AI-DRIVEN ROAD TRAFFIC REGULATION SYSTEM USING YOLOv8 AND LSTM

This study presents the autonomous traffic regulation system, which provides a solution to modern traffic issues by combining artificial intelligence, reinforcement learning, and computer vision. The system continuously monitors real-time traffic conditions, detects congestion, and automatically adjusts signal timing to optimize traffic flow. The results show that traffic signals that respond to current conditions significantly boost system performance, decrease average waiting times, and effectively reduce traffic jams compared to traditional systems that rely on fixed schedules. The system's capability to adjust to changing traffic patterns, made possible by its ongoing learning process, enables it to operate efficiently in constantly evolving environments. The autonomous traffic regulation system makes use of both computer vision and reinforcement learning technologies. The combination of these two allows the system to track and identify vehicles, which, in turn, helps it improve its effectiveness over time. Due to these characteristics, the autonomous traffic regulation system is a solution for smart city settings. This system represents a meaningful progress in the development of intelligent transportation systems in urban areas. With future enhancements such as real-time traffic prediction, vehicle communication, and prioritization of emergency services, the system can reach even higher levels of performance and achieve safe, intelligent urban transportation.

[1] Linux Foundation, Acumos AI: Open Source Artificial Intelligence Platform, Linux Foundation Publication, 2020. [2] C. R. Goenawan et al., See the Unseen: Grid-Wise Drivable Area Detection Dataset and Network using LiDAR, Remote Sensing, vol. 16, no. 20, p. 3777, 2024. [Online]. Available: https://doi.org/10.3390/rs16203777 [3] C.R.Goenawan , K.T.Wijaya , S.-H.Kong, Enhancing Performance of 3D Point Cloud Completion Networks using Consistency Loss, Neurocomputing, 2024, 129037, ISSN 0925-2312.                   https://doi.org/10.1016/j.neucom.2024.129037 [4] A. Geron, Hands-On Machine Learning with Scikit-Learn and Tensor Flow, 1st ed. California, USA: O’Reilly Media, 2017. [5] R. Anyoha, The History of Artificial Intelligence, Harvard University: Science in the News, 2017. [6] A. Esteva et al., Dermatologist-level classification of skin cancer with deep neural networks, Nature, vol. 542, no. 7639, pp. 115–118, 2017. [7] Y. Wu et al., Google’s neural machine translation system: Bridging the gap between human and machine translation, arXiv preprint arXiv:1609.08144, 2016. [8] J. Schmidhuber, Deep learning in neural networks: An overview, Neural Networks, vol. 61, pp. 85–117, 2015. [9] C. Wang et al., Webpage depth view ability prediction using deep sequential neural networks, IEEE Transactions on Knowledge and Data Engineering, 2018. [10] D. Sculley et al., Hidden technical debt in machine learning systems, Advances in Neural Information Processing Systems, pp. 2503–2511, 2015. [11] S. Zhao et al., Packaging and Sharing Machine Learning Models via the Acumos AI Open Platform, AT&T Research Labs, New Jersey, USA, 2019. [12] M. Vartak et al., Model db: A System for Machine Learning Model Management, in Proceedings of the Workshop on Human-In-the-Loop Data Analytics. ACM, 2016. https://doi.org/10.1145/2939502.2939516 [13] AT&T and Linux Foundation, Acumos AI: An Open Source AI Machine Learning Platform, 2017. [Online]. Available: https://www.acumos.org/