Record Details     « New Search       Brief Record Full Record MARC Record     Bibliographic Data 001    UPD-00014951850 005    20090717042358.0 008    080209s1996 ph a 000 eng 041 ## $aeng 040 ## $aDML 042 ## $aDMLUC 090 ## $aLG 995 1996 E6$bA53 100 1# $aAndales, Zaldy A. 245 13 $aAn auto-orienting error-recovering fuzzy navigation system for initially unknown environments :$bdesign and simulation /$cZaldy A. Andales. 260 ## $c1996. 300 ## $a185 leaves :$b ill. 500 ## $aComputer print-out. 502 ## $aThesis (M.S.)--University of the Philippines, Diliman. 520 ## $aThe main objective of the study was to device and integrate fuzzy logic-based auto-orienting and error-recovering schemes for a fuzzy navigation system of an Automated Guided Vehicle (AGV) placed in an initially unknown, stationary and flat environment. As a testbed for simulation, AGV and environment simulators were created. As a basis for the fuzzy navigation rules, the AGV assumed four inputs (three ultrasonic range sensor values and the target direction) and two outputs (AGV direction and speed). A single photo sensor was assumed for detecting targets. A fuzzy navigation system was implemented employing 81 IF-THEN rules. For the auto-orienting scheme, it was assumed that multiple targets were placed near perimeter walls, and a fuzzy wall-following system was successfully implemented to automatically detect all targets and map out their locations. For error-recovery, a fuzzy spiral search system was successfully devised for recovery from shift and dislocation errors. This strategy required the AGV to scan its vicinity by following a spiral path until the target was found. A fuzzy wall-following search system, on the other hand, successfully provided recovery from timeouts and trap errors. Thus all errors were effectively recovered from with minimal assumed sensors installed. These schemes were integrated with the normal fuzzy navigation system to form the Integrated Fuzzy Navigation System (IFNS). Switching between schemes was managed by an over-all task planner. To evaluate the IFNS, an A* Heuristic Search-based Configuration Space System was implemented (ACSS). The ACSS, being an optimal path finding technique served as the standard. Results showed that the path efficiency was 84.9899% with respect to the optimal. However, by employing sub-goals or intermediated targets as guides, the path length efficiency improved to 94.7859%. Sub-goals also proved to be effective in eliminating trap and timeout situations. 650 #0 $aAutomated guided vehicles systems. 650 #0 $aFuzzy logic. 650 #0 $aFuzzy systems. 905 ## $aFI 852 ## $aUP$bDENG-II$hLG 995 1996 E6 $iA53$rRoom-Use Only$xE2-33t 905 ## $aUP 852 ## $aUP$bDARCHIVES$hLG 995 1996 E6 $iA53$rRoom-Use Only$xUARD-8605t 852 ## $aUP$bDENG-II$hLG 995 1996 E6 $iA53$rRoom-Use Only$xE2-33t 842 ## $aThesis           Physical Location University of the Philippines Diliman Main Library: University Archives LG 995 1996 E6 / A53 Diliman: College of Engineering Library II LG 995 / 1996 E6 A53 Diliman: College of Engineering Library II LG 995 1996 E6 / A53           Digital Copy Not Available          Add to Book Cart  |   Download MARC