Can audio-visual speech recognition outperform acoustically enhanced speech recognition in automotive environment?

The use of visual features in the form of lip movements to improve the performance of acoustic speech recognition has been shown to work well, particularly in noisy acoustic conditions. However, whether this technique can outperform speech recognition incorporating well-known acoustic enhancement techniques, such as spectral subtraction, or multi-channel beamforming is not known. This is an important question to be answered especially in an automotive environment, for the design of an efficient human-vehicle computer interface. We perform a variety of speech recognition experiments on a challenging automotive speech dataset and results show that synchronous HMM-based audio-visual fusion can outperform traditional single as well as multi-channel acoustic speech enhancement techniques. We also show that further improvement in recognition performance can be obtained by fusing speech-enhanced audio with the visual modality, demonstrating the complementary nature of the two robust speech recognition approaches.

Multiple cameras for audio-visual speech recognition in an automotive environment

Audio-visualspeechrecognition, or the combination of visual lip-reading with traditional acoustic speechrecognition, has been previously shown to provide a considerable improvement over acoustic-only approaches in noisy environments, such as that present in an automotive cabin. The research presented in this paper will extend upon the established audio-visualspeechrecognition literature to show that further improvements in speechrecognition accuracy can be obtained when multiple frontal or near-frontal views of a speaker's face are available. A series of visualspeechrecognition experiments using a four-stream visual synchronous hidden Markov model (SHMM) are conducted on the four-camera AVICAR automotiveaudio-visualspeech database. We study the relative contribution between the side and central orientated cameras in improving visualspeechrecognition accuracy. Finally combination of the four visual streams with a single audio stream in a five-stream SHMM demonstrates a relative improvement of over 56% in word recognition accuracy when compared to the acoustic-only approach in the noisiest conditions of the AVICAR database.

Low-cost hardware speech enhancement for improved speech recognition in automotive environments

Voice recognition is one of the key enablers to reduce driver distraction as in-vehicle systems become more and more complex. With the integration of voice recognition in vehicles, safety and usability are improved as the driver’s eyes and hands are not required to operate system controls. Whilst speaker independent voice recognition is well developed, performance in high noise environments (e.g. vehicles) is still limited. La Trobe University and Queensland University of Technology have developed a low-cost hardware-based speech enhancement system for automotive environments based on spectral subtraction and delay–sum beamforming techniques. The enhancement algorithms have been optimised using authentic Australian English collected under typical driving conditions. Performance tests conducted using speech data collected under variety of vehicle noise conditions demonstrate a word recognition rate improvement in the order of 10% or more under the noisiest conditions. Currently developed to a proof of concept stage there is potential for even greater performance improvement.

FPGA implementation of dual-microphone delay-and-sum beamforming for in-car speech enhancement and recognition

In an automotive environment, the performance of a speech recognition system is affected by environmental noise if the speech signal is acquired directly from a microphone. Speech enhancement techniques are therefore necessary to improve the speech recognition performance. In this paper, a field-programmable gate array (FPGA) implementation of dual-microphone delay-and-sum beamforming (DASB) for speech enhancement is presented. As the first step towards a cost-effective solution, the implementation described in this paper uses a relatively high-end FPGA device to facilitate the verification of various design strategies and parameters. Experimental results show that the proposed design can produce output waveforms close to those generated by a theoretical (floating-point) model with modest usage of FPGA resources. Speech recognition experiments are also conducted on enhanced in-car speech waveforms produced by the FPGA in order to compare recognition performance with the floating-point representation running on a PC.

