Governing Lethal Behavior in Autonomous Robots

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

The catalyst layer used in a fuel cell is an archetype of multi-component composite materials used as electrodes in many electrochemical devices. In this talk, we will describe and link together a number of studies of the fuel cell catalyst layer that we have been carried out over the past 10 to 15 years. The first of these was a modeling effort that was initially carried out to describe the geometrically constrained limits on utilization of dispersed catalyst in the layer. To answer our original question, the catalyst components were broken down to constituents (Pt nanoparticles, carbon particles, ionomer) and interactions between components were used to build a hierarchical model of the catalyst layer. The interactions were defined in the context of colloidal interaction theory using descriptions of electronic properties of the constituents derived from experimental measurements of refractive index/dielectric properties across a wide range of frequencies. The model satisfactorily captured some aspects of the structure of the catalyst layer, e.g. the tendency toward segregation of components, and provides a good bound to the utilization of catalysts but failed to provide a quantitatively accurate description of certain transport-related physical properties. In the next phase of the work, the connectivity of ionomer in the catalyst layer was probed using NMR relaxation and diffusion measurements on water in the catalyst layer as a function of water content. Finally, a detailed study of the distribution of components within the catalyst layer, the catalyst layer porosity and the adsorption of ionomer in catalyst layers, all as functions of composition, was undertaken. These three chapters of work correspond to three different probes usng models and experiments of many types to yield a more nuanced picture of the catalyst layer. Future prospects and needs will be presented in the context of still-missing information and where we might look for it. Acknowledgement We gratefully acknowledge the support of this work by the US Army Research Office and the Naval Research Research Office. The significant contributions of Professor J. Adin Mann of Case Western Reserve university and Dr. Karren More of Oak Ridge National Lab are also appreciated.

: Small radial inflow turbines have various applications in industry and are successfully being used as a major component of gas turbines and turbochargers. The performance of the turbine is of great importance for the success of these systems. Although the design of radial inflow turbines has improved in the last few years, the detailed aerodynamic study of these components still needs considerable attention. Better understanding of the flow processes involved in these machines will provide a good basis for the design of improved components and for this purpose a joint research programme has been undertaken by Cranfield, the US Army Research Office and Turbomach (San Diego). The main objective of the study was to investigate the flow processes involved in the radial inflow turbine using experimental methods. The work presented in this report describes the various steps undertaken to perform these studies together with results. The project involved two phases, the first measurements downstream of the rotor and the second measurements between the nozzle guide vanes and rotor inlet. The scope of the study was limited to two man years during which considerable success was achieved in obtaining the desired measurement but more time to further reduce this data and to compare with computational results, which was outside the scope of the present study, would have been advantageous. The studies using laser anemometry undertaken downstream of the turbine rotor indicated a swirling flow with a region of counter-rotating flow in the centre. A cobra probe was used to compare with the laser anemometry results. Both measuring techniques show similar trends of flow velocity and flow angle at various running conditions.... Radial Turbines, Laser Anemometry.

Jagdish Chandra (ed) 1984 Philadelphia: Siam viii + 191 pp price £23.80 ISBN 0 89871 052 9 Chaotic motion is the pseudo-random behaviour of systems governed by deterministic classical equations of motion. Nonlinear dynamics and associated chaotic motion have attracted much research interest in recent years, particularly because the results are applicable to a wide range of problems in science and engineering.

