Design of a gesture controlled robotic gripper arm using neural networks

The aim of this work is to propose a method to build an efficient Bangla voice controlled robotic gripper mechanism using Neural Networks. Robots are becoming an essential part of many industries and fields. Presently, various ways are used to control one. The most user friendly one of them is controlling it by voice commands. Though voice controlled robots are becoming a popular concept now, construction of Bangla voice controlled robots is still a new idea. Controlling the robot with voice commands along with visual feeds helps the robot to operate easily and more accurately. This robot consists of three modules: speech command recognition module, object classifier module and robotic gripper arm module. At first, the robot takes voice commands on which objects to grab and displace; then it finds the object using the object classifier module. And finally it grabs and displaces the object using the robotic gripper arm module. The speech recognition module and the object classifier module uses two distinct neural networks along with additional hardware to perform their tasks. This paper presents the design and fabrication process of the robot discussed so that robots can be made using this design that works under different situations. This robot can be used to perform tasks with a high efficiency on both industrial and domestic levels.