As artificial intelligence (AI) technology advances, its applications expand into various fields, including robotic motion control. Self-balancing robots commonly employ PID or LQR control and typically require extensive parameter tuning for proper execution. An emerging alternative is using deep reinforcement learning algorithms, leveraging on Lightweight Neural Networks (NNs) and Catastrophic Forgetting Mitigation, in robot control. This paper details the process and outcomes of training a Q-network within a virtual environment via a deep Q-network (DQN) algorithm, preventing the network from getting overtrained and deploying it on a microcontroller to achieve a self-balancing robot. Additionally, it describes methods for facilitating interactions between the self-balancing robot and the real-world environment during training.

In this paper, we present an optimized implementation of Convolutional Neural Network (CNN) for road sign recognition on an FPGA platform, utilizing dynamic quantization and pruning techniques. Traditional CNN models are computationally intensive and require significant memory, which poses challenges for deployment on resource-limited hardware such as FPGA. To address these challenges, a hybrid approach was proposed combining dynamic quantization with pruning, allowing layer-specific precision adjustment while removing unnecessary network connections. This method significantly reduces the computational complexity and memory usage while maintaining high accuracy in road sign recognition tasks. The quantized and pruned model is implemented on the Tang Primer 25K FPGA, demonstrating efficient hardware utilization and real-time performance. Experimental results show that the proposed approach substantially reduces power consumption and resource utilization, with minimal impact on accuracy, making it highly suitable for embedded systems and edge computing applications.

Auto-infusion systems stand as instrumental entities in the medical sector, facilitating a streamlined and automated approach towards patient infusion processes. The evolution of these systems has been significantly influenced by the integration of Internet of Things (IoT) technologies [1], heralding a new era of enhanced reliability, precision, and functionality. This research aims to elucidate the development and operation of an IoT-based auto-infusion system equipped with a myriad of features optimized for real-time monitoring and responsiveness. Central to the system's operation is a peristaltic pump, meticulously regulated by a stepper motor, ensuring precise control over the infusion process. The system boasts a multitude of monitoring and alarm functionalities, such as flow rate detection, obstruction alerts, leakage notifications, low fluid level warnings, and bubble detection features. These capabilities ensure the provision of a robust and secure operational environment, minimizing potential risks and enhancing the reliability of the infusion process. An added innovative facet of the system is the incorporation of a UPS (Uninterruptible Power Supply), ensuring continuous operation even in scenarios afflicted by power outages, thereby enhancing the resilience and reliability of the infusion system.

Acoustic levitation, an emerging technology that employs sound waves to suspend solid particles and liquid droplets in mid-air, holds immense promise. Its potential applications span diverse fields, including environmental science, mechanical engineering, biochemistry, food science, and cancer therapy. The method involves manipulating matter using sound waves to create novel forms of energy and material. However, standing wave acoustic levitation faces challenges, particularly unstable soundwaves that hinder stationary levitation. This study delves into the intricacies of the standing wave acoustic levitation system. Additionally, it compares this method with near-field and parametric array levitation techniques. Each approach utilizes sound waves to create “invisible fingers” that hold objects, but they have distinct advantages and limitations. To address instability, two standing wave acoustic levitation systems are constructed and tested: a simpler version and an expert-level setup. The goal is to enhance accuracy and stability. Ultimately, this research aims to unlock the full potential of acoustic levitation.

This paper investigates the impact of sustainable development and climate change mitigation efforts by evaluating renewable energy technologies for reducing carbon emissions in residential buildings in Singapore. It begins with an extensive literature review, covering key initiatives such as the Green Mark scheme, household energy consumption patterns, and various government policies including the Carbon Tax, the Mandatory Energy Labeling Scheme (MELS), and the Minimum Energy Performance Standards (MEPS). Additionally, it addresses the embodied carbon emissions associated with residential buildings. The study then examines the existing renewable energy technologies in Singapore, focusing on Solar Photovoltaic (PV) systems, Wind Turbines, Biomass Energy, and Energy Storage solutions. Using HOMER PRO software, the research evaluates these technologies in terms of data on efficiency, costs, Net Present Cost (NPC), and the number of batteries required, incorporating sensitivity data for different sizes of residential flats through Multiple Criteria Data Analysis. Simulations conducted with RETSCREEN software for both single-family homes and multi-unit housing take into account various factors such as weather patterns, energy consumption, and financial implications. These simulations provide a comprehensive assessment of the potential for these technologies to significantly reduce carbon emissions in residential buildings, aligning with Singapore's sustainability goals and contributing to global climate change mitigation efforts.

