Abstract

Recent breakthroughs in both NLP and machine learning have let loose completely new opportunities toward more fluent interactions between humans and robots. Traditional robot control interfaces usually rely on structured programming or resort to using specialized commands that are challenging and require users to be heavily trained in their use. The state-of-the-art techniques in NLP-from transformer-based language models to novel few-shot learning approaches-endow robots with a great ability to comprehend and generate human-like language, thus opening more intuitive modes of communication, lowering programming complexity, and broadening their potential scope of application.

The proposed session intends to bring together researchers and practitioners who are pushing the boundaries of NLP-driven robotic systems. This session will present new approaches to context-aware language understanding, including adaptive dialogue management and real-time natural language instructions for robots. We explore how systems can interpret voice commands in real-time, disambiguate natural language, continuously learn from environmental cues and user feedback. We will discuss the key emerging paradigms that are opening up large language models within the robotic architectures to let robots foresee the needs of users rather than just understand user instructions, and thus collaborate with humans in a much more cooperative way.

Another important topic will concern the capability for increased robot autonomy and safety. In this regard, semantic understanding allows robots to contextualize situations much more effectively and identify misinterpretation, instantly correcting their actions to avoid certain mistakes or dangers. We conclude with an ethical and societal review of the use of language-endowed robots in a wide variety of environments-from industrial plants and hospital wards to open spaces-applying guarantees that these NLP developments engender a correct, transparent, and ethical use of such systems.

They will go away with state-of-the-art insights, real-world examples, and future perspectives that help design, develop, and deploy advanced human-robot collaboration systems that tap into the full potential of NLP.

Biography

 I have recently attained my Doctorate from Harokopio University. I am employed as an AI specialist at the Presidency of Governance in Greece. I have authored and continue to publish scholarly articles pertaining to artificial intelligence, secure communications, and multidimensional assessment of high-risk AI systems in accordance with the EU AI Act.

Abstract

The adoption of autonomous vehicles (AVs) depends heavily on robust perception systems trained on diverse datasets. However, a significant challenge is the scarcity of rare yet critical classes like ambulances in publicly available datasets. This class imbalance impacts the ability of AVs to detect and respond to emergency vehicles accurately, posing safety risks in high-stakes scenarios. Synthetic data generation effectively addresses this limitation by leveraging advanced simulation platforms such as CARLA, LGSVL, and AirSim.

These simulation tools allow the creation of highly customizable datasets with rare classes in diverse conditions, including varying lighting, weather, and urban environments. By generating synthetic scenarios tailored to underrepresented objects like ambulances, we can enhance training datasets without the constraints of real-world data collection. This approach improves object detection accuracy and performance in edge cases critical to AV safety.

In this session, we will explore how synthetic data generation can bridge the gap left by public datasets and provide practical insights into integrating simulation-based data into AV development pipelines. We will focus on the importance of balancing datasets for classes like ambulances and demonstrate the impact of this augmentation on model performance. Attendees will understand how synthetic data improves AV reliability, paving the way for safer and more efficient autonomous driving systems. This discussion underscores the transformative role of simulation in addressing real-world challenges in AV perception.

Biography

Chandra Jaiswal holds a bachelor’s degree in computer science and engineering, an MBA, and has completed a PhD in AI and Data Science at NCAT Greensboro, USA. With over 15 years of experience in supply chain management, he is a seasoned Distribution System Analyst who excels in integrating advanced technologies such as AI, Computer Vision, and Robotics to optimize supply chain operations. His contributions to robotics have also added significant value to Autonomous Augmented Reality (AR) and Virtual Reality (VR) systems, showcasing his ability to bridge cutting-edge innovations with practical applications. Chandra’s leadership and expertise have modernized supply chain processes, enhanced operational efficiency, and positioned him as a forward-thinking innovator in supply chain and autonomous systems.

Abstract
Artificial Intelligence and Robotics are pivotal to the future well-being of society and a key catalyst for economic growth. They open up boundless possibilities for sustainable development and the betterment of humanity. Humanitarian technology encompasses the application of advanced technologies to enhance society and improve the lives of individuals at the base of the pyramid (BoP) and in least developed countries (LDCs). This emerging field strives to provide innovative solutions to unaddressed social issues, alleviating suffering, promoting human dignity, and saving lives. Whether it involves providing clean water, enhancing healthcare delivery, or assisting in disaster response, humanitarian technology has the potential to make a substantial impact on people’s lives. Currently, 7 billion individuals reside below the poverty line of USD 1.90 per day, while 3.35 billion live below the poverty line of USD 5.50 per day. Universities globally should prioritize exploring innovative applications of AI and robotics for humanitarian and development purposes.

The speaker will delve into how Humanitarian Robotics can enhance efficiency in emergency relief missions, risk reduction, disaster management and support longer-term rural health programmes in LDCs. Additionally, the discussion will explore how engineers and designers can repurpose mature technologies and solutions to address pressing humanitarian needs more swiftly, cost-effectively, and efficiently. The speaker will also provide real-world examples from current projects to demonstrate how impact-oriented humanitarian technology projects can be integrated into formal curricula to enhance student learning and inspire future generations of humanitarian engineers.

Abstract

This study focuses on parameter optimization and detection signal comparison of planar coil particle detection sensors, and conducts in-depth analysis of different core materials. Through finite element simulation analysis and experimental verification, the influence of coil turns, diameter, thickness and other parameters on magnetic field distribution and particle detection signals was systematically explored, and the sensor chip structure design was optimized. The study found that a planar coil with 60 turns and an inner diameter of 1.2 mm performed best. At the same time, the enhancement effect of different magnetic core materials (including iron core, silicon steel sheet and permalloy) on the magnetic field and detection signal was investigated. It was found that the permalloy magnetic core can significantly improve the sensor performance. Copper particles have effective detection capabilities. This research provides theoretical and experimental basis for optimizing the detection sensor of metal particle contaminants in oil, and has important academic value and application prospects.

Biography

Yuqing Sun received the M.Sc. and Ph.D. degrees in marine engineering from Dalian Maritime University, Dalian, China, in 1989 and 1996, respectively. He is currently a Professor with the Marine Engineering College, Dalian Maritime University. His current research interests include mechatronics, ship’s auxiliary machinery, hydraulic drive and control, and refrigeration technique.