Abstract: In this work we present a new processdistribution approach for sensor data fusion (SDF) systems, called FuseDis (fusion distribution using tessellated space). It is based on a hybrid partitioning approach, aimed at managing computational burden and achieving scalability.
First, a functional decomposition dicides SDF functionality into taskgroups, vectorizing operations. Second, a partitioning of the dataspace, based on geographic attributes of the data, is applied to parallelize the processing.
A tesselation applied to the plot-space imlicitly defines for each tile the set of candidate-tracks yielding useful correlations with the plots in a tile. Some tracks may occur as correlation-candidates for multiple tiles. Conflicts caused by correlations of such tracks with plots in different tiles are solved by combining correlations of the involved tracks and plots into independent association problems
Abstract: In this article we consider the issue of optimal control in collaborative multi-agent systems with stochastic dynamics. The agents have a joint task in which they have to reach a number of target states. The dynamics of the agents contains additive control and additive noise, and the autonomous part factorizes over the agents. Full observation of the global state is assumed. The goal is to minimize the accumulated joint cost, which consists of integrated instantaneous costs and a joint end cost. The joint end cost expresses the joint task of the agents. The instantaneous costs are quadratic in the control and factorize over the agents. The optimal control is given as a weighted linear combination of single-agent to single-target controls. The single-agent to single-target controls are expressed in terms of diffusion processes. These controls, when not closed form expressions, are formulated in terms of path integrals, which are calculated approximately by Metropolis-Hastings sampling. The weights in the control are interpreted as marginals of a joint distribution over agent to target assignments. The structure of the latter is represented by a graphical model, and the marginals are obtained by graphical model inference. Exact inference of the graphical model will break down in large systems, and so approximate inference methods are needed. We use naive mean field approximation and belief propagation to approximate the optimal control in systems with linear dynamics. We compare the approximate inference methods with the exact solution, and we show that they can accurately compute the optimal control. Finally, we demonstrate the control method in multi-agent systems with nonlinear dynamics consisting of up to 80 agents that have to reach an equal number of target states.
Abstract: Gas distribution models can provide comprehensive information about a large
number of gas concentration measurements, highlighting, for example, areas of unusual
gas accumulation. They can also help to locate gas sources and to plan where
future measurements should be carried out. Current physical modeling methods,
however, are computationally expensive and not applicable for real world scenarios
with real-time and high resolution demands. This chapter reviews kernel methods
that statistically model gas distribution. Gas measurements are treated as random
variables, and the gas distribution is predicted at unseen locations either using a
kernel density estimation or a kernel regression approach. The resulting statistical apmodels
do not make strong assumptions about the functional form of the gas distribution,
such as the number or locations of gas sources, for example. The major
focus of this chapter is on two-dimensional models that provide estimates for the
means and predictive variances of the distribution. Furthermore, three extensions
to the presented kernel density estimation algorithm are described, which allow to
include wind information, to extend the model to three dimensions, and to reflect
time-dependent changes of the random process that generates the gas distribution
measurements. All methods are discussed based on experimental validation using
real sensor data.
Abstract: To study gas dispersion, several statistical gas distribution modelling approaches have
been proposed recently. A crucial assumption in these approaches is that gas distribution models are
learned from measurements that are generated by a time-invariant random process which can capture
certain fluctuations in the gas distribution. More accurate models can be obtained by modelling
changes in the random process over time. In this work we propose a time-scale parameter that relates
the age of measurements to their validity to build the gas distribution model in a recency function.
The parameters of the recency function define a time-scale and can be learned. The time-scale
represents a compromise between two conflicting requirements to obtain accurate gas distribution
models: using as many measurements as possible and using only very recent measurements. We
have studied several recency functions in a time-dependent extension of the Kernel DM+V. Based
on real-world experiments and simulations of gas dispersal (presented in this paper) we demonstrate
that TD Kernel DM+V improves the obtained gas distribution models in dynamic situations. This
represents an important step towards statistical modelling of evolving gas distributions.