Artificial reflexes

An alternative is to model the process of locomotion as a black box with multiple inputs and outputs. Tomovic [82] proposed a nonnumerical control method, called artificial reflex control (ARC). ARC (Figure 15.4) refers to a skill-based expert using rules that have an Then structure [83]. The locomotion is presented as a sequence of discrete events. Each of these discrete events can be described in terms of a sensory pattern and a motor activity. A sensory pattern occurring during particular motor activity can be recognized with the use of human-made (artificial or natural) sensors. The specific discrete event is called... 

Artificial reflex control Expert system using rule-based control. 

Artificial reflex control A sensory-driven control algorithm based on knowledge representation... 

Kyberd and Chappell (1994) use a system they call hierarchical artificial reflexes to automate the control process. In their multifunctional hand, they take the operator out of Ihe loop and use onboard processing and sensors in the hand to tell die hand what grasp pattern to adopt. The operator only provides a conventional single degree of freedom open or close EMG signal. The idea is that by allowing the processor to take control, it reduces the mental loading on the operator. A major factor in the success or failure of these devices is confidence in the mechanism on the part of the user, so as to relinquish control to the artificial reflex. 

Aperia, A, 10 3 Aplysia abdominal ganglion R15cell,22-ll Apte, S.M., 62 5 ARC (artificial reflex control),... 

---