A novel mobile robot navigation method based on combined feature based scan matching and fastslam algorithm

Özgür, Ayhan
The main focus of the study is the implementation of a practical indoor localization and mapping algorithm for large scale, structured indoor environments. Building an incremental consistent map while also using it for localization is partially unsolved problem and of prime importance for mobile robot navigation. Within this framework, a combined method consisting of feature based scan matching and FastSLAM algorithm using LADAR and odometer sensor is presented. In this method, an improved data association and localization accuracy is achieved by feeding the SLAM module with better incremental pose information from scan matching instead of raw odometer output. This thesis presents the following contributions for indoor localization and mapping. Firstly a method combining feature based scan matching and FastSLAM is achieved. Secondly, improved geometrical relations are used for scan matching and also a novel method based on vector transformation is used for the calculation of pose difference. These are carefully studied and tuned based on localization and mapping performance failures encountered in different realistic LADAR datasets. Thirdly, in addition to position, orientation information usage in line segment and corner oriented data association is presented as an extension in FastSLAM module. v The method is tested with LADAR and odometer data taken from real robot platforms operated in different indoor environments. In addition to using datasets from the literature, own datasets are collected on Pioneer 3AT experimental robot platform. As a result, a real time working localization algorithm which is pretty successive in large scale, structured environments is achieved.


A Conditional coverage path planning method for an autonomous lawn mower
Karol, Ardıç; Konukseven, Erhan İlhan; Koku, Ahmet Buğra; Department of Mechanical Engineering (2016)
Randomized and deterministic coverage path planning methods are widely used in autonomous lawn mowers. Random planning cannot guarantee a complete coverage, whereas, many deterministic techniques are not solely eligible for unstructured outdoor environments, since they highly suffer from wheel slippage or numerical drift. Besides, complete coverage techniques either demands high computational power or expensive sensor hardware. A genuine, Conditional Coverage Path Planning (CCPP) method, which satisfies com...
A framework to embed a spatial statistics toolbox in open-source GIS software: kernel density estimation example
Cavur, M.; Düzgün, Hafize Şebnem (2017-01-01)
It is widely known that geographic information systems (GIS) should include more spatial data analysis (SDA) techniques. The issues of which techniques should be included and how statistical analysis can be integrated with GIS are still widely debated. This paper focuses on the development of a framework that implements R SDA techniques in the uDig. For this purpose, a simple interface is designed between two open-source software applications, uDig and the R statistical software package. The tight coupling ...
A temporal neural network model for constructing connectionist expert system knowledge bases
Alpaslan, Ferda Nur (Elsevier BV, 1996-04-01)
This paper introduces a temporal feedforward neural network model that can be applied to a number of neural network application areas, including connectionist expert systems. The neural network model has a multi-layer structure, i.e. the number of layers is not limited. Also, the model has the flexibility of defining output nodes in any layer. This is especially important for connectionist expert system applications.
A Probabilistic approach to sparse multi scale phase based stereo
ULUSOY PARNAS, İLKAY; Halıcı, Uğur; HANCOCK, EDWIN (2004-04-30)
In this study, a multi-scale phase based sparse disparity algorithm and a probabilistic model for matching are proposed. The disparity algorithm and the probabilistic approach are verified on various stereo image pairs.
A deep learning approach for the transonic flow field predictions around airfoils
Duru, Cihat; Alemdar, Hande; Baran, Özgür Uğraş (2022-01-01)
Learning from data offers new opportunities for developing computational methods in research fields, such as fluid dynamics, which constantly accumulate a large amount of data. This study presents a deep learning approach for the transonic flow field predictions around airfoils. The physics of transonic flow is integrated into the neural network model by utilizing Reynolds-averaged Navier–Stokes (RANS) simulations. A detailed investigation on the performance of the model is made both qualitatively and quant...
Citation Formats
A. Özgür, “A novel mobile robot navigation method based on combined feature based scan matching and fastslam algorithm,” M.S. - Master of Science, Middle East Technical University, 2010.