Computational approaches to motor learning by imitation
Computational approaches to motor learning by imitation
复制标题
DOI：
10.1098/rstb.2002.1258
复制
发表时间：
2003-03-29
影响因子：
6.3
通讯作者：
Billard, A
中科院分区：
文献类型：
--
作者：
Schaal, S;Ijspeert, A;Billard, A
动作模仿需要一套复杂的机制，将观察到的教师的动作映射到自己的动作装置上。相关问题包括运动识别、姿态估计、姿态跟踪、身体对应、从外部空间到自我空间的坐标转换、观察到的运动与先前学习的运动的匹配、不受观察限制的冗余自由度的解析、用于模拟的合适的运动表示、运动控制的模块化等。所有这些主题本身都是计算和神经生物学科学中活跃的研究问题，以至于将它们结合到一个完整的模拟系统中仍然是一个艰巨的任务-事实上，人们可以争辩说，我们需要理解完整的知觉-动作环路。作为解开模仿复杂性的一种策略，本文将纯粹从计算的角度来审查模仿，即我们将回顾已提出的解决部分模仿问题的统计和数学方法，并讨论它们的优缺点和基本原理。鉴于本期特刊侧重于对其他贡献的动作识别，本文将主要强调模仿的运动方面，假设感知系统已经识别了所演示运动的重要特征并创建了相应的空间信息。基于运动控制在控制策略方面的形式化及其相关的性能标准，可以产生有用的模仿学习分类，以阐明不同的方法和未来的研究方向。
Movement imitation requires a complex set of mechanisms that map an observed movement of a teacher onto one's own movement apparatus. Relevant problems include movement recognition, pose estimation, pose tracking, body correspondence, coordinate transformation from external to egocentric space, matching of observed against previously learned movement, resolution of redundant degrees-of-freedom that are unconstrained by the observation, suitable movement representations for imitation, modularization of motor control, etc. All of these topics by themselves are active research problems in computational and neurobiological sciences, such that their combination into a complete imitation system remains a daunting undertaking-indeed, one could argue that we need to understand the complete perception-action loop. As a strategy to untangle the complexity of imitation, this paper will examine imitation purely from a computational point of view, i.e. we will review statistical and mathematical approaches that have been suggested for tackling parts of the imitation problem, and discuss their merits, disadvantages and underlying principles. Given the focus on action recognition of other contributions in this special issue, this paper will primarily emphasize the motor side of imitation, assuming that a perceptual system has already identified important features of a demonstrated movement and created their corresponding spatial information. Based on the formalization of motor control in terms of control policies and their associated performance criteria, useful taxonomies of imitation learning can be generated that clarify different approaches and future research directions.