Five experiments are presented in this thesis. They investigate the impact of the interpreter's language exposure on the interpreter's fluency in simultaneous interpreting. The measurement of the interpreter's fluency involves a quantitative procedure developed by Kirsner and his colleagues (2002). The procedure is based on computer-assisted analyses of pause and speech segment durations and associated variables. The quantitative measures are employed to determine whether or not different levels of the language exposure influence fluency in simultaneous interpreting. The impact of the interpreter's language exposure is elaborated upon within the framework of dynamic systems theory. Experiment 1 is a pilot case study investigating the applicability of the methodology developed by Kirsner and his colleagues (2002, 2005) to the research in fluency in simultaneous interpreting. In Experiments 2-5 this methodology is extended to investigate the impact of the participants' language exposure on fluency in simultaneous interpreting involving various language pair combinations. Experiment 2 investigates the interpreter students' exposure to their third working language, Norwegian. It has been found that the participant's fluency in the experimental tasks critically depends on the amount of out-of-classroom language exposure, involving the language exposure to Swedish, a language typologically closely related to Norwegian. Experiment 3 investigates the impact of the interpreter students' language exposure gained during the stay abroad in Germany on their fluency in interpretation from/into German. Significant gains in fluency due to the stay abroad have been found between the groups of participants and their respective controls who studied German at their respective home universities. Experiments 4 and 5 explore the impact of the on-going language exposure to the interpreter's second language in the conditions of discontinuity in practicing simultaneous interpreting. The participants with the language pair combinations Finnish/Swedish (experiment 4) and English/Russian (experiment 5) have discontinued practicing simultaneous interpreting, yet enjoy a continuous exposure to their working languages. The participants have exhibited superior fluency measures compared to their respective controls (beginner and advanced students), and inferior fluency measures compared to the control group of professional interpreters. Data analysis of all the participants under all experimental conditions have not yielded any pause duration distributions specific to simultaneous interpreting.

This thesis provides an overview of work conducted to investigate human spatial hearing in situations involving multiple concurrent sound sources. Much is known about spatial hearing with single sound sources, including the acoustic cues to source location and the accuracy of localisation under different conditions. However, more recently interest has grown in the behaviour of listeners in more complex environments. Concurrent sound sources pose a particularly difficult problem for the auditory system, as their identities and locations must be extracted from a common set of sensory receptors and shared computational machinery. It is clear that humans have a rich perception of their auditory world, but just how concurrent sounds are processed, and how accurately, are issues that are poorly understood. This work attempts to fill a gap in our understanding by systematically examining spatial resolution with multiple sound sources. A series of psychophysical experiments was conducted on listeners with normal hearing to measure performance in spatial localisation and discrimination tasks involving more than one source. The general approach was to present sources that overlapped in both frequency and time in order to observe performance in the most challenging of situations. Furthermore, the role of two primary sets of location cues in concurrent source listening was probed by examining performance in different spatial dimensions. The binaural cues arise due to the separation of the two ears, and provide information about the lateral position of sound sources. The spectral cues result from location-dependent filtering by the head and pinnae, and allow vertical and front-rear auditory discrimination. Two sets of experiments are described that employed relatively simple broadband noise stimuli. In the first of these, two-point discrimination thresholds were measured using simultaneous noise bursts. It was found that the pair could be resolved only if a binaural difference was present; spectral cues did not appear to be sufficient. In the second set of experiments, the two stimuli were made distinguishable on the basis of their temporal envelopes, and the localisation of a designated target source was directly examined. Remarkably robust localisation was observed, despite the simultaneous masker, and both binaural and spectral cues appeared to be of use in this case. Small but persistent errors were observed, which in the lateral dimension represented a systematic shift away from the location of the masker. The errors can be explained by interference in the processing of the different location cues. Overall these experiments demonstrated that the spatial perception of concurrent sound sources is highly dependent on stimulus characteristics and configurations. This suggests that the underlying spatial representations are limited by the accuracy with which acoustic spatial cues can be extracted from a mixed signal. Three sets of experiments are then described that examined spatial performance with speech, a complex natural sound. The first measured how well speech is localised in isolation. This work demonstrated that speech contains high-frequency energy that is essential for accurate three-dimensional localisation. In the second set of experiments, spatial resolution for concurrent monosyllabic words was examined using similar approaches to those used for the concurrent noise experiments. It was found that resolution for concurrent speech stimuli was similar to resolution for concurrent noise stimuli. Importantly, listeners were limited in their ability to concurrently process the location-dependent spectral cues associated with two brief speech sources. In the final set of experiments, the role of spatial hearing was examined in a more relevant setting containing concurrent streams of sentence speech. It has long been known that binaural differences can aid segregation and enhance selective attention in such situations. The results presented here confirmed this finding and extended it to show that the spectral cues associated with different locations can also contribute. As a whole, this work provides an in-depth examination of spatial performance in concurrent source situations and delineates some of the limitations of this process. In general, spatial accuracy with concurrent sources is poorer than with single sound sources, as both binaural and spectral cues are subject to interference. Nonetheless, binaural cues are quite robust for representing concurrent source locations, and spectral cues can enhance spatial listening in many situations. The findings also highlight the intricate relationship that exists between spatial hearing, auditory object processing, and the allocation of attention in complex environments.