This study explores how Artificial Intelligence (AI) can be used to enhance diversity and inclusion in engineering education. By using AI, schools can find and reduce biases, tailor learning to individual needs, and support students from underrepresented backgrounds. AI can also improve accessibility and help teachers adopt more inclusive methods. We discuss the benefits and challenges of using AI in this context, showing its potential to make engineering education more equitable.

Artificial Intelligence (AI) is transforming the educational landscape by offering personalized learning experiences that cater to individual student needs. This report delves into the psychological aspects of AI -enhanced learning experiences, focusing on their impact on student motivation, engagement, cognitive load, and emotional well-being. By analyzing various case studies and existing literature, we explore the benefits and challenges associated with integrating AI into educational settings. The findings indicate that AI-driven tools can significantly enhance motivation and engagement by providing tailored learning paths and real-time, relevant feedback. Additionally, AI can help manage cognitive load and offer emotional support, fostering a supportive and effective learning environment. However, challenges such as dependency on technology, data privacy concerns, and potential stress must be addressed to fully harness the benefits of AI in education. This report underscores the importance of strategic implementation and continuous evaluation of AI tools to ensure they contribute positively to students' psychological well-being and academic success.

This paper investigates how Collaborative Learning environments enriched by AI technologies have emerged as a pivotal advancement in modern education, promising enhanced learning experiences and outcomes. AI facilitates personalized learning pathways, fosters engagement through interactive tools, and supports peer interaction in dynamic educational settings. However, integrating AI into educational practices poses concerning challenges. Despite these challenges, AI holds immense potential to revolutionize collaborative learning by optimizing educational resources, promoting student engagement, and preparing learners for future challenges.

Signals are crucial in conveying information across various fields. This paper presents new methods for processing electromyographic (EMG) signals to create AI systems that decode muscle activity for arm movement control. Using an advanced dataset, the study focuses on enhancing prosthetic control and rehabilitation technologies through sophisticated signal processing and machine learning techniques. Various preprocessing steps improved signal quality, and a diverse set of features was extracted and classified. The results highlight the potential for more intuitive and responsive robotic arm movements, contributing to better prosthetic and rehabilitation solutions.

This paper investigates how clinical technologies focusing in radiology enriched by AI technologies have emerged as a pivotal advancement in the field of radiology, offering advancements that enhance both diagnostic accuracy and operational efficiency. These AI systems improve diagnostic performance by identifying subtle patterns that might be overlooked by human radiologists. Additionally, AI integration with Picture Archiving and Communication Systems (PACS) streamlines image management, automating routine tasks such as image analysis and report generation, which boosts workflow efficiency and productivity. Despite these benefits, challenges such as data privacy, algorithm transparency, and the need for continuous validation of AI models remain pertinent. As AI technology evolves, it holds the promise of further enhancing radiological practices, improving patient outcomes, and contributing to a more efficient healthcare system.

This paper investigates how AI-driven tools and platforms are integrated into engineering curricula to ready future engineers for the digital age. It emphasizes the advantages of improved learning experiences and tailored education, stressing the significance of updating curriculum content and training faculty. It also addresses challenges such as ethical concerns and integration complexities, highlighting the necessity for strategic planning to optimize AI's role in encouraging innovation and preparing students for careers in technology.

This paper explores the integration of AI-based assessment techniques in engineering education, highlighting their potential to enhance personalized feedback, improve learning outcomes, and streamline assessment processes. It examines various AI tools and their applications in automated grading, adaptive testing, and intelligent tutoring systems. Additionally, it addresses the challenges of implementing AI in educational settings, including fairness, data privacy, and integration with existing systems. The paper concludes with a discussion on strategic planning and continuous improvement to optimize AI's role in fostering an effective assessment framework for engineering education.

The self-balancing robot represents a significant advancement in the realm of mobile robotic platforms. This paper introduces the design of an intelligent embedded system aimed at managing the direction and speed of the stepper motor that drives the self-balancing robot. Controlling the speed and direction of the stepper motor is crucial for maintaining the stability of the two-wheeled robot. Stability is achieved by keeping the robot in an upright walking position. The proposed smart embedded system is engineered to handle sensing, control, and actuation functions.

