Evaluating fuzzy voice commands by user guided visual attention for controlling a robot manipulator

This paper proposes a method for evaluating fuzzy voice commands (FVCs) by user guided visual attention. FVCs inherit fuzzy linguistic information due to the inclusion of fuzzy linguistic terms such as “very little.” The contextual meaning of such information depends on the environmental arrangement. Therefore, visual attention system (VAS) is utilized to evaluate the fuzzy linguistic information based on the environmental conditions. It is implemented by assuming that the corresponding distance value for a particular FVC depends on the spatial arrangement of the surrounding objects in the environment. A fuzzy logic based voice command evaluation system (VCES) is proposed to assess the fuzzy linguistic information in user commands. In addition, the user has been given the freedom to adjust command evaluation process by changing the attention level. It is achieved by using attentive voice commands such as “move carefully.” A situation of object manipulation by navigating the tip of a robot manipulator through the user´s working space is demonstrated to illustrate the system where a PA-10 robot manipulator is used.