Texture-Independent Vision-Based Closed-Loop Fuzzy Controllers for NavigationTasks

摘要

The paper deals with vision-based closed-loop control schemes for collisionavoidance as well as maintenance of clearance in a-priori unknown textured environments. These control schemes are base on fuzzy logic and employ a visual motion cue, they call the Visual Threat Cue (VTC) that provides some measure for a relative change in range as well as clearance between 3D surface and a fixated observer in motion. It is a collective measure obtained directly from the raw data of gray level images, is independent of the 3D surface texture and needsno optical flow information, 3D reconstruction, segmentation, feature tracking or preprocessing. This motion cue is scale-independent, rotation independent and is measure in (time-1) units. Design of a closed-loop conventional controller for vision based navigation tasks pose a problem as the system is complex and ill-defined. On the other hand fuzzy control which is closer in spirit to human thinking and can implement linguistically expressed heuristic contrl policies directly without any knowledge about the dynamics of the complex process. The fuzzy controllers were tested in real time using a 486-based Personal Computer and a camera capable of undergoing 6-DOF motion. Results are highly encouraging.