Search

Proactive assistance in human–robot collaboration remains a challenging objective, as the spatial–temporal coordination of the human–robot motion must be considered in conjunction with the object and environmental context. In this paper, we propose an environment-adaptive probabilistic interaction primitive method using learning-from-demonstration. In particular, we propose a novel phase estimation algorithm called Single-axis Uniform Interval Interpolation, which alleviates the restriction of Gaussian or uniform distribution of phase variables. In addition, the environmental constraints in human–robot interactive skills are learned via the regression between environmental parameters and the weight vectors. The proposed method is implemented in a proactive robotic system for typical industrial-motivated human–robot collaborative scenarios, such as assistive push-button assembly and human–robot collaborative object covering. The experimental result validates the effectiveness of the proposed approach. [Display omitted] • A novel phase estimation algorithm for non-Gaussian or non-uniform phase variables. • Interactive Probabilistic Movement Primitives are extended to be environment-adaptive. • A system is implemented for enabling proactive robotic assistance behaviors. [ABSTRACT FROM AUTHOR]

A key prerequisite to make user instruction of work tasks by interactive demonstration effective and convenient is situated multi-modal interaction aiming at an enhancement of robot learning beyond simple low-level skill acquisition. We report the status of the Bielefeld GRAVIS-robot system that combines visual attention and gestural instruction with an intelligent interface for speech recognition and linguistic interpretation to allow multi-modal task-oriented instructions. With respect to this platform, we discuss the essential role of learning for robust functioning of the robot and sketch the concept of an integrated architecture for situated learning on the system level. It has the long-term goal to demonstrate speech-supported imitation learning of robot actions. We describe the current state of its realization to enable imitation of human hand postures for flexible grasping and give quantitative results for grasping a broad range of everyday objects. [Copyright &y& Elsevier]

This paper develops a general policy for learning the relevant features of an imitation task. We restrict our study to imitation of manipulative tasks or gestures. The imitation process is modeled as a hierarchical optimization system, which minimizes the discrepancy between two multi-dimensional datasets. To classify across manipulation strategies, we apply a probabilistic analysis to data in Cartesian and joint spaces. We determine a general metric that optimizes the policy of task reproduction, following strategy determination. The model successfully discovers strategies in six different imitative tasks and controls task reproduction by a full body humanoid robot. [Copyright &y& Elsevier]