京都大学 / 新制・課程博士 / 博士(情報学) / 甲第24980号 / 情博第853号 / 新制||情||143(附属図書館) / 京都大学大学院情報学研究科システム科学専攻 / (主査)教授 田中 利幸, 教授 下平 英寿, 准教授 本多 淳也 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

This cross-departmental initiative bridges Business and IT to establish a nexus for technological innovation, academic research, and tangible retail application, particularly in the realms of Robotics, Sensor Technology, Service Technology, Data Science driven by Artificial Intelligence within a retail setting. Conceived as both a living lab and an innovation hub, this project embodies a fully-operational retail store of the future, furnished with cutting-edge technologies and resourced by experts across varied disciplines and research domains. The overarching objective centers on facilitating knowledge generation and transfer among students, faculty, retailers, and technology companies. By doing so, the lab endeavors to foster collaborative solutions to aptly address the pressing challenges currently being faced by the retail industry, paving the way for sustainable, innovative developments for the future.

No description available.

In this thesis, analyses of the moisture budget in the 5th ECMWF reanalysis (ERA5) and in phase 6 of the Coupled Model Intercomparison Project (CMIP6) historical simulations and future projections are leveraged to explore the maintenance and response to climate change of the hydroclimate in the Mediterranean region. Recent and future projected changes in the Mediterranean, a climate-change hot and dry spot. The transition toward a warmer and drier climate has substantial social and economic implications; hence, it is critical to invest in understanding the Mediterranean hydroclimate change over the coming decades. One of our goals is to complement previous work by further decomposing the mean flow into contributions by the zonal-mean flow, which is dominated by the mean meridional circulation, and by zonally anomalous circulations and/or moisture, namely the stationary eddies. In present-day climate, as depicted by ERA5, annual-mean net evaporation (negative P-E) over the ocean and net precipitation (positive P-E) over land are primarily due to submonthly transient eddies converging moisture originating from the sea into the surrounding land. Overall, total stationary eddies reinforce the transient tendency over the ocean but oppose it over land, with the zonal-mean meridional circulation exerting a minor drying tendency limited to the region's southernmost latitudes. These large-scale features are captured quite well in the ensemble-mean of ten CMIP6 models analyzed in this work. The same CMIP6 model subset is thus used to study the response of the Mediterranean hydroclimate at the end of the 21st century under the SSP5-8.5 scenario. According to the CMIP6 multi-model mean, the climatological annual mean P – E is projected to decrease drastically by the end of the 21st century, both over northern Mediterranean land regions as well as the sea. These changes are not due to the transient eddies, which exert an overall negative but weak tendency, but are driven by changes in the time-mean flow. Consistent with the mean climatological moisture budget, this drying arises from the zonally anomalous circulation term, that is, enhanced zonally anomalous descent and lower-level diverging wind patterns over the Mediterranean region. Our results highlight the importance of circulation changes within the Mediterranean region and their impacts on the hydrological cycle. However, the thermodynamic adjustment is never trivial in our region and is deserving of further investigation. In particular, we are exploring if and to what extent an extended scaling, which is based only on climatological quantities and changes in surface properties, performs better than the simple Clausius-Clapeyron scaling by including thermodynamic changes in advection. Our analyses indeed show that, both in the annual mean and through the seasonal cycle, the extended scaling better captures the full thermodynamic component, which, unlike the simple scaling, predicts a wettening over the ocean. While not fully accounting for the magnitude nor the extent of this wettening, the extended scaling outperforms the simple scaling. %Areas of better agreement also include the nearby land regions, including France and the Iberian Peninsula, where the extended scaling predicts a stronger drying. Throughout the target region, the differences between the two scalings primarily arise from the contribution of the terms involving the gradients of fractional changes in local relative humidity and near-surface temperature changes. Even if largely cancelling, these two terms give rise to a pattern grossly characterized by moistening over the ocean and drying over neighbouring regions. Overall, the results of this thesis highlight how changes in the hydrological cycle in the Mediterranean region result from a complex interplay between different mechanisms, arising from both thermodynamic and dynamical changes. In particular, our results emphasize how the overall drying tendency in the region is primarily due to zonally asymmetric circulation changes rather than by changes in the mean meridional circulation, and is augmented by changes in transient eddies and those arising through the simple thermodynamic wet-get-wetter mechanism, and is partly opposed over the ocean and reinforced over the land regions by thermodynamic changes in advection. By shedding light on all of the involved mechanisms, this work advances our understanding of the factors that make the Mediterranean region a climate-change hot and dry spot.