Addition of fillers to polymers has long been used as a strategy to enhance the properties of polymers. In the last two decades, however, a class of materials termed as "polymer nanocomposites" (PNC) has evolved to refer to blends of polymers and inorganic fillers in which the inorganic component has one or more dimensions below 100 nm.1-3 Examples of such materials include mixtures of a variety of homopolymers and multicomponent polymers blended with fillers such as clays, nanotubes, Fullerenes, polyhedral oligomeric silsesquioxanes (POSS), nanoscopic silica particles, and so forth.2-7 Initial interest and research in such materials arose from the significant property enhancements reported for extremely low loadings of nanofillers. These reports fueled intense research into the mechanical, electrical, barrier, and fire retardancy properties of PNCs and have led to some notable commercial technologies utilizing PNCs.1, 8-10 More recent efforts have expanded the above class of studies and materials by examining the properties of nanocomposites involving polymer blends and block copolymers, where the potential to create multifunctional materials possessing novel electrical, magnetic, and optical properties have been reported.11, 12 However, despite the scale of intense research on PNCs, much of the initial promise of PNCs has not yet borne out. In this viewpoint article, I present an (admittedly) personal perspective of some of the issues confronting the field of PNCs, focusing especially on the opportunities by which theory, modeling, and computer simulations from polymer physics may aid in realizing the full potential of PNCs. Many of the common nanofillers used in PNCs are unfortunately characterized by strong van der waals interparticle attractions, which promote their agglomeration.13 Moreover, the effective interactions between nanofillers are also influenced by the polymer-filler interactions, and if the latter are unfavorable, it leads to conditions conducive to the aggregation of the fillers. Aggregation and clustering of the fillers typically leads to a significant reduction in the interfacial contact between the fillers and the polymer and hence a deterioration of the property enhancements that can be attained in PNCs. Consequently, achieving stable, dispersed configurations of nanofillers in the polymer matrix is a prime requirement to realize the full property potential of PNCs. Many experimental strategies have been explo-red to overcome the above challenge.14 These have ranged from using polymer-filler combinations, which are known to exhibit favorable affinities to functionalizing the fillers by self-assembling surfactant and grafting polymers to promote favorable polymer-filler interactions (and provide a steric barrier against agglomeration). Also, successful strategies using external fields to promote nonequilibrium dispersion of the nanofiller have been demonstrated.15 From a theoretical perspective, the outstanding question confronting the design of such strategies is "For specified combination of matrix polymer(s), filler(s), the functionalizing moieties and/or external fields, can we predict the equilibrium and nonequilibrium structural characteristics of the PNC dispersion?" Anisotropy of fillers: Most of the PNCs explored for applications involve filler particles such as clays, carbon nanotubes, nanofibers, and so forth, which exhibit considerable shape anisotropy. Indeed, such fillers prove most suitable for exploiting significant polymer-filler interfacial contact at even extremely low loadings of the filler. In contrast, very few of the aforementioned theories and simulations deal with the phase behavior and structure of mixtures of anisotropic fillers dispersed in single and multicomponent polymers.27-29 The rich literature of colloidal physics suggests that the equilibrium characteristics of anisotropic colloids can be expected to be significantly different and richer compared with their isotropic counterparts.30 Moreover, situations combining anisotropic fillers with multicomponent polymers such as block copolymers can be expected to reveal a rich interplay between polymer self-assembly and colloidal phase behavior. It will be of significant interest to PNC research to translate some of the theoretical advances achieved in the context of spherical fillers to the corresponding situations involving anisotropic fillers. Role of Functionalization: As mentioned earlier, functionalization of the fillers is a common strategy to overcome the issue of agglomeration of fillers. However, in much of the existing theories, the role of functionalization is treated either empirically as a modified polymer-filler interaction18, 23 or in more detailed formalism as a modification of the effective polymer-mediated potentials between the fillers. The first approach can be deemed reasonable for small oligomeric or surfactant-like functionalizers, but suffers from the drawback that it leaves open the question of the relationship between the polymer-filler interaction and the physicochemical properties of the functionalizing group. The second approach has typically been effected for the situation of grafted polymer functionalizers.31 It is, however, a pairwise interaction approach that works only for the homogeneous phase of the polymer matrix and in addition neglects multibody effects, which may become important when either long grafting polymers or higher concentrations of fillers are used. Overall, it would be desirable to develop a suite of simulations and/or modeling tools that, similar to the nongrafted case, can obviate such assumptions while still being able to predict the structure of the PNC dispersion. Structure under nonequilibrium and external field conditions: Despite the best experimental strategies, nonequilibrium effects resulting from filler aggregation and/or external fields are bound to remain important for many applications of PNCs, especially for situations involving anisotropic fillers. Some unresolved questions include: "How does the structure of a PNC dispersion evolve upon dispersing the fillers in the polymer matrix?," "Can quantities such as the fractal dimension and cluster size distributions be predicted for specified polymer-filler combinations?," "How does externally applied shear, electric, and magnetic fields (and combinations thereof) impact upon the nonequilibrium state of the dispersion?" Although computer simulations may shed light on some of the relevant issues, the time and length scales, which can be probed by such means may not necessarily overlap with experimental regimes, and there is a need for development of appropriate theoretical models to address pertinent issues. There have been some notable theoretical advances in this regard in the context of colloidal science,32, 33 and a preliminary assessment of their applicability to the field of PNCs would be extremely beneficial. However, issues unique to PNCs, such as the significant anisotropy of the fillers, potentially long-ranged interparticle interactions (mediated by the polymers), the dynamical and rheological response of the polymer matrix, are expected to prove important and provide fertile and challenging areas for theoretical research. Detailed structural characteristics: Many experimental studies of the mechanical, electrical, and rheological behavior of PNCs have suggested an intimate connection between enhanced macroscopic properties and the percolation behavior of the fillers.10, 34-36 Although such features can be directly extracted in computer simulations, existing theoretical frameworks mostly concern with the phase behavior and/or structural features at a pair