Rivalry within and between strategic networks : an investigation of the United States Automotive Industry

As a consequence of the increased incidence of collaborative arrangements between firms, the competitive environment characterising many industries has undergone profound change. It is suggested that rivalry is not necessarily enacted by individual firms according to the traditional mechanisms of direct confrontation in factor and product markets, but rather as collaborative orchestration between a number of participants or network members. Strategic networks are recognised as sets of firms within an industry that exhibit denser strategic linkages among themselves than other firms within the same industry. Based on this, strategic networks are determined according to evidence of strategic alliances between firms comprising the industry. As a result, a single strategic network represents a group of firms closely linked according to collaborative ties. Arguably, the collective outcome of these strategic relationships engineered between firms suggest that the collaborative benefits attributed to interorganisational relationships require closer examination in respect to their propensity to influence rivalry in intraindustry environments. Derived in large from the social sciences, network theory allows for the micro and macro examination of the opportunities and constraints inherent in the structure of relationships in strategic networks, establishing a relational approach upon which the conduct and performance of firms can be more fully understood. Research to date has yet to empirically investigate the relationship between strategic networks and rivalry. The limited research that has been completed utilising a network rationale to investigate competitive patterns in contemporary industry environments has been characterised by a failure to directly measure rivalry. Further, this prior research has typically embedded investigation in industry settings dominated by technological or regulatory imperatives, such as the microprocessor and airline industries. These industries, due to the presence of such imperatives, are arguably more inclined to support the realisation of network rivalry, through subscription to prescribed technological standards (eg., microprocessor industry) or by being bound by regulatory constraints dictating operation within particular market segments (airline industry). In order to counter these weaknesses, the proposition guiding research - Are patterns of rivalry predicted by strategic network membership? – is embedded in the United States Light Vehicles Industry, an industry not dominated by technological or regulatory imperatives. Further, rivalry is directly measured and utilised in research, thus distinguishing this investigation from prior research efforts. The timeframe of investigation is 1993 – 1999, with all research data derived from secondary sources. Strategic networks were defined within the United States Light Vehicles Industry based on evidence of horizontal strategic relationships between firms comprising the industry. The measure of rivalry used to directly ascertain the competitive patterns of industry participants was derived from the traditional Herfindahl Index, modified to account for patterns of rivalry observed at the market segment level. Statistical analyses of the strategic network and rivalry constructs found little evidence to support the contention of network rivalry; indeed, greater levels of rivalry were observed between firms comprising the same strategic network than between firms participating in opposing network structures. Based on these results, patterns of rivalry evidenced in the United States Light Vehicle Industry over the period 1993 – 1999 were not found to be predicted by strategic network membership. The findings generated by this research are in contrast to current theorising in the strategic network – rivalry realm. In this respect, these findings are surprising. The relevance of industry type, in conjunction with prevailing network methodology, provides the basis upon which these findings are contemplated. Overall, this study raises some important questions in relation to the relevancy of the network rivalry rationale, establishing a fruitful avenue for further research.

The effectiveness of marketing communication within the immersive environment

Immersive environments are part of a recent media innovation that allow users to become so involved within a computer-based simulated environment that they feel part of that virtual world (Grigorovici, 2003). A specific example is Second Life, which is an internet-based, three-dimensional immersive virtual world in which users create an online representation of themselves (an avatar) to play games and interact socially with thousands of people simultaneously. This study focuses on Second Life as an example of an immersive environment, as it is the largest adult freeform virtual world, home to 12 million avatars (IOWA State University, 2008). Already in Second Life there are more than 100 real-life brands from a range of industries, including automotive, professional services, and consumer goods and travel, among others (KZero, 2007; New Business Horizons, 2009). Compared to traditional advertising media, this interactive media can immerse users in the environment. As a result of this interactivity, users can become more involved with a virtual environment, resulting in prolonged usage over weeks, months and even years. Also, it can facilitate presence. Despite these developments, little is known about the effectiveness of marketing messages in a virtual world context. Marketers are incorporating products into Second Life using a strategy of online product placement. This study, therefore, explores the perceived effectiveness of online product placement in Second Life in terms of effects on product/brand recall, purchase intentions and trial. This research examines the association between individuals’ involvement with Second Life and online product placement effectiveness, as well as the relationship between individuals’ Second Life involvement and the effectiveness of online product placement. In addition, it investigates the association of immersion and product placement involvement. It also examines the impact of product placement involvement on online product placement effectiveness and the role of presence in affecting this relationship. An exploratory study was conducted for this research using semi-structured in-depth interviews face-to-face, email-based and in-world. The sample comprised 24 active Second Life users. Results indicate that product placement effectiveness is not directly associated with Second Life involvement, but rather effectiveness is impacted through the effect of Second Life involvement on product placement involvement. A positive relationship was found between individuals’ product placement involvement and online product placement effectiveness. Findings also indicate that online product placement effectiveness is not directly associated with immersion. Rather, it appears that effectiveness is impacted through the effect of immersion on product placement involvement. Moreover, higher levels of presence appear to have a positive impact on the relationship between product placement involvement and product placement effectiveness. Finally, a model was developed from this qualitative study for future testing. In terms of theoretical contributions, this study provides a new model for testing the effectiveness of product placement within immersive environments. From a methodological perspective, in-world interviews as a new research method were undertaken. In terms of a practical contribution, findings identified useful information for marketers and advertising agencies that aim to promote their products in immersive virtual environments like Second Life.

