Wyniki 1-4 spośród 4 dla zapytania: authorDesc:"JANUSZ BĘDKOWSKI"

Parallel implementation of hybridICP data registration


  In this paper new implementation of On-Line data registration method is shown. Algorithm is using composition of classical approaches point to point and point to plane to achieve better convergence compared to single methods. To improve the performance the parallel computing based on NVIDIA CUDA capabilities is used mainly for k-nearest neighborhood search. Many research has been done concerning 3D data registration. It is easy to find on opinion that point to plane method is better in case of accuracy than point to point method. In theory it is true, but in real application such mobile robot moving in INDOOR environment equipped with commercial available 3D measurement system there are several exceptions. In this paper we are focused on real application and we demonstrate disadvantages of point to plane method that affect the aligning accuracy. The problems are related to the approximation accuracy that appear in data containing stairs, corners etc. The main contributions of this paper are: - new implementation of hybridICP data registration algorithm based on composition of classical approaches point to point and point to plane, - improvement based on parallel computation applied mainly for k-nearest neighbor search - empirical evaluation based on data set delivered by mobile robot equipped with commercial available 3D laser measurement system working in INDOOR environment. Related Work Most range data registration techniques are variants on the iterative closest point (ICP) algorithm, proposed by Chen and Medioni in [3] and Besl and McKay in [4]. We can find also an alternative solution to ICP for data alignment [12]. Point to plane approach is known to be the most accurate [1, 11, 13, 15, 17]. The combination of the point to point and point to plane algorithms into a single probabilistic framework is introduced in [2], it allows for the noise measurement and other probabilistic techniques to increase robustness in cont[...]

Multirobot symulator ze sterownikiem robotów opartym na szybkim hybrydowym klasyfikatorze DTFAM

Czytaj za darmo! »

Prezentujemy innowacyjny szybki klasyfikator dla czujnika LSM SICK 200 dla robota mobilnego poruszającego się w nieznanym środowisku. Przypadkowość jest tu determinowana przez przypadkowe położenie ruchomych obiektów w postaci spacerujących ludzi. Zaprojektowano symulator umożliwiający przeprowadzenie eksperymentów w wirtualnym budynku, w którym porusza się do 100 osób. Symulator umożliwia [...]

Application of semantic maps for mobile robot simulation


  In this paper a new concept of using semantic map for robot operator training purpose is described. The approach consists of 3D laser data acquisition, semantic elements extraction (using image processing techniques) and transformation to rigid body simulation engine, therefore the State Of the Art related to those research topics will be discussed. The combination of a 2D laser range finder with a mobile unit was described as the simulation of a 3D laser range finder in [1]. In this sense we can consider that several researches are using so called simulator of 3D laser range finder to obtain 3D cloud of points [2]. The common 3D laser simulator is built on the basis of a rotated 2D range finder. The rotation axis can be horizontal [3], vertical [4] or similarly to our approach (the rotational axis lies in the middle of the scanners field of view). Semantic information extracted from 3D laser data is recent research topic of modern mobile robotics. In [5] a semantic map for a mobile robot was described as a map that contains, in addition to spatial information about the environment, assignments of mapped features to entities of known classes. In [6] a model of an indoor scene is implemented as a semantic net. This approach is used in [7] where robot extracts semantic information from 3D models built from a laser scanner. In [8] the location of features is extracted by using a probabilistic technique (RANSAC). Also the region growing approach [9] extended from [10] by efficiently integrating k-nearest neighbor (KNN) search is able to process unorganized clouds of points. The semantic map building is related to SLAM (Simultaneous Localization And Mapping) problem [11]. Most of recent SLAM (Simultaneous Localization And Mapping) techniques use camera [12], laser measurement system [13] or even registered 3D laser data [14]. Concerning the registration of 3D scans described in [15] we can find several techniques solving this imp[...]

Zastosowanie metod inteligencji obliczeniowej do analizy struktury defektowej półprzewodników wysokorezystywnych

Czytaj za darmo! »

W artykule przedstawiono wyniki badań centrów defektowych w półizolującym monokrysztale fosforku indu (InP), przeprowadzonych metodą niestacjonarnej spektroskopii fotoprądowej PITS (Photo-Induced Transient Spectroscopy) z zastosowaniem maszyny wektorów nośnych SVM (Support Vector Machine) [1,2] oraz maszyny wektorów istotnych RVM (Relevance Vector Machine) [3] do aproksymacji powierzchni widmowej Laplace’a, otrzymanej poprzez numeryczne przekształcenie relaksacyjnych przebiegów fotoprądu zmierzonych w zakresie temperatur 30...320K. Linie grzbietowe fałd występujących na powierzchni widmowej określają temperaturowe zależności szybkości emisji nośników ładunku z centrów defektowych [4-6]. Celem zastosowania nowych metod aproksymacji jest stworzenie możliwości wyznaczania p[...]

 Strona 1