CAREER: A Framework for Revealing How Locomotor Control Emerges from the reciprocal Interactions of Neural and Mechanical Systems

动物在几乎所有的环境中都能以惊人的稳定性和敏捷性移动。要做到这一点，它们依赖于相互作用的神经和机械系统，这些系统必须在传感、驱动和环境相互作用的物理学背景下运行。现代数据采集工具提供了前所未有的对实现控制的潜在神经、肌肉和机械信号的访问。然而，这些信号本身并不是理解感觉信息如何转化为运动输出的框架。这个CAREER项目利用几种新兴的实验和分析技术来展示运动动力学如何从这种感觉运动转换中产生。这种提取感觉运动控制原理的框架基于高分辨率神经肌肉记录、信息论和降维数据分析以及系统识别和控制理论的形式语言。这个建议的实验框架是可转移到其他生活和工程系统组成的许多子系统连接的反馈，如基因或蛋白质网络，细胞力学，或人口动力学。从这项工作中产生的运动控制原理将与BRAIN计划和类似的计划协同作用，通过提供背景来解释对大脑结构，解剖学和连接的深入，详细的理解。他们还将解决机器人和神经工程新时代的挑战。迫切需要神经技术，使灵活的运动，同时体现在物理系统。更广泛地说，这项工作将使基于研究的本科学习转化为科学和教育产品。它将为格鲁吉亚理工学院不断增长的生命系统物理课程带来神经科学的组成部分。在VIP计划中，学生将通过参与组织成垂直指导系统的研究性学习，在生物学，物理学和工程学的界面上接受培训。最终，大脑和身体如何控制运动的研究是一个可访问的背景下，参与公众和进一步融合物理学和生物学的教育，甚至在研究型大学的墙壁外。该研究使用三个实验平台（两个动物和一个机器人）。它们提供了对动物在使用不同运动模式时所面临的共同神经和机械处理挑战的见解。该研究计划包括记录和改变一个几乎完整的运动程序（所有神经肌肉命令的附件）与尖峰水平的分辨率，而动物的行为在虚拟现实环境中。PI的工作还探讨了不同的控制架构和机动性的感觉运动决定因素的性能后果。该计划将利用运动的跨学科研究，为格鲁吉亚理工学院的生命系统物理学创建一个科学的铸造厂，重点是物理学方法如何为理解神经信号提供背景。铸造厂的核心将是一个垂直整合的本科生研究团队，将完成基于研究的教育到课堂启发研究的循环。为了将影响转移到大学之外，研究人员将与当地高中教师合作，根据高速成像、表面肌电图和运动研究的可及性，创新新的教育工具。

Animals move with remarkable stability and agility through nearly every environment. To do so, they rely on interacting neural and mechanical systems that must operate in the context of the physics of sensing, actuation, and environmental interactions. Modern data acquisition tools provide unprecedented access to the underlying neural, muscular, and mechanical signals that implement control. However, these signals alone are not a framework for understanding how sensory information transforms into motor outputs. This CAREER project leverages several emerging experimental and analytical techniques to show how locomotor dynamics emerge from this sensorimotor transform. This framework for extracting principles of sensorimotor control is based on combining high-resolution neuromuscular recordings, information theoretic and dimensionality-reduction data analytics, and the formal language of system identification and control theory. The experimental framework of this proposal is transferrable to other living and engineered systems composed of many subsystems connected by feedback such as gene or protein networks, cell mechanics, or population dynamics. Principles of locomotor control that emerge from this work will synergize with the BRAIN initiative and similar programs by providing context through which to interpret a deep, detailed understanding of brain structure, anatomy, and connectivity. They will also address challenges in the new era of robotics and neural engineering. There is a pressing need for neuro-technologies that enable versatile movement while embodied in physical systems. More broadly this work will enable translation of research-based undergraduate learning into scientific and education products. It will bring a neuroscience component to the growing Physics of Living Systems curriculum at Georgia Tech. Within the VIP program, students will receive training at the interface of biology, physics, and engineering by engaging in research-based learning that is organized into a vertical mentoring system. Ultimately the study of how brain and body control movement is an accessible context in which to engage the public and further converge education of physics and biology even outside the walls of the research university.The research uses three experimental platforms (two animals and one robot). They provide insights into the shared neural and mechanical processing challenges animals face despite using different modes of locomotion. The research program includes recording and altering a nearly complete motor program (the set of all neuromuscular commands to appendages) with spike-level resolution while an animal behaves in a virtual reality environment. The PI's work also explores the performance consequences of different control architectures and the sensorimotor determinants on maneuverability. The program will use the interdisciplinary study of movement to create a scientific foundry for the Physics of Living Systems at Georgia Tech with an emphasis on how physics approaches provide a context for understanding neural signals. The core of the foundry will be a vertically integrated research team of undergraduates that will complete the loop of research-based education to classroom-inspired research. To transfer the impact outside the university, the researchers will team with local high school teachers to innovate new education tools based on the accessibility of high-speed imaging, surface electromyography, and the study of movement.