Sensory-Motor Coordination based Movement Learning and Dynamic Adaptation in Human Friendly Robot

在第三年，我们改进了第二年提出的基于神经元的运动发生器的适用性，使其应用于更广泛的领域，特别是与人类运动行为的交互。然而，人与下肢辅助器具之间的同步缺乏同步和主动控制机制。因此，需要一种认知模型，能够产生和整合从微观控制到宏观规划，从短期适应到长期优化的流动性。重要的是，认知模型可以分为微观、中观和宏观;这些概念对应于感知、知觉和认知（从神经科学的角度）以及短期、中期和长期适应（从生态心理学的角度）。这表明多腿运动需要五个智能功能，表现为1）注意力模块，2）自适应运动控制模块，3）对象识别模块，4）环境地图构建模块，以及5）最佳运动规划模块。所提出的神经认知模型从多视角整合了这些智能功能。我们进行了实验的实时重新规划和行为协调在粗糙的地形和动态环境的水平。结果，所提出的模型已发表在两个高影响力的期刊和两个会议，和一个期刊论文正在审查。

In the third year, we improve the appicability of the proposed neuro-based locomotion generator that has been done in the second year, to more general field of application, especially in interaction with human movement behavior. However, the synchronization between human and lower-limb assistive device has lack in synchronization and active control mechanism. Therefore, a cognitive model is required that can engender and integrate mobility from microscopic control to macroscopic planning and from short-term adaptation to long-term optimization. Importantly, cognitive models can be discussed as micro-, meso-, and macro-scopic; these concepts correspond to sensing, perception, and cognition (in terms of neuroscience) and short-, medium-, and long-term adaptation (in terms of ecological psychology). This suggests that multi-legged locomotion requires five intelligent functions, which manifest as 1) an attention module, 2) an adaptive locomotion control module, 3) an object recognition module, 4) an environmental map building module, and 5) an optimal motion planning module. The proposed neuro-cognitive model integrates these intelligent functions from a multi-scopic point of view. We conducted experiments on real-time re-planning and behavior coordination in rough terrain and dynamic environments level. As the result, the proposed model has been published in two high impact journal and two conferences, and one journal paper under review.