Sistem ve Kontrol Mühendisliği

Permanent URI for this community

https://digitalarchive.library.bogazici.edu.tr/handle/123456789/11617

Browse

Browsing Sistem ve Kontrol Mühendisliği by Subject "Artificial intelligence."

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Comparison of path planning algorithms
    (Thesis (M.S.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2007., 2007.) Geleri, Fuat.; Akın, H. Levent.
    Path planning problems arise in many different fields such as; robotics, assembly analysis virtula prototyping, pharmaceutical drug design, manufacturing, and computer animation. Path planning algorithms aim to solve problems that involve computing a continuous sequence, a path, of configurations between an initial and goal configuration. Planning of a path involves some constraints, such as computing a collision-free path. We compared various path palnning and navigation algorithms. as reactive algorithm, an improved version of Artificial Potential Field (APF) algorithm is used. In robot coordination this algorithm is the superior algorithm. It coordinates 250 robots easily. whereas deliberative algorithms, such as Rapidly-exploring Random Tree Connect (RRT Connect) algorithm, can only coordinate 40 robots with high costs. The other deliberative algorithms, Rapidly-exploring Random Tree (RRT), Probebilistic Roadmap (PRM) and Lazy Probabilistic Roadmap (Lazy PRM), could not coordinate more than 20 robots within feasible resource and time limits in our tests. In robot coordination reactive algorithms are more succesful, but, when the environment contains local minima, using a deliberative algorithm iv inevitable. In path palnning for multiple robots, decentralized approaches, or partially grouping of the robots show better performances. As the number of the controlled robots in the environmental increases, using decentralized approaches becomes a requirement, because the amount of the required time and the resources increases exponentially in centralized approaches, but linearly in decentralized approaches. partially grouping of the robots gives the best performance results, because the resource requirements increase nearly linear, and nearby robots are controlled in centralized manner
  • Thumbnail Image
    Item
    Design and simulation of a mobile tour-guide robot for museum guidance
    (Thesis (M.S.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2006., 2006.) Sarı, Serdar.; Akın, H. Levent.
    This thesis aims at building a realistic design of a fully autonomous social mobile robot act as a tour guide to the visitors in museums and that can give information about the museum and the exhibits. It should define all the parts robot need such as locomotion mechanism, sensors, batteries, motors, body, control algorithms and methods required to accomplish its objective. Three dimensional modeling of the robot according to this design, building a virtual museum and dynamic agents, based on a real museum, in a simulation environment for further research and progress are also accomplished. Also, to verify the design, a fuzzy controller based on a layered behavior-based approach, has developed to control the robot. The behaviors used by the robot include going to target point, avoiding obstacle, avoiding human in front of it by speaking and waiting the visitors behind who are following him.
  • Thumbnail Image
    Item
    Design and simulation of a multi-agent autonomous robot system for industrial facilities
    (Thesis (M.S.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2006., 2006.) Özkan, Sarp Baran.; Akın, H. Levent.
    In this study we aim to observe the usage of multi-agent autonomous robots in industrial facilities and the related alternative application details of this usage. The purpose of the utilization of multi-agent autonomous robots in manufacturing plants is to pass over the bottlenecks existing in the usage of Automated Guided Vehicle (AGV) systems, which are being used in transportation of materials in such systems. Multi-agent autonomous robots are a current research issue in the field of robotics. Unlike AGV systems, determining their routes without using fixed transmitters, autonomous robots will be able to move liberally between loading and unloading points. Mobility in manufacturing environment very frequently causes the blockage of the route of AGV systems or gives way to the impossibility of using such environment for other activities. In case of interruption of paths by fixed obstacles or dynamic obstacles described as workers, robots will be able to continue towards their targeted destination either by maneuvers of avoidance or by determining a new route. Consequently this situation will provide a new facility for a more flexible manufacturing. For modeling the environment in our study, we have applied Webots, a 3-D simulation program that, both with its graphical interface and its C, C++ and Java programming units, is used to create a test environment for the applications of units, various production environment layouts, robotic elements trials, robot architecture and different path planning and task assignment strategies. In our studies, robots and production environment have been modeled and programmed. Conclusions of simulations have all been evaluated to measure the proposed system’s responsiveness to flexible manufacturing. The files and instructions necessary to run the simulation program are available in the appendix CD which is attached inside the back cover of the thesis. Information related to the content and usage of the appendix CD is in the Read.Me text file.
  • Thumbnail Image
    Item
    Forecasting stock market return using artificial neural networks
    (Thesis (M.S.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2006., 2006.) Doğan, Volkan.; Gürgen, Fikret.; Okay, Nesrin.
    Stock market’s return prediction is an important concept in emergent markets like Turkey and Brazil. In this thesis, I used artificial neural networks architecture to model and predict stock markets. Istanbul stock exchange indices National-100 and National-30 are used for domestic market, Brazilian stock exchange index, BVSP, is used as the international market. Input space is divided into clusters with statistical clustering technique Expectation Maximization. ANN’s structure Mixture of Experts is used and local experts are assigned to each cluster. While local experts are learning their region of interest, in parallel, gating experts combine the outputs of them to model overall structure. Besides, future returns are predicted based on patterns obtained from past trainings. In financial time series modeling using ANN, I used past returns in simulations. Since we know two investigated markets are highly volatile and chaotic, the volatility factor is added to the analysis as well. Volatility calculated from RiskMetrics™ [30] is also included in the models to capture different dynamic features of the data. Results of our simulations are evaluated by using previously defined and widely accepted performance measures. Another interesting result is gained during our simulations; two countries of different macroeconomic structures show similarities. Interaction between Turkish and Brazilian stock markets is not surprising; during the financial liberalization of the 1980-1990s, Turkey imported many constitutional laws from Latin American economies especially from Brazil.