IMITATION LEARNING AND ANCHORING THROUGH CONCEPTUAL SPACES

In order to have a robotic system able to effectively learn by imitation and not merely reproduce the movements of a human teacher, the system should have the capability to deeply understand the perceived actions to be imitated. This paper deals with the development of a cognitive architecture for learning by imitation in which a rich conceptual representation of the observed actions is built. The purpose of the following discussion is to show how the same conceptual representation can be used both in a bottom-up approach, in order to learn sequences of actions by imitation learning paradigm, and in a top-down approach, in order to anchor the symbolical representations to the perceptual activities of the robotic system. Experiments concerned with the problem of teaching a humanoid robotic system simple manipulative tasks are reported.     

Proximity-Coupling-Induced Significant Enhancement of Coercive Field and Curie Temperature in 2D van der Waals Heterostructures

Magnetism in 2D has long been the focus of condensed matter physics due to its important applications in spintronic devices. A particularly promising aspect of 2D magnetism is the ability to fabricate 2D heterostructures with engineered optical, electrical, and quantum properties. Recently, the discovery of intrinsic ferromagnetisms in atomic thick materials has provided a new platform for investigations of fundamental magnetic physics. In contrast to 2D CrI₃ and Cr₂Ge₂Te₆ insulators, itinerant ferromagnetic Fe₃GeTe₂ (FGT), which has a larger intrinsic perpendicular anisotropy, higher Curie temperature (T_C), and relatively better stability, is a promising candidate for achieving permanent room-temperature ferromagnetism through interface or component engineering. Here, it is shown that the ferromagnetic properties of FGT thin flakes can be modulated through coupling with a FePS₃. The magneto-optical Kerr effect results show that the T_C of FGT is improved by more than 30 K and that the coercive field is increased by ≈100% due to the proximity coupling effect, which changes the spin textures of FGT at the interface. This work reveals that antiferromagnet/ferromagnet coupling is a promising way to engineer the magnetic properties of itinerant 2D ferromagnets, which paves the way for applications in advanced magnetic spintronic and memory devices.     

Experimental Evaluation of Trilateration-Based Outdoor Localization with LoRaWAN

Long Range Wide Area Network (LoRaWAN) in the Internet of Things (IoT) domain has been the subject of interest for researchers. There is an increasing demand to localize these IoT devices using LoRaWAN due to the quickly growing number of IoT devices. LoRaWAN is well suited to support localization applications in IoTs due to its low power consumption and long range. Multiple approaches have been proposed to solve the localization problem using LoRaWAN. The Expected Signal Power (ESP) based trilateration algorithm has the significant potential for localization because ESP can identify the signal’s energy below the noise floor with no additional hardware requirements and ease of implementation. This research article offers the technical evaluation of the trilateration technique, its efficiency, and its limitations for the localization using LoRa ESP in a large outdoor populated campus environment. Additionally, experimental evaluations are conducted to determine the effects of frequency hopping, outlier removal, and increasing the number of gateways on localization accuracy. Results obtained from the experiment show the importance of calculating the path loss exponent for every frequency to circumvent the high localization error because of the frequency hopping, thus improving the localization performance without the need of using only a single frequency.     

Viscoelasticity and fracture toughness of blended epoxy resins containing two monomers with different molecular weights

In the present work, we investigated the thermo-viscoelasticity and fracture toughness of various cured blends of two epoxy monomers with different molecular weights: 380 (Epikote 828) and 900 (Epikote 1001). The blended resins were easily prepared, and miscibility (no phase separation) in the blended resins was expected. The composition of the blended epoxy resins ranged from 0 to 100% by weight of the Epikote 1001. The measured damping factor and dynamic loss modulus in the glass-transition confirmed that each blended resin had a single phase, i.e., they were miscible. The fracture toughness at room temperature increased modestly with the Epikote 1001 content over the whole range (0–100 wt%). We found that below the glass-transition temperature, the macromolecular modifications enabled tailoring of the fracture toughness while maintaining the glassy bending modulus and with little change in the glass-transition temperature.     

A new energy efficient single-stage flash drying system integrated with heat recovery applications in industry

Abstract

A new system developed here conducts the thermal management of the flue gas from the cement industry and employs this heat for the drying of raw materials before reaching to the preheating section. As of now, an additional amount of heat is used to provide the drying section with hot air while this proposed configuration recovers the heat from the same plant and employs it for drying purpose. This approach also results in cost saving as this configuration decreases the cost of heating the ambient air to provide the drying section with hot air. The entire system is simulated using the Aspen Plus industrial software. A comprehensive thermodynamic analysis is conducted for each component of the proposed system keeping more focus on the drying section. Numerous parametric and sensitivity studies are conducted to investigate the system performance, pressure drop and raw materials drying. The energy and exergy efficiencies of the drying process are found to be 55.6% and 24.17%. On the basis of the present results achieved, one can say that this novel configuration can be proved as a benchmark for the cement industry and has the capability to reduce the cost as well.     

