Search

A recent trend in the IoT is to shift from traditional cloud-centric applications towards more distributed approaches embracing the fog and edge computing paradigms. In autonomous robots and vehicles, much research has been put into the potential of offloading computationally intensive tasks to cloud computing. Visual odometry is a common example, as realtime analysis of one or multiple video feeds requires significant on-board computation. If this operations are offloaded, then the on-board hardware can be simplified, and the battery life extended. In the case of self-driving cars, efficient offloading can significantly decrease the price of the hardware. Nonetheless, offloading to cloud computing compromises the system's latency and poses serious reliability issues. Visual odometry offloading requires streaming of video-feeds in real-time. In a multivehicle scenario, enabling efficient data compression without compromising performance can help save bandwidth and increase reliability., Part of ISBN 978-4-907626-41-9QC 20240315

This article describes a framework for load shedding techniques using dynamic pricing and multi-agent system. The islanded microgrid uses solar panels and battery energy management system as a source of energy to serve remote communities who have no access to the grid with a randomized type of power in terms of individual load. The generated framework includes modeling of solar panels, battery storage and loads to optimize the energy usage and reduce the electricity bills. In this work, the loads are classified as critical and non-critical. The agents are designed in a decentralized manner, which includes solar agent, storage agent and load agent. The load shedding experiment of the framework is mapped with the manual operation done at Kisiju village, Pwani, Tanzania. Experiment results show that the use of pricing factor as a demand response makes the microgrid sustainable as it manages to control and monitor its supply and demand, hence, the load being capable of shedding its own appliances when the power supplied is not enough., QC 20200722

Offloading computationally intensive tasks such as lidar or visual odometry from mobile robots has multiple benefits. Resource constrained robots can make use of their network capabilities to reduce the data processing load and be able to perform a larger number tasks in a more efficient manner. However, previous works have mostly focused on cloud offloading, which increases latency and reduces reliability, or high-end edge devices. Instead, we explore the utilization of FPGAs at the edge for computational offloading with minimal latency and high parallelism. We present the potential for modelling feature-based odometry in VHDL and utilizing FPGA implementations., QC 20200626

Three-dimensional representations and maps are the key behind self-driving vehicles and many types of advanced autonomous robots. Localization and mapping algorithms can achieve much higher levels of accuracy with dense 3D point clouds. However, the cost of a multiple-channel three-dimensional lidar with a 360°field of view is at least ten times the cost of an equivalent single-channel two-dimensional lidar. Therefore, while 3D lidars have become an essential component of self-driving vehicles, their cost has limited their integration and penetration within smaller robots. We present an FPGA-based 3D lidar built with multiple inexpensive RPLidar A1 2D lidars, which are rotated via a servo motor and their signals combined with an FPGA board. A C++ package for the Robot Operating System (ROS) has been written, which publishes a 3D point cloud. The mapping of points from the two-dimensional lidar output to the three-dimensional point cloud is done at the FPGA level, as well as continuous calibration of the motor speed and lidar orientation based on a built-in landmark recognition. This inexpensive design opens a wider range of possibilities for lower-end and smaller autonomous robots, which can be able to produce three-dimensional world representations. We demonstrate the possibilities of our design by mapping different environments., QC 20200617

This work presents a tri-band planar antenna array envisioned for assimilation into future 5G mm-wave wireless communication applications. The proposed assembly comprises of four antenna elements excited by a 1×4 T-junction power divider/Combiner and Defected Ground Structure (DGS). The antenna array presented in this work is modeled on Rogers RT/Duroid 5880 substrate with an overall size of 14 × 60 × 0.787 mm3. The frequency bands covered by the antenna system are centered at 29, 37.37 and 41 GHz with 10 dB bandwidths of 1.25, 1.3 and 2.14 GHz respectively. The peak gain attained by the antenna array is 11.8 dB. The proposed design with compact, simple and low profile assembly along with good radiation characteristics ascertains to be appropriate for mobile terminals and handheld wireless devices., Part of ISBN 9781728113227QC 20230920

