Donald A. Sofge

A blended population approach to cooperative coevolution for decomposition of complex problems

Cooperative coevolutionary architectures provide a framework for solving complex problems by decomposing them into constituent subproblems, solving the subproblems, and then reintegrating the solutions. This paper presents a blended cooperative coevolution model which offers advantages over traditional evolutionary algorithms and currently-used cooperative coevolutionary architectures.

Neural network based process optimization and control

The authors present, compare, and contrast several neurocontrol systems for controlling and optimizing the manufacture of thermoplastic composite structures. Attention is given to direct-inverse neurocontrol for manufacturing and adaptive-critic-based manufacturing neurocontrol. In particular, adaptive critic neurocontrol systems that are capable of online learning and process optimization have been demonstrated in simulation.<<ETX>>

Reactive Motion Planning for Unmanned Aerial Surveillance of Risk-Sensitive Areas

This paper proposes a reactive motion-planning approach for persistent surveillance of risk-sensitive areas by a team of unmanned aerial vehicles (UAVs). The planner, termed PARCov (Planner for Autonomous Risk-sensitive Coverage), seeks to: i) maximize the area covered by sensors mounted on each UAV; ii) provide persistent surveillance; iii) maintain high sensor data quality; and iv) reduce detection risk. To achieve the stated objectives, PARCov combines into a cost function the detection risk with an uncertainty measure designed to keep track of the regions that have been surveyed and the times they were last surveyed. PARCov reduces the uncertainty and detection risk by moving each quadcopter toward a low-cost region in its vicinity. By reducing the uncertainty, PARCov is able to increase the coverage and provide persistent surveillance. Moreover, a nonlinear optimization formulation is used to determine the optimal altitude for flying each quadcopter in order to maximize the sensor data quality while minimizing risk.

A generalized graph-based method for engineering swarm solutions to multiagent problems

We present two key components of a principled method for constructing modular, heterogeneous swarms. First, we generalize a well-known technique for representing swarm behaviors to extend the power of multiagent systems by specializing agents and their interactions. Second, a novel graph-based method is introduced for designing swarm-based behaviors for multiagent teams. This method includes engineer-provided knowledge through explicit design decisions pertaining to specialization, heterogeneity, and modularity. We show the representational power of our generalized representation can be used to evolve a solution to a challenging multiagent resource protection problem. We also construct a modular design by hand, resulting in a scalable and intuitive heterogeneous solution for the resource protection problem.

Onboard Flow Sensing For Downwash Detection and Avoidance On Small Quadrotor Helicopters

Abstract : Small rotary-wing UAVs are susceptible to gusts and other environmental disturbances that a ect inow at their rotors. Inow variations cause unexpected aerodynamic forces through changes in thrust conditions and unmodeled blade apping dynamics. This pa- per introduces an onboard, pressure-based ow measurement system developed for a small quadrotor helicopter. The probe-based instrumentation package provides spatially dis- tributed airspeed measurements along each of the aircraft- xed axes. Lateral and vertical windspeed estimates enable the development of disturbance-tolerant ight control strate- gies. The focus of this paper is vertical ow disturbances such as those caused by the downwash of a second vehicle. Real-time velocity measurements are incorporated into a recursive Bayesian estimator to localize a nearby rotorcraft using its downwash. A path planner developed for proximity ight is demonstrated through indoor ight testing with multiple vehicles to safely guide an instrumented quadrotor around the downwash of nearby rotorcraft.

A New Approach to Organizations: Stability and Transformation in Dark Social Networks

Uncovering information from well-defined organizations for social network analysis is straightforward, but such analyses of social networks have not led to valid predictions about their actions or stability. For dark social networks, which comprise illicit drug gangs or terrorists, uncovering information to compute a social network analysis is more difficult to solve. The authors used a new theory that is based on the conservation of information to assess organizations and dark social networks, concluding that social network analyses that are properly constrained should be invaluable for bookkeeping (storing information recovered from neighborhood canvasses such as with the Defense Advanced Research Projects Agencys Tactical Ground Reporting); for theory (e.g., angiogenesis, in which a tumor takes over the infrastructure of a body; a criminal street gang such as MS-13 takes control of its territory from city authorities); and for benchmarking (e.g., comparing operational performance of models with case studies or random graphs to assure equivalence between models). The results outline a path forward to advance the theory of organizations for enterprise change and continuity.

COLLABORATING WITH HUMANOID ROBOTS IN SPACE

One of the great challenges of putting humanoid robots into space is developing cognitive capabilities for the robots with an interface that allows human astronauts to collaborate with the robots as naturally and efficiently as they would with other astronauts. In this joint effort with NASA and the entire Robonaut team, we are integrating natural language and gesture understanding, spatial reasoning incorporating such features as human–robot perspective taking, and cognitive model-based understanding to achieve a high level of human–robot interaction. Building greater autonomy into the robot frees the human operator(s) from focusing strictly on the demands of operating the robot, and instead allows the possibility of actively collaborating with the robot to focus on the task at hand. By using shared representations between the human and robot, and enabling the robot to assume the perspectives of the human, the humanoid robot may become a more effective collaborator with a human astronaut for achieving mission objectives in space.

A Survey of Quantum Programming Languages: History, Methods, and Tools

Quantum computer programming is emerging as a new subject domain from multidisciplinary research in quantum computing, computer science, mathematics (especially quantum logic, lambda calculi, and linear logic), and engineering attempts to build the first non-trivial quantum computer. This paper briefly surveys the history, methods, and proposed tools for programming quantum computers circa late 2007. It is intended to provide an extensive but non-exhaustive look at work leading up to the current state-of-the-art in quantum computer programming. Further, it is an attempt to analyze the needed programming tools for quantum programmers, to use this analysis to predict the direction in which the field is moving, and to make recommendations for further development of quantum programming language tools.

Conservation of Information: A New Approach to Organizing Human-Machine-Robotic Agents under Uncertainty

For many years, social scientists have struggled to make sense of the shift between individual and group perception, the difference between observation and action, and the meaning of interdependence. Interactions between these factors produce stable worldviews that contain more illusory than actual connections to reality. We attribute these struggles to scientists embedded in the social fabric, the lack of a measurement theory, and the difficulty of testing new theory with human subjects, groups and organizations. Our work includes field research with observations of citizen organizations advising the Department of Energy (DOE) on its environmental cleanup; laboratory simulations of DOE field results; stock market data; and computational modeling (coupled differential equations, control theory, AI, Gaussian distributions, uncertainty models, Fourier transform pairs, continuous and discrete wavelets). Results from laboratory experiments and stock markets agree with our theory, but many questions remain, forming a high-risk research plan. Our objective is to incorporate interdependent uncertainty into computational intelligence to better instantiate autonomy or decentralized control for mixed human-machine systems.

Structural health monitoring using neural network based vibrational system identification

Composite fabrication technologies now provide the means for producing high-strength, low-weight panels, plates, spars and other structural components which use embedded fiber optic sensors and piezoelectric transducers. These materials, often referred to as smart structures, make it possible to sense internal characteristics, such as delaminations or structural degradation. We use neural network based techniques for modeling and analyzing dynamic structural information for recognizing structural defects. This yields an adaptable system which gives a measure of structural integrity for composite structures.<<ETX>>