| 1. |
-
2021 Vehicle Dynamics seminar
- 2021
-
Proceedings (redaktörskap) (övrigt vetenskapligt/konstnärligt)abstract
- The seminar is held annually. The full title of this year's seminar was "2021 Vehicle Dynamics seminar -- for Future Mobility ...and not only Lateral".
|
|
| 2. |
- Aaboen, Lise, 1978, et al.
(författare)
-
Relationship development patterns of university-based start-ups
- 2021
-
Ingår i: A Research Agenda for the Entrepreneurial University. ; , s. 207-228
-
Bokkapitel (övrigt vetenskapligt/konstnärligt)abstract
- In the present chapter, we focus on how start-ups become embedded in the university and industry context(s) in order to suggest a research agenda for a more systemic approach to university and industry actors when studying start-up development in entrepreneurial universities. More specifically, the story should not end in business formation but an entrepreneurial university also has roles in the further development of start-ups originating both from the university and industry. We base our suggested research agenda on a case study of the automotive and transport cluster in Western Sweden, which is established as both an academic context and an industry context. We relied both on secondary data about the case as such as well as primary data about the relationship development patterns of the 9 start-ups in the cluster. We identified 5 different patterns of how start-ups become embedded. Arguing that the development and success of university-based start-ups have to be understood in terms of how they embed in their context(s) entails several important implications for further studies connected to additional details of the development patterns, the roles universities play in the networking and embedding of start-ups and the design of collaborative platforms for collaboration between various actors.
|
|
| 3. |
- Aarrestad, Thea, et al.
(författare)
-
Fast convolutional neural networks on FPGAs with hls4ml
- 2021
-
Ingår i: Machine Learning: Science and Technology. - : IOP Publishing. - 2632-2153. ; 2:4
-
Tidskriftsartikel (refereegranskat)abstract
- We introduce an automated tool for deploying ultra low-latency, low-power deep neural networks with convolutional layers on field-programmable gate arrays (FPGAs). By extending the hls4ml library, we demonstrate an inference latency of 5 mu s using convolutional architectures, targeting microsecond latency applications like those at the CERN Large Hadron Collider. Considering benchmark models trained on the Street View House Numbers Dataset, we demonstrate various methods for model compression in order to fit the computational constraints of a typical FPGA device used in trigger and data acquisition systems of particle detectors. In particular, we discuss pruning and quantization-aware training, and demonstrate how resource utilization can be significantly reduced with little to no loss in model accuracy. We show that the FPGA critical resource consumption can be reduced by 97% with zero loss in model accuracy, and by 99% when tolerating a 6% accuracy degradation.
|
|
| 4. |
|
|
| 5. |
- Abayazid, Fady, et al.
(författare)
-
A New Assessment of Bicycle Helmets: The Brain Injury Mitigation Effects of New Technologies in Oblique Impacts
- 2021
-
Ingår i: Annals of Biomedical Engineering. - : Springer Science and Business Media LLC. - 1573-9686 .- 0090-6964. ; 49:10, s. 2716-2733
-
Tidskriftsartikel (refereegranskat)abstract
- New helmet technologies have been developed to improve the mitigation of traumatic brain injury (TBI) in bicycle accidents. However, their effectiveness under oblique impacts, which produce more strains in the brain in comparison with vertical impacts adopted by helmet standards, is still unclear. Here we used a new method to assess the brain injury prevention effects of 27 bicycle helmets in oblique impacts, including helmets fitted with a friction-reducing layer (MIPS), a shearing pad (SPIN), a wavy cellular liner (WaveCel), an airbag helmet (Hövding) and a number of conventional helmets. We tested whether helmets fitted with the new technologies can provide better brain protection than conventional helmets. Each helmeted headform was dropped onto a 45° inclined anvil at 6.3 m/s at three locations, with each impact location producing a dominant head rotation about one anatomical axes of the head. A detailed computational model of TBI was used to determine strain distribution across the brain and in key anatomical regions, the corpus callosum and sulci. Our results show that, in comparison with conventional helmets, the majority of helmets incorporating new technologies significantly reduced peak rotational acceleration and velocity and maximal strain in corpus callosum and sulci. Only one helmet with MIPS significantly increased strain in the corpus collosum. The helmets fitted with MIPS and WaveCel were more effective in reducing strain in impacts producing sagittal rotations and a helmet fitted with SPIN in coronal rotations. The airbag helmet was effective in reducing brain strain in all impacts, however, peak rotational velocity and brain strain heavily depended on the analysis time. These results suggest that incorporating different impact locations in future oblique impact test methods and designing helmet technologies for the mitigation of head rotation in different planes are key to reducing brain injuries in bicycle accidents.