The depletion of water table coupled with the decrease in rainfall due to climate change has pushed international communities to look for better ways to save water. Furthermore, access to the water table in disconnected communities in off-grid locations is a challenge to many African countries. These challenges negatively affect farming activities where irrigation plays an important role. In this paper, we propose a sustainable and smart DC microgrid irrigation system using multi-Agent systems along with Internet of Things technologies. The system is composed of a microgrid connected water pumping system and Internet of Things enabled sensors for irrigation. An algorithm that uses an agent-based approach to regulate energy demand from the PV system and controls irrigation is also introduced. The features of the proposed system have been compared against features for related systems., QC 20200427

The KTH Royal Institute of Technology and the University of Dar es Salaam (UDSM) have successfully developed a new educational model, 'challenge driven education' (CDE). The Model comprises the following building blocks: Integrating societal challenges into the regular curriculum, close collaboration with external stakeholders 'the challenge owners' through the design thinking methodology, international and cross disciplinary student teams that develop solutions to the formulated challenges, teacher training modules that support the pedagogical learning and interaction with external stakeholders. The formulated challenges as well as the proposed solutions shall contribute to the sustainable development goals (SDGs) as formulated by the UN agenda 2030.The ambition of the program is to develop student and teacher skills that can tackle the challenges of our time, like the sustainable development goals. The program is developed through a strategic partnership of KTH Royal Institute of Technology and the College of ICT of University of Dar es Salaam, UDSM supported by a STINT grant (The Swedish foundation for international cooperation in research and higher education). External stakeholders like the Dar es Salaam water and sewage authority, DAWASA, have defined challenges like clean water supply of Dar es Salaam. Students and teachers develop actionable problem statements in a design thinking process together with DAWASA, the external stakeholder. Student teams of UDSM and KTH develop proposed solutions to the formulated challenges in their final year project. Implementation of the proposed solutions by external stakeholders generates steps towards a sustainable development. In similar fashion, students from UDSM come to Stockholm KTH to participate in challenge based courses developing solutions of the sustainability challenges of the Swedish society. The key contribution to society is students that have become confident of their skills and ability to work for a sustai, QC 20200610

This article presents a twelve-element based wideband microstrip linear array antenna. In the array design, the elements are studded with stubs to enhance the bandwidth. Resonances are achieved in the broader band by introducing the slotted ground. The design is equipped with superstrate at a specific distance from the array antenna to improve the gain and efficiency. The proposed work has a good percentage bandwidth of 43.28% with stable radiation characteristics. The introduction of superstrate causes the proposed structure to enhance its peak gain to 7.6 dBi and the efficiency to 77%. The presented array covers the high-frequency spectrum from 24.343 GHz to 37.79 GHz, thus making it a good choice for millimeter wave applications., QC 20200427

The development of autonomous vehicles has seen considerable advances over the past decade. However, specific challenges remain in the area of autonomous waterborne navigation. Two key aspects in autonomous surface vehicles are sensor calibration and segmentation of water surface. Cameras and other sensors in a car or drone can be installed accurately in a specific position and orientation. In a large vessel, this is not always possible, as sensors might be installed around the vessel or on masts. Taking advantage of the medium in which these vehicles operate, the water plane can be used as a reference for different sensors to calibrate their orientation. This allows more accurate localization of obstacles of objects. State-of-the-art deep learning techniques have been successfully applied for water surface segmentation in open sea. However, in other environments such as small rivers or lakes with still waters, a different approach might enable more accurate water surface estimation. We propose a method to estimate the water plane based on the detection of local symmetry planes that naturally occur when objects are reflected in the water. By using a point cloud generated with a stereo camera, we are able to accurately estimate the water level and, at the same time, calibrate the camera position and improve the localization of obstacles. We assume that an approximate position and orientation of the camera is known with respect to the sea level. We demonstrate the efficiency of our method with data obtained in the Aura river, Finland, with our prototype vessel facing both the riverside and the center of the river., Part of ISBN 9781728150093QC 20230921

