A Simple Learning Strategy for HighSpeed （一个简单的高速的学习策略）.pdfVIP

A Simple Learning Strategy for High-Speed Quadrocopter Multi-Flips ¨ Sergei Lupashin, Angela Schollig, Michael Sherback, Raffaello D’Andrea Abstract— We describe a simple and intuitive policy gradient method for improving parametrized quadrocopter multi-flips by combining iterative experiments with information from a first-principles model. We start by formulating an N-flip maneuver as a five-step primitive with five adjustable pa- rameters. Optimization using a low-order first-principles 2D vertical plane model of the quadrocopter yields an initial set of parameters and a corrective matrix. The maneuver is then repeatedly performed with the vehicle. At each iteration the state error at the end of the primitive is used to update the maneuver parameters via a gradient adjustment. The method is demonstrated at the ETH Zurich Flying Machine Arena testbed on quadrotor helicopters performing and improving on flips, double flips and triple flips. I. INTRODUCTION Our objective is to use a low-order, first-principles model of a quadrocopter in order to be able to perform and improve upon single, double and triple flips. In particular we desire a formulation of a flip primitive such that it is able to return the quadrocopter exactly to its initial state, plus a radians change in rotation about one of its principal axes. In addition, we seek an approach that avoids complex online computations and does not require or attempt to track an a priori known feasible trajectory. Miniature quadrotor helicopters in both indoor and outdoor environments are a popular and challenging autonomous aerial research platform. Several established quadrocopter research gr