This paper proposes an approach to enable a robot to learn social skills to interact with humans in social settings. Our approach is based on integrating a cognitive control architecture with imitation learning and human intention recognition. The originally developed cognitive control architecture was expanded to include behavior generalization, behavior generation, and human intention recognition. Our approach provides a framework for a robot to be able to estimate human intention through common features of human gestures through observation couple with past experiences stored in the long-term memory in the form of learned social knowledge, and to cognitively generate appropriate motion behaviors or modify current behavior using arm and hand. Key components of our approach are a probabilistic arm and hand gesture recognition and cognitive control modules which integrate the gesture recognition with human intention recognition and motion behavior generation. Several experiments were carried out on humanoid robots to validate the proof of concept.

This research investigates adults' understanding of differences in the basic nature of intelligence exhibited by humans and by machines such as computers and robots. We tested these intuitions by asking participants to make predictions about the behaviors of different entities in situations where actions could be based on either goal-directed intentional thought or more mechanical nonintentional thought. Across several studies, adults made more intentional predictions about the behavior of humans than about the behavior of robots or computers. Although initial experiments demonstrated that participants made very similar predictions for computers and anthropomorphic robots, when asked to track robots' attention to objects, participants began to predict more intentional behaviors for the robot. A multiple regression demonstrated that differential behavioral predictions about mechanical and human entities were associated with ratings of goal understanding but not overall intelligence of current computers/rob...

The goal of this paper is to develop a cognitive developmental approach for a humanoid robot so that it can provisionally discover self-affordance relations between certain arm limb movements and corresponding motor units by exploring the outcomes of its random arm movements while in a crawling position. Learning of the right and the left arm affordances is based on self-exploration and a set of experience similarly to how a human baby discovers action-effect relations of own arm movements. We address the early development of self-affordances, similar to infants, which encodes the relationships between actions, objects, and the effect on the environment.

This paper proposes a computational framework for humanoid robots to learn complex behaviors through combining robotic self-exploration and demonstrations of humans. A modified Rapidly-growing Random Tree (RRT)-Connect algorithm is used for exploration, a Linear Global Model (LGM) is used for recording demonstrations, a spatial-temporal extension of Isomap algorithm is used for dimension reduction which enables the exploration in a low-dimensional latent space, and the log likelihood function of the distribution of sampled data in the joint space is used to project the data in the latent space back to the joint space. An experiment of imitating a conducting behavior is carried out to demonstrate the effectiveness of this framework.

This paper introduces a new approach to develop robots that can learn general affordance relations from their experiences. Our approach is a part of larger efforts to develop a cognitive robot and has two components: (a) the robot models affordances as statistical relations among actions, object properties and the effects of actions on objects, in the context of a goal that specifies preferred effects and outcomes, (b) to exploit the general-knowledge potential of actual experiences, the robot engages in internal rehearsal by playing out virtual scenarios grounded in yet different from actual experiences. To the extent the robot accurately appreciates affordance relations, the robot can autonomously predict the outcomes of its behaviors before executing them. Internal rehearsal-based outcome production in turn facilitates planning of a sequence of behaviors toward successful task execution. We also report simulation results of internal rehearsal-based traversability affordance learning of a humanoid robot.

In this paper, we present a method for motion generation from automatically derived behaviors for a humanoid robot. Behaviors are derived automatically by using the underlying spatio-temporal structure in motion. The derived behaviors are stored in a robot's long-term (or procedural) memory. New motions are generated from the derived ones with a search mechanism. In our approach, vision, speech recognition, short-term memory and decision-making operate in parallel with long-term memory in a unique architecture. This organization is intended for autonomous robot control and learning.

A human-robot interface for a mobile robot was extended to include a discrete geodesic dome called a Sensory EgoSphere (SES) The SES is a two-dimensional data structure, centered on the robot's coordinate frame. The SES provides the robot's perspective of a remote environment via images, sonar, and laser range finder representations. It was proposed that the SES would enhance the general interface usability by decreasing perceived workload and increasing situational awareness. A human factors evaluation was performed to evaluate the established hypothesis. Novice users participated in the evaluation. The purpose of this paper is to review some of the evaluation results.