The agricultural and farming industries have been widely influenced by the disruption of the Internet of Things. The impact of the IoT is more limited in countries with less penetration of mobile internet such as sub-Saharan countries, where agriculture commonly accounts for 10 to 50% of their GPD. The boom of low-power wide-area networks (LPWAN) in the last decade, with technologies such as LoRa or NB-IoT, has mitigated this providing a relatively cheap infrastructure that enables low-power and long-range transmissions. Nonetheless, the benefits that LPWAN technologies enable have the disadvantage of low-bandwidth transmissions. Therefore, the integration of Edge and Fog computing, moving data analytics and compression near end devices, is key in order to extend functionality. By integrating artificial intelligence at the local network layer, or Edge AI, we present a system architecture and implementation that expands the possibilities of smart agriculture and farming applications with Edge and Fog computing and LPWAN technology for large area coverage. We propose and implement a system consisting on a sensor node, an Edge gateway, LoRa repeaters, Fog gateway, cloud servers and end-user terminal application. At the Edge layer, we propose the implementation of a CNN-based image compression method in order to send in a single message information about hundreds or thousands of sensor nodes within the gateway's range. We use advanced compression techniques to reduce the size of data up to 67% with a decompression error below 5%, within a novel scheme for IoT data., Part of ISBN 9781728132891QC 20210920

This paper ideates a novel texture descriptor that retains its classification accuracy under varying conditions of image orientation, scale, and illumination. The proposed Overlapped Multi-oriented Tri-scale Local Binary Pattern (OMTLBP) texture descriptor also remains insensitive to additive white Gaussian noise. The wavelet decomposition stage of the OMTLBP provides robustness to photometric variations, while the two subsequent stages - overlapped multi-oriented fusion and multi-scale fusion - provide resilience against geometric transformations within an image. Isolated encoding of constituent pixels along each scale in the joint histogram enables the proposed descriptor to capture both micro and macro structures within the texture. Performance of the OMTLBP is evaluated by classifying a variety of textured images belonging to Outex, KTH-TIPS, Brodatz, CUReT, and UIUC datasets. The experimental results validate the superiority of the proposed method in terms of classification accuracy when compared with the state-of-the-art texture descriptors for noisy images., QC 20200421

Kernel correlation filters (KCF) demonstrate significant potential in visual object tracking by employing robust descriptors. Proper selection of color and texture features can provide robustness against appearance variations. However, the use of multiple descriptors would lead to a considerable feature dimension. In this paper, we propose a novel low-rank descriptor, that provides better precision and success rate in comparison to state-of-the-art trackers. We accomplished this by concatenating the magnitude component of the Overlapped Multi-oriented Tri-scale Local Binary Pattern (OMTLBP), Robustness-Driven Hybrid Descriptor (RDHD), Histogram of Oriented Gradients (HoG), and Color Naming (CN) features. We reduced the rank of our proposed multi-channel feature to diminish the computational complexity. We formulated the Support Vector Machine (SVM) model by utilizing the circulant matrix of our proposed feature vector in the kernel correlation filter. The use of discrete Fourier transform in the iterative learning of SVM reduced the computational complexity of our proposed visual tracking algorithm. Extensive experimental results on Visual Tracker Benchmark dataset show better accuracy in comparison to other state-of-the-art trackers., QC 20191031

Kernel correlation filters (KCF) demonstrate significant potential in visual object tracking by employing robust descriptors. Proper selection of color and texture features can provide robustness against appearance variations. However, the use of multiple descriptors would lead to a considerable feature dimension. In this paper, we propose a novel low-rank descriptor, that provides better precision and success rate in comparison to state-of-the-art trackers. We accomplished this by concatenating the magnitude component of the Overlapped Multi-oriented Tri-scale Local Binary Pattern (OMTLBP), Robustness-Driven Hybrid Descriptor (RDHD), Histogram of Oriented Gradients (HoG), and Color Naming (CN) features. We reduced the rank of our proposed multi-channel feature to diminish the computational complexity. We formulated the Support Vector Machine (SVM) model by utilizing the circulant matrix of our proposed feature vector in the kernel correlation filter. The use of discrete Fourier transform in the iterative learning of SVM reduced the computational complexity of our proposed visual tracking algorithm. Extensive experimental results on Visual Tracker Benchmark dataset show better accuracy in comparison to other state-of-the-art trackers.

