In recent years, Unmanned Aerial Vehiclesud(UAVs) are emerged as an attractive technology for differentudtypes of military and civil applications which have gainedudimportance in academic researches. In these emerging researchudareas, UAV autonomy gets a great part and mainly it refersudthe ability for automatic take-off, landing and path planningudof UAVs. In this paper, we focused of the path planning ofudUAVs for controlling a number of waypoints in the missionudarea. If the area is large and the number of points that must beudchecked is greater, then it is not possible to check every possibleudsolution, therefore, we have to use some efficient algorithms, likeudgenetic algorithms (GAs), to calculate the path. However if theudnumber of these points exceeds a certain number, then we haveudto use some additional accelerating mechanisms to speed up theudcalculation time. Typically two techniques are used for speedingudup: parallelization and distribution of calculation. In this paperudgenetic algorithm is parallelized on CUDA architecture byudusing Graphical Processing Units (GPUs). Experimental resultsudshowed that this approach produces efficient solutions in a shortudtime.
展开▼
