A DRFSM DEDS algorithm is used to coordinate the movement of the robot sensor and the probe. Feedback is provided to the robot arm, based on visual observations, so that the object under consideration can be explored. This DRFSM was generated by GIJoe, a graphical tool for building FSMs that was modified to build DRFSMs[5]. The output of GIJoe was compiled and linked with a text-only driver for testing and then linked directly to the experimental sensing code.
The DEDS control algorithm will also guide the probe to the relevant parts of the objects that need to be explored in more detail (curves, holes, complex structures, etc.) Thus, the DEDS controller will be able to model, report, and guide the robot and the probe to reposition intelligently in order to recover the structure and shape parameters. The data and parameters derived from the sensing agent are fed into the CAD system for designing the geometry of the part under inspection.
The first step in running this experiment was setting the lighting conditions as desired (same conditions when constructing the reflectance map table), then initializing the robot and the camera and set them to initial positions. The experiment starts by taking images for the object from two positions, to generate two sets of contours to be fed into the stereo module for depth estimation. Using the stereo module with the assistance of the reflectance map table and the camera calibration module, an initial set of world coordinates for these contours is generated. Next, the DRFSM DEDS machine drives the probe and the robot arm holding the camera to inspect the object using the information generated from the stereo module and the relation between the object's contours.
The controlling DEDS machine performed as expected. Results for one of the two models used is shown in Figure 6.