Use of neural networks in robot positioning of large flexible redundant manipulators

Deflection compensation of flexible boom structures in robot positioning is usually done using tables containing the magnitude of the deflection with inverse kinematics solutions of a rigid structure. The number of table values increases greatly if the working area of the boom is large and the required positioning accuracy is high. The inverse kinematics problems are very nonlinear, and if the structure is redundant, in some cases it cannot be solved in a closed form. If the structural flexibility of the manipulator arms is taken into account, the problem is almost impossible to solve using analytical methods. Neural networks offer a possibility to approximate any linear or nonlinear function. This study presents four different methods of using neural networks in the static deflection compensation and inverse kinematics solution of a flexible hydraulically driven manipulator. The training information required for training neural networks is obtained by employing a simulation model that includes elasticity characteristics. The functionality of the presented methods is tested based on the simulated and measured results of positioning accuracy. The simulated positioning accuracy is tested in 25 separate coordinate points. For each point, the positioning is tested with five different mass loads. The mean positioning error of a manipulator decreased from 31.9 mm to 4.1 mm in the test points. This accuracy enables the use of flexible manipulators in the positioning of larger objects. The measured positioning accuracy is tested in 9 separate points using three different mass loads. The mean positioning error decreased from 10.6 mm to 4.7 mm and the maximum error from 27.5 mm to 11.0 mm.

Economic Geographical Analysis of The Finnish Paper Industry

The purpose of this academic economic geographical dissertation is to study and describe how competitiveness in the Finnish paper industry has developed during 2001–2008. During these years, the Finnish paper industry has faced economically challenging times. This dissertation attempts to fill the existing gap between theoretical and empirical discussions concerning economic geographical issues in the paper industry. The main research questions are: How have the supply chain costs and margins developed during 2001–2008? How do sales prices, transportation, and fixed and variable costs correlate with gross margins in a spatial context? The research object for this case study is a typical large Finnish paper mill that exports over 90 % of its production. The economic longitudinal research data were obtained from the case mill’s controlled economic system and, correlation (R2) analysis was used as the main research method. The time series data cover monthly economic and manufacturing observations from the mill from 2001 to 2008. The study reveals the development of prices, costs and transportation in the case mill, and it shows how economic variables correlate with the paper mills’ gross margins in various markets in Europe. The research methods of economic geography offer perspectives that pay attention to the spatial (market) heterogeneity. This type of research has been quite scarce in the research tradition of Finnish economic geography and supply chain management. This case study gives new insight into the research tradition of Finnish economic geography and supply chain management and its applications. As a concrete empirical result, this dissertation states that the competitive advantages of the Finnish paper industry were significantly weakened during 2001–2008 by low paper prices, costly manufacturing and expensive transportation. Statistical analysis expose that, in several important markets, transport costs lower gross margins as much as decreasing paper prices, which was a new finding. Paper companies should continuously pay attention to lowering manufacturing and transporting costs to achieve more profitable economic performance. The location of a mill being far from markets clearly has an economic impact on paper manufacturing, as paper demand is decreasing and oversupply is pressuring paper prices down. Therefore, market and economic forecasting in the paper industry is advantageous at the country and product levels while simultaneously taking into account the economic geographically specific dimensions.

Geographical Studies of Underwater Light Dynamics in the Coastal Archipelago of SW Finland, Baltic Sea

The underwater light field is an important environmental variable as it, among other things, enables aquatic primary production. Although the portion of solar radiation that is referred to as visible light penetrates water, it is restricted to a limited surface water layer because of efficient absorption and scattering processes. Based on the varying content of optical constituents in the water, the efficiency of light attenuation changes in many dimensions and over various spatial and temporal scales. This thesis discusses the underwater light dynamics of a transitional coastal archipelago in south-western Finland, in the Baltic Sea. While the area has long been known to have a highly variable underwater light field, quantified knowledge on the phenomenon has been scarce, patchy, or non-existent. This thesis focuses on the variability in the underwater light field through euphotic depths (1% irradiance remaining), which were derived from in situ measurements of vertical profiles of photosynthetically active radiation (PAR). Spot samples were conducted in the archipelago of south-western Finland, mainly during the ice-free growing seasons of 2010 and 2011. In addition to quantifying both the seasonal and geographical patterns of euphotic depth development, the need and usability of underwater light information are also discussed. Light availability was found to fluctuate in multiple dimensions and scales. The euphotic depth was shown to have combined spatio-temporal dynamics rather than separate changes in spatial and temporal dimensions. Such complexity in the underwater light field creates challenges in data collection, as well as in its utilisation. Although local information is needed, in highly variable conditions spot sampled information may only poorly represent its surroundings. Moreover, either temporally or spatially limited sampling may cause biases in understanding underwater light dynamics. Consequently, the application of light availability data, for example in ecological modelling, should be made with great caution.