|
|
| 6. |
- Abbasi, Rosa, et al.
(författare)
-
Deductive Verification of Floating-Point Java Programs in KeY
- 2021
-
Ingår i: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). - Cham : Springer International Publishing. - 1611-3349 .- 0302-9743. ; 12652 LNCS, s. 242-261
-
Konferensbidrag (refereegranskat)abstract
- Deductive verification has been successful in verifying interesting properties of real-world programs. One notable gap is the limited support for floating-point reasoning. This is unfortunate, as floating-point arithmetic is particularly unintuitive to reason about due to rounding as well as the presence of the special values infinity and ‘Not a Num-ber’ (NaN). In this paper, we present the first floating-point support in a deductive verification tool for the Java programming language. Our support in the KeY verifier handles arithmetic via floating-point decision procedures inside SMT solvers and transcendental functions via axiomatization. We evaluate this integration on new benchmarks, and show that this approach is powerful enough to prove the absence of floating-point special values—often a prerequisite for further reasoning about numerical computations—as well as certain functional properties for realistic benchmarks.
|
|
| 7. |
- Abbaspour, S., et al.
(författare)
-
Real-Time and Offline Evaluation of Myoelectric Pattern Recognition for the Decoding of Hand Movements
- 2021
-
Ingår i: Sensors. - : MDPI AG. - 1424-8220. ; 21:16
-
Tidskriftsartikel (refereegranskat)abstract
- Pattern recognition algorithms have been widely used to map surface electromyographic signals to target movements as a source for prosthetic control. However, most investigations have been conducted offline by performing the analysis on pre-recorded datasets. While real-time data analysis (i.e., classification when new data becomes available, with limits on latency under 200-300 milliseconds) plays an important role in the control of prosthetics, less knowledge has been gained with respect to real-time performance. Recent literature has underscored the differences between offline classification accuracy, the most common performance metric, and the usability of upper limb prostheses. Therefore, a comparative offline and real-time performance analysis between common algorithms had yet to be performed. In this study, we investigated the offline and real-time performance of nine different classification algorithms, decoding ten individual hand and wrist movements. Surface myoelectric signals were recorded from fifteen able-bodied subjects while performing the ten movements. The offline decoding demonstrated that linear discriminant analysis (LDA) and maximum likelihood estimation (MLE) significantly (p < 0.05) outperformed other classifiers, with an average classification accuracy of above 97%. On the other hand, the real-time investigation revealed that, in addition to the LDA and MLE, multilayer perceptron also outperformed the other algorithms and achieved a classification accuracy and completion rate of above 68% and 69%, respectively.
|
|
| 8. |
- Abdel Hady, Ahmed Wagih Abdallah, 1990, et al.
