Fusion of voice, gesture, and human-computer interface controls for remotely operated robot

This paper presents an overview of a robot teleoperation system using voice, gesture, and human-computer interface (HCI) controls. The system consists of three basic software components including (a) acquisition and recognition of control commands from multiple inputs, (b) client-server network communication, and (c) command fusion and execution by a robot and its arm. The inputs for recognition of control commands come from (1) wired or wireless microphones, (2) wired orientation sensors mounted on human arms, and (3) HCI devices, such as a mouse, a keyboard or a text file with the sequence of control commands. The set of gesture commands is based on the US Navy lexicon for navigating aircrafts on the ground. Fusion of multiple commands is performed by (a) analyzing time delays and (b) assigning different priorities to commands and the clients issuing those commands. Consistent and conflicting commands are considered before a selected command is executed by a robot. For an emergency control, a video signal is sent to a monitoring station.