Multiple Gaussian process models for direct time series forecasting

This paper focuses on the problem of time series forecasting using the Gaussian process models. The Gaussian process model is a nonparametric model and the output of the model has Gaussian distribution with mean and variance. The multiple Gaussian process models as every step ahead predictors are used for time series forecasting in accordance with the direct approach. The separable least‐squares approach that combines the genetic algorithm with the linear least‐squares method is applied to train these Gaussian process models. Simulation results are shown to illustrate the effectiveness of the proposed direct forecasting method. © 2011 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

16 Kb/s high quality voice encoding for satellite communication networks

This paper describes applications of adaptive predictive coding (APC) with maximum likelihood quantization (MLQ) which can cover a wide range of coding rates from 4.8 to 16 kb/s for low C/N satellite communication systems, such as maritime, aeronautical mobile and thin-route satellite communication systems, and also for speech and data integration, including digital circuit multiplication equipment (DCME) in business communication systems, such as INTELSAT business services (IBS). A 16 kb/s APC–MLQ hardware codec has been implemented by NEC–7720 DSP chips and the performance has been confirmed in subjective quality of speech through conversational tests. The objective performance has also been evaluated for non-voice signals, such as single and multi-frequency tones, and 1200 and 2400 b/s voiceband data signals. The APC-MLQ codec can transmit the voice-band data at 1200 b/s over two asynchronous tandem links and at 2400 b/s over one link. It was noted that the APC-MLQ codec is superior in speech performance at 16 kb/s to a narrow-band companded FM and meets requirements for low C/N satellite communication systems. For voice and data integration into 16 kb/s for 64 kb/s links, we propose a multi-media multiplexing for low C/N digital satellite communication systems and also a small-scale circuit multiplication system for business use. In these systems, a variable rate coding of APC-MLQ from 4.8 to 16 kb/s can be effectively introduced for voice and data integration.     

Real-time control software for transaction processing based on colored safe Petri Net model

A Petri Net-based real-time control software system for transaction processing is proposed to improve flexibility and maintainability in real-time application software. This software system is named Station Controller (SCR). Transaction processing control programs can be understandably described using Control Net (C-net) which is a problem-oriented language based on a colored safe Petri Net model. The described C-net program is directly executed by a net-interpreter installed on a 16-bit micro-computer system. In this paper, some features of the SCR are described and a successful application of the SCR is illustrated.     

Non-parametric identification of continuous-time Hammerstein systems using Gaussian process model and particle swarm optimization

This paper deals with a non-parametric identification of continuous-time Hammerstein systems using Gaussian process (GP) models. A Hammerstein system consists of a memoryless non-linear static part followed by a linear dynamic part. The identification model is derived using the GP prior model which is described by the mean function vector and the covariance matrix. This prior model is trained by the separable least-squares (LS) approach combining the linear LS method with particle swarm optimization to minimize the negative log marginal likelihood of the identification data. Then the non-linear static part is estimated by the predictive mean function of the GP, and the confidence measure of the estimated non-linear static part is evaluated by the predictive covariance function of the GP. Simulation results are shown to illustrate the proposed method.     

Experimental investigation into influence of negative attitudes toward robots on human–robot interaction

Negative attitudes toward robots are considered as one of the psychological factors preventing humans from interacting with robots in the daily life. To verify their influence on humans‘ behaviors toward robots, we designed and executed experiments where subjects interacted with Robovie, which is being developed as a platform for research on the possibility of communication robots. This paper reports and discusses the results of these experiments on correlation between subjects’ negative attitudes and their behaviors toward robots. Moreover, it discusses influences of gender and experience of real robots on their negative attitudes and behaviors toward robots.

