The High Redundancy Actuator (HRA project investigates the use of large numbers of small actuation elements to achieve fault tolerance. The large number of components involved poses a unique challenge from a control perspective. This paper presents the two main options to control the HRA: using robust control (passive fault tolerance), and reconfigurable control (active fault tolerance). The robust controller is designed using H℞ methods, and handles the different system behaviours of the HRA with only small changes to the closed-loop system. In contrast, control reconfiguration detects the fault and changes the control laws accordingly. Multi-Agent System (MAS) concepts are used to apply localised multiple-model control and fault detection on an individual element level. The results of both approaches are compared to illustrate the trade-off between the complexity of the control approach and the resulting performance under different fault situations.