Chahinez Ounoughi, the PhD student of the Department of Software Science, will defend her PhD thesis "Urban Traffic: Data Fusion and Vehicle Flow Prediction in Smart Cities" on March 12, 2024, starting at 14:00. The defense will take place in room ICT-638 (Akadeemia tee 15a, ICT building of TalTech) and can be also followed via Zoom.
Ph.D. thesis "Urban Traffic: Data Fusion and Vehicle Flow Prediction in Smart Cities" proposes a holistic solution to urban traffic congestion, promising significant improvements in traffic prediction and management efficiency, utilizing advanced deep-learning architecture and data fusion techniques.
The research introduces four critical contributions to traffic management and Intelligent Transportation Systems (ITS). The first contribution presents a highly efficient hybrid deep neural network solution in terms of time while maintaining high accuracy in predicting traffic speeds within road segments. This method streamlines the process of identifying congestion-prone areas, allowing timely interventions to manage traffic flow effectively.
Once congestion is detected, it is essential to alleviate congestion and improve traffic flow. The second contribution proposes an enhanced, sustainable, and proactive traffic signal control system driven by noise prediction in response to this need. By integrating sustainability considerations and proactive decision-making, this innovative method optimizes traffic signals to reduce congestion and enhance overall traffic management efficiency.
To harness the wealth of available traffic data, the third contribution delves into a systematic exploration of data fusion techniques within the realm of Intelligent Transportation Systems (ITS). This comprehensive review identifies methodologies that effectively integrate data from various sources.
The fourth contribution introduces an innovative data fusion technique that integrates features from both traffic and environmental sensors. This integration represents a significant advancement, as it enriches the precision of traffic prediction models and enables well-informed decisions in traffic light management. By dynamically adjusting signal timings based on real-time traffic conditions and predicted congestion levels, this approach optimizes traffic flow and contributes to the effective management of urban transportation systems. The author argues that these contributions offer valuable insights for transportation planning and policymaking, ultimately enhancing urban citizens' quality of life and the operational efficiency of transportation systems.
The thesis is published in the Digital Collection of TalTech Library.
Supervisor: Prof. Sadok Ben Yahia, TalTech.
Oponents:
- Ass. Prof. Mahdi Zargayouna, University Gustave Eiffel, France;
- Prof. Mauro Vallati, University of Huddersfield, Great Britain.
Meeting ID: 936 2926 6947
Passcode: 599544