AUTHORS: Peter Odry, Imre Rudas
Download as PDF
ABSTRACT: In the implementation phase of a task, we expect it to be as reliable and safe as possible with the best possible parameters, regardless of operating conditions; for example, if we build a structure, or write a software, or simply want to get from one point to another in a big city for a given time, etc. Operation under these conditions is called robust operation. Finding a robust solution is one of the key strategic issues in today's accelerated world. There is no time today to develop a tool, to build it and test it and then modify it, then apply this cycle several times in every possible load environment, but we are looking for mathematical methods to solve this optimization process in a simulation environment. The winning strategy is not simply about the selection of the optimization method, but also about the definition of the adequate quality (fitness), the robustness of the resulting optimum or sensitivity analysis, and the uncertainty analysis of several parameters. Modern engineering problems are often composed by objectives that must be taken into account simultaneously for better design performance. Normally, these objectives are conflicting, i.e., an improvement in one of them does not lead, necessarily, to better results for the other ones. To overcome this difficulty, many methods to solve multi-objective optimization problems (MOP) have been proposed. The simulation model includes environmental or mission parameters that are not part of the parameters to be optimized but their variation creates different scenarios. A multi-scenario simulation can be created with the typical values of these parameters where the optimum is searched for all scenarios at the same time. Such optimum is more robust than one achieved through a process using separate scenarios since the intended use of the robot is represented by the multi-scenarios. A common goal for system design is robustness: the ability of a system to operate correctly in various conditions and fail gracefully in an unexpected situation. This paper deals with two different research domains, where the goal of finding the robust solution is presented. First, Szabad(ka)-II hexapod walker robot as a complex mechanical structure characterized by three motors per leg is analyzed. This robot is particularly suitable for testing the robustness of the closed loop system. During the design process, the minimization of the mechanical complexity was carefully addressed with the aim to reduce the unwieldy appearance. In the second part of the paper, robust methods applied in tomography are discussed. Achieving robustness in tomographic methods is very important due to the presence of different measurement noises. The applied inverse calculation methods are sensitive to noise effects, moreover, the obtained results are characterized by uncertainties due to the unknown dynamics of different noise sources.
KEYWORDS: optimal solution, linear and heuristic optimalization, robust solution, robotics, tomographyREFERENCES:
1] I. Kecskes, E. Burkus, F. Bazso, P. Odry,
Model validation of a hexapod walker robot,
Robotica Vol. 35 No. 2, 2015, pp. 419-462.
 I. Kecskes, P. Odry, Optimization of PI and fuzzy-PI controllers on simulation model of Szabad(ka)-II walking robot, Int. J. Adv. Robot. Syst. 11 (2014) 186. doi:10.5772/59102.
 Laslo Tadic, Miodrag Popovic, Peter Odry: 'Fuzzified Gabor Filter for License Plate Detection' International Scientific Journal Engineering Applications of Artificial Intelligence, 2016, DOI: 10.1016/j.engappai.2015.09.009
 Magnetic tomography of ferromagnetic materials dept structure, PNo : 1600165
 Data acquisition and processing method to assembly soft tomography examinations: PNo : 1500616
 István Kecskés, Ervin Burkus, Zoltán Király, Ákos Odry, Péter Odry: Comparatition of motor controllers using a simplified robot leg: PID vs fuzzy logic; MCSI, august 2017; Corfu
 Cariou C., Berducat M., Gobor Z. and Seiferth B.: (2017), Mobile Robot Trajectory Planning Under Kinematic and Dynamic Constraints for Partial and Full Field Coverage. J. Field Robotics. doi:10.1002/rob.21707
 Ákos Odry; Róbert Fullér; Imre J Rudas; Péter Odry: Kalman filter for mobile-robot attitude estimation: Novel optimized and adaptive solutions, Mechanical System and Signal Processing (ISSN: 0888-3270) 110: pp. 569-589. (2018); DOI: 10.1016/j.ymssp.2018.03.053;