Showing papers by "Masahiro Tomono published in 2016"
01 Aug 2016
TL;DR: BLSMI is a combination of methods composed of a kernel-based dependence estimator and noise reduction by bootstrap aggregating, which can handle richer features and robustly estimate dependence and performed best in terms of calibration accuracy.
Abstract: The goal of this study is to achieve automatic extrinsic calibration of a camera-LiDAR system that does not require calibration targets. Calibration through maximization of statistical dependence using mutual information (MI) is a promising approach. However, we observed that existing methods perform poorly on outdoor data sets. Because of their susceptibility to noise, objective functions of previous methods tend to be non-smooth, and gradient-based searches fail in local optima. To overcome these issues, we introduce a novel dependence estimator called bagged least-squares mutual information (BLSMI). BLSMI is a combination of methods composed of a kernel-based dependence estimator and noise reduction by bootstrap aggregating (bagging), which can handle richer features and robustly estimate dependence. We compared ours with previous methods using indoor and outdoor data sets, and observed that our method performed best in terms of calibration accuracy. While previous methods showed degraded performance on outdoor data sets because of the local optima problem, our method exhibited high calibration accuracy both on indoor and outdoor data sets.
24 citations
TL;DR: This paper proposes a novel localization approach that can be applied to sidewalks based on existing 2D street maps and employs a computationally efficient estimator of squared-loss mutual information, through which it achieves near real-time performance.
Abstract: Recently, localization methods based on detailed maps constructed using simultaneous localization and mapping have been widely used for mobile robot navigation. However, the cost of building such maps increases rapidly with expansion of the target environment. Here, we consider the problem of localization of a mobile robot based on existing 2D street maps. Although a large amount of research on this topic has been reported, the majority of the previous studies have focused on car-like vehicles that navigate on roadways; thus, the efficacy of such methods for sidewalks is not yet known. In this paper, we propose a novel localization approach that can be applied to sidewalks. Whereas roadways are typically marked, e.g. by white lines, sidewalks are not and, therefore, road boundary detection is not straightforward. Thus, obtaining exact correspondence between sensor data and a street map is complex. Our approach to overcoming this difficulty is to maximize the statistical dependence between the sensor data ...
4 citations