Computing with Words — A Framework for Human-Computer Interaction

Abstract

In this paper we explore the possibility of using computation with words (CWW) systems and CWW-based human-computer interface (HCI) and interaction to enable efficient computation and HCI. The application selected to demonstrate the problems and potential solutions is in the context of autonomous driving. The specific problem addressed is of a machine instructed by human word commands to execute the task of parking two manned or unmanned cars in a two-car garage using CWW. We divide the interaction process into two steps: (1) feasibility verification and (2) execution. In order to fulfill the task, we begin with verifications of feasibility in terms of assessing whether the garage is unoccupied, checking general ballpark dimensions, inspecting irregular shapes, and classifying the cars that need to be parked, in terms of size, types of vehicles, ranges of acceptable tolerances needed if the cars are manned or not, and means of collision avoidance. The execution of the autonomous driving part is directed by sensory non-numeric fuzzy information that indicates distances from walls or obstacles. The execution algorithm uses a sequence of driving instructions aimed at using the available space in a simple and efficient way without resorting to elaborate numerical calculations, such as making sure that the car is within 2 inches of the wall. The system and its usability are qualitatively analyzed. The analysis shows that the approach has a potential for reducing computational complexity and improving system usability.