School of Science and Engineering, University of the Sunshine Coast, Sippy Downs, QLD 4558, Australia.
Data61, The Robotics and Autonomous Systems Group, CSIRO, Fortitude Valley, QLD 4006, Australia.
Integr Comp Biol. 2021 Nov 17;61(5):1769-1782. doi: 10.1093/icb/icab037.
Tail movement is an important component of vertebrate locomotion and likely contributes to dynamic stability during steady-state locomotion. Previous results suggest that the tail plays a significant role in lizard locomotion, but little data are available on tail motion during locomotion and how it differs with morphological, ecological, and phylogenetic parameters. We collected high-speed vertical climbing and horizontal locomotion video data from 43 lizard species from four taxonomic groups (Agamidae, Gekkota, Scincidae, and Varanidae) across four habitats. We introduce a new semi-automated and generalizable analysis pipeline for tail and spine motion analysis including markerless pose-estimation, semi-automated kinematic recognition, and muti-species data analysis. We found that step length relative to snout-vent length (SVL) increased with tail length relative to SVL. Examining spine cycles agnostic to limb stride phase, we found that ranges of inter-tail bending compared with inter-spine bending increased with relative tail length, while ranges of tail deflection relative to spine deflection increased with relative speed. Considering stepwise strides, we found the angular velocity and acceleration of the tail center of mass increased with relative speed. These results will provide general insights into the biomechanics of tails in sprawling locomotion enabling biomimetic applications in robotics, and a better understanding of vertebrate form and function. We look forward to adding more species, behaviors, and locomotor speeds to our analysis pipeline through collaboration with other research groups.
尾巴运动是脊椎动物运动的一个重要组成部分，可能有助于在稳态运动中提供动态稳定性。先前的结果表明，尾巴在蜥蜴运动中起着重要作用，但关于运动中的尾巴运动以及它如何与形态、生态和系统发育参数不同的数据很少。我们从四个分类群（鬣蜥科、壁虎科、石龙子科和巨蜥科）的 43 种蜥蜴中收集了高速垂直攀爬和水平运动的视频数据，涵盖了四个栖息地。我们引入了一种新的半自动化和可推广的尾巴和脊柱运动分析管道，包括无标记姿势估计、半自动运动识别和多物种数据分析。我们发现，相对于头长（SVL）的步长随着尾巴相对于 SVL 的长度而增加。在不考虑肢体跨步相位的情况下检查脊柱周期，我们发现相对于脊柱弯曲的尾巴弯曲范围随着相对尾巴长度的增加而增加，而尾巴相对于脊柱的挠度范围随着相对速度的增加而增加。考虑到逐步跨步，我们发现尾巴质心的角速度和加速度随着相对速度的增加而增加。这些结果将为伸展运动中尾巴的生物力学提供一般见解，为机器人仿生学应用提供支持，并更好地理解脊椎动物的形态和功能。我们期待通过与其他研究小组的合作，在我们的分析管道中添加更多的物种、行为和运动速度。