There is a desire to provide small, lightweight mobile interaction devices. This paper describes a PDA-based teleoperation interface for mobile robot. This interface differs from previous PDA teleoperation interfaces in that it uses only touch based interactions rather than stylus interaction. Three screens were developed: an imagery-only screen, a sonar and laser range finder based screen, and image with sensory overlay screen. This paper presents the interface motivation and design and discusses issues that arise when employing a PDA as a touch screen. This paper also describes the planned human factors evaluation.

Teams of humans and robots working together can provide effective solutions to problems. In such applications, effective human-robot teaming relies on being able to communicate information about the current perception or understanding of the environment. In this paper, human-robot teaming on navigational tasks is discussed. The role of the human user will be to specify the goal point(s) for the robot, and also to interact with the robot in the event of perceptual errors. A novel navigation method, Egocentric Navigation (ENav), has been developed based on egocentric representations. In this paper, two knowledge sharing methods are described which exploit the characteristics of ENav. In the first method, robots share Landmark EgoSpheres, and in the second method the robot shares it Sensory EgoSphere with the human user for visual perception correction.

Shows how a Sensory EgoSphere (SES), a robot-centric geodesic dome that represents the short term memory of a mobile robot, could enhance a human-robot interface. It is proposed that the addition of this visual representation of the sensor data on a mobile robot enhances the effectiveness of a human-robot interface. The SES migrates information presentation to the user from the sensing level to the perception level. The composition of the vision with other sensors on the SES surrounding the robot gives clarity and ease of interpretation. It enables the user to better visualize the present circumstances of the robot. The human-robot interface (HRI) will be implemented through a graphical user interface (GUI) which contains the SES, command prompt, compass, environment map, sonar and laser display. This paper proposes that the SES increases situational awareness and allows the human supervisor to accurately ascertain the present perception (sensory input) of the robot and use this information to assist the robot in getting out of difficult situations.

In this research, a taxonomy is introduced to cover important considerations for human-robot interactions. As an application of passive human-robot interaction, two modalities for localizing humans based on sound source localization and infrared motion detection were developed and integrated with the face-tracker system of a humanoid ISAC (intelligent soft arm control), in order to direct ISACs attention and to prevent it from being quickly distracted. The sound source localization and passive infrared motion detection systems are used to provide the face-tracker system with candidate regions for finding a face. In order to avoid the situation where the robot appears to be "hyperactive" and cannot give sufficient attention to a newly discovered face, these sensing modules should not directly gain control of the tracking if the system has recently acquired a new face. Our goal is to allow a human to redirect the attention of the system but give the system a method to ignore the distraction if recently engaged.

Describes a biologically-inspired control architecture for the McKibben actuated limbs of a humanoid robot. The antagonistically driven joints are actuated using a biological control model observed in the measurement of human muscle electromyograms (EMG) during reaching movements in the vertical plane. The paradigm uses the summation of tonic and phasic EMG signals to activate the human muscles. The humanoid robot's muscles, actuated by pressure control, are controlled with feedforward pressure patterns analogous to the tonic and phasic activation in the human model. Proprioceptive feedback is utilized in the control architecture to correct for misperceived loading conditions and time variance of the actuators. The control architecture, initial experimental results, and experiments are discussed in the paper. A result of this control paradigm is the realization of actuation with lower stiffness and therefore safer operation for human-humanoid interaction. It is expected that such a motion of the humanoid will closely resemble human motion and will facilitate a more human-friendly human-robot interaction.

In order for a humanoid robot to be accepted in society and perform as an intelligent human assistant or companion, it must exhibit robust human-humanoid interaction (HHI). Additionally, the humanoid must be able to recognize if it is working properly, and if not, why. We describe our development of a framework for HHI. This framework includes a high-level agent called the Self Agent; part of its role is to maintain information about the status of the humanoid. To accomplish this, we have developed a technique called system status evaluation that the humanoid uses to detect and localize failures. We present the results of the use of SSE on a humanoid system called ISAC.

