Dynamic Maneuvers in Legged Locomotion

快速启动、停止和改变方向等动态动作是许多哺乳动物在高速追逐或飞行情况下的重要行为。在这个项目中，我们将通过理论分析、数值模拟和使用机器人舞蹈机进行实验相结合的方式来研究这些动作。这项工作的重点将放在非重复动作上，而不是在疾驰或小跑时发生的周期性运动。冲量-动量模型，结合弹道轨迹，将作为我们理论分析的基础。进化机器人策略，包括遗传算法，将与高速动态模拟相结合，以进化动态动作。我们将设计并建造一台相对较小的实验性舞蹈机，以进一步研究此类系统的动力学。

Dynamic maneuvers such as rapid starts, stops, and changes of directions are important behaviors for many mammals in high-speed pursuit or flight situations. In this project we will study these maneuvers by means of a combination of theoretical analysis, numerical simulation, and experiment using robotic galloping machines. The focus of the work will be on non-repetitive maneuvers, as opposed to the cyclic motions that occur when galloping or trotting. An impulse-momentum model, combined with ballistic trajectories, will be used as the basis for our theoretical analysis. Evolutionary robotics strategies, including genetic algorithms, will be used in combination with high-speed dynamic simulation to evolve the dynamic maneuvers. We will design and build a relatively small experimental galloping machine to further investigate the dynamics of such systems.