Experimental and Computational Investigation of Closed Loop Flight Control in the Hawkmoth Manduca Sexta

Tyson Hedrick博士将研究动物如何在有限的不确定执行器和具有固有和可变延迟的感觉系统的情况下实现各种稳定的运动行为。具体来说，我们将在闭环、自由飞行环境下对Manduca sexta的飞行控制进行研究，采用计算和实验相结合的方法，利用工程控制理论对系统进行分析。取食的飞蛾在花前盘旋，从中吸取花蜜，因此必须跟踪花的位置和方向。当飞蛾在花朵附近盘旋时，在轻微微风的扰动下，甚至是飞蛾自己的翅膀产生的向下气流的扰动下，它们的花朵也会振荡。因此，飞蛾必须发展出一种有效的跟踪行为，这种行为可以在实验室条件下被激发出来，在实验室条件下，飞蛾跟踪机械驱动的人造花的运动。这允许直接实验操作和测量系统输入（花的位置）和输出（飞蛾的位置）在生物相关的整个生物体行为。此外，已知花的追踪具有光学基础而不是触觉基础，并且光学路径受到50至150毫秒的大延迟，或2至6次翼拍。飞行中的方向也是通过光学途径部分感知的，并且受到同样的延迟。这些延迟可能会支配控制器的功能，以至于神经和感觉系统的限制需要比单独考虑的生物力学系统所需的阻尼更大的阻尼。这些不同因素的影响将通过扩展先前开发的Manduca sexta飞行的动态模拟来评估；该计算模型将用于根据飞蛾的空气动力学和最大肌肉功率输出的物理可能性来评估推断的飞行控制传递函数。这些结果将为研究自由行为动物的闭环控制建立一个新的系统，并将对寻求将控制理论应用于各种尺度的生物系统的研究人员产生兴趣。这项拟议的研究产生的更广泛的影响包括通过实验室助理的方式，让高中生和本科生接触到动手的有机生物学实验方法。这项研究的结果将在PI的实验室网站上公布，在国内和国际会议上发表，并在科学期刊上发表。最后，本研究的结果将应用于微型飞行器的飞行和控制以及其他生物仿生工程的努力。

Dr. Tyson Hedrick will study how animals achieve a variety of stable locomotor behaviors with a limited set of actuators of uncertain performance and a sensory system with inherent and variable delays. Specifically, the flight control of the hawkmoth Manduca sexta will be examined in a closed loop, free flight context using a combination of computational and experimental methods to analyze the system using engineering control theory. Feeding hawkmoths hover in front of the flower from which they draw nectar, and therefore must track the position and orientation of the flower. The flowers from which the hawkmoth feeds also tend to oscillate under the perturbation of a slight breeze or even the downward airflow generated by the moth''s own wings as it hovers near the flower. Thus, the hawkmoth has had to develop an effective tracking behavior, which can be elicited under laboratory conditions where the moth tracks the movements of a mechanically actuated artificial flower. This allows direct experimental manipulation and measurement of both the system input (flower position) and output (moth position) in a biologically relevant whole organism behavior. Furthermore, flower tracking is known to have an optical rather than tactile basis, and optic pathways are subject to large delays of 50 to 150 milliseconds, or 2 to 6 wingbeats. Orientation in flight is also partially sensed via optical pathways and is subject to these same delays. These delays may dominate the function of the controller to the extent that limitations in the neural and sensory systems require damping beyond what would be necessary for the biomechanical system considered separately. The influence of these different factors will be evaluated via extension of a dynamic simulation of the flight of Manduca sexta previously developed; this computational model will be used to evaluate the inferred flight control transfer functions in light of what is physically possible given the moth''s aerodynamics and maximum muscle power output. These results will establish a new system for examining closed loop control in freely behaving animals and will be of interest to researchers seeking to apply control theory to biological systems at a variety of scales. The broader impacts resulting from this proposed research include the exposure of high school and undergraduate students to hands-on, experimental approaches in organismal biology by way of laboratory assistanceships. Results from this research will be made available on the PI''s laboratory web site, presented at national and international conferences, and published in scientific journals. Finally, the results of this research will have applications to the flight and control of micro-air vehicles and other bio-mimentic engineering efforts.