The balance of the supply of renewable energy sources with electricity demand will become increasingly difficult with further penetration of renewable energy sources. Traditionally, large stationary batteries have been used to store renewable energy in excess of electricity demand and dispatch the stored energy to meet future electricity demand. Cool thermal energy storage is a feasible renewable energy balancing solution that has economic and environmental advantages over utility scale stationary lead-acid batteries. Two technologies, ice harvesters and internal-melt ice-on-coil cool thermal energy storage, have the capability to store excess renewable energy and use the energy to displace electricity used for building cooling systems. When implemented by a utility, cool thermal energy storage can replace large utility scale batteries for renewable energy balancing in utility regions with high renewable energy penetration. The California Independent System Operator (CAISO) region and the Electric Reliability Council of Texas (ERCOT) are utility regions with large solar and wind resources, respectively, that can benefit from installation of cool thermal energy storage systems for renewable energy balancing. With proper scheduling of energy dispatched from cool thermal energy storage, these technologies can be effective in displacing peak power capacity for the region, in displacing traditional building cooling equipment, and in recovering renewable energy that would otherwise be curtailed. / Master of Science

The existence of pathwise stationary solutions of this stochastic partial differential equation (SPDE, for abbreviation) is demonstrated. In Part II, a connection between certain kind of state constrained controlled Forward-Backward Stochastic Differential Equations (FBSDEs) and Hamilton-Jacobi-Bellman equations (HJB equations) are demonstrated. The special case provides a probabilistic representation of some geometric flows, including the mean curvature flows. Part II includes also a probabilistic proof of the finite time existence of the mean curvature flows.

This thesis aims to contribute towards a better understanding of the optimal coordination of monetary and fiscal policy in complex economic environments. We analyze the characteristics of optimal dynamics in an economy in which neither prices nor wages adjust instantaneously and lump-sum taxes are unavailable as a source of government finance. We then propose that monetary and fiscal policy should be coordinated to satisfy a pair of simple `specific targeting rules', a rule for inflation and a rule for the growth of real wages. We show that such simple rule-based conduct of policy can do remarkably well in replicating the dynamics of the economy under optimal policy following a given shock. We study optimal policy coordination in the context of an economy where a constant proportion of agents lacks access to the asset market. We find that the optimal economy moves along an analogue of a conventional inflation-output variance frontier in response to a government spending shock, as the population share of non-Ricardian agents rises. The optimal output response rises, while inflation volatility subsides. There is little evidence that increased government spending would crowd in private consumption in the optimal economy. We investigate the optimal properties and wider implications of a macroeconomic policy framework aimed at meeting an unconditional debt target. We show that the best stationary policy in terms of an unconditional welfare measure is characterized by highly persistent debt dynamics, less history-dependence in the conduct of policy, less reliance on debt finance and more short-term volatility following a government spending shock compared with the non-stationary `timelessly optimal' plan.

Ce travail présente des modèles d'écoulement en temps continu de flux de trafic routier. En premier lieu, il s'agit de modèles microscopiques de poursuite. Un modèle par systèmes d'équations différentielles couplées est proposé, basé sur le temps inter-véhiculaire. Ce modèle intègre un temps de réaction et des possibilités d'anticipation pour chaque véhicule. Les paramètres sont estimés par maximum de vraisemblance dans un modèle statistique à deux niveaux. Des simulations permettent de caractériser le comportement d'une file de véhicules. Dans une approche stochastique, un modèle d'évolution de la distance inter-véhiculaire est étudié à l'aide du processus Markovien de saut zero-range. L'introduction d'un temps de réaction tend à produire des ondes cinématiques. D'autre part, un modèle d'écoulement de trafic par le processus Markovien de saut des misanthropes est proposé. Il s'agit d'une modélisation au niveau mésoscopique, adaptée à la simulation de flux de trafic sur un réseau / This work presents different continuous-time traffic flow models. Microscopic models are considered first. A model by coupled differential equation system is proposed, based on the time gap. It incorporates a reaction time parameter and some anticipation possibilities, for each vehicle. The parameters are estimated by maximum likelihood over a two-level statistical model. Simulations allow to characterise the behaviour of a vehicles line. In a stochastic approach, a model of the distance gap evolution is studied with a zero-range process. The introduction of a reaction time parameter produces kinematics waves. On the other hand, traffic flow model by a misanthropes process is proposed. It is a mesoscopic approach, adapted to the simulation of traffic flow on a network

No description available.

This thesis is about the intelligent compression of large 3D point cloud datasets. The non-parametric method that we describe simultaneously generates a continuous representation of the workspace surfaces from discrete laser samples and decimates the dataset, retaining only locally salient samples. Our framework attains decimation factors in excess of two orders of magnitude without significant degradation in fidelity. The work presented here has a specific focus on gathering and processing laser measurements taken from a moving platform in outdoor workspaces. We introduce a somewhat unusual parameterisation of the problem and look to Gaussian Processes as the fundamental machinery in our processing pipeline. Our system compresses laser data in a fashion that is naturally sympathetic to the underlying structure and complexity of the workspace. In geometrically complex areas, compression is lower than that in geometrically bland areas. We focus on this property in detail and it leads us well beyond a simple application of non-parametric techniques. Indeed, towards the end of the thesis we develop a non-stationary GP framework whereby our regression model adapts to the local workspace complexity. Throughout we construct our algorithms so that they may be efficiently implemented. In addition, we present a detailed analysis of the proposed system and investigate model parameters, metric errors and data compression rates. Finally, we note that this work is predicated on a substantial amount of robotics engineering which has allowed us to produce a high quality, peer reviewed, dataset - the first of its kind.