Voice-controlled object tracking smart robot

Abstract

Object tracking robots, if can be controlled smartly through voice, can be of tremendous help for the physically handicapped people. In this paper, we try to implement a voice-controlled object tracking robot. A speech recognition system is used to recognize a set of predefined commands such as forward, backward, left, right and rotation at a particular turning angle. The robot navigates its way as per the voice-command signal, while also tracking the desired object. The voice-command signal processing is carried out in real-time, using an on-time cloud server that converts it to text form. The command signal text is then transferred to the robot via Bluetooth network to control its differential drive. The prototype smart robot consists of three sub-systems: speech recognition system, object tracking system and the differential-drive based movement's control system. The accuracy and efficiency of the speech recognition system is examined through a set of experiments. The effect of factors such as noise and distance etc. are examined, with encouraging results. The prototype robot can recognize the voice-commands within a Bluetooth range, i.e, 10 m. Possible extensions are also discussed that could lead to a wide range of further applications.