Naganaboyina Vinod Kumar, Ch. Hema Vardhan, Gattu Mohan Goud, M. Mohan Krishna Reddy, U. Devendra

DOI: https://doi.org/10.5281/zenodo.23454964

Urban traffic management increasingly requires systems that can reason over heterogeneous, time-varying information rather than rely only on static maps or historical averages. This paper presents the GenAI-Powered Smart Traffic Assistant, a prototype intelligent transportation platform that combines real-time location, traffic and weather inputs, historical traffic data, machine-learning-based congestion prediction, computer-vision vehicle detection, route recommendation, alerts, explainable AI (XAI), and role-aware dashboards in a unified workflow. The system is derived from a semester project report and its associated prototype deployment. The design follows a layered client-server architecture with a web/mobile oriented interface, application services, machine-learning components, and persistent traffic data. A structured development methodology is used to integrate data collection, preprocessing, prediction, recommendation, visualization, security, and maintenance. Rather than claiming benchmark accuracy that is not reported in the source project, the paper evaluates the prototype through functional coverage, architecture consistency, safety constraints, and explicit non-functional targets. The resulting framework provides a practical foundation for multimodal traffic assistance and a pathway toward live CCTV integration, IoT sensing, adaptive signal control, continuous model learning, and smart-city interoperability.

artificial intelligence, intelligent transportation systems, traffic prediction, route recommendation, computer vision, YOLO, weather-aware analytics, explainable AI, GenAI, smart cities

This paper presented a GenAI-Powered Smart Traffic Assistant that integrates real-time traffic information, weather context, location, historical records, computer vision, congestion prediction, route recommendation, alerts, explainability and role-based monitoring in a unified architecture. The system translates the supplied project report into a research-oriented formulation and preserves the project’s central design principle: heterogeneous traffic evidence should be converted into understandable, timely assistance rather than treated as isolated data feeds. The work is best viewed as an integrated prototype framework with a clear path toward empirical research. Its main value lies in the architecture, data-flow organization and extensibility of the assistant. The next step is a controlled evaluation on public and locally collected datasets with reproducible model training, weather-aware ablations, route-level metrics, latency measurements and human-centered evaluation of explanations and alerts. With these additions, the project can evolve from a demonstration platform into a defensible intelligent transportation research system.

1] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You Only Look Once: Unified, Real-Time Object Detection,” in Proc. IEEE CVPR, Las Vegas, NV, USA, 2016, pp. 779-788, doi: 10.1109/CVPR.2016.91.
[2] A. Bewley, Z. Ge, L. Ott, F. Ramos, and B. Upcroft, “Simple Online and Realtime Tracking,” in Proc. IEEE ICIP, 2016, pp. 3464-3468, doi: 10.1109/ICIP.2016.7533003.
[3] Y. Li, R. Yu, C. Shahabi, and Y. Liu, “Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting,” in Proc. ICLR, 2018. 5 GENAI-POWERED SMART TRAFFIC ASSISTANT
[4] B. Yu, H. Yin, and Z. Zhu, “Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting,” in Proc. IJCAI, 2018, pp. 3634-3640, doi: 10.24963/ijcai.2018/505.
[5] Z. Wu, S. Pan, G. Long, J. Jiang, and C. Zhang, “Graph WaveNet for Deep Spatial-Temporal Graph Modeling,” in Proc. IJCAI, 2019, pp. 1907-1913, doi: 10.24963/ijcai.2019/264.
[6] C. Zheng, X. Fan, C. Wang, and J. Qi, “GMAN: A Graph Multi-Attention Network for Traffic Prediction,” in Proc. AAAI, vol. 34, no. 1, 2020, pp. 1234-1241, doi: 10.1609/aaai.v34i01.5477.
[7] H. Zhou, S. Zhang, J. Peng, S. Zhang, J. Li, H. Xiong, and W. Zhang, “Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting,” in Proc. AAAI, vol. 35, no. 12, 2021, pp. 11106-11115, doi: 10.1609/aaai.v35i12.17325.
[8] B. Lim, S. O. Arik, N. Loeff, and T. Pfister, “Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecasting,” Int. J. Forecasting, vol. 37, no. 4, pp. 1748-1764, 2021, doi: 10.1016/j.ijforecast.2021.03.012.
[9] N. Kumar and M. Raubal, “Applications of Deep Learning in Congestion Detection, Prediction and Alleviation: A Survey,” Transp. Res. Part C, vol. 133, p. 103432, 2021, doi: 10.1016/j.trc.2021.103432.
[10] M. Shaygan, C. Meese, W. Li, X. G. Zhao, and M. Nejad, “Traffic Prediction Using Artificial Intelligence: Review of Recent Advances and Emerging Opportunities,” Transp. Res. Part C, vol. 145, p. 103921, 2022, doi: 10.1016/j.trc.2022.103921.
[11] W. Jiang and J. Luo, “Graph Neural Network for Traffic Forecasting: A Survey,” Expert Syst. Appl., vol. 207, p. 117921, 2022, doi: 10.1016/j.eswa.2022.117921.
[12] M. Afandizadeh et al., “Deep Learning Algorithms for Traffic Forecasting: A Comprehensive Review and Comparison with Classical Ones,” J. Adv. Transp., vol. 2024, Art. no. 9981657, 2024, doi: 10.1155/2024/9981657.
[13] S. Bouktif, A. Cheniki, A. Ouni, and H. El-Sayed, “Deep Reinforcement Learning for Traffic Signal Control with Consistent State and Reward Design Approach,” Knowl.-Based Syst., vol. 267, p. 110440, 2023, doi: 10.1016/j.knosys.2023.110440.
[14] F. Rasheed, K.-L. A. Yau, R. M. Noor, C. Wu, and Y.-C. Low, “Deep Reinforcement Learning for Traffic Signal Control: A Review,” IEEE Access, vol. 8, pp. 208016-208044, 2020, doi: 10.1109/ACCESS.2020.3034141.
[15] G. Garbinčius and M. Makulavičius, “Urban Traffic Flow Simulation in Various Weather Conditions Using Long Short-Term Memory-Based Machine Learning,” Transport, 2026, doi: 10.3846/transport.2026.26773. Source materials: supplied GenAI-Powered Smart Traffic Assistant project report; supplied AstraMedica reference PDF; project prototype URL supplied by the requester.

Naganaboyina Vinod Kumar, Ch. Hema Vardhan, Gattu Mohan Goud, M. Mohan Krishna Reddy, U. Devendra (2026). GenAI-Powered Smart Traffic Assistant: A Multimodal AI Framework for Real-Time Traffic Prediction, Route Recommendation, and Explainable Urban Mobility. International Journal of Computer Science Engineering Techniques, 10(5), 371–376. ISSN: 2455-135X. DOI: https://doi.org/10.5281/zenodo.23454964