This is the final part of the two- part series paper, where the simulation results namely leg’s trajectory, path tracing, obsacles detected, path constructed, movements of three motors and test like two-element chain three- element chain, are presented. I have left r&d, so will not correct update or complete papers. The earlier version of two parts, which was able to get saved from computer crash is uploaded.

This study shows the application of AI image generators with fine-tuning to the Kansei engineering design process.  Kansei engineering is a method for surveying and analyzing users’ latent feelings and preferences to product and service development (Nagamachi, 1991, 2011).. There are several common difficulties in applying Kansei engineering, including a lack of variety in evaluation samples, the need for non-designers to participate in the design process, and the fact that designers are often too busy to fully engage in the design process. In order to address these challenges, we have applied an AI image generator, StableDiffusion  (Rombach et al. 2021),to the Kansei engineering design process. The “Hypernetworks” framework to fine-tune StableDiffusion in order to make better representation of the shape and appearance of milk cartons. Based on previous Kansei engineering study on milk cartons (Ishihara et al., 1996) , several attempts are shown to use AI-generated designs to create innovative and visually appealing milk carton designs, including designs incorporating red and blue colors and abstract shapes, and a design inspired by the abstract paintings of Jackson Pollock. The authors conclude that the use of AI technology can be a valuable tool in boosting innovative design through Kansei engineering, and that further research is needed to explore the potential for interaction between humans and AI in the design process.

This paper presents a methodology of adapting complete open-source digital tooling, ISA, IPs and manufacture-able PDKs to tapeout a minimalist RISC-V based SoC.

Forest inventory forms the foundation of forest management. Remote sensing (RS) is an efficient means of measuring forest parameters at scale. Remotely sensed species classification can be used to estimate species abundances, distributions, and to better approximate metrics such as above ground biomass.  State of the art methods of RS species classification rely on deep learning models such as convolutional neural networks (CNN). These models have 2 major drawbacks: they require large samples of each species to classify well and they lack explainablity. Therefore, rare species are poorly classified causing poor approximations of their associated parameters. We show that the classification of rare species can be improved by as much as 8 F1-points using a neuro-symbolic (NS) approach that combines CNNs with a NS framework. The framework allows for the incorporation of domain knowledge into the model through the use of mathematically represented rules, improving model explainability.

Data transfer infrastructures composed of Data Transfer Nodes (DTN) are critical to supporting distributed computing and storage capabilities for clouds, data repositories, and complexes of supercomputers and instruments. The infrastructure’s throughput profile, estimated as a function of the connection round trip time using Machine Learning (ML) methods, is an indicator of its operational state, and has been utilized for monitoring, diagnosis and optimization purposes. We show that the inherent statistical variations and precision of the throughput profiles estimated by ML methods can be exploited for unauthorized use of DTNs’ computing and network capacity. We present a game theoretic formulation that captures the cost-benefit trade-offs between an attacker that attempts to hide under the profile’s statistical variations and a provider that counters by using additional measurements at an added cost. The Nash equilibrium conditions of this game provide qualitative insights and bounds for the success probabilities of the attacker and provider, based on the generalization equation of the ML-estimate. We present experimental results that illustrate scenarios wherein a significant portion of DTN computing capacity is compromised without being detected by an attacker that exploits the ML estimate properties.

This is the second part of two-part series where the novel hybrid design of 1.5 MVA 25kV traction powered by 415 kW solar grid, capable of fully supporting the entire load of the section of the Govt./DISCOM railway feeder has been presented. In this second Part-II, the per-unit system, zero sequence network diagram, controller design, fault calculations (L-L, L-G), idmt relay settings, safety measures and guaranteed scheme of relay trip coordination with 1.5 MVA 132kV/25kV State utility owned feeder and proposed 1.5 MVA 25kV feeder has been explained. The whole proposal is also simulated and the results are in the dataset. railway distribution feeder along with various compensators is explained. The whole proposal I have left r&d, so will not correct update or complete papers. The earlier version of two parts, which was able to get saved from computer crash is uploaded.

Driving requires continuous decision making from a driver taking into account all available relevant information. Automating driving tasks also automates the related decisions. However, humans are very good at dealing with bad quality, fuzzy, informal and incomplete information, whereas machines generally require solid quality information in a formalized format. Therefore, the development of automated driving functions relies on the availability of machine-usable information. A digital twin contains quality controlled information collected and augmented from different sources, ready to be supplied to such an automated driving function. An information model that describes all conceivably relevant information is necessary. To this end, a list of requirements that such an information model should meet is proposed and each requirement is argued for. Based on the anticipated services and applications that such a system should support, a collection of requirements for system architecture is derived. Information modeling is performed for selected relevant information groups. A system architecture has been proposed and validated with three different implementations, addressing several different applications to support decisions at a highway tunnel construction site in Austria and throughout the Test Bed Lower Saxony in Germany.

