CRCNS Research Project: Collaborative Research: Experimental, Numerical, and Robotic Study of the Role of Dynamic Load Sensing in Legged Locomotion

昆虫是高度移动的步行者，它们的肌肉输出适应不断变化的环境。它们腿上的力传感器可以检测到增加和减少的力，这可以帮助它们协调腿部并补偿外力，例如在不平坦的地面上行走时。这个计算神经科学合作研究（CRCNS）项目调查了传感器的动态，使它们能够检测力的增加和减少，测量这些动态在身体上的差异或相似之处，并探索这些动态如何影响行走的控制。竹节虫和蟑螂腿的记录将为昆虫运动的神经机器人模型提供信息。该模型是一个六足机器人，具有类似昆虫的力传感器。机器人的控制系统基于昆虫的神经系统，使研究人员能够测试动态力感知如何影响肌肉控制。学习如何将动态力传感集成到机器人的控制系统中可能会导致更成功的农业，采矿和勘探步行机器人，并可能揭示人类在运动过程中如何利用动态力传感。该项目还将通过为西弗吉尼亚州的中学生和高中生举办机器人研讨会，以及针对女性和第一代大学生的教学和指导活动，增加STEM方面的培训。为了更好地理解动态负荷感知（dF/dt）对运动神经控制的重要性，研究者将对竹节虫和蟑螂的力感知进行研究，因为竹节虫和蟑螂具有外力感知器官——campaniform sensilla (CS)，适合于实验。有证据表明，CS支持整个腿部肌肉的协同激活，增强它们以克服姿势扰动并调节它们以限制肌肉力量。为了更好地理解这种协同作用的组织，研究人员将构建昆虫腿的神经力学模型，其运动学和动力学将为它们的传入和传出记录提供必要的背景。研究人员假设，当站立阶段开始时，整个腿部的CS被调整为快速招募肌肉协同作用，使运动输出适应站立期间负荷的变化，并通过在站立阶段结束时发出减少力的信号来支持腿间协调。为了将CS模型浸入到行走身体的机械环境中，并进行在体内不可能进行的实验，研究人员将把该模型应用于昆虫drosophbot的动态缩放神经机器人模型，并在其行走时记录所有腿上的应变传感器。将比较机器人和动物之间的致动器扭矩和CS响应，以完善研究人员的动态力传感和协同肌肉募集模型。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Insects are highly mobile walkers who adapt their muscle output to their ever‐changing environment. Force sensors in their legs detect increasing and decreasing forces, which may help them coordinate their legs and compensate for external forces, for example, when walking on an uneven surface. This Collaborative Research in Computational Neuroscience (CRCNS) project investigates the dynamics of the sensors that enable them to detect force increases and decreases, measuring the differences or similarities in these dynamics across the body, and exploring how these dynamics affect the control of walking. Recordings from stick insect and cockroach legs will inform a neurorobotic model of insect locomotion. The model is a six‐legged robot with force sensors like those in insects. The robot's control system is based on the insect nervous system, enabling the investigators to test how dynamic force sensing affects muscle control. Learning how to integrate dynamic force sensing into the robot's control system may lead to more successful walking robots for agriculture, mining, and exploration, and may shed light onto how humans utilize dynamic force sensing during locomotion. This project will also increase training in STEM through robotics workshops for West Virginia middle- and high-school students, and teaching and mentoring activities focused on women and first-generation college students.To better understand the importance of sensing dynamic load (dF/dt) to the neural control of locomotion, the investigators will study the force sensing of stick insects and cockroaches, which are amenable to experimentation due to their external force sensing organs, campaniform sensilla (CS). Evidence suggests that CS support the synergistic activation of muscles throughout the leg, enhancing them to overcome postural perturbations and regulating them to limit muscle forces. To better understand the organization of such synergies, the investigators will construct a neuromechanical model of the insect leg, whose kinematics and dynamics will provide necessary context for their afferent and efferent recordings. The investigators hypothesize that CS throughout the leg are tuned to rapidly recruit muscle synergies when the stance phase begins, to adapt motor outputs to variations in load during stance, and to support interleg coordination by signaling decreasing forces as the stance phase ends. To immerse the CS model into the mechanical context of a walking body and perform experiments that would be impossible in vivo, the investigators will apply the model to their dynamically‐scaled neuro‐robotic model of an insect, Drosophibot, and record from strain sensors on all legs as it walks. Actuator torques and CS responses will be compared between the robot and animal to refine the investigators' models of dynamic force sensing and synergistic muscle recruitment.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.