Devesh K. Jha

Information Fusion of Passive Sensors for Detection of Moving Targets in Dynamic Environments

This paper addresses the problem of target detection in dynamic environments in a semi-supervised data-driven setting with low-cost passive sensors. A key challenge here is to simultaneously achieve high probabilities of correct detection with low probabilities of false alarm under the constraints of limited computation and communication resources. In general, the changes in a dynamic environment may significantly affect the performance of target detection due to limited training scenarios and the assumptions made on signal behavior under a static environment. To this end, an algorithm of binary hypothesis testing is proposed based on clustering of features extracted from multiple sensors that may observe the target. First, the features are extracted individually from time-series signals of different sensors by using a recently reported feature extraction tool, called symbolic dynamic filtering. Then, these features are grouped as clusters in the feature space to evaluate homogeneity of the sensor responses. Finally, a decision for target detection is made based on the distance measurements between pairs of sensor clusters. The proposed procedure has been experimentally validated in a laboratory setting for mobile target detection. In the experiments, multiple homogeneous infrared sensors have been used with different orientations in the presence of changing ambient illumination intensities. The experimental results show that the proposed target detection procedure with feature-level sensor fusion is robust and that it outperforms those with decision-level and data-level sensor fusion.

Feature level sensor fusion for target detection in dynamic environments

This paper addresses the problem of target detection in dynamic environments. A key challenge here is to simultaneously achieve high probabilities of correct detection with low false alarm rates under limited computation and communication resources. To this end, a procedure of binary hypothesis testing is proposed based on agglomerative hierarchical feature clustering. The proposed procedure has been experimentally validated in the laboratory setting on a mobile robot for target detection by using multiple homogeneous (with different orientations) infrared sensors in the presence of changing ambient light intensities. The experimental results show that the proposed target detection procedure with feature-level sensor fusion outperforms those with decision-level sensor fusion.

Depth estimation in Markov models of time-series data via spectral analysis

Symbol sequences are generated from observed time series data to construct probabilistic finite state automata (PFSA) models that capture the evolution of the dynamical system under consideration. One of the key challenges here is to estimate the relevant history or depth (i.e., the size of temporal memory) of the symbol sequences; in this context, spectral decomposition of the one-step transition matrix has been recently proposed for depth estimation. This paper compares the performance of depth estimation by spectral analysis with that of other commonly used metrics (e.g., log-likelihood, entropy rate and signal reconstruction) for analysis of symbolic dynamic systems. For experimental validation of the proposed concept, time-series data of fatigue damage evolution in a polycrystalline alloy, collected on a laboratory apparatus, have been discretized to generate symbol sequences. The depths, estimated by the spectral decomposition method, are then compared with those obtained by other metrics, and the results are found to be in close agreement. Furthermore, unsupervised clustering of time-series data, obtained for a number of test specimens in the fatigue-test experiments, demonstrates the efficacy of the proposed depth estimation method as well as the accuracy of depth estimation via spectral analysis and PFSA model construction.

Topology optimisation for energy management in underwater sensor networks

In general, battery-powered sensors in a sensor network are operable as long as they can communicate sensed data to a processing node. In this context, a sensor network has two competing objectives: (1) maximisation of the network performance with respect to the probability of successful search for a specified upper bound on the probability of false alarms, and (2) maximisation of the network’s operable life. As both sensing and communication of data consume battery energy at the sensing nodes of the sensor network, judicious use of sensing power and communication power is needed to improve the lifetime of the sensor network. This paper presents an adaptive energy management policy that will optimally allocate the available energy between sensing and communication at each sensing node to maximise the network performance subject to specified constraints. Under the assumptions of fixed total energy allocation for a sensor network operating for a specified time period, the problem is reduced to synthesis of an optimal network topology that maximises the probability of successful search (of a target) over a surveillance region. In a two-stage optimisation, a genetic algorithm-based meta-heuristic search is first used to efficiently explore the global design space, and then a local pattern search algorithm is used for convergence to an optimal solution. The results of performance optimisation are generated on a simulation test bed to validate the proposed concept. Adaptation to energy variations across the network is shown to be manifested as a change in the optimal network topology by using sensing and communication models for underwater environment. The approximate Pareto-optimal surface is obtained as a trade-off between network lifetime and probability of successful search over the surveillance region.