An active vision system enables the selective capture of important information for various applications. With an active vision system, the control of gaze direction toward a visual target can simplify many of the robot's interactions with the world. This work presents the implementation of a camera head control system for a mobile robot, called HelpMate. Helpmate's binocular head is composed of two Sony color cameras mounted on two independent pan-tilt heads. The camera head control behavior is designed to be similar to that of human eyes, which include saccade, smooth pursuit, vergence and vestibulo-ocular reflex. Saccade allows the robot to quickly direct its gaze to the target, smooth pursuit keeps both cameras fixated on the moving target, vergence makes sure that the two cameras are looking at the same object, and vestibulo-ocular reflex compensates the camera motion for the robot body's movement. Control methods have been designed and tested through experimentation. By using color segmentation, HelpMate can detect and locate a human holding a specific colored object, who is approaching within its range. Once the target is located, the two cameras are fixated, and HelpMate's body turns toward the target. The camera head then smoothly tracks the moving target and guides the robot to follow the target as it moves.

Modern manufacturing system have to deal with dynamic situations such as customer specification changes, machine break-down, emergency orders and other kinds of disturbances. This requires the manufacturing system to be adaptive to handle such situations. The holonic concept has been proposed as an efficient paradigm for developing such an adaptive manufacturing system. There are still some critical issues to be investigated such as how to define the holons for a given application context and make them reusable, what should be an appropriate system architecture and how to design the system cooperation mechanisms for better system performance. The Intelligent Machine Architecture (IMA) is a software architecture developed at Vanderbilt University. It provides a framework for developing reusable components and constructing reconfigurable, extensible systems. Although it is mainly used for robots, it is also a suitable development tool for designing adaptive manufacturing systems. The authors explain how to build reusable, reconfigurable and extensible holonic manufacturing systems based on the IMA framework.

The paper proposes a concept for a next generation manufacturing system, which is globally distributed rather than geographically limited. This new manufacturing system is possible because of recent changes in manufacturing technology. To work as a global unit, companies need to work closely together to improve efficiency by sharing resources, experts, tools, and training. To support these collaborative works, an Internet-based remote assembling system (RAS) is proposed. This system is remotely controlled through the Internet to allow an operator at a remote location to operate resources at an assembling workcell. For example, a Web browser connects the operator to the Web server at the assembling workcell to download an Internet-based design application, allowing the operator to select the product that needs to be assembled. After the selection, a command to assemble the product is created and sent to the assembling workcell. Along with this concept, preliminary results based on simulation are shown.

This paper proposes a simple method for localization using an electronic compass. Electronic compasses are often used to detect the heading of mobile robots. However, electronic compasses have one drawback when used inside a building: they can easily be disturbed by electromagnetic sources (e.g., power lines) or large ferro-magnetic structures (e.g., bookshelves). However, this paper introduces another indoor application of electronic compasses. We take advantage of the magnetic field disturbances by using them as distinctive place recognition signatures. We first gather information about the changing heading as our robot travels along the hallway outside the lab, and then store this information. As the robot traverses the hallway, it gathers the information from the electronic compass and matches it with the pre-stored data. If a match is found, the robot can determine its current position. We use a sequential least-squares approximation approach for matching the signature. The simulation results will show that the robot can distinguish its location by using these signatures.

