Hexacopter Flight Performance Comparison with CCA vs. WCA Control Allocation

Abstract

This paper presents an extensive discussion between two control allocation methods for hexacopters. The first one is the so-called classical control allocation (CCA) based on a pseudo-inverse matrix, and the second one is the weighted control allocation (WCA) based on a weighted pseudo-inverse matrix. Control allocation is about mapping the virtual control vector $\nu=[T,\ L,\ M,\ N]^{\top}$ made of thrust and roll, pitch and yaw torque, respectively, to the individual motors' speed. The CCA method is studied under motor saturation circumstances, in which CCA can lead to a crash. The WCA method on the other hand displays better performance than CCA for: a) motor saturation, b) vehicle's mass changes, 3) flight-controller poorly-tuned parameters, and 4) motor-fault tolerance. The performance comparison between these two control allocation techniques is made through simulations and real-flight tests.