Path planning in GPS-denied environments via collective intelligence of distributed sensor networks

ABSTRACT This paper proposes a framework for reactive goal-directed navigation without global positioning facilities in unknown dynamic environments. A mobile sensor network is used for localising regions of interest for path planning of an autonomous mobile robot. The underlying theory is an extension of a generalised gossip algorithm that has been recently developed in a language-measure-theoretic setting. The algorithm has been used to propagate local decisions of target detection over a mobile sensor network and thus, it generates a belief map for the detected target over the network. In this setting, an autonomous mobile robot may communicate only with a few mobile sensing nodes in its own neighbourhood and localise itself relative to the communicating nodes with bounded uncertainties. The robot makes use of the knowledge based on the belief of the mobile sensors to generate a sequence of way-points, leading to a possible goal. The estimated way-points are used by a sampling-based motion planning algorithm to generate feasible trajectories for the robot. The proposed concept has been validated by numerical simulation on a mobile sensor network test-bed and a Dubin’s car-like robot.

Path planning in GPS-denied environments: A collective intelligence approach

This paper proposes a framework for reactive goal-directed navigation without global positioning facilities in unknown environments. A mobile sensor network is used for localization of regions of interest for path planning of an autonomous mobile robot in the absence of global positioning facilities. The underlying theory is an extension of a generalized gossip algorithm that has been recently developed in a language-measure-theoretic setting. The gossip algorithm has been used to propagate local decisions of target detection over a mobile sensor network and thus, it generates a belief for the target detected over the network. The proposed concept has been validated through numerical experiments with a mobile sensor network and a point mass robot.

Classification of Two-Phase Flow Patterns by Ultrasonic Sensing

This paper presents a methodology for classification of twophase flow patterns in fluid systems, which takes the measurements of an in situ ultrasonic sensor as inputs. In contrast to the common practice of having an array of ultrasonic detectors, the underlying algorithm requires only a single sensor hardware in combination with an integrated software of signal conditioning, feature extraction, and pattern classification. The proposed method is noninvasive and can be implemented in a variety of industrial applications (e.g., petrochemical processes and nuclear power plants). This concept of flow pattern classification is experimentally validated on a laboratory test apparatus. [DOI: 10.1115/1.4007555]

Fractal analysis of crack initiation in polycrystalline alloys using surface interferometry

Microstructural degradation is a predominant source of damage in polycrystalline alloys that are commonly used in diverse applications. For early diagnosis and prognosis of failures, it is essential to understand the mechanisms of damage growth specifically in the crack initiation phase, which is still an intriguing phenomenon for scientists due to sensing inaccuracies and modeling uncertainties. Measurements of gradually evolving deformations on the material surface during crack initiation provide early warnings of forthcoming widespread damage. In this paper, a surface interferometer is used to generate 3-D surface profiles of polycrystalline alloy specimens under oscillating load. The concepts of fractal geometry are used to quantify the changes in the 3-D surface profiles as early indicators of damage evolution in the crack initiation phase.

Data-driven anytime algorithms for motion planning with safety guarantees

This paper presents a learning-based (i.e., data-driven) approach to motion planning of robotic systems. This is motivated by controller synthesis problems for safety critical systems where an accurate estimate of the uncertainties (e.g., unmodeled dynamics, disturbance) can improve the performance of the system. The state-space of the system is built by sampling from the state-set as well as the input set of the underlying system. The robust adaptive motion planning problem is modeled as a learning-based approach evasion differential game, where a machine-learning algorithm is used to update the statistical estimates of the uncertainties from system observations. The system begins with a conservative estimate of the uncertainty set to ensure safety of the underlying system and we relax the robustness constraints as we get better estimates of the unmodeled uncertainty. The estimates from the machine learning algorithm are used to refine the estimates of the controller in an anytime fashion. We show that the values for the game converges to the optimal values with known disturbance given the statistical estimates on the uncertainty converges. Using confidence intervals for the unmodeled disturbance estimated by the machine learning estimator during the transient learning phase, we are able to guarantee safety of the robotic system with the proposed algorithms during transience.

Data-driven robot gait modeling via symbolic time series analysis

This paper addresses data-driven mode modeling and Bayesian mode estimation in hidden-mode hybrid systems (HMHS). For experimental validation in a laboratory setting, an HMHS is built upon a six-legged T-hex robot that makes use of a library of gaits (i.e., the modes of walking) to perform different motion maneuvers. To accurately predict the behavior of the robot, it is important to first infer the gait being used by the robot. The walking robots motion behavior can then be modeled as a transition system based on the pattern of switching among these gaits. In this paper, a symbolic time-series-based statistical learning method has been adopted to construct the generative models of the gaits. Efficacy of the proposed algorithm is demonstrated by laboratory experimentation to model and then infer the hidden dynamics of different gaits for the T-hex walking robot.