correlation level. It may be of interest to the PNC experimental community to develop a theoretical framework, which can predict the the influence of different interactions, functionalization, and external fields upon more detailed structural characteristics such as the percolation characteristics37, 38 fractal and cluster dimensions of the fillers in the polymer matrix. Connection to atomistic details: Much of the above discussion focused on some open questions in the context of structure of PNCs as examined from a "coarse-grained" polymer physics perspective. Although such efforts are valuable for providing design guidelines, deducing mechanistic underpinnings and suggesting new experimental strategies, questions involving the detailed chemistry of the filler, polymer and functional groups cannot typically be answered in such frameworks. This situation is compounded by the fact that much of the coarse-grained modeling of PNCs use potentials and interaction parameters, which cannot also be directly determined from experimental measurements. The latter contrasts with the analogous situation in multicomponent polymer melts and solutions where experimental solubility measurements provide direct information on the parameters used in the corresponding coarse-grained theories.39 Addressing this shortcoming in the context of PNCs requires the development of efficient computer simulation tools and methodologies, which can render the connection between the chemistry of the components and the coarse-grained parameters more quantitative. The research field of "coarse-graining methodologies" is undergoing intense development in the context of simple and polymeric fluids.40-42 Although some notable "multiscale" efforts have also occurred in the context of PNCs,43, 44 much work remains to be done in surmounting the time, length scale challenges accompanying the interactions, structure, and properties of PNCs. Detailed structural information: Much of the aforementioned macroscopic theories were developed for model structural representations such as random or completely aligned dispersions while ignoring PNC specific features such as aggregation and/or complex structural arrangement of the fillers. Consequently, it is not surprising that such ideal models cannot quantitatively predict the properties of PNCs. One potential area for future research is the development of macroscopic models capable of accommodating detailed structural information, such as experimentally obtainable pair correlation functions of the fillers or cluster sizes and fractal dimension of aggregates, while rendering quantitative predictions regarding the macroscopic properties of PNCs. This area has recently seen some notable advances, especially arising from numerical continuum mechanical approaches.47-49 It may be envisioned that the results of such efforts can be used to guide the refinement of classical models to accommodate the PNC specific features. Interfacial effects: An important feature that distinguishes PNCs from traditional composites is the presence of significant amounts of polymer-filler interfaces and its influence upon the macroscopic properties. To highlight the resulting issues, I point out the status with respect to the "simpler" situation of polymer films supported on a solid substrate. In such contexts, experiments have clearly demonstrated that the polymer-surface interactions can lead to properties for the polymer layer, which are in general markedly different from that of the bulk polymeric material.50 Moreover, for properties such as glass transition temperatures,51 aging dynamics,52, 53 and elastic moduli,54 the interfacial effects have been shown to persist to extremely long length scales relative to the physical dimensions of the polymer. In the context of properties that exhibit only a short ranged interfacial effects (such as solubility of penetrants and mechanical properties), the aforementioned approaches may suffice to accommodate the influence of polymer-filler interfaces upon macroscopic properties. However, to endow a predictive value to theories, quantitative connections between the physicochemical polymer-filler interactions and the properties of the interfacial layer (or the parameters in the boundary conditions) in such models needs to be established. Although computer simulations may aid in this effort, in many cases, new formalisms may be necessary to discern quantities such as "local" mechanical properties,55 "local" conductivities, "local" penetrant diffusivities, and so forth, in such efforts. Once the connections between polymer-filler interactions and the interfacial properties are firmly established, "homogenization theories" similar to that used for the macroscopic properties of composites and dispersed media may be pursued. However, such approaches still need to be augmented to account for the possibility of complex structural arrangement of the fillers (an issue discussed earlier) and the overlap of interfacial layers which may consequently arise. As discussed earlier, some interfacial properties (such as the glass transition temperatures) are known to display long ranged variations influenced by the surface of the fillers, and in such cases, boundary condition and/or three-phase model approaches are likely to prove inadequate for capturing the influence of polymer-filler interfaces upon the macroscopic properties of the system. Refining the existing models to account for such properties represents an extremely challenging task, because in many cases there is a lack of understanding of the physics underlying such long ranged variations in polymer thin films. Moving to PNCs brings even more complications, such as the possibility for overlap of interfacial layers arising from different particles (such overlaps become very likely for long-ranged effects), leading to significant multibody interaction effects. Overall, this is an area for which the directions of theoretical modeling are far less clear (to me), suggesting that radically new ideas may be necessary to make progress in understanding and predicting the properties of polymer thin films and PNCs. The above represents an opinionated selection of some of the open questions confronting theoretical modeling of polymer nanocomposite systems. While active research is in progress in a number of the above-mentioned areas, successful fruition in many of the issues requires a confluence of ideas from different fields, viz., colloid and polymer physics, statistical and continuum mechanics, modeling and simulation approaches, and most importantly, experiments and theories. While technology and experiments are likely to provide more focus to the issues and theoretical efforts, progress in the above delineated issues is expected to have significant fundamental implications for many related areas of polymer and colloidal science, and may also potentially lead to uses and applications of PNCs above and beyond what has been realized or even dreamed about. This research program on polymer nanocomposites has been graciously supported by research grants from the National Science Foundation, American Chemical Society, Air Force Research Laboratories, Robert A. Welch Foundation, and the US Army Research Office under grant W911NF-07-1-0268. The author thanks Dr. Victor Pryamitsyn for useful comments which helped to shape the ideas expressed in this viewpoint. The author also acknowledges Profs. Glenn Fredrickson, Kenneth Schweizer, John Torkelson, Peter Green, Donald Paul, Jack Douglas, Sanat Kumar, Richard Vaia, and Ramanan Krishnamoorti for many discussions over the years while patiently educating him on the issues in polymer nanocomposites.