This thesis deals with the Simultaneous Localisation and Mapping algorithm as it pertains to the deployment of mobile systems in unknown environments. Simultaneous Localisation and Mapping (SLAM) as defined in this thesis is the process of concurrently building up a map of the environment and using this map to obtain improved estimates of the location of the vehicle. In essence, the vehicle relies on its ability to extract useful navigation information from the data returned by its sensors. The vehicle typically starts at an unknown location with no a priori knowledge of landmark locations. From relative observations of landmarks, it simultaneously computes an estimate of vehicle location and an estimate of landmark locations. While continuing in motion, the vehicle builds a complete map of landmarks and uses these to provide continuous estimates of the vehicle location. The potential for this type of navigation system for autonomous systems operating in unknown environments is enormous. One significant obstacle on the road to the implementation and deployment of large scale SLAM algorithms is the computational effort required to maintain the correlation information between features in the map and between the features and the vehicle. Performing the update of the covariance matrix is of O(n�) for a straightforward implementation of the Kalman Filter. In the case of the SLAM algorithm, this complexity can be reduced to O(n�) given the sparse nature of typical observations. Even so, this implies that the computational effort will grow with the square of the number of features maintained in the map. For maps containing more than a few tens of features, this computational burden will quickly make the update intractable - especially if the observation rates are high. An effective map-management technique is therefore required in order to help manage this complexity. The major contributions of this thesis arise from the formulation of a new approach to the mapping of terrain features that provides improved computational efficiency in the SLAM algorithm. Rather than incorporating every observation directly into the global map of the environment, the Constrained Local Submap Filter (CLSF) relies on creating an independent, local submap of the features in the immediate vicinity of the vehicle. This local submap is then periodically fused into the global map of the environment. This representation is shown to reduce the computational complexity of maintaining the global map estimates as well as improving the data association process by allowing the association decisions to be deferred until an improved local picture of the environment is available. This approach also lends itself well to three natural extensions to the representation that are also outlined in the thesis. These include the prospect of deploying multi-vehicle SLAM, the Constrained Relative Submap Filter and a novel feature initialisation technique. Results of this work are presented both in simulation and using real data collected during deployment of a submersible vehicle equipped with scanning sonar.

Thesis (Ph.D.)--University of Texas at Dallas, 2007. / Includes vita. Includes bibliographical references (leaves 94-103)

Zugl.: Dresden, Techn. Univ., Diss., 2004

University, Diss., 2006--Karlsruhe.

Zugl.: Darmstadt, Techn. Univ., Diss., 2008

Universiẗat, Hochsch. für Wirtschafts-, Rechts- und Sozialwiss., Diss.--St. Gallen, 1997.

Thesis (Ph. D.)--University of Texas at El Paso, 2009. / Title from title screen. Vita. CD-ROM. Includes bibliographical references. Also available online.

Techn. Universiẗat, Diss., 2000--Berlin.