Drone technology has undoubtedly become an integral part of our everyday life in recent years. The business and industrial use of unmanned aerial vehicles (UAVs) can provide advantageous solutions in many areas of life, and they are also optimal for emergency situations and for accessing hard-to-reach places. However, their application poses numerous technological and regulatory challenges to be overcome. One of the weak links in the operation of UAVs is the limited availability of energy. In order to address this issue, the authors developed a novel trajectory planning method for UAVs to optimize energy consumption during flight. First, an “energy map” was created, which was the basis for trajectory planning, i.e., determining the energy consumption of the individual components. This was followed by configuring the 3D environment including partitioning of the work space (WS), i.e., defining the free spaces, occupied spaces (obstacles), and semi-occupied/free spaces. Then, the corresponding graph-like path(s) were generated on the basis of the partitioned space, where a graph search-based heuristic trajectory planning was initiated, taking into account the most important wind conditions including velocity and direction. Finally, in order to test the theoretical results, some sample environments were created to test and analyze the proposed path generations. The method eventually proposed was able to determine the optimal path in terms of energy consumption.

The genetic algorithms (GAs), as the optimization methods, are expansively used in mobile robot techniques. In this paper a previously developed, multi-agent based map building process optimizing GA [1] will be revised in form of GA algorithm optimization and GA's parameter tuning. The results of the algorithm-improving will be compared and analyzed in the conclusion.

Drone technology has undoubtedly become an integral part of our everyday life these days. The business and industrial use of unmanned aerial vehicles (UAV) can provide advantageous solutions in many areas of life, and they are also optimal for emergency situations and for accessing hard-to-reach places. However, their application poses numerous technological and regulatory challenges to be overcome. One of the weak links in the operation of UAVs is the limited availability of energy. In order to address this issue authors developed a novel trajectory planning method for UAVs to optimize the energy necessity during flight. First, an “energy map” was created, which was the basis for trajectory planning, i.e., determining the energy consumption of the individual components. This was followed by configuring the 3D environment including partitioning of the work space (WS), in fact, defining the free spaces, occupied spaces (obstacles) and semi-occupied/free spaces. Then the corresponding graph-like path(s) were generated on the basis of the partitioned space, where a graph search-based heuristic trajectory planning was initiated, taking into account the most important wind conditions including velocity and direction. Finally, in order to test the theoretical results, some sample environments were created to test and analyse the proposed path generations. The method eventually proposed was able to determine the optimal path in terms of energy consumption.

Any well defined environment can be extended to an Intelligent Space (iSpace) if it becomes able to collect information through ambient sensors, cameras, agents, traditional and special interfaces to process the knowledge at local level, to take actions at local level and/or in cooperation with other units and to send an alert to a higher level surveillance system. The iSpace can, thus, offer a frame for different intelligent applications implemented on the computer system of the space. This paper gives an overview about several educational initiations and research results of the iSpace laboratory built at the Institute of Mechatronics and Vehicle Engineering, Óbuda University. The presented results cover a wide range of topics including the development of an intelligent eto robot-room; MobiRobi, the voice controlled mobile robot; a user-friendly home surveillance system; and a new modification strategy of the illumination applied in the room in order to improve the color vision of defected persons.

In this paper face recognition and the tracking process of the selected face are described. In the beginning, face recognition is detailed in a static environment, where different effective face recognition methods are shown, such as LBP and HOG combination. A method of selecting, following, and tracking a face in a dynamic environment is further discussed. For this, a CNN was used (trained) for correct recognition instead of the initially tested HOG method, as it proved more adaptable to shifts in perspective and lighting on faces. The methods have also been evaluated in practice using the drone described in chapters II.B, and III.

Until now, the region where the mobile agent was capable to travel has been established like the result of subtraction between the areas of the “workspace, WS” minus “configured obstacles, CB”. This region is known like “aligned free space” (AFS). Let we suppose the following path planning methods: BUG algorithm, TBA algorithm, SFP algorithm and so on. The common feature of these algorithms is, that each of them, planning the path in near close to the obstacle. In these cases, if we will not consider the localization's accuracy, it can happen, that mobile agent will hits with obstacle. In this paper, first of all the navigation's area (NA) will be defined, and then will be suggested, how is it possible to create a navigation's area with assistance of fuzzy set. Actually the fuzzy set will be assigned to the “Voronoi diagram” of the reduced (aligned) free space (AFS) of the environment, or it is possible to reverse: that is in maximum of the fuzzy set can be created the centre line (Voronoi diagram) of the reduced free space.