Oxidative dehydrogenation of methanol over a ruthenium-containing polyoxomolybdate supported on metal oxides chemically modified with silane coupling agent

A salt of ruthenium-containing polyoxomolybdate anion ([Ru₂Mo₁₄O₅₀]¹⁰⁻ or [Ru₂Mo₁₄O₅₂]¹⁴⁻; Ru₂Mo₁₄) was synthesized by mixing ruthenium chloride and sodium molybdate in a buffer solution (acetic acid–ammonium acetate) at pH 5. The polyanion, Ru₂Mo₁₄, was loaded on a silica carrier chemically modified with a silane coupling agent having a 1,2-diaminoethyl group (DAPS–SiO₂). X-ray diffraction, FT-IR and surface area measurements revealed that Ru₂Mo₁₄ was highly dispersed on DAPS–SiO₂ up to 20 wt.% loading. The highly dispersed Ru₂Mo₁₄ exhibited a high level of activity for oxidative dehydrogenation of methanol to formaldehyde with a high selectivity, whereas combustion of methanol occurred on the bulky salt of Ru₂Mo₁₄ and ruthenium catalyst supported on a SiO₂ carrier.     

Partial oxidation of methane using a microscale non-equilibrium plasma reactor

A flow-type, microscale, non-equilibrium plasma reactor was developed for partial oxidation of methane without a catalyst. A wide range of oxygen and methane mixtures was directly processed without dilution or explosion at ambient temperature because the microscale plasma reactor removes excess heat generated by partial oxidation, thereby maintaining a reaction field at temperatures near room temperature. Consequently, the least reactive methane was excited by high-energy electrons, whereas successive destruction of reactive oxygenates was minimized simultaneously within the extremely confined environment. A highly reactive and quenching environment is thereby obtained within a single reactor: these are paradoxical conditions in conventional thermochemical processes. A major product among liquid oxygenates was methanol, whose selectivity reached 34% at 30% of methane conversion. Selectivity of oxygenates such as methanol and formaldehyde depends strongly on the fragmentation pattern of methane dissociation by electron impact. Maximum selectivity of oxygenates, which is estimated from numerical simulation of a filamentary microdischarge, reaches 60% when the applied electric field corresponds to the breakdown field of methane (80 Td, 1 Td = 10⁻¹⁷ V cm²). The discharge current increases markedly with an applied electric field, but the selectivity of oxygenates decreases as the field strength increases.     

Effect of polymerization conditions on the syndiotactic-specificity in radical polymerization of N-isopropylacrylamide and fractionation of the obtained polymer according to the stereoregularity

Radical polymerization of N-isopropylacrylamide (NIPAAm) was examined in the presence of hexamethylphosphoramide (HMPA). The addition of an excess amount of HMPA induced syndiotactic-specificity that gradually enhanced as the feed monomer was consumed. The syndiotacticity of the obtained poly(NIPAAm)s was improved by increasing the [HMPA]₀/[NIPAAm]₀ ratio to five and prolonging the polymerization time to 96 h (racemo=72%). It was also revealed that more stereoregulated poly(NIPAAm) could be fractionated by reprecipitating the resulting polymers from hexane-THF mixture. This result suggested that more stereoregulated poly(NIPAAm) showed a lower solubility than less stereoregulated poly(NIPAAm)s. Furthermore, unusual hysteresis was observed in transmittance analysis of an aqueous solution of the fractionated syndiotactic poly(NIPAAm).     

Effects of arousing attention on distracted driver’s following behaviour under uncertainty

This paper proposes issuing alerts to arouse driver attention when the traffic situation ahead is uncertain. Here, “uncertain” means that there exist cues suggesting something is happening in traffic ahead, but it is unclear to the driver and the driver assistance system whether an emergency situation will really occur in the near future, because traffic ahead cannot be seen directly. In such a situation, the driver has to pay attention ahead to prepare a quick response in case an emergency occurs. An experiment with a high-fidelity driving simulator was conducted to investigate the effectiveness of an attention arousing system when the driver was distracted. The results showed that the attention arousing system contributed to increasing time headway (THW) to the lead vehicle, where THW is the headway distance divided by the host vehicle speed and is often used to evaluate the rear-end collision risk. Such safety compensating behaviour was observed not only under distracted conditions, but also under non-distracted conditions. However, the compensating behaviour was not excessive under non-distracted conditions. The attention arousing was effective to maintain safety when the lead vehicle decelerated rapidly. The reaction time was long when the drivers were performing a distracting task. However, because the drivers had increased THW beforehand when conducting the distracting secondary task, the delay in reaction did not worsen the situation.