International Journal of Humanoid Robotics

A hand-designed internal representation of the world cannot deal with unknown or uncontrolled environments. Motivated by human cognitive and behavioral development, this paper presents a theory, an architecture, and some experimental results for developmental robotics. By a developmental robot, we mean that the robot generates its "brain" (or "central nervous system," including the information processor and controller) through online, real-time interactions with its environment (including humans). A new Self-Aware Self-Effecting (SASE) agent concept is proposed, based on our SAIL and Dav developmental robots. The manual and autonomous development paradigms are formulated along with a theory of representation suited for autonomous development. Unlike traditional robot learning, the tasks that a developmental robot ends up learning are unknown during the programming time so that the task-specific representation must be generated and updated through real-time "living" experiences. Experimental results with SAIL and Dav developmental robots are presented, including visual attention selection, autonomous navigation, developmental speech learning, range-based obstacle avoidance, and scaffolding through transfer and chaining.

Control for and interaction with humanoid robots is often restrictive due to limitations of the robot platform and the high dimensionality of controlling systems with many degrees of freedom. We focus on the problem of providing a "skill-level interface" for a humanoid robot. Such an interface serves as (i) a modular foundation for structuring task-oriented control, (ii) a parsimonious abstraction of motor-level control (e.g. PD-servo control), and (iii) a means for grounding interactions between humans and robots through common skill vocabularies. Our approach to constructing skill-level interfaces is two-fold. First, we propose a representation for a skill-level interface as a "behavior vocabulary," a repertoire of modular exemplar-based memory models expressing kinematic motion. A module in such a vocabulary encodes a flow field (or gradient field) in joint angle space that describes the "flow" of kinematic motion for a particular skill-level behavior, enabling prediction from a given kinematic configuration. Second, we propose a data-driven method for deriving behavior vocabularies from time-series data of human motion using spatio-temporal dimension reduction and clustering. Results from evaluating an implementation of our methodology are presented along with the application of derived behavior vocabularies as predictors towards on-line humanoid trajectory formation and off-line motion synthesis.

Sensing is a key element for any intelligent robotic system. This paper describes the current progress of a project in the Intelligent Robotics Research Center at Monash University that has the aim of developing a synergistic set of sensory systems for a humanoid robot. Currently, sensing modes for colour vision, stereo vision, active range, smell and airflow are being developed in a size and form that is compatible with the humanoid appearance. Essential considerations are sensor calibration and the processing of sensor data to give reliable information about properties of the robot's environment. In order to demonstrate the synergistic use of all of the available sensory modes, a high level supervisory control scheme is being developed for the robot. All time-stamped sensor data together with derived information about the robot's environment are organized in a blackboard system. Control action sequences are then derived from the blackboard data based on a task description. The paper presents details of each of the robot's sensory systems, sensor calibration, and supervisory control. Results are also presented of a demonstration project that involves identifying and selecting mugs containing household chemicals. Proposals for future development of the humanoid robot are also presented.

This paper presents an overview of our work towards building socially intelligent, cooperative humanoid robots that can work and learn in partnership with people. People understand each other in social terms, allowing them to engage others in a variety of complex social interactions including communication, social learning, and cooperation. We present our theoretical framework that is a novel combination of Joint Intention Theory and Situated Learning Theory and demonstrate how this framework can be applied to develop our sociable humanoid robot, Leonardo. We demonstrate the robot's ability to learn quickly and effectively from natural human instruction using gesture and dialog, and then cooperate to perform a learned task jointly with a person. Such issues must be addressed to enable many new and exciting applications for robots that require them to play a long-term role in people's daily lives.

In this paper, we propose a control strategy allowing us to perform the transition of velocities included in [0 m/s; 1 m/s] for the dynamic walking of a virtual under-actuated robot (RABBIT) without reference trajectories. This strategy of control enables us to carry out the transition from stop towards walking and the reverse process. The interest of this method is that, on the one hand, the intrinsic dynamics of the system are exploited by using a succession of active and passive phases and, on the other hand, the control strategy is very simple to implement on-line. Moreover, we apply this method by taking into account the technological limitations related to experimentation on the real robot such as dry and viscous frictions, maximum torque, and maximum power.

In the past fifty years, efforts in classical AI have focussed on computerizing human intelligence. Naturally, computerized human intelligence is not a proof of machine or robot intelligence because the programs underlying computerized human intelligence are still made by humans. So far, there is no computer nor robot which is creative enough to master its own language and to compose a text expressing its intentions. Thus, it is time to shift our research focus from computerizing human intelligence to developing machine intelligence. A first and necessary step towards this goal is to make machines or robots learn, manipulate, understand and create both elementary and composite meanings encoded in a natural language such as English. Elementary and composite meanings could be acquired through both sample texts and images. Hence, we propose a learning-synthesis-analysis framework which aims to enable a robot, or computer, to understand and convey meaning through texts or images. The main contribution of this paper is to lay out a sound foundation on which interdisciplinary research could effectively progress toward the development of machines or robots that understand meaning through texts or images.