Reverse engineering and finite element analysis of a 3-wheeler automotive structure

Auto rickshaws (3-wheelers) are the most sought after transport among the urban and rural poor in India. The assembly of the vehicle involves assemblies of several major components. The L-angle is the component that connects the front panel with the vehicle floor. Current L-angle part has been observed to experience permanent deformation failure over period of time. This paper studies the effect of the addition of stiffeners on the L-angle to increase the strength of the component. A physical model of the L-angle was reversed engineered and modelled in CAD before static loading analysis were carried out on the model using finite element analysis. The modified L-angle fitted with stiffeners was shown to be able to withstand more load compare to previous design.

Inclusion of tapered tubes in enhancing the crash performance of automotive frontal structures

This paper treats the design and analysis of an energy absorbing system. Experimental tests were conducted on a prototype, and these tests were used to validate a finite element model of the system. The model was then used to analyze the response of the system under dynamic impact loading. The response was compared with that of a similar system consisting of straight circular tubes, empty and foam-filled conical tubes. Three types of such supplementary devices were included in the energy absorbing system to examine the crush behavior and energy absorption capacity when subjected to axial and oblique impact loadings. The findings were used to develop design guidelines and recommendations for the implementation of tapered tubes in energy absorbing systems. To this end, the system was conceptual in form such that it could be adopted for a variety of applications. Nevertheless, for convenience, the approach in this study is to treat the system as a demonstrator car bumper system used to absorb impact energy during minor frontal collisions.

SOI sensing technologies for harsh environment

This paper reviews and addresses certain aspects of Silicon-On-Insulator (SOI) technologies for a harsh environment. The paper first describes the need for specialized sensors in applications such as (i) domestic and other small-scale boilers, (ii) CO2 Capture and Sequestration, (iii) oil & gas storage and transportation, and (iv) automotive. We describe in brief the advantages and special features of SOI technology for sensing applications requiring temperatures in excess of the typical bulk silicon junction temperatures of 150oC. Finally we present the concepts, structures and prototypes of simple and smart micro-hotplate and Infra Red (IR) based emitters for NDIR (Non Dispersive IR) gas sensors in harsh environments. © 2012 IEEE.

