We present the tensegrity aerial vehicle, a design of collision-resilient rotor robots with icosahedron tensegrity structures. The tensegrity aerial vehicles can withstand high-speed impacts and resume operation after collisions. To guide the design process of these aerial vehicles, we propose a model-based methodology that predicts the stresses in the structure with a dynamics simulation and selects components that can withstand the predicted stresses. Meanwhile, an autonomous re-orientation controller is created to help the tensegrity aerial vehicles resume flight after collisions. The re-orientation controller can rotate the vehicles from arbitrary orientations on the ground to ones easy for takeoff. With collision resilience and re-orientation ability, the tensegrity aerial vehicles can operate in cluttered environments without complex collision-avoidance strategies. Moreover, by adopting an inertial navigation strategy of replacing flight with short hops to mitigate the growth of state estimation error, the tensegrity aerial vehicles can conduct short-range operations without external sensors. These capabilities are validated by a test of an experimental tensegrity aerial vehicle operating with only onboard inertial sensors in a previously-unknown forest.
翻译:我们展示了紧张的空中飞行器,这是一个具有相撞时态紧张结构的抗撞旋转机器人的设计。紧张的空中飞行器能够承受高速撞击,并在碰撞后恢复运行。为了指导这些空中飞行器的设计过程,我们提出了一个基于模型的方法,通过动态模拟预测结构中的压力,并选择能够承受预测压力的部件。与此同时,建立了一个自主的重新定向控制器,以帮助紧张的空中飞行器在碰撞后恢复飞行。调整控制器可以将地面上的车辆从任意方向旋转到容易起飞的车辆。如果具有碰撞复原力和再定向能力,紧张的空中飞行器可以在封闭的环境中运作,而没有复杂的避免碰撞战略。此外,通过采用惯性导航战略,用短跳来取代飞行,以减缓国家估计误差的增长,紧张的空中飞行器可以在没有外部传感器的情况下进行短程操作。这些能力通过实验性紧张性航空飞行器运行的试验得到验证,而试验时只能在已知的森林内惯性传感器上进行。