This is a four-part paper series, which elaborates the novel hybrid design where the rated capacity of 415 kW solar power plant producing guaranteed 1.66 MWh per day, owned by a private firm supplies electrical energy to its privately served consumers and also feeds 25 kV railway distribution feeder located 20 Km away from the premises. This 415V transmission is boosted and feeds the lagging feeding post of 40 km long traction line which has hybrid 132 kV/25 kV transmission from DISCOM or Govt. owned substation, supporting train speed upto 160km/hr. These four part paper series are complete guide, enough to motivate investors and successfully demonstrates relay trip coordination, leading the role of one section of feeding post from 132 kV/25kV utility and handling of other safety and fault conditions. In this Part-1, the solar plant design with anti-islanding, guaranteed uninterrupted power supply, typical railway distribution feeder, design of proposed feeder - its connection to feeder, mounting pole, booster transformer, circuit breakers, isolators, lightening arresters, has been explained. The whole proposal is also simulated and the results are in the dataset. I have left r&d, so will not correct update or complete papers. The earlier version of two parts, which was able to get saved from computer crash is uploaded.

We have observed a boom in video streaming over the Internet, especially during the Covid-19 pandemic, that could exceed the network resource availability. In addition to upgrading the network infrastructure, finding a way to smartly adapt the streaming system to the network and users’ conditions to satisfy clients’ perceptions is exceptionally critical, too. This paper proposes a new QoE-aware adaptive streaming scheme over HTTP - ABRA - to make flexible adaptations based on the network and the client’s current status. Besides, we propose a technique that can keep the buffer at an average high for more than 10s. We were limiting the phenomena of rebuffering due to unexpected and unpredictable bandwidth changes. The algorithm keeps the quality of subsequent versions at a constant level even when the average bitrate decreases, increasing the QoE. Experimental results show that our method can improve QoE from 7.86% to 20.41% compared to state-of-the-art methods. ABRA can provide better performance in terms of QoE score in all buffer conditions compared to the existing solutions while maintaining a minimum secured buffer level for the worst case.

Motivated by the large number of wearables offering geolocation, human mobility mining has emerged as an novel research field  within AI. The study of mobility creates increasingly predictable models in which it is easy to find patterns of behaviour. However, this data is not publicly available and access to it is restricted to large telecommunications operators. In this context, this paper aims to solve one of the main problems of human mobility databases, i.e. the scarcity of data for the generation of human mobility models. For this purpose, Generative adversarial network (GANs) have been proposed to generate synthetic time-series mobility data. Moreover, several neural network models are proposed to assess the impact of synthetic data generation on the prediction of human mobility. Our results show that the use of synthetic data improves predictions of human mobility compared to models based on available measured data.

This is the third part of four-part series paper, where the control dynamics built on transtibial mathematical model is explained. The trajectory development and control dynamics equations with respect to obstacles detected is explained with popular algorithms employed in robotics and self-driving vehicles but . In this paper, it is mathematically and practically proved that Lagrangian of the kinetic and potential energy of artificial leg can never be equal to zero and is always equal to non-conservative forces and thus, any model based on the assumption of conservative forces is fully invalid practically. I have left r&d, so will not correct update or complete papers. The earlier version of two parts, which was able to get saved from computer crash is uploaded.

This paper proposes the use of an event camera as a component of a vision system that enables counting of fast-moving objects - in this case, falling corn grains. These type of cameras transmit information about the change in brightness of individual pixels and are characterised by low latency, no motion blur, correct operation in different lighting conditions, as well as very low power consumption. The proposed counting algorithm processes events in real time. The operation of the solution was demonstrated on a stand consisting of a chute with a vibrating feeder, which allowed the number of grains falling to be adjusted. The objective of the control system with a PID controller was to maintain a constant average number of falling objects. The proposed solution was subjected to a series of tests to determine the correctness of the developed method operation. On their basis, the validity of using an event camera to count small, fast-moving objects and the associated wide range of potential industrial applications can be confirmed.

People have become increasingly aware of the environment in which they live in recent years. This awareness has prompted the development of a reliable environmental monitoring system. This paper presents an energy-efficient and low-power sensor node that could be used for outdoor Internet of Things (IoT) applications to monitor real-time environment conditions such as humidity and temperature remotely. The proposed sensor node could directly connect with a cellular base station over GSM/GPRS quad-bands 850MHz,900MHz,1800MHz, and 1900MHz without needing any Access Point(AP) in the network. The sensor node consists of a central controller based on Atmega328, a communication module on SIM800, DHT22 temperature, and a humidity sensor with a solar-powered charging system. The system is specially designed for robust IoT outdoor applications such as environment monitoring in wilds, large-scale farms, forests, etc. For optimal usage of the power of the battery, an energy-saving mechanism is proposed based on deep sleep scheduling, thus considerably reducing the power consumption and prolonging the up-time of the sensor node.

