FIBR: Neuromechanical Systems Biology

在过去的半个世纪里，从分子马达到运动过程中的大脑图像，我们见证了前所未有的发现。尽管取得了如此非凡的进展，但不同规模的众多结构如何相互作用，产生动物的标志性行为--运动--仍然难以捉摸。虽然运动可以解构为一个简单的级联--从大脑到肌肉，再到外部世界--但这种单向框架未能纳入各层次之间和内部网络相互作用所产生的基本动态特性。最大的挑战是阐明允许动物运动的复杂的动力系统网络。这项研究将通过对奔跑的昆虫进行物理干扰，用虚拟地形挑战它们，在测量运动、力和神经肌肉信号(http://polypedal.berkeley.edu/NSB).)的同时退化和重写神经代码，从而揭示基本的控制结构该项目利用动力系统方法与我们引人注目的六足腿机器人作为控制实验，比在动物中更容易操作参数，但将物理模型置于真实环境中。即使生物学家采取多层次、多维度的方法，该项目所陈述的挑战也不能仅在生物学领域内满足。这一挑战需要多学科的努力，以匹配数学模型、数值模拟、物理模型以及生物实验中的数据。该项目的团队吸收了所有这些学科的专家。广泛的影响：对于更广泛的社区，该项目将提供适用于基因调节、新陈代谢和心脏病学以及神经科学和运动的动态模型的计算工具。这些发现将刺激神经机械系统生物学的一个新领域，启发工程学中的新控制器，导致假肢和人造肌肉的开发，并进一步设计出第一个具有真正与动物相媲美的性能的搜救机器人。最重要的是，该项目将培养出一种新型的年轻科学家，他们能够领导这一新兴的多学科前沿。这些发现将通过出版、座谈会、基于网络的旗舰课程和教育机器人竞赛来分享。

The last half-century has witnessed unprecedented discoveries from molecular motors to brain images during movement. Despite such extraordinary advances, how the multitudes of structures at different scales interact to generate an animals' hallmark behavior - locomotion - remains elusive. While locomotion can be deconstructed into a simple cascade - from brain to muscles to work on the external world - such a unidirectional framework has failed to incorporate essential dynamic properties created by an interaction of networks among and within levels. The grand challenge is to elucidate the complex network of dynamical systems that allows animal locomotion. This research will uncover fundamental control architectures by physically perturbing running insects, challenging them with virtual terrains, degrading and rewriting the neural codes while measuring motion, forces and neuromuscular signals (http://polypedal.berkeley.edu/NSB). The project capitalizes on a dynamical systems approach in concert with our remarkable hexapedal, legged robot as a controlled experiment, manipulating parameters more easily than in animals, but subjecting the physical model to real environments. The project's stated challenge cannot be met within the domain of Biology alone, even if biologists take a multi-level, multidimensional approach. The challenge demands a multidisciplinary effort to match data across mathematical models, numerical simulations, physical models, as well as biological experiments. The project's team incorporates experts across the full spectrum of these disciplines.Broader Impacts: To the broader community, the project will deliver computational tools applicable to dynamical models of gene regulation, metabolism, and cardiology, as well as neuroscience and locomotion. Findings will stimulate a new field of neuromechanical systems biology, inspire novel controllers in engineering, lead to development of prostheses and artificial muscles and further the design of the first search and rescue robot that has performance truly comparable to animals. Most importantly, the project will produce a new type of young scientist who can lead this emerging multidisciplinary frontier. Discoveries will be shared through publication, symposia, a web-based flagship course, and educational robotics contests.