A compact chipless RFID tag with robust readable features is presented in this paper. The tag is made up of novel concentric butterfly slot resonators. Bit data is encoded in the frequency signature of the tag. Each slot corresponds to a resonance peak representing a bit ‘1’, whereas an absence of the peak signifies a bit ‘0’. Proposed resonator design demonstrates insensitivity to different polarization and incident angles of the linearly polarized impinging electromagnetic wave. The tag operates in the frequency band of 4.7-9.7 GHz, limited within the license-free ultra wide-band. Rogers RT/duroid (R) 5880 substrate is used to realize a 10-bit capacity design that spans 14 times 14 sq. mm resulting in a bit density of 5.1 bits/sq. cm.

This work presents a compact, quad-band planar antenna intended for assimilation into flexible and conformal devices. The CPW-fed semicircular shaped prototype with rake-shaped slots is designed, realized and characterized experimentally. The frequency bands covered by the proposed radiator are centered at 2.5, 3.7, 5.5 and 8 GHz with measured impedance bandwidths of 16%, 13.5%, 11.8% and 14.63% respectively. The proposed antenna is thus enabled to support WLAN, ISM, Bluetooth, WiMAX, LTE and X-band applications. The antenna exhibits a significant gain. The radiation characteristics of the proposed radiator are measured in concave and convex bent shapes at various radii to analyze its flexibility. Performance of the antenna remains almost unaffected in the bent situation. Measurements demonstrate good coherence with simulations. The compactness and good performance of the design both in bent and unbent conditions proves it to be the better contender for future multiband conformal wireless applications.

In the knowledge-based society, the legacy education system does not provide the needed skills for creative engineers especially enhancing student innovation and entrepreneurship capacity. Triple-helix model is a concept that aims to bond universities, industry and government in a bid to create innovations. In Europe, integrating research, education, and innovation together in a comprehensive manner has been the major driving force for local and European university development, as example in the form of European Institute Innovation Technology (EIT). At KTH, there are activities that alien the Teaching and Learning Activities (TLA) with different task group with the aim of creating a mutual innovation capacity to contribute solutions for major social challenges. Some of these task groups are Cross-Cultural Faculty Development for Challenge Driven Education, Global learning and digital platform and open innovation platform for learning. The progress and the success are measured by the number of joint student teams and their skills, knowledge development with the follow-up workshop, and the ongoing research and results of the socio-oriented projects. To enhance TLA and the teaching and learning practices, we have developed new curriculums (MSc. and PhD) for our partners to spark innovation and entrepreneurship where the students interact with Open Lab activities. The assessments show that the enrolled students have gained creative skills in dealing with engineering problem and consolidate their knowledge to improve the future TLA and the Intended Learning Outcome (ILO).

It remains a challenge to run Deep Learning in devices with stringent power budget in the Internet-of-Things. This paper presents a low-power accelerator for processing Convolutional Neural Networks on the embedded devices. The power reduction is realized by exploring data reuse in three different aspects, with regards to convolution, filter and input features. A systolic-like data flow is proposed and applied to rows of Processing Elements (PEs), which facilitate reusing the data during convolution. Reuse of input features and filters is achieved by arranging the PE array in a 3D tiled architecture, whose dimension is 3 x 14 x 4. Local storage within PEs is therefore reduced and only cost 17.75 kB, which is 20% of the state-of-the-art. With dedicated delay chains in each PE, this accelerator is reconfigurable to suit various parameter settings of convolutional layers. Evaluated in UMC 65 nm low leakage process, the accelerator can reach a peak performance of 84 GOPS and consume only 136 mW at 250 Mhz., QC 20181210

The smart grid is a new paradigm that aims to modernize the legacy power grid. It is based on the integration of ICT technologies, embedded system, sensors, renewable energy and advanced algorithms for management and optimization. The smart grid is a system of systems in which communication technology plays a vital role. Safe operations of the smart grid need a careful design of the communication protocols, cryptographic schemes, and computing technology. In this article, the authors describe current communication technologies, recently proposed algorithms, protocols, and architectures for securing smart grid communication network. They analyzed in a unifying approach the three principles pillars of smart-gird: Sensors, communication technologies, and security. Finally, the authors elaborate open issues in the smart-grid communication network., QC 20190415

