1 Department of Mechanical Engineering, University of Washington , Seattle, WA , USA.
2 Department of Biology, University of Washington , Seattle, WA , USA.
J R Soc Interface. 2019 Apr 26;16(153):20190035. doi: 10.1098/rsif.2019.0035.
The halteres of flies are mechanosensory organs that serve a crucial role in the control of agile flight, providing sensory input for rapid course corrections to perturbations. Derived from hind wings, halteres are actively flapped and are thus subject to a variety of inertial forces as the fly undergoes complex flight trajectories. Previous analyses of halteres modelled them as a point mass, showing that Coriolis forces lead to subtle deflections orthogonal to the plane of flapping. By design, these models could not consider the effects of force gradients associated with a mass distribution, nor could they reveal three-dimensional spatio-temporal patterns of strain that result from those forces. In addition, diversity in the geometry of halteres, such as shape and asymmetries, could not be simply modelled with a point mass on a massless rod. To study the effects of mass distributions and asymmetries, we examine the haltere subject to both flapping and body rotations using three-dimensional finite-element simulations. We focus on a set of simplified geometries, in which we vary the stalk and bulb shape. We find that haltere mass distribution gives rise to two unreported deformation modes: (i) halteres twist with a magnitude that strongly depends on stalk and bulb geometry and (ii) halteres with an asymmetric mass distribution experience out-of-plane bending due to centrifugal forces, independent of body rotation. Since local strains at the base of the haltere drive deformations of mechanosensory neurons, we combined measured neural encoding mechanisms with our structural analyses to predict the spatial and temporal patterns of neural activity. This activity depends on both the flapping and rotation dynamics, and we show how the timing of neural activity is a viable mechanism for rotation-rate encoding. Our results provide new insights in haltere dynamics and show the viability for timing-based encoding of fly body rotations by halteres.
蝇类的平衡棒是机械感觉器官，在敏捷飞行的控制中起着至关重要的作用，为快速对微扰进行航线修正提供了感觉输入。平衡棒由后翅衍生而来，通过主动拍打而运动，因此当蝇经历复杂的飞行轨迹时，会受到各种惯性力的作用。之前对平衡棒的分析将其建模为一个质点，表明科里奥利力会导致与拍打平面正交的微妙偏转。由于设计原因，这些模型无法考虑与质量分布相关的力梯度的影响，也无法揭示由这些力产生的三维时空应变模式。此外，平衡棒的几何形状（例如形状和不对称性）的多样性无法通过无质量杆上的质点简单建模。为了研究质量分布和不对称性的影响，我们使用三维有限元模拟研究了平衡棒在拍打和身体旋转下的情况。我们专注于一组简化的几何形状，在这些形状中，我们改变了柄和球状体的形状。我们发现平衡棒的质量分布会产生两种以前未报道的变形模式：（i）平衡棒的扭转，其幅度强烈依赖于柄和球状体的几何形状；（ii）由于离心力的作用，具有不对称质量分布的平衡棒会经历离面弯曲，而与身体旋转无关。由于平衡棒基部的局部应变会驱动机械感觉神经元的变形，因此我们将测量的神经编码机制与结构分析相结合，以预测神经活动的时空模式。这种活动既取决于拍打又取决于旋转动态，我们展示了神经活动的时间如何成为旋转速率编码的可行机制。我们的结果提供了平衡棒动力学的新见解，并展示了基于时间的编码对蝇体旋转的可行性。
Zotero 插件