As robotics and automation applications extend to the service sector, researchers have to increasingly deal with performing robotic actions in uncertain and unstructured environments. A traditional solution to this problem models uncertainty about the effects of actions by probabilities conditioned on the state of the environment, making it possible to select plans that have the highest probability of success in a given situation. Reactive systems use another approach to handling uncertainty, by employing a set of predefined situation-response rules that make it possible to move toward the goal from any situation, whether expected or unexpected. This paper describes a planner that combines the two approaches. A proactive component generates plans that are biased toward picking the most reliable action in a given situation, and a reactive component can alter the selected actions based on unexpected situations that may arise in uncertain environments. Action selection is driven by a spreading activation mechanism on a probabilistic network that encodes the domain knowledge. A decision-theoretic framework incorporates quantitative goal utilities and action costs into the action selection mechanism. Experiments conducted demonstrate the ability of the planner to plan with hard and soft domain constraints and action costs, modify plans as a reaction to unexpected changes in the environment or goal utilities, and plan in situations with multiple conflicting goals.

This paper presents a simple but effective method for integrating sensors and actuators in both real and virtual environments using an interactive simulator over the Internet. Our research explores the combination of visual feedback and force feedback to enhance our 3D collision avoidance approach for a dual-arm humanoid robot. The robot control server uses 3D vision-based position estimates of obstacles, such as a human hand or objects on the table, to compute a collision-free path in real-time. The dual-arm robot is capable of performing multiple interactive tasks while avoiding collisions between its own arms or with other objects. The interactive simulator provides visual feedback to the user and sends the user's commands to the robot server and the stereo camera server via the Internet. Thus, it is well-designed for robot teleassistance applications. Force feedback effects are generated, using an inexpensive commercially available joystick, to enhance the sensation of the real environment while interacting with the simulation. Finally, experimental results using our collision avoidance approach are presented to demonstrate the performance of this system.

We describe our progress on the development of ROBIN, a robotic inspector. ROBIN is a structure-climbing robot designed for man-made environments. It is intended to carry cameras and other sensors onto man-made structures such as bridges, buildings, aircraft and ships for inspection. The robot has two vacuum fixtures connected by a 4 degree-of-freedom articulated mechanism that together allow it to walk across surfaces and will permit transition between adjacent surfaces. ROBIN is novel in several areas. It is the only climbing robot that uses McKibben type pneumatic muscles for movement. It is also novel in its use of a subsumption architecture controller in a climbing robot. ROBIN is one of the few climbing robots that is equipped with a mechanism that is capable of transitions between surfaces or from a horizontal surface to a vertical surface below.

In this paper we present the design and implementation of a novel low-cost active vision system for robotic applications. The system is comprised of two parts: A 4-DOF trinocular active camera head and a DSP-based image acquisition and processing board. Design issues and tradeoffs are discussed. The performance of the system is evaluated by running an edge-based object tracking algorithm that is also presented in this paper. Experimental results show that the system is capable of tracking objects at 30 frames/sec.

Intelligent control techniques have emerged to overcome some deficiencies in conventional control methods in dealing with complex real-world systems. These problems include knowledge adaptation, learning, and expert knowledge incorporation. In this paper, a hybrid network that combines fuzzy inferencing and neural networks is used to model and to control complex dynamic systems. The network takes advantage of the learning algorithms developed for neural networks to generate the knowledge base used in fuzzy inferencing. The network as used to model and to control a robot arm with flexible pneumatic actuator. Comparison with a nonlinear control technique used for the robot joints is also presented.

In this paper a system that successfully detects stationary objects in indoor scenes is presented. The doors and ceiling lamps have been chosen to be detected in the initial phase of the authors' system. The scene was divided into four different regions, The edges resulting from the walls/ceiling and the walls/floor intersections are the boundaries that separate the different regions. The use of the log-polar mapping transforms the problem of finding four edges of unknown orientations, that separated the different regions, into finding four horizontal edges. The log-polar map was performed such that it possesses the varying resolution property. The objects were detected using model-matching in the log-polar domain. The varying resolution property of the mapping simplified the model-matching problem. The detection (of edges and then objects) was very fast due to the small size of the log-polar image (64/spl times/64) pixels.>

This paper reviews the state of the art in service robots for the disabled and the elderly people. The classification is based upon system hardware configurations. Key R&D issues in the field of service robots for the disabled and elderly are discussed. The ISAC-HERO system being developed at the Intelligent Robotics Laboratory at Vanderbilt University is presented as an example of an intelligent robotic aid system. Then emerging trends in the areas of user interface, learning and adaptation are presented.>

Vanderbilt University

Design and implementation of a reflex con- troller for a robotic aid system called ISAC (Intelli- gent Soft Arm Control system) based on the frame- work of a layered control architecture is described. In particular, the implementation of a collision avoid- ance strategy is analyzed, and an analogy with the human reaction is presented along with a mathemati- cal interpretation. Finally, a task-dependent reflex ac- tion strategy is introduced to handle the cases where not enough information is available on the potentially harmful situation endangering the user.

