Fuzzy Control of a Robotic Finger Actuated by Shape Memory Alloy Wires

This paper presents the development of a robotic finger driven by nonconventional actuators, consisting of thin shape memory alloy (SMA) wires. In order to monitor and control the angles formed by each phalanx, a specific system for capturing and interpreting digital images was implemented. By image processing, this system is capable to determine the angles without the need for installation of phalanx rotation sensors, leading to weight and volume reduction of the prototype. For this artificial vision system, a simple camera with a fuzzy logic control technique was used, which was very effective in monitoring the position of the robotic finger.