This paper explores the transformative potential of gamification and immersive learning experiences, enhanced by artificial intelligence (AI), in modern education. Gamification leverages game design elements to boost engagement, motivation, and learning outcomes, while immersive technologies such as virtual reality (VR) and augmented reality (AR) create interactive, experiential learning environments. AI plays a pivotal role by personalizing learning experiences, adapting content to individual needs, and providing real-time feedback. This study reviews existing literature presents case studies of successful implementations, and discusses the benefits and challenges associated with these technologies. By integrating AI with gamification and immersive learning, educators can create dynamic, engaging, and effective educational experiences. The paper also addresses ethical considerations, accessibility issues, and future research directions, ultimately highlighting the significant impact of AI-driven gamification and immersive learning on the future of education.

Wearable devices can enhance quality of life by allowing patients to live independently while monitoring vital signs remotely. The reliance on batteries, however, is a limitation, as devices stop working once the battery is depleted. Energy harvesting from human and environmental sources presents a solution, providing continuous power for wearables and portable devices. This project focuses on low-voltage energy harvesting methods like photovoltaic cells, thermoelectric generators, piezoelectric materials, and magnetic induction to power IoT systems, with stored energy in batteries.

This proposed work aims to explore how Augmented Reality (AR) and Artificial Intelligence (AI) can revolutionize the experience of individuals struggling with Body Dysmorphic Disorder (BDD), a unique mental health condition characterized by a distorted view of their appearance. Research indicates that a significant percentage of individuals with BDD who undergo plastic surgery are dissatisfied with their results. By integrating AR and AI technologies, this project seeks to develop innovative therapeutic interventions that can provide realistic visualizations and personalized feedback. These technologies can help individuals understand their perceptions and promote healthier self-images without resorting to surgical procedures. Additionally, the project will investigate the potential of AI-driven cognitive-behavioral therapy (CBT) tools to offer customized treatment plans. Through user-centric design and collaboration with mental health professionals, the project aims to create a holistic approach that not only addresses the visual aspects of BDD but also tackles the underlying cognitive distortions. Ultimately, this research could pave the way for more effective and non-invasive treatment options, significantly improving the quality of life for those affected by BDD.

This brief presents an asymmetrical dead-time control driver (ASDTCD) for synchronous buck converter operating in the continuous conduction mode. Dead-time control is an important metric for improving the efficiency of switching mode power regulator. Without an additional circuit, the proposed ASDTCD can generate dead time by controlling the slope for the output signal of the driver. The proposed ASDTCD utilizes the transition between triode region and saturation region for the power transistor to avoid body-diode conduction and shoot-through current while minimizing the switching loss. Thus, high-speed body-diode conduction sensor is avoided; thereby, reducing the power consumption and saving silicon area. In addition, the body-diode conduction time control accuracy is also enhanced. Less than 1-ns body-diode conduction time has been achieved without bringing in shoot-through current across 10-450-mA load range. With less than 0.5% of the total input power consumed, the proposed ASDTCD takes less than 1% of the power transistor area. This design is implemented in the 0.18-μm CMOS process.

In this paper, it introduces a proposed hybrid control system which can simultaneously optimize deadtime and reverse inductor current with both the proposed Switched Capacitor Delay Deadtime Controller (SCD-DTC) and Unbalanced Input Pair Zero Current Detector (UIP-ZCD). Furthermore, the total silicon chip area of the hybrid control system occupies an area of 1.44mm2. The estimated power efficiency is 95.8% which has taken into account of losses due to wire bonding, package leads, PCB traces and other parasitic effects. The VIN_BUCKis 2.8-3.3V and being regulated to a VOUTvalue of 1.8V while driving 5-30mA of load current. The proposed SCD-DTC implemented in the buck converter minimizes the duration of body diode conduction to be <; 0.1ns.

A high resolution Vernier Ring Time-to-digital Converter is presented in this paper. Body bias is applied to its delay cells to obtain a finer delay difference between two delay chains. The delay cells and arbiters are implemented in a ring structure, thus allowing a large input time interval to be measured. The digital circuit nature of this converter is also attractive for low power and small area design. The simulation results reveal a 3 ps resolution, a -0.22/0.11 LSB differential nonlinearity (DNL) and a 9-bit range. The prototype chip is fabricated in 65 nm CMOS process consuming 0.44 mW with a 1.2 V power supply and occupies an area of 0.014 mm2.