Design and performance comparison of fuzzy logic based tracking controllers

Several camera tracking controllers based on fuzzy logic principles have been designed and tested in software simulation in the software technology at the Johnson Space Center. The fuzzy logic based controllers utilize range measurement and pixel positions from the image as input parameters and provide pan and tilt gimble rate commands as output. Two designs of the rule base and tuning process applied to the membership functions are discussed in light of optimizing performance for a v-bar approach testcase. The two controller designs are compared in terms of responsiveness, and ability to maintain the object in the field-of-view of the camera.<
>