In uncertain and dynamic environments, decentralized data fusion (DDF) techniques have been widely used to estimate the states and the uncertainty levels over large mission spaces in a robust and scalable way. In data fusion frameworks using distributed sensor networks, undetected sensor failures can degrade the quality of fusion results of the entire system. Therefore, DDF methods which are robust to inconsistent data are needed. In this paper, a fault-tolerant Bayesian DDF method using Gaussian mixture models is developed. The probability of agent reliability states, which represent consistency of local estimates that agents share with their neighbors, are modeled as weights of mixture models and estimated together with the target process. The target process and reliability states are updated in a decentralized Bayesian way, exploiting the properties of Gaussian mixture models. To prevent the hypothesis explosion problem of Gaussian mixture models, a mixture compression method considering the physical meaning of mixture weights is utilized. A numerical simulation on a 2D dynamic target tracking problem is presented to verify performance of the suggested algorithm and compared with existing DDF methods. It is shown that the suggested algorithm gives more compact fusion results compared to existing fault-tolerant DDF method.

In search applications, autonomous unmanned vehicles must be able to efficiently reacquire and localize mobile targets that can remain out of view for long periods of time in large spaces. As such, all available information sources must be actively leveraged - including imprecise but readily available semantic observations provided by humans. To achieve this, this work develops and validates a novel collaborative human-machine sensing solution for dynamic target search. Our approach uses continuous partially observable Markov decision process (CPOMDP) planning to generate vehicle trajectories that optimally exploit imperfect detection data from onboard sensors, as well as semantic natural language observations that can be specifically requested from human sensors. The key innovation is a scalable hierarchical Gaussian mixture model formulation for efficiently solving CPOMDPs with semantic observations in continuous dynamic state spaces. The approach is demonstrated and validated with a real human-robot team engaged in dynamic indoor target search and capture scenarios on a custom testbed.

Kalman filters are routinely used for many data fusion applications including navigation, tracking, and simultaneous localization and mapping problems. However, significant time and effort is frequently required to tune various Kalman filter model parameters, e.g. process noise covariance, pre-whitening filter models for non-white noise, etc. Conventional optimization techniques for tuning can get stuck in poor local minima and can be expensive to implement with real sensor data. To address these issues, a new “black box” Bayesian optimization strategy is developed for automatically tuning Kalman filters. In this approach, performance is characterized by one of two stochastic objective functions: normalized estimation error squared (NEES) when ground truth state models are available, or the normalized innovation error squared (NIS) when only sensor data is available. By intelligently sampling the parameter space to both learn and exploit a nonparametric Gaussian process surrogate function for the NEESINIS costs, Bayesian optimization can efficiently identify multiple local minima and provide uncertainty quantification on its results.