The integration of Information and Communication Technologies in the power grid has given rise to new applications such as Advanced Metering Infrastructure. However, these technologies have also exposed the grid to new forms of security attacks. It is challenging to analyse and find solutions for these attacks due to the complexity of the grid system. There is a need of simple mechanisms to model and study the security implications of new power applications. This paper introduces a simulation model to evaluate the impact of security attacks on the Advanced Metering Infrastructure of a microgrid. The model is evaluated by demonstrating three security attack scenarios which lead to theft of power theft, privacy loss, and power outage., QC 20211020

Smart-grid is a complex system that incorporates distributed control, communication, optimization, and management functions in addition to the legacy functions such as generation, storage, and control. The design and test of new smart-grid algorithms require an efficient simulator. Agent based simulation platforms are the most popular tools that work well in the control and monitoring functionalities of the power electric network such as the microgrid. Most existing simulation tools necessitate either simulated or static data. In this paper, we propose a hardware-in-loop simulator for dc-microgrid. The simulator reads the power generated by the PV panels and the battery SoC using Raspberry PI. A physical agent that MRS on Raspberry PI sends the real-time data to a dc-microgrid simulator that runs on a PC. As a proof of concept, we implemented a load-shedding algorithm using the proposed system., QC 20190917

A compact, robust, chipless radio frequency identification (RFID) tag is proposed. Resonant elements patterned in a concentric fashion encode data in the spectral domain employing frequency shift encoding. The proposed tag encodes 28.25 data bits over a miniscule physical footprint of 25 x 25 mm(2). The formulated scheme is demonstrated to be viable for encoding of temporal variables. The electromagnetic performance of the presented design is investigated for different laminates: Rogers RT/duroid (R) 5880 and Taconic TLX-0. Multiple tag prototypes employing a variety of substrates are realized and evaluated for in-laboratory performance. The proposed design is compared with existing work reported in literature. Code density of 4.52 bits/cm(2) has been successfully achieved. The tag design operates from 3.07 to 10.6 GHz and is readily realizable on flexible laminates. Smart retail, intelligent packaging, adaptive ticketing, and similar time-related applications can be materialized using the proposed tag., QC 20181001

Control and monitoring processes in the electric field are fundamental aspects in smart grid. Different attributes, systems and mechanisms are required in handling activities from transmission, distribution and consumer side of the electrical network based on advanced technologies. In Tanzania, there still exists manual operations on distribution side and very little on the transmission side. Agent based distributed system has been seen to work successfully which involves the use of artificial intelligence and autonomous actions during the operations and controlling activities. Therefore, in this manner, the generated model based on REPAST simulation tool for solar driven DC microgrid will be achieved to show features of control and monitoring the grid including load shedding technique and demand responses based on pricing. Furthermore, the design of the prototype based on the stated features will be implemented. The advantages and impact of the demonstration can be deployed and used in any community for the application of any renewable energy systems., Part of proceedings: ISBN 9781538641637QC 20211020

I Nowadays, health care at home becomes more and more important, there are also some product which can measure the ECG, EMG with wearable device. However, these devices are not so accurate for diagnosis because of the low sample rate and less channel on body. In this project, we design a wearable system with 8-channels AFE and use Wi-Fi module to transfer the data to cloud so that we can measure the ECG or EMG more accurate at home, almost at the same sample rate and channels at the hospital. And then the cloud is built for received the data and a real-time display can help doctor monitor the patients' condition remotely. template is essential with some standard steps., QC 20190613

A client-server architecture style is one of the common approaches enabling separation of concerns in distributed systems. In the Internet of Things architecture, this approach exists in different configuration of sensors, actuators, gateways in the Fog layer and servers in the Cloud. This configuration affects the degree of interoperability, scalability and other functional and non-functional system requirements. In this paper, we reflect on best practices in the web and REST style to address IoT challenges; one of the constraints in REST, Code on Demand, is used for IoT to enhance the flexibility and interoperability of resource constrained clients at the perception layer. Scripts written in a domain specific language, DoS-IL, are organized and stored at the Fog layer for sensor and actuators nodes to request and execute the incoming script. A generic application layer protocol and RESTful server are presented along with experimental results., QC 20181108