: US Army Research Office has funded this 2-year basic research effort by ICx-Agentase to study new techniques of modifying native enzymes with polymers via atom transfer radical polymerization to increase their solubility and utility in organic solvents and to increase their stability in body. Protein-initiated ATRP would enable us to overcome many problems in conventional technology that modifies proteins with polyethylene glycol or pegylation. These problems include lack of control on number of polymer chains attached per protein, low yield, tedious purifications, and less-than-optimal stabilization. The first goal of this study is to develop new technique/s to modify enzymes with hydrophobic polymers and dissolve high concentrations of modified enzymes in organic solvent and apply them in self decontaminating paints and coatings. The second goal of the study was to study the potential of using of ATRP modified proteins for therapeutic applications in vivo. Chymotrypsin and BChE conjugates synthesized by ATRP were demonstrated and were highly active and much more stable in mouse serum as compared to conventionally pegylated enzymes.

Annals of the New York Academy of SciencesVolume 410, Issue 1 p. 323-337 THERMAL AND SHOT NOISE EFFECTS ON NONLINEAR OSCILLATORS* E. Ben-Jacob, E. Ben-Jacob Department of Physics and Astronomy Tel Aviv University Tel Aviv, Israel Institute of Theoretical Physics, University of California, Santa Barbara, California 93106.Search for more papers by this authorD. J. Bergman, D. J. Bergman Department of Physics Ohio State University Columbus, Ohio 43210 Permanent address: Department of Physics and Astronomy, Tel Aviv University, Tel Aviv, Israel.Search for more papers by this authorB. J. Matkowsky, B. J. Matkowsky Department of Engineering Sciences and Applied Mathematics Technological Institute Northwestern University Evanston, Illinois 60201Search for more papers by this authorZ. Schuss, Z. Schuss Department of Engineering Sciences and Applied Mathematics Technological Institute Northwestern University Evanston, Illinois 60201 Permanent address: Department of Mathematics, Tel Aviv University, Tel Aviv, Israel.Search for more papers by this author E. Ben-Jacob, E. Ben-Jacob Department of Physics and Astronomy Tel Aviv University Tel Aviv, Israel Institute of Theoretical Physics, University of California, Santa Barbara, California 93106.Search for more papers by this authorD. J. Bergman, D. J. Bergman Department of Physics Ohio State University Columbus, Ohio 43210 Permanent address: Department of Physics and Astronomy, Tel Aviv University, Tel Aviv, Israel.Search for more papers by this authorB. J. Matkowsky, B. J. Matkowsky Department of Engineering Sciences and Applied Mathematics Technological Institute Northwestern University Evanston, Illinois 60201Search for more papers by this authorZ. Schuss, Z. Schuss Department of Engineering Sciences and Applied Mathematics Technological Institute Northwestern University Evanston, Illinois 60201 Permanent address: Department of Mathematics, Tel Aviv University, Tel Aviv, Israel.Search for more papers by this author First published: July 1983 https://doi.org/10.1111/j.1749-6632.1983.tb23331.xCitations: 4 * This research was supported, in part, by grants from the US Department of Energy, no. DE-AC02-78ERO-4650, from the US Army Research Office, no. DAAG29-79-C-0183, from the US Air Force Office for Scientific Research, no. 78–3602C, from the Israel Academy of Sciences, no. 707, and from the National Council for Research and Development, Kernfor-schungszentrum, Karlsruhe, West Germany. One of us (BJM) gratefully acknowledges the support of a John Simon Guggenheim Memorial Foundation Fellowship. AboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Citing Literature Volume410, Issue1Fifth International Conference on Collective PhenomenaJuly 1983Pages 323-337 RelatedInformation

