Input-to-task redundancy and dynamic control allocation for multi-agent systems

This paper introduces the notion of input-to-task redundancy in multi-agent systems and extends the dynamic control allocation paradigm to input-constrained multi-agent systems. In particular, it is shown that a proper selection of a subset of the system outputs can entail a weak redundancy condition with respect to a given task to be executed, thereby providing additional degrees of freedom to cope with potential input constraints. Moreover, we propose a way of quantitatively characterizing the additional freedom in the controllability of the system by means of a redundancy degree suitably defined. The efficacy of the approach has been tested and validated by numerical simulations in a platooning application.