Sensor Planning for Mobile Robot Localization---A Hierarchical Approach Using a Bayesian Network and a Particle Filter

In this paper, we propose a hierarchical approach to solving sensor planning for the global localization of a mobile robot. Our system consists of two subsystems: a lower layer and a higher layer. The lower layer uses a particle filter to evaluate the posterior probability of the localization. When the particles converge into clusters, the higher layer starts particle clustering and sensor planning to generate an optimal sensing action sequence for the localization. The higher layer uses a Bayesian network for probabilistic inference. The sensor planning takes into account both localization belief and sensing cost. We conducted simulations and actual robot experiments to validate our proposed approach.     

Mathematical modeling of meniscus profile and melt flow in electromagnetic casters

The paper describes a mathematical model which predicts the meniscus shape and melt flow in an electromagnetic caster. Computations were carried out for the two types of caster (with and without a screen to shape the magnetic field) in commercial use. The dependence of meniscus shape on parameters such as inductor geometry, placement, current, and frequency, as well as screen properties and placement, was determined. Calculated velocities showed agreement with measurements of other investigators on a physical model and an actual caster. The effect of an auxiliary low frequency inductor on melt flow was also computed.     

Measurements of the (n, np+d) activation cross sections with 13.4–14.9 MeV neutrons using a fusion neutronics source/JAERI

The activation cross sections for 20 (n, np+d) reactions were measured in the energy range between 13.4 and 14.9 MeV by the activation method. The mass-separated isotopes of ⁸⁷Sr, ^95,100Mo, ¹⁰⁴Ru, ¹⁰⁶Pd, ^113,116Cd, ^118,119,120Sn, ^123,128,130Te, ^184,186W, and ^189,190Os were irradiated. The 16 cross sections, excepting those for ¹¹⁸Sn, ¹²⁸Te and ^184,186W, were obtained for the first time. The d–T neutron source of the fusion neutronics source (FNS) at the Japan Atomic Energy Research Institute (JAERI) was used for irradiation. All cross section values were determined relative to that of the ²⁷Al (n, ) ²⁴Na reaction (ENDF/B-VI). To measure weakly induced activities, an efficiency calibration technique with a well-type HPGe detector was applied. The present results were compared with the comprehensive evaluated data in the JENDL-3.3, the JENDL-Activation File, the ENDF/B-VI and the FENDL/A-2.0. Most of the data in the JENDL-3.3 and the JENDL-Activation File were in good agreement with the present result. However, relative to our values, 13 of the 20 evaluated data in FENDL/A-2.0 were overestimated more than 2 times or underestimated by less than one tenth.     

Comparison of performance and emissions of a supercharged dual-fuel engine fueled by hydrogen and hydrogen-containing gaseous fuels

This study investigated the engine performance and emissions of a supercharged engine fueled by hydrogen (H₂), and three other hydrogen-containing gaseous fuels such as primary fuels, and diesel as pilot fuel in dual-fuel mode. The energy share of primary fuels was about 90% or more, and the rest of the energy was supplied by diesel fuel. The hydrogen-containing fuels tested in this study were 13.7% H₂-content producer gas, 20% H₂-content producer gas and 56.8% H₂-content coke oven gas (COG). Experiments were carried out at a constant pilot injection pressure and pilot quantity for different fuel-air equivalence ratios and at various injection timings. The experimental strategy was to optimize the pilot injection timing to maximize engine power at different fuel-air equivalence ratios without knocking and within the limit of the maximum cylinder pressure. Better thermal efficiency was obtained with the increase in H₂ content in the fuels, and neat H₂ as a primary fuel produced the highest thermal efficiency. The fuel-air equivalence ratio was decreased with the increase in H₂ content in the fuels to avoid knocking. Thus, neat H₂-operation produced less maximum power than other fuels, because of much leaner operations. Two-stage combustion was obtained; this is an indicator of maximum power output conditions and a precursor of knocking combustion. The emissions of CO and HC with neat H₂-operation were 98-99.9% and NOx about 85-90% less than other fuels.     

MICROSTRUCTURAL STUDY OF CYCLIC STRAIN HARDENING BEHAVIOUR IN BIAXIAL STRESS STATES AT ELEVATED TEMPERATURE

This paper describes the cyclic strain hardening behaviour and dislocation structures of material in biaxial low cycle fatigue at elevated temperatures. In this study, push-pull, reversed torsion and combined push-pull/reversed torsion tests were carried out using a type 304 stainless steel in air. While there was no significant difference between the cyclic stress amplitudes in the push-pull and reversed torsion tests on a von Mises' base, combination tests exhibited a 40% increase in stress amplitude. Most of the dislocations in the first two types of test adopted ladder or maze structures, while in the later case cells were found. Changing the loading mode at a certain cycle, for example, from push-pull to reversed torsion, revealed that stress amplitude depended mainly on the concurrent applied strain mode and furthermore, that the strain mode before the interchange had little or no effect on the stress amplitude after the interchange. Tests were also performed in order to examine how prestrained material hardened in the three different loading modes, with the following results: prestrained material in push-pull or in reversed torsion exhibited an anisotropic stress response, while the material in the combined tests exhibited an isotropic response. These cyclic responses are discussed in connection with the dislocation structure.     

Distinct specificity in the binding of inositol phosphates by pleckstrin homology domains of pleckstrin, RAC-protein kinase, diacylglycerol kinase and a new 130 kDa protein

Psychological assessments made at the time of bone marrow transplantation (BMT) were examined in 31 patients at 1-2 years posttransplantation and factors associated with survival were explored. Psychological assessments were carried out soon after admission to hospital for the BMT and about 3-4 weeks later. Cox regression survival analysis revealed that an interactive biopsychological model could explain survival status. Shorter survival was associated with mismatched marrow grafts (p = 0.04), prior experience with chemoradiotherapy (p-0.02), disease stage (p = 0.04), higher symptom distress during BMT (p = 0.008), less hopefulness (p = 0.005), and more acceptance of the situation (p = 0.02). These psychological/psychosomatic and personality characteristics may act directly by interacting with immune function or indirectly by leading to other behaviors known to affect survival in cancer patients. Enhancing more effective coping strategies and altering high symptom-related distress during BMT seem important means not only for improving psychosocial adjustment (quality of life), as has been shown in the literature, but also for increasing survival (quantity of life).     

Multilevel QAM Modulation Techniques for Digital Microwave Radios

A pilot carrier injection method is described together with feedback balance coding which reduces spectral power near the carrier. Robustness of carrier recovery using the pilot carrier injection method is theoretically estimated. The estimation suggests that recovered carrier SNR higher than 40 dB can be expected even under muitipath fading with notch depth of 45 dB located just at the carrier frequency. Signatures for multipath fading are estimated for a 64-QAM system with transversal equalizers as a countermeasure. Measured signatures agree reasonably well with the calculated ones. Dependences of signatures on modulation level, transversal equalizer tap number, and rolloff rate are also shown.