots is also pointed out. On the basis of the discussion; the problems in the case of designing robot-human cooperation systems are pointed out and identified from the viewpoints of ergonomics and human factors. A few examples of results from the ergonomic and cognitive engineering research are briefly described.

Localization and approaching to the human by mobile home robot

Abstract: This paper introduces a robotic system that uses a PTZ camera to find the human in the room and locate its position. Then the robot is guided to go near to the human. This includes human detection and localization, robot localization and robot navigation. Human's head and face is detected by using a combination of motion detection, shape analysis and color information. The 3D position of human's head and face is estimated by using the depth from focus method. For robot localization we use a very fast measuring system that utilizes ultrasonic and infrared signals simultaneously. For robot navigation we use a map of environment consisting of doors, walls and static objects. After locating human and robot a visibility graph is searched for the shortest path and obstacle avoidance is done in a reflexive method using ultrasonic sensors.

Trajectory planning for hand-over between human and robot

Abstract: In the future, there will be many cases where robots should perform cooperative tasks with humans in daily life. We focus on a hand-over motion as an example of cooperative work between a human and a robot, and propose an algorithm which enables a robot to perform a human-like motion. First we analyze trajectories and velocity patterns of a hand-over motion performed by two humans. From this experimental results we notice that a receiver's motion, during hand-over has some typical characteristics. We then confirm that a human-like motion can be generated using these characteristics. Finally, we plan the robot's motion considering these results. Initially, two kinds of potential fields are used to generate a motion command which leads the robot along a trajectory similar to that followed by the human. In addition, more precise motion is considered at the end of the hand-over operation to guarantee accurate positioning and to soften the shock of contact. Experimental results using 2-link planar robot arm show the validity of the proposed method.

Personality through faces for humanoid robots

Abstract: Empirical studies have shown the great extent to which a face characterizes personality. For this reason, among others, the face of a humanoid robot is paramount. In fact, a social contract exists between the robot and each human it interacts with. We present the design issues we have encountered in designing a face for the humanoid robot Cog. The dimensions, organic form and mechanical constraints must be worked out to attain our goals. Our implementation of the design process is novel with respect to our use of solid modeling and stereo lithography processes and has proven to be essential to our quest for an iconic human-like face.

Growing RBF structures using self-organizing maps

Abstract: We present a novel growing RBF network structure using SOM in this paper. It consists of SOM and RBF networks respectively. The SOM performs unsupervised learning and also the weight vectors belonging to its output nodes are transmitted to the hidden nodes in the RBF networks as the centers of RBF activation functions, as a result one to one correspondence relationship is realised between the output nodes in SOM and the hidden nodes in RBF networks. The RBF networks perform supervised training using delta rule. Therefore, the current output errors in the RBF networks can be used to determine where to insert a new SOM unit according to the rule. This also makes it possible to make the RBF networks grow until a performance criterion is fulfilled or until a desired network size is obtained. The simulations on the two-spirals benchmark are shown to prove the proposed networks have good performance.

Fuzzy theoretical interactions between consciousness and emotions

Abstract: Emotional processes in the paper are defined as both synergistic processes of consciousness with stimuli and expressive processes caused for movements of the mind affected with consciousness. Models of the former are organized by system control models and the latter expressed from the introduced psychogenic particles driven by consciousness. It is proved that the probability space formed by psychogenic particles becomes the probability informational resource of expressions. Fuzzy decision problems with emotions are definable on the hypothesis that emotions are fuzzy. Expressions of a hatred by random walking of an autonomous vehicle (AV) are found from the application example of emotion process models with the AV avoiding obstacles.

Chaotic characteristic in human hand movement

Abstract: This paper reveals the chaotic properties of the voluntary single arm and two arms movements in various conditions based on the movement frequency and movement modes (single arm mode, in-phase mode and anti-phase mode). The chaotic analysis of rhythmic forearm movement were performed, and it was confirmed that the chaotic characteristics represented by the correlation dimension and the largest Lyapunov exponents were included in the rhythmic forearm movements. Furthermore, it was clarified that there existed the difference in the values of the Lyapunov exponents and fractal dimensions between right-left forearms and consequently the left forearm movements was more complex than the right forearm movements.

3D shape evaluation in virtual space - effectiveness of gazing line-action linkage system

Abstract: It is indicated that there is the high possibility of disagreement between the evaluation of shape in virtual space and that in real space. We (1999) previously proposed the system for shape evaluation that integrates the visual information in virtual space and the tactile and gazing line-motion linkage information in real space. In this paper, the preferences of 3D shape images in the proposed virtual space under two experimental conditions were compared with that of the real photo-forming products made from the same data by the sensory evaluation of paired comparison and the questionnaire analysis. It was found that the preference for shapes was consistent in both spaces in the relation of the preference among shapes based on the Bradley-Terry model in sensory evaluation. This shows that the proposed system provides almost the same environment as the real space for 3D shape evaluation. The results of questionnaire also indicate that the proposed system is enough to evaluate the 3D shapes in virtual space.