Addressing the pressing need for real-time solutions in Intelligent Transportation Systems (ITS), this project emerges at the intersection of innovation and necessity. Motivated by the unique challenges of managing large data volumes and maintaining robust communication in fast-moving vehicle environments, the project turns to LiDAR technology. The objectives encompass leveraging Mobile Edge Computing (MEC) to reduce computational complexity and utilizing predictive beamforming to fortify communication links between vehicles. Together, these strategies aim to enhance real-time navigation, traffic management, and urban mobility, driving the future of responsive, efficient, and intelligent transportation.
Elevated LiDARs (ELiDs)
Elevated LiDAR (ELiD) systems represent a groundbreaking approach to LiDAR technology in the realm of Intelligent Transportation Systems (ITS), particularly for autonomous vehicles (AVs). Traditional on-board LiDAR sensors, while providing rich data, come with significant drawbacks such as high costs, extensive power consumption, and a limited field-of-view (FoV). The ELiD system tackles these challenges by relocating LiDAR sensors to elevated positions, such as lamp-posts or building sides, and moving processing units to the cloud or the edge. This elevated perspective not only reduces the costs and power-hungry components associated with on-board LiDARs but also enhances the sensing range, requiring fewer sensors. Furthermore, the ELiD system offers additional applications, such as infrastructure monitoring, making it a versatile and efficient solution. By reimagining the placement and processing of LiDAR technology, ELiD systems open new avenues for more accessible, environmentally friendly, and effective implementation of AV technology, aiming to reshape modern urban mobility.
Integrating Elevated LiDAR with Mobile Edge Computing
The integration of ELiD with MEC represents a strategic alignment in AV technology, facilitating real-time applications. ELiD offers an elevated perspective, allowing for extensive environmental scanning without the limitations of traditional on-board sensors. When coupled with MEC, the data processing shifts from remote cloud servers to localized edge servers, providing a low-latency response vital for real-time AV control systems. This collaboration enables dynamic adaptation to rapidly changing environments, a critical factor in traffic management and urban mobility. By leveraging MEC’s distributed processing capabilities with ELiD’s aerial sensing, the need for excessive hardware on the vehicle is minimized, reducing costs and energy consumption. This unique combination fosters an efficient, agile, and cost-effective solution that advances the state of ITS, showcasing innovation in both sensing and computation.
Enhancing Vehicle Communication through Deep Learning:
Elevated LiDAR (ELiD), when paired with predictive beamforming, has a transformative effect on communication within Intelligent Transportation Systems (ITS). The introduction of deep learning algorithms amplifies ELiD’s ability to adapt to rapid changes in vehicle movements. These algorithms analyze the fine-grained data collected by ELiD, allowing for predictive models that actively compensate for changes in the dynamic environment of autonomous driving. This deep learning-enabled prediction facilitates beamforming, creating a resilient and precise communication link between vehicles and Mobile Edge Computing (MEC) units. Together, ELiD, beamforming, and deep learning craft an integrated communication solution that eliminates latency problems and fortifies links, contributing to a seamless and intelligent transportation experience.
Enhancing Latency and Real-Time Performance
Through the innovative integration of Elevated LiDAR (ELiD) with deep learning-driven beamforming, our project has effectively reduced latency and enhanced real-time responsiveness within Intelligent Transportation Systems (ITS). By minimizing network delays and optimizing bandwidth allocation, we’ve ensured robust and adaptive real-time communication in dynamic environments. These advancements demonstrate our approach’s potential to significantly improve efficiency and responsiveness in modern transportation systems.
Selected Publications
- M. C. Lucic, H. Ghazzai, A. Alsharoa and Y. Massoud, “A Latency-Aware Task Offloading in Mobile Edge Computing Network for Distributed Elevated LiDAR,” 2020 IEEE International Symposium on Circuits and Systems (ISCAS), Seville, Spain, 2020.
- O. Rinchi, A. Alsharoa, and I. Shatnawi “Deep-Learning-Based Accurate Beamforming Prediction Using LiDAR-Assisted Network”, in Proc. of the 34-th IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC’23), Toronto, ON, Canada, Sept. 2023.