Negative Information for Occlusion Reasoning in Dynamic Extended Multiobject Tracking

Abstract

A novel approach to utilize negative information to improve the precision and accuracy of extended multiobject tracking is presented. The parameterized probability density of object tracks undetected in sensor data is updated via inferences about the conditions necessary to result in occlusion of the undetected object. Negative information is also leveraged to inform track existence and data association, both of which contribute to a more sensible belief of the local dynamic scene. Simulation and experimental results are presented from autonomous driving scenarios, demonstrating that the use of negative information leads to a more complete, accurate, precise, and intuitive belief of the local scene, enabling high-level tasks that would otherwise be impractical.

Hough Transform-based Road Boundary Localization

“…The proposed method deals with the occlusion problem, and then provides stable performance in crowded situations. The Hough transform method is already used in [12] for road-boundary detection, but a camera is used in [12]. This method is unsuitable for use in laser scanners and is quite different from the method proposed in this study.…”

“…In [12][13][14][15][16][17], the algorithms are designed to overcome the limitation for the cases of object alignment. In [12][13][14], a specific model is proposed for estimating objects in the occluded area.…”

“…In [12][13][14], a specific model is proposed for estimating objects in the occluded area. Koch [15] and Ahmed et al [16] used data fusion to solve the problem of tracking the occluded object using a single laser scanner.…”

Optimal Continuous State POMDP Planning With Semantic Observations: A Variational Approach

“…For example, placing a target object at the edge of a camera's view could generate either a positive (true detection) or negative (false miss) detection from the identification algorithm, with some probability for each outcome. When these probabilities are cast in a likelihood model, they can take advantage of negative information in Bayesian reasoning for target tracking [23] [24], as well as act as a generative model of semantic observations in planning problems [25].…”

Estimation of Blood Oxygen Content Using Context-Aware Filtering

“…Other related problems are sensor scheduling [10], [21] and filtering with intermittent measurements [20] but these works only handle continuous measurements. In robotic applications, techniques have been developed to use both kinds of measurements through information fusion [1], [4], [24] and bounded distributions [18]; however, the former works do not incorporate system dynamics whereas the latter approach results in unwieldy distributions that require approximations. Finally, a similar problem to the one addressed in this paper can be solved using particle filters [3]; yet, particle filters may suffer from particledeprivation problems in high-dimensional spaces whereas the context-aware filter provides exact estimates regardless of dimensionality.…”