When a system controlled by a low-frequency digital feedback controller experiences a sudden large disturbance such as an impact, its dynamic performance and stability may deteriorate. Inspired by various reflex actions in human body, analog reflex actions are designed to compensate the effects of large disturbances to digital controllers. The human reflexes are triggered by motor neurons as soon as sensory neurons detect stimuli above a threshold without any decision process involving the brain. In this approach, large disturbances are attenuated more rapidly by external sensor-actuator modules connected with only analog components. A series of simulations is performed to design the reflex actions and its findings are articulated in this paper. The system performance is evaluated by measuring the peak error and the settling time. The paper presents a discussion on various techniques to determine appropriate reflex actions with only minimal information about the disturbance profiles.

This content is only available via PDF.
You do not currently have access to this content.