Although the concept of engineered molecular communication has been around for quite some time, practical approaches with truly biocompatible setups are still scarce. However, molecular communication has a large potential in future medical applications and may be a solution to size constraints of antenna based transmission systems. In this work we therefore present a testbed using biocompatible magnetic nanoparticles. Based on previous work, all testbed components have been improved regarding performance and size, making a large step forwards regarding miniaturisation and a data transmission approach. In addition, a setup for localised twodimensional sensing of magnetic nanoparticles is presented. All improvements are evaluated individually and combined to achieve a net data rate of more than 6 bit/s, significantly higher than any other comparable biocompatible setup.

In medium to high frequency power transformers, apart from minimizing the power loss in the core and copper, the design of its thermal circuit plays an important role in containing the temperature rise in different parts. It is extremely complex because there exist multiple loss centers with different heat removal characteristics. Moreover, the power loss is concentrated around a small core volume and the rest of it is available for convection and radiation. This article proposes to use a hybrid core configuration to improve the heat removal feature of the transformer. This article establishes through practical validation on suitability i.e., the magnetic compatibility of cores in the hybrid magnetic circuit. It further explores the application prospect of the proposed concept where the thermal performances have been compared using different core types in two different application domains. In the first case, the core could always remain energized to its rated value, and in the second design, both the windings could always draw their respective rated current.

This paper presents a fully integrated second-order level-crossing sampling data converter for sensing sparse signals with real-time detection of turning points. Compared with con- ventional level-crossing sampling ADCs which sample at fixed voltage levels, the proposed circuits update tracking thresholds using linear prediction, which forms sloped sampling levels. The prediction and calculation are done digitally by modifying the digital control logic of a conventional SAR ADC. The system sepa- rates the sampling and quantization processes and only selects the turning points in the input waveform for quantization. The output of the proposed data converter consists of both the quantized amplitudes of the selected sampling points and the timestamps between the selected sampling points. The main advantages of the proposed circuits are data savings and power savings for the following digital signal processing or communication circuits. The test chip was fabricated using a 180nm CMOS process. When sensing sparse signals such as ECG signals, the proposed ADC saves 30% compared to a conventional SAR ADC with an equal quantizer resolution and achieves a compression factor of 6.17.

Accurate assessment of pleural line is crucial for the application of lung ultrasound (LUS) in monitoring lung diseases, thereby aim of this study is to develop a quantitative analysis method of pleural line. At first, the novel cascaded deep learning model based on convolution and multilayer perceptron was proposed to locate and segment the pleural line in LUS images, whose results were applied for quantitative analysis of textural and morphological features, respectively. By using gray-level co-occurrence matrix and self-designed statistical methods, eight textural and three morphological features were generated to characterize the pleural lines. Furthermore, the machine learning-based classifiers were employed to evaluate the lesion degree of pleural line in LUS images. We prospectively evaluated 5390 LUS images acquired from 31 pneumonia patients. Experimental results demonstrated that the extraction and assessment effect has reached 0.87 (i.e., Dice) and 94.47% (i.e., Accuracy), respectively, and the comparison with previous methods found statistical significance (P<0.001 for all). Thus, the proposed method has great application potential for assessment of pleural line on LUS images and aid lung disease diagnosis and treatment.

Diabetes mellitus (DB) is the most challenging and fastest-growing global public health challenge. An estimated 10.5% of the global adult population suffers from diabetes, and almost half of them are undiagnosed. The growing at-risk population exacerbated the shortage of health resources, with an estimated 10.6% and 6.2% of adults worldwide having impaired glucose tolerance (IGT) and impaired fasting glycemia (IFG), respectively. All the current diabetes screening methods are invasive and opportunistic and must be conducted in a hospital or a laboratory by trained professionals. At-risk subjects might remain undetected for years and miss the precious time window for early intervention in preventing or delaying the onset of diabetes and its complications. This study was conducted at KK Womenâ\euro™s and Childrenâ\euro™s Hospital of Singapore, and five hundred participants were recruited (mean age 38.73 ± 10.61 years; mean BMI 24.4 ± 5.1 kg/m2). The blood glucose levels, for most participants, were measured before and after 75g of sugary drink using both the conventional glucometer (Accu-Chek Performa) and the wrist-worn wearable. The results obtained from the glucometer were used as the ground truth measurements. We propose leveraging photoplethysmography (PPG) sensors and machine learning techniques to incorporate this into an affordable wrist-worn wearable device to detect elevated blood glucose levels (⩾ 7.8mmol/L) non-invasively. Multiple machine learning models were trained and assessed with 10-fold cross-validation using subject demographic data and critical features extracted from the PPG measurements as predictors. Support vector machine (SVM) with a radial basis function kernel has the best detection performance with an average accuracy of 84.7%, a sensitivity of 81.05%, a specificity of 88.3%, a precision of 87.51%, a geometric mean of 84.54% and F-score of 84.03%. Hence, PPG measurements can be utilized to identify subjects with elevated blood glucose measurements and assist in the screening of subjects for diabetes risk.