Protonex wins contract to develop portable unit

Interactive robots not only need to react in predefined or deterministic scenarios but also learn and adapt in real-time, mirroring cognitive flexibility akin to human intelligence. Achieving this autonomy entails developing cognitive architectures that integrate reactive, deliberative and emergent capabilities. Thus, this paper presents MERLIN2, a hybrid cognitive architecture to generate, control, plan, and monitor behaviors in autonomous robots. This architecture combines reactive, deliberative, and emergent components, aiming to enhance adaptability in dynamic environments and make intelligent real-time decisions, thereby improving autonomy and performance. MERLIN2 comprises a deliberative system, based on a knowledge base and a symbolic planner; and a behavioral system composed of reactive components and several emergent components. It addresses core cognitive aspects like action selection, perception, memory, learning, reasoning, and explainability. MERLIN2 is evaluated in a simulated world and in the real world Carry My Luggage task from the RoboCup@Home. Therefore, the experimentation presented in this article showcases the architecture as a valid solution for autonomous robots.

In this paper, we implement a computational model of a neuromodulatory system in an autonomous robot. The model is based on a set of anatomical and physiological properties of the mammalian dopamine system, one of several diffuse ascending systems of the brain. The output of this system acts as a value signal, which modulates widely distributed synaptic changes in sensory and motor areas. During reward conditioning, the model learns to generate tonic and phasic responses, which are consistent with potential roles as reward predictions and prediction errors. Different sets of neural units generate precisely timed signals that exert positive effects (predictive) and negative effects (if a predicted reward is omitted) on neuroplasticity. We test the learning and behavior of the robot in different environmental contexts, and observe changes in the development of neural connections within the neuromodulatory system that depend on the robot's interaction with the environment. Simulation of a computational model responsive to rewarding stimuli leads to the emergence of conditioned appetitive behaviors. These studies represent a step towards investigating the behavior of autonomous robots controlled by biologically based neuromodulatory systems.