High-current inductors for high-power automotive DC-DC converters

This thesis is focused on the investigation of magnetic materials for high-power dcdc converters in hybrid and fuel cell vehicles and the development of an optimized high-power inductor for a multi-phase converter. The thesis introduces the power system architectures for hybrid and fuel cell vehicles. The requirements for power electronic converters are established and the dc-dc converter topologies of interest are introduced. A compact and efficient inductor is critical to reduce the overall cost, weight and volume of the dc-dc converter and optimize vehicle driving range and traction power. Firstly, materials suitable for a gapped CC-core inductor are analyzed and investigated. A novel inductor-design algorithm is developed and automated in order to compare and contrast the various magnetic materials over a range of frequencies and ripple ratios. The algorithm is developed for foil-wound inductors with gapped CC-cores in the low (10 kHz) to medium (30 kHz) frequency range and investigates the materials in a natural-convection-cooled environment. The practical effects of frequency, ripple, air-gap fringing, and thermal configuration are investigated next for the iron-based amorphous metal and 6.5 % silicon steel materials. A 2.5 kW converter is built to verify the optimum material selection and thermal configuration over the frequency range and ripple ratios of interest. Inductor size can increase in both of these laminated materials due to increased airgap fringing losses. Distributing the airgap is demonstrated to reduce the inductor losses and size but has practical limitations for iron-based amorphous metal cores. The effects of the manufacturing process are shown to degrade the iron-based amorphous metal multi-cut core loss. The experimental results also suggest that gap loss is not a significant consideration in these experiments. The predicted losses by the equation developed by Reuben Lee and cited by Colonel McLyman are significantly higher than the experimental results suggest. Iron-based amorphous metal has better preformance than 6.5 % silicon steel when a single cut core and natural-convection-cooling are used. Conduction cooling, rather than natural convection, can result in the highest power density inductor. The cooling for these laminated materials is very dependent on the direction of the lamination and the component mounting. Experimental results are produced showing the effects of lamination direction on the cooling path. A significant temperature reduction is demonstrated for conduction cooling versus natural-convection cooling. Iron-based amorphous metal and 6.5% silicon steel are competitive materials when conduction cooled. A novel inductor design algorithm is developed for foil-wound inductors with gapped CC-cores for conduction cooling of core and copper. Again, conduction cooling, rather than natural convection, is shown to reduce the size and weight of the inductor. The weight of the 6.5 % silicon steel inductor is reduced by around a factor of ten compared to natural-convection cooling due to the high thermal conductivity of the material. The conduction cooling algorithm is used to develop high-power custom inductors for use in a high power multi-phase boost converter. Finally, a high power digitally-controlled multi-phase boost converter system is designed and constructed to test the high-power inductors. The performance of the inductors is compared to the predictions used in the design process and very good correlation is achieved. The thesis results have been documented at IEEE APEC, PESC and IAS conferences in 2007 and at the IEEE EPE conference in 2008.

An architectural approach to autonomics and self-management of automotive embedded electronic systems

Embedded electronic systems in vehicles are of rapidly increasing commercial importance for the automotive industry. While current vehicular embedded systems are extremely limited and static, a more dynamic configurable system would greatly simplify the integration work and increase quality of vehicular systems. This brings in features like separation of concerns, customised software configuration for individual vehicles, seamless connectivity, and plug-and-play capability. Furthermore, such a system can also contribute to increased dependability and resource optimization due to its inherent ability to adjust itself dynamically to changes in software, hardware resources, and environment condition. This paper describes the architectural approach to achieving the goals of dynamically self-configuring automotive embedded electronic systems by the EU research project DySCAS. The architecture solution outlined in this paper captures the application and operational contexts, expected features, middleware services, functions and behaviours, as well as the basic mechanisms and technologies. The paper also covers the architecture conceptualization by presenting the rationale, concerning the architecture structuring, control principles, and deployment concept. In this paper, we also present the adopted architecture V&V strategy and discuss some open issues in regards to the industrial acceptance.

Improving MTM-UAS to predetermine automotive maintenance times

The preventive knowledge of serviceability times is a critical factor for the quantification of after-sales services costs of a vehicle. Predetermined motion time system are frequently used to set labor rates in industry by quantifying the amount of time required to perform specific tasks. The first such system is known as Methods-time measurement (MTM). Several variants of MTM have been developed differing from each other on their level of focus. Among them MTM-UAS is suitable for processes that average around 1-3 min. However experimental tests carried out by the authors in Elasis (Research Center of FIAT Group) demonstrate that MTM-UAS is not the optimal approach to measure serviceability times. The reason is that it doesn't take into account ergonomic factors. In the present paper the authors propose to correct the MTM-UAS method including in the task analysis the study of human postures and efforts. The proposed approach allows to estimate with an "acceptable" error the time needed to perform maintenance tasks since the first phases of product design, by working on Digital Mock-up and human models in virtual environment. As a byproduct of that analysis, it is possible to obtain a list of maintenance times in order to preventively set after-sales service costs. © 2012 Springer-Verlag.