RESUMEN
We propose a new linear RGB-D simultaneous localization and mapping (SLAM) formulation by utilizing planar features of the structured environments. The key idea is to understand a given structured scene and exploit its structural regularities such as the Manhattan world. This understanding allows us to decouple the camera rotation by tracking structural regularities, which makes SLAM problems free from being highly nonlinear. Additionally, it provides a simple yet effective cue for representing planar features, which leads to a linear SLAM formulation. Given an accurate camera rotation, we jointly estimate the camera translation and planar landmarks in the global planar map using a linear Kalman filter. Our linear SLAM method, called L-SLAM, can understand not only the Manhattan world but the more general scenario of the Atlanta world, which consists of a vertical direction and a set of horizontal directions orthogonal to the vertical direction. To this end, we introduce a novel tracking-by-detection scheme that infers the underlying scene structure by Atlanta representation. With efficient Atlanta representation, we formulate a unified linear SLAM framework for structured environments. We evaluate L-SLAM on a synthetic dataset and RGB-D benchmarks, demonstrating comparable performance to other state-of-the-art SLAM methods without using expensive nonlinear optimization. We assess the accuracy of L-SLAM on a practical application of augmented reality.
RESUMEN
This paper introduces a design and implementation of a base station, capable of positioning sensor nodes using an optical scheme. The base station consists of a pulse laser module, optical detectors and beam splitter, which are mounted on a rotation-stage, and a Time to Digital Converter (TDC). The optical pulse signal transmitted to the sensor node with a Corner Cube Retro-reflector (CCR) is reflected to the base station, and the Time of Flight (ToF) data can be obtained from the two detectors. With the angle and flight time data, the position of the sensor node can be calculated. The performance of the system is evaluated by using a commercial CCR. The sensor nodes are placed at different angles from the base station and scanned using the laser. We analyze the node position error caused by the rotation and propose error compensation methods, namely the outlier sample exception and decreasing the confidence factor steadily using the recursive least square (RLS) methods. Based on the commercial CCR results, the MEMS CCR is also tested to demonstrate the compatibility between the base station and the proposed methods. The result shows that the localization performance of the system can be enhanced with the proposed compensation method using the MEMS CCR.
RESUMEN
Tracking the locations and identities of moving targets in the surveillance area of wireless sensor networks is studied. In order to not rely on high-cost sensors that have been used in previous researches, we propose the integrated localization and classification based on semi-supervised learning, which uses both labeled and unlabeled data obtained from low-cost distributed sensor network. In our setting, labeled data are obtained by seismic and PIR sensors that contain information about the types of the targets. Unlabeled data are generated from the RF signal strength by applying Gaussian process, which represents the probability of predicted target locations. Finally, by using classified unlabeled data produced by semi-supervised learning, identities and locations of multiple targets are estimated. In addition, we consider a case when the labeled data are absent, which can happen due to fault or lack of the deployed sensor nodes and communication failure. We overcome this situation by defining artificial labeled data utilizing characteristics of support vector machine, which provides information on the importance of each training data point. Experimental results demonstrate the accuracy of the proposed tracking algorithm and its robustness to the absence of the labeled data thanks to the artificial labeled data.
RESUMEN
Surface modified carbon strip electrode with Bi nanopowder was suggested for a simultaneous analysis of Zn, Cd, and Pb ions by a square wave anodic stripping voltammetry, and the influence of the modifying Bi mass and particle size on the trace metal response was investigated. The Bi nanopowders with various particle size distributions were synthesized by an optimization of the gas condensation condition, in which a refractory crucible was applied for the evaporation of volatile Bi, and then immobilized on the surface of a working electrode. The result of the stripping measurements shows that when the modifying mass and the particle size of the Bi powder were in the range of 2 to 5 microg/cm2 and less than 300 nm, respectively, a well-developed and reproducible stripping response was obtained. The proposed "mercury-free" carbon strip electrode, modified with Bi nanopowder, is conveniently usable and directly applicable to a trace metal analysis without a pre-deposition of Bi and complicated surface polishing steps.