Abstract not found

ing away from the specific coding of the spinal fields, the examples from neurobiology provide the framework for a motor control system based on a small number of additive primitives (or basis behaviors) sufficient for a rich output movement repertoire. Our previous work (Matari'c 1995, Matari'c 1997), inspired by the same biological results, has successfully applied the idea of basis behaviors to control of mobile robots 6 by fitting it directly into the modular behavior-based control paradigm. Applictions of schema theory (Arbib 1992) to behavior-based mobile robots (Arkin 1987) have employed a similar notion of composable behaviors, stemming from foundations in neuroscience (Arbib 1981, Arbib 1989). The idea of using such primitives for articulator control has been recently studied in robotics. Williamson (1996) and Marjanovi'c, Scassellati & Williamson (1996) developed a 6 DOF (degrees of freedom) robot arm controller. While in the biological and mobile robotics work primitives c...

We present a comprehensive survey of robot Learning from Demonstration (LfD), a technique that develops policies from example state to action mappings. We introduce the LfD design choices in terms of demonstrator, problem space, policy derivation and performance, and contribute the foundations for a structure in which to categorize LfD research. Specifically, we analyze and categorize the multiple ways in which examples are gathered, ranging from teleoperation to imitation, as well as the various techniques for policy derivation, including matching functions, dynamics models and plans. To conclude we discuss LfD limitations and related promising areas for future research.

This paper is concerned with the evaluation of a model of human imitation of arm movements. The model consists of a hierarchy of articial neural networks, which are abstractions of brain regions involved in visuo-motor control. These are the spinal cord, the primary and pre-motor cortexes (M1 & PM), the cerebellum, and the temporal cortex. A biomechanical simulation is developed which models the muscles and the complete dynamics of a 37 degree of freedom humanoid. Input to the model are data from human arm movements recorded using video and marker-based tracking systems. The model's performance is evaluated for reproducing reaching movements and oscillatory movements of the two arms. Results show a high qualitative and quantitative agreement with human data. In particular, the model reproduces the well known features of reaching movements in humans, namely the bellshaped curves for the velocity and quasi-linear hand trajectories. Finally, the model's performance is compared to that o...

This paper describes the use of task primitives in robot learning from observation. A framework has been developed that uses observed data to initially learn a task and then the agent goes on to increase its performance through repeated task performance (learning from practice). Data that is collected while a human performs a task is parsed into small parts of the task called primitives. Modules are created for each primitive that encode the movements required during the performance of the primitive, and when and where the primitives are performed. The feasibility of this method is currently being tested with agents that learn to play a virtual and an actual air hockey game. 1

We want to build robots capable of rich social interactions with humans, including natural communication and cooperation. This work explores how imitation as a social learning and teaching process may be applied to building socially intelligent robots, and summarizes our progress toward building a robot capable of learning how to imitate facial expressions from simple imitative games played with a human, using biologically inspired mechanisms. Our approach is heavily influenced by the ways human infants learn to communicate with their caregivers and understand the actions of others in intentional terms. Among the key ideas that we draw from work on the development of human social intelligence, the most crucial is the hypothesis that in human infants, imitative interactions, starting with facial mimicry, are a significant stepping-stone in developing appropriate social behavior, learning to predict other’s actions, and ultimately, understanding the intensions of others. 1

Abstract. Motor control is a complex problem and imitation is a powerful mechanism for acquiring new motor skills. In this paper, we describe perceptuo-motor primitives, a biologically-inspired notion for a basis set of perceptual and motor routines. Primitives serve as a vocabulary for classifying and imitating observed human movements, and are derived from the imitator’s motor repertoire. We describe a model of imitation based on such primitives and demonstrate the feasibility of the model in a constrained implementation. We present approximate motion reconstruction generated from visually captured data of typically imitated tasks taken from aerobics, dancing, and athletics. 1

This paper describes experiments performed with forty subjects wearing an eye-tracker and watching and imitating videos of finger, hand, and arm movements. For all types of stimuli, the subjects tended to fixate on the hand, regardless of whether they were imitating or just watching. The results lend insight into the connection between visual perception and motor control, suggesting that: 1) people analyze human arm movements largely by tracking the hand or the end-point, even if the movement is performed with the entire arm, and 2) when imitating, people use internal innate and learned models of movement, possibly in the form of motor primitives, to recreate the details of whole-arm posture and movement from end-point trajectories. Keywords: Perceptual-motor interaction; Eye-tracking; Movement imitation Theme: Motor Systems and Sensorimotor Integration Topic: Control of Posture and Movement 1 Introduction Imitation is one of the most ubiquitous forms of human learning. What appea...

This paper appeared in Trends in Cognitive Science, Vol. 2, No. 3, March 1998, 82-87.)

This paper presents a framework that allows an agent to use observed data to initially learn a predefined set of primitives and the conditions under which they are used. A method is included for the agent to increase its performance while operating in the environment. The details of implementing this framework on agents that play air hockey in simulation and on an actual table will be presented. Issues involved with using observation data and primitives to increase the learning rate of agents are discussed.