NSF/SBE-BSF:Integration of kinesthetic and tactile information in perception, action, and learning

简单的动作，如打开罐子或点燃火柴，取决于抓住不同形状的物体的能力，以及在不同方向上施加规则良好的力和运动的能力。当我们拿着一杯水时，手指施加的力直接作用于玻璃杯表面，我们的大脑协调这些力以防止滑动并保持玻璃杯的方向。计划中的研究将把控制工程和机器人学的计算方法与神经科学的技术和理论结合起来，以了解大脑在举起和操纵物体时如何控制抓地力，以及它如何估计物体的机械性能。调查将考虑对象属性的不确定性(例如，对象的硬度、滑度、质地等)如何影响对对象的把握，以及如何通过人为增加触觉信息来增强基本操作技能。这些研究的结果将影响几个领域，包括(1)神经科学(了解我们用手控制和操纵物体的能力)；(2)技术(向远程操作和机器人辅助手术等机器人设备的用户展示力量信息)；(3)神经康复(用于开发智能机器人假肢和新的练习，以促进丢失的手动技能的恢复)。这些研究将建立美国和以色列科学家之间的合作互动，他们拥有包括机器人、控制、运动系统、神经科学和神经康复在内的一系列专业知识。在灵巧的操作过程中，让手指合作需要整合不同类型的信息。手指有丰富的触觉传感器阵列，可以产生与物体相互作用相关的信号。然而，为了让大脑知道这些力的大小和方向，它还必须知道手指在空间中的方向。这一信息是由其他来源获得的，这些来源构成了物体在空间中的位置感觉的基础。抓取的控制也可以通过肌肉中的动觉感应器来获得信息，而这些感应器又与关节在空间中的构型感觉相结合。这个项目利用了一项新技术，可以在手指的皮肤上进行受控拉伸。安装在机器人机械手上的皮肤拉伸设备将用于在受试者执行操作任务时应用各种受控扰动。扰动将被用来增加或减少触觉和动觉反馈之间的一致性，并研究大脑如何适应在不可预测和可预测的力中保持稳定抓取的能力。该方法将理论和实验相结合，以解决感知和控制操纵和抓地力方面的多种信息源的整合问题。该项目得到了美国国家科学基金会和美国-以色列双国科学基金会的合作支持。

Simple acts such as opening a jar or lighting a match depend on the ability to grasp objects with different shapes and to apply well-regulated forces and movements in different directions. When we hold a glass of water, the forces applied by the fingers are directed against the surface of the glass, and our brain coordinates these forces to prevent slippage and maintain the orientation of the glass. The planned studies will combine computational methods derived from control engineering and robotics with techniques and theories from neuroscience to understand how the brain controls grip forces when lifting and manipulating objects and how it estimates an object's mechanical properties. The investigations will consider how grasping an object is affected by uncertainties about the object properties (e.g., it's hardness, slipperiness, texture, etc) and how fundamental manipulation skills may be enhanced by artificially augmenting tactile information. The outcomes of these studies will influence several domains including (1) neuroscience (understanding our ability to control and manipulate objects with our hands); (2) technology (for presenting force information to users of robotic devices such as teleoperation and robot-assisted surgery); and (3) neurorehabilitation (for developing intelligent robotic prostheses and new exercises to promote the recovery of lost manual skills). These studies will establish a collaborative interaction between scientists in the United States and Israel having a range of expertise that includes robotics, control, motor systems neuroscience and neurorehabilitation.Getting the fingers to cooperate during dexterous manipulation requires the integration of different types of information. The fingers have a rich array of tactile sensors that generate signals related to the interaction with the object. However, for the brain to know the magnitude and direction of these forces, it must also know the orientation of the fingers in space. This information is obtained by other sources that form the basis for the sense of position of the body in space. The control of grasp can also be informed by kinesthetic force sensors in the muscles that in turn are integrated with the sense of the joint's configuration in space. This project takes advantage of a new technology that allows applying controlled stretches to the skin of the fingers. Skin stretch devices mounted on a robotic manipulator will be used to apply a variety of controlled perturbations while subjects are performing manipulation tasks. Perturbations will be applied to increase or decrease the consistency between tactile and kinesthetic feedback, and to investigate how the brain adapts the ability to maintain a stable grasp during both unpredictable and predictable forces. The approach will combine theory and experiments to tackle the integration of multiple information sources in perception and in controlling manipulation and grip forces.This project is being supported by a partnership between the National Science Foundation and the U.S.-Israel Binational Science Foundation.