Risk assessment of technological processes (chemical and power plants, electro-mechanical systems) is a complex process that requires enumeration of all possible failure modes, their probability of occurrence, and their consequences. Traditionally such studies have been performed by a committee of expert engineers with diverse backgrounds. This paper discusses the use of qualitative modeling techniques based on deriving behavior from structural descriptions and causal reasoning to aid automating and enhancing the risk analysis process. Hierarchical schemes are used for describing component structure, and system functionality is derived from a set of primitive functions and parameters defined for the domain. The system uses these models to automatically generate fault and event networks for hypothesized fault situations specified by users.

No abstract available.

This article describes NetGraph, an object-oriented toolset for risk assessment applications. There are two significant aspects of NetGraph functionality. Firstly, NetGraph is a graphical knowledge acquisition tool for risk assessment, designed for robust structuring and efficient use of domain knowledge. A component library facility in NetGraph helps the domain expert to store and easily retrieve frequently encountered objects in complex systems. Time spent in designing fault trees is significantly reduced by the help of a cut/paste facility that allows the expert to easily define new fault trees by retrieving subtrees (components) of previously defined fault trees and putting them together. A second aspect of NetGraph is fault tree analysis. Fault trees designed in NetGraph can be used by either experts or operators to analyze fault behavior and probability of complex systems. A hypertext interface is also being developed to provide efficient communication of risks to individuals.

This article describes an architecture for an intelligent computer-assisted instruction (ICAI) system for use in engineering domains. A qualitative model is utilized for representing the physical characteristics and operating methodologies of the systems tutored; planning techniques are employed for instruction using a text-based tutor; reasoning about the engineering model and about the student behavior during instruction is accomplished using a production system. Examples are drawn from two specific implementations: tutorials about (1) a hot water heater and (2) a power distribution system. Specific recommendations are given for implementing ICAI systems in engineering domains.

This paper reports on work being conducted for NASA to develop a simulation expert system called NESS that can assist the user to run digital simulations of dynamic systems and interpret the output data to determine system characteristics. The paper describes our design principle, system architecture and knowledge base. Finally, in order to demonstrate NESS' simulation and analysis capabilities, a session log involving a printwheel position controller is shown.

In response to the need for comprehensive assessment of environmental quality, the development of a large-scale computer model called SEAS was initiated by the US Environmental Protection Agency in 1972. Although SEAS has provided a framework for the collection, storage, and treatment of environmental data, it has increasingly become apparent that such large-scale models have several major limitations that preclude their usefulness as strategic environmental management tools. These major limitations are reviewed, key design specifications are offered, and a design process to ameliorate some of the limitations of using such models for environmental management are proposed.

An agenda for technology assessment in the field of materials was developed through a process of dialogue among experts by use of a procedure known as a "science forum." Materials scientists and technologists formed one group of experts, and social scientists formed another group. A scenario of a "constrained economy" provided a background for the forecast of science and technology for the remainder of this century. From this forecast, "candidates for technology assessment" were proposed and were rated as to urgency by both groups of experts. Two statistical procedures were used to estimate the degree of urgency for technology assessment (TA), as seen by each group of experts. An impressive degree of correlation was found between the ratings of the two groups and between the ranks as measured by the two statistical procedures. Five of the 21 candidates for technology assessment were judged to be "prime," and four more were judged to be "secondary." While the study was done for materials technologies it should be applicable also to other fields of technology.