Performance evaluation of an adaptive self-organizing frequency reuse approach for OFDMA downlink

Orthogonal frequency division multiple access (OFDMA) is extensively utilized for the downlink of cellular systems such as long term evolution (LTE) and LTE advanced. In OFDMA cellular networks, orthogonal resource blocks can be used within each cell. However, the available resources are rare and so those resources have to be reused by adjacent cells in order to achieve high spectral efficiency. This leads to inter-cell interference (ICI). Thus, ICI coordination among neighboring cells is very important for the performance improvement of cellular systems. Fractional frequency reuse (FFR) has been widely adopted as an effective solution that improves the throughput performance of cell edge users. However, FFR does not account for the varying nature of the channel. Moreover, it exaggerates in caring about the cell edge users at the price of cell inner users. Therefore, effective frequency reuse approaches that consider the weak points of FFR should be considered. In this paper, we present an adaptive self-organizing frequency reuse approach that is based on dividing every cell into two regions, namely, cell-inner and cell-outer regions; and minimizing the total interference encountered by all users in every region. Unlike the traditional FFR schemes, the proposed approach adjusts itself to the varying nature of the wireless channel. Furthermore, we derive the optimal value of the inner radius at which the total throughput of the inner users of the home cell is as close as possible to the total throughput of its outer users. Simulation results show that the proposed adaptive approach has better total throughput of both home cell and all 19 cells than the counterparts of strict FFR, even when all cells are fully loaded, where other algorithms in the literature failed to outperform strict FFR. The improved throughput means that higher spectral efficiency can be achieved; i.e., the spectrum, which is the most precious resource in wireless communication, can be utilized efficiently. In addition, the proposed algorithm can provide significant power saving, that can reach 50% compared to strict FFR, while not penalizing the throughput performance.

     

Cotton Leaf Diseases Recognition Using Deep Learning and Genetic Algorithm

Globally, Pakistan ranks 4 in cotton production, 6 as an importer of raw cotton, and 3 in cotton consumption. Nearly 10% of GDP and 55% of the country's foreign exchange earnings depend on cotton products. Approximately 1.5 million people in Pakistan are engaged in the cotton value chain. However, several diseases such as Mildew, Leaf Spot, and Soreshine affect cotton production. Manual diagnosis is not a good solution due to several factors such as high cost and unavailability of an expert. Therefore, it is essential to develop an automated technique that can accurately detect and recognize these diseases at their early stages. In this study, a new technique is proposed using deep learning architecture with serially fused features and the best feature selection. The proposed architecture consists of the following steps: (a) a self-collected dataset of cotton diseases is prepared and labeled by an expert; (b) data augmentation is performed on the collected dataset to increase the number of images for better training at the earlier step; (c) a pre-trained deep learning model named ResNet101 is employed and trained through a transfer learning approach; (d) features are computed from the third and fourth last layers and serially combined into one matrix; (e) a genetic algorithm is applied to the combined matrix to select the best points for further recognition. For final recognition, a Cubic SVM approach was utilized and validated on a prepared dataset. On the newly prepared dataset, the highest achieved accuracy was 98.8% using Cubic SVM, which shows the perfection of the proposed framework..     

Sensor location optimization for fault diagnosis with a comparison to linear programming approaches

The critical importance of sustaining fault diagnosis, as a major system tool, is unquestionable if the high performance and reliability of increasingly complex engineering systems is to be sustained over time and across a wide operating range. However, it is quite difficult to retain the joint ability of fault detection and isolation as it requires a strong system architecture. That is why, before designing an industrial supervision system, the determination of a system’s monitoring ability based on technical specifications is important as finding the source of the failure is not trivial in systems with a large number of components and complex component relationships. This paper presents an efficient and cost-effective fault detection and isolation (FDI) scheme that evolved from an earlier work [1]. FDI specifications are translated into constraints of the optimization problem considering that the whole set of analytical redundancy relations has been generated, under the assumption that all candidate sensors are installed and later on tested by an optimization algorithm using binary and relaxed versions of linear and nonlinear programming. By doing so, critical information about the presence or absence of a fault is gained in the shortest possible time, with not only confirmation of the findings but also an accurate unfolding in time of the finer details of the fault, thus completing the overall diagnostic picture of the system under test. The proposed scheme is evaluated extensively on a two-tank process used in industry, exemplified by a benchmarked laboratory-scale coupled-tank system.     

Design and Experimental Evaluation of a Database-Assisted V2V Communications System Over TV White Space

In the past years, many works have demonstrated the applicability of Coarse-Grained Reconfigurable Array (CGRA) accelerators to optimize loops by using software pipelining approaches. They are proven to be effective in reducing the total execution time of multimedia and signal processing applications. However, the run-time reconfigurability of CGRAs is hampered overheads introduced by the needed translation and mapping steps. In this work, we present a novel run-time translation technique for the modulo scheduling approach that can convert binary code on-the-fly to run on a CGRA. We propose a greedy approach, since the modulo scheduling for CGRA is an NP-complete problem. In addition to read-after-write dependencies, the dynamic modulo scheduling faces new challenges, such as register insertion to solve recurrence dependences and to balance the pipelining paths. Our results demonstrate that the greedy run-time algorithm can reach a near-optimal ILP rate, better than an off-line compiler approach for a 16-issue VLIW processor. The proposed mechanism ensures software compatibility as it supports different source ISAs. As proof of concept of scaling, a change in the memory bandwidth has been evaluated. In this analysis it is demonstrated that when changing from one memory access per cycle to two memory accesses per cycle, the modulo scheduling algorithm is able to exploit this increase in memory bandwidth and enhance performance accordingly. Additionally, to measure area and performance, the proposed CGRA was prototyped on an FPGA. The area comparisons show that a crossbar CGRA (with 16 processing elements and including an 4-issue VLIW host processor) is only 1.11 × bigger than a standalone 8-issue VLIW softcore processor.