Copyright © 2025 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{175821,
        author = {Lav Dixit and Yash Tyagi and Harsh Tyagi and Varsha and Sakshi Bhardwaj},
        title = {Vision Drive: Smart Object Detection for Autonomous Vehicles},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {11},
        pages = {3982-3986},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=175821},
        abstract = {Autonomous vehicles depend on sophisticated object detection technology to facilitate efficient and safe driving. Such technologies combine sensor fusion methods, machine learning, and computer vision techniques to detect and track multiple objects, such as pedestrians, obstacles, and other vehicles, in real-time. This paper examines state-of-the-art object detection models and their contribution to boosting the perception skills of autonomous vehicles.
Deep learning Object Detection Algorithms, such as convolutional neural networks (CNNs) and transformer models, have greatly enhanced detection performance. Yet, occlusion, illumination changes, and computational cost remain essential challenges. We examine current developments in algorithms such as YOLO, Faster R-CNN, and SSD, examining their advantages, disadvantages, and applicability to autonomous driving scenarios.
Sensor fusion that combines information from cameras, LiDAR, and radar is essential in enhancing object detection precision. This paper investigates multi-sensor fusion techniques that advance perception by limiting false positives while enhancing decision-making in dense driving environments. We also emphasize current developments in multimodal learning and sensor data processing for autonomous vehicles.
In spite of tremendous advances, real-world deployment of object detection systems is hampered by robustness, real-time processing, and adversarial attacks. Future research directions, such as edge AI, self-supervised learning, and explainable AI, are proposed in this paper to enhance the safety and reliability of autonomous driving. Overcoming these challenges will be crucial to fully autonomous and safe vehicle navigation.},
        keywords = {Autonomous Vehicles, Object Detection, YOLO, CNNs, Sensor Fusion, Deep Learning, Real-Time Processing, LiDAR, Radar, Machine Learning, Autonomous Driving, Edge AI, