Distribution Based Performance Analysis of Control Allocation Strategies Under Failure Conditions

In this paper, means of consolidated analysis are presented for control allocation algorithms. Control allocation algorithms solve the problem of computing the control vector given the pseudo control vector. A large variety of algorithms exist such as pseudo inverse based, linear programming based or direct allocation algorithms. In order to choose the appropriate one for each individual application, a thorough investigation is necessary. In the case of systems with redundant actuators a systematic analysis of actuator failures must be performed. This paper introduces a novel approach of generating the distribution of certain performance metrics among all failure cases. The system on which the control allocation was tested is an eVTOL aircraft with twelve propellers.