I. Introduction
Recently, the distributed consensus tracking control of multi-agent systems (MASs) has attracted great attention due to the extensive applications in practical engineering, and considerable achievements in the field of approximation-based adaptive consensus tracking control have emerged for MASs with uncertainties, by incorporating neural networks (NNs) or fuzzy logic systems (FLSs) approximators into conventional backstepping technique [1]–[2]. Concretely, in [1], based on NNs, an adaptive consensus control method is proposed for a class of nonlinear multi-agent systems with state time-delay. In [2], the adaptive fault tolerant consensus control of high-order multiagent systems with input quantization and time-varying parameters is studied by using NNs-based approximators. However, all the aforementioned dynamics are low-power (i.e., the system powers are equal to one). In contrast to low-power dynamics, most notably, power-chained nonlinear MASs (i.e., the system powers are positive odd integers greater than one) are generalization of aforementioned ones. For this family, a parametric adaptive control technique called adding-one-power-integrator is originally proposed in [3]. Unfortunately, the existing works [4]–[7] about power-chained nonlinear dy-namics only consider the scenario of single agent. On the other hand, switched dynamics is a current research highlights topic, due to the versatility in the modelling of complex engineering systems [8]–[10].