Tilt-Prioritized Quadrocopter Attitude Control
IEEE Transactions on Control Systems and Technology(2020)
摘要
This paper presents an attitude tracking control law and control allocation strategy for quadrocopters which prioritizes the vehicleu0027s ability to achieve a desired translational acceleration. The quadrocopteru0027s attitude error is split into a reduced attitude error, which describes the misalignment of the thrust direction, and a yaw error, which describes the orientation error about the thrust direction. A model-based proportional-derivative control law is derived, where the proportional action is in terms of the reduced attitude and yaw errors, and the derivative action is in terms of the quadrocopteru0027s angular velocity error. Almost global asymptotic convergence of the reduced attitude error is established, and the convergence of the yaw error is proven. It is further shown that the attitude control law decouples the reduced attitude error dynamics from the yaw error dynamics. A control allocation strategy is derived which exploits the decoupling in order to prioritize the correction of the reduced attitude over the yaw error. The proposed control strategy is computationally lightweight and therefore well-suited to run on board quadrocopters at high rates. Experimental results demonstrate improved error recovery and position tracking performance.
更多查看译文
关键词
Aerial robotics,attitude control,control allocation,nonlinear control,quadrocopter
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络