Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{198351,
        author = {Ruturaj Jadhav and Aakash Hingrajiya and Soham Gore and Veer Jain},
        title = {Predictive Urban Navigation Using Hybrid TCN–STGCN Traffic Forecasting and Reinforcement Learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {9412-9415},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=198351},
        abstract = {Urban navigation systems commonly rely on static or short-term traffic predictions, limiting their ability to handle dynamic congestion caused by multiple users selecting similar routes. This project presents a hybrid traffic routing framework that integrates deep learning–based traffic prediction with graph-based path planning and reinforcement learning into a unified system. A Temporal Convolutional Network (TCN) and Spatio-Temporal Graph Convolutional Network (ST-GCN) are employed to model temporal and spatial traffic patterns and predict future edge-level travel times. These predictions are incorporated into an OpenStreetMap-based road network, where edge weights represent predicted travel time, enabling efficient route computation using the A* algorithm under various constraints such as avoiding tolls or signals. To address the limitation of deterministic routing, a Proximal Policy Optimization (PPO) framework with parameter sharing is introduced to select among multiple candidate routes generated by A*, considering both predicted and realized congestion. The system is evaluated on a Mumbai subgraph and demonstrates improved traffic distribution and more balanced routing compared to static approaches. The results highlight the effectiveness of combining predictive modeling, graph search, and reinforcement learning for adaptive urban traffic routing. Index Terms— Urban Traffic Forecasting, Intelligent Transportation Systems, Context-Aware LSTM, Parking Prediction, Reinforcement Learning, Smart Mobility.},
        keywords = {Urban Traffic Forecasting, Intelligent Transportation Systems, Context-Aware LSTM, Parking Prediction, Reinforcement Learning, Smart Mobility},
        month = {April},
        }