The matrix sensing problem is an important low-rank optimization problem that has found a wide range of applications, such as matrix completion, phase synchornization/retrieval, robust principal component analysis (PCA), and power system state estimation. In this work, we focus on the general matrix sensing problem with linear measurements that are corrupted by random noise. We investigate the scenario where the search rank r is equal to the true rank [Formula: see text] of the unknown ground truth (the exact parametrized case), as well as the scenario where r is greater than [Formula: see text] (the overparametrized case). We quantify the role of the restricted isometry property (RIP) in shaping the landscape of the nonconvex factorized formulation and assisting with the success of local search algorithms. First, we develop a global guarantee on the maximum distance between an arbitrary local minimizer of the nonconvex problem and the ground truth under the assumption that the RIP constant is smaller than [Formula: see text]. We then present a local guarantee for problems with an arbitrary RIP constant, which states that any local minimizer is either considerably close to the ground truth or far away from it. More importantly, we prove that this noisy, overparametrized problem exhibits the strict saddle property, which leads to the global convergence of perturbed gradient descent algorithm in polynomial time. The results of this work provide a comprehensive understanding of the geometric landscape of the matrix sensing problem in the noisy and overparametrized regime. Funding: This work was supported by grants from the National Science Foundation, Office of Naval Research, Air Force Office of Scientific Research, and Army Research Office.

This issue of Computer is based on two workshops in distributed processing held at Brown University August 17-19, 1976, and August 3-5, 1977. Sponsored by the Army Research Office, the National Science Foundation, and the Office of Naval Research, the workshops attempted to define what distributed processing means and to develop a taxonomy of distributed processing applications and techniques. Achievements to date and outstanding research problems were examined in an attempt to find either commonality of problems and solutions or substantial differences.

This paper deals with experimental and theoretical investigations aimed at increasing the basic understanding of the interaction of acoustic waves with random media. The ranges of frequency are 0.85–3.0 MHz and 5.0–6.0 MHz. The random medium consists of a suspension of polystyrene spheres (mean diameter 0.589 mm, standard deviation 0.066 mm) in a solution of water and sugar. For different concentrations, experimental results for attenuation and pulse broadening for transmission and backscattering, and attenuation as a function of the receiving angle for transmission only are presented. In the range of 0.85–1.1 MHz, the total scattering cross section of the particles depends on the 13th power of frequency. This dependence is due to the role of shear waves inside the particles. The power spectrum density of a burst of sine wave backscattered from the random medium is measured. The experimental and theoretical results agree well. The complex propagation constant of the coherent wave in the medium with large concentrations of particles is measured. It deviates from those predicted for a low‐density medium. For the theoretical calculations, the size distribution of the particles was considered. [Work supported by CNPq in Brazil and the U.S. Army Research Office and the Office of Naval Research.]

One approach to the design of intelligent autonomous robots is through evolutionary computation. In this approach, the robot’s behavior is evolved through a process of simulated evolution, applying the Darwinian principles of survival-of-the-fittest and inheritance-with-variation to the development of the robot’s control programs. In previous studies, we illustrated this approach on problems of learning individual behaviors for autonomous mobile robots. Our previous work has focused on tasks which were reasonably complex, but which required only a single behavior. In order to scale this approach to more realistic scenarios, we now consider methods for evolving complex sets of tasks. Our approach has been to extend the basic evolutionary learning method to encompasses co-evolution, that is, the simultaneously evolution of multiple behaviors. This paper addresses alternative designs within this basic paradigm. Specifically, we focus on dependencies among the learning agents, that is, what a given learning agent needs to know about other agents in the system. By using domain knowledge, it is possible to reduce or eliminate interactions among the agents, thereby reducing the effort required to co-evolve these agents as well as reducing the impediments to learning caused by these interactions.

Autonomous robots need to learn intelligent behaviour by interacting with the real world and adapting to changing environments. In order to study the adaptive behaviour of intelligent systems, an autonomous humanoid robot has been developed which understands goals, senses its environment in context with the goal and operates independently by means of experience gathering and a self-generating will.