A dynamic multi-axis control allocation scheme for real-time applications

Abstract

The Dynamic Weighting Control Allocator (DWCA) was introduced by Sadien et al. (Control Engineering Practice, 2019) to solve the allocation problem raised by the yaw control of an on-ground aircraft. It handles the classical tradeoff between virtual reference inputs realisation and control power minimisation. But it also offers three specific features. First, the actuators reach saturation almost simultaneously, which allows an efficient recovery in case of failure. Then, several industrial requirements are taken into account, such as implementation ease, low computational cost and compatibility with certification constraints. And finally, the actuators can be prioritised, the last ones being used sparingly (to avoid overheating, fatigue, maintenance cost…) only when the virtual reference inputs cannot be realised by the first ones only. But despite its attractiveness, the DWCA is limited to the 1-dimensional case, which means that only one degree of freedom can be controlled. This is not sufficient to deal with today's challenges in the aeronautical and automotive fields, for example, where certain control problems are inherently multi-dimensional. With the progressive advent of autonomous cars, new concepts of light hybrid or electric vehicles are, for example, emerging, where both hydraulic and electric actuators are mounted on each wheel to improve the combined management of speed and steering. In this context, this paper introduces a non-trivial generalisation of the DWCA, which retains the various characteristics and advantages of the initial version and can be used in the multi-dimensional case.

Keywords:

• Multi-dimensional control allocation
• weighted pseudo-inverse with dynamic weight adaptation
• actuator prioritisation

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research is supported by Airbus Operations S.A.S., the National Association of Research and Technology (ANRT) and the French Ministry of Higher Education, Research and Innovation under the CIFRE contract 2016/1058.