(författare)
-
Micro/nano-structural evolution in spruce wood during soda pulping
- 2021
-
Ingår i: Holzforschung. - : Walter de Gruyter GmbH. - 1437-434X .- 0018-3830. ; 75:8, s. 754-764
-
Tidskriftsartikel (refereegranskat)abstract
- Alkaline delignification of wood tissue is the core of the global pulping technology and the most prominent large-scale separation of the main wood components. This work aims at improved understanding of the interplay between the topochemistry of alkaline pulping and the associated morphological changes. Morphology and chemical structure of partially soda-delignified wood chips were studied combining X-ray tomography (XRT), X-ray diffraction analysis and compositional characterization (lignin and carbohydrate content). The XRT studies of wet samples (providing 3D structural information without interfering drying effects), allowed observation of the cell wall separation as an increasing amount of lignin was removed with the increasing pulping time. Comparison between the microstructure of the surface and the central parts of the treated chips showed a more delignified microstructure at the surface, which highlights the dependence of the delignification process on the mass transport (hydroxide ions and lignin fragments) through the wood tissue. The crystallite size of cellulose increased in the <200> crystal planes during the early stage of pulping while there was little effect on the <110> plane.
|
|
| 9. |
- Abdollahi, Mehdi, 1985, et al.
(författare)
-
Impact of processing technology on macro- and micronutrient profile of protein-enriched products from fish backbones
- 2021
-
Ingår i: Foods. - : MDPI AG. - 2304-8158. ; 10:5
-
Tidskriftsartikel (refereegranskat)abstract
- Impacts of processing technology (mechanical separation and pH-shift processing) on protein recovery from salmon, herring and cod backbones and the content of macro- and micronutrients in the recovered protein enriched products were investigated. Mechanical separation led to higher protein recovery compared with the pH-shift process and using both techniques, recovery ranked the species as herring > salmon > cod. However, the pH-shift process up-concentrated protein from herring and salmon backbones more efficiently than mechanical separation by removing more fat and ash. This consequently reduced n-3 PUFA and vitamin D content in their protein isolates compared with the backbones and mechanically separated meat (MSM). Cod protein isolate, however, contained higher levels of these nutrients compared with MSM. Mechanical separation concentrated vitamins E and C in salmon MSM but not for cod and herring. Opposite, pH-shift processing reduced levels of these two vitamins for cod and herring backbones, while vitamins D and C were reduced for salmon. For minerals, selenium, calcium, magnesium, and potassium were lower in protein isolates than MSM, while copper, zinc, iron and manganese were similar or higher. Overall, there is a major potential for upcycling of fish backbones to food ingredients, but processing technology should be carefully balanced against the desired nutrient profile and final application area.
|
|
| 10. |
- Abdurrokhman, Iqbaal, 1991
(författare)
-
Binary mixtures and cationic modification of protic ionic liquids. Local structure and transport properties
- 2021
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Protic ionic liquids are a subclass of ionic liquids, which in many cases are obtained by mixing an equimolar amount of Brønsted acid and base. Protic ionic liquids possess an exchangeable proton that gives them distinct features, e.g. the potential to conduct protons and the ability to form extended hydrogen bonds, which is different from the case of their aprotic counterparts. The protic ionic liquids considered in this thesis are limited to those based on the imidazolium and triazolium cation, which (compared to aprotic ionic liquids) are less developed and have been investigated to a lesser extent. The work included in this thesis has been devoted to investigate strategies that could possibly promote a fast proton motion. These strategies have primarily consisted in mixing a protic ionic liquid with a molecular solvent or another parented ionic liquid and in modifying the cationic structure by either attaching alkyl chains of different length to the imidazolium cation or synthesizing new protic ionic liquids derived from another heterocyclic ring such as triazole. In both cases, a detailed characterization by means of vibrational spectroscopy could provide valuable information about intermolecular interactions. An important finding from the studies conducted in this thesis is that altering the cation's structure or adding a co-solvent leads to a significant change in chemical and transport properties. This thesis also includes an extensive work devoted to develop a new solvent-free method to synthesize dry and pure protic ionic liquids, in the laboratory scale. An analytical approach to accurately quantify their purity using nuclear magnetic spectroscopy has also been developed. In addition, the ionic conductivity of the investigated protic ionic liquid systems has been evaluated by use of impedance spectroscopy, from room temperature up to 140 °C. The results included in my work indicate that the two investigated strategies can potentially be used, each with its limitations, to develop thermochemically stable proton conducting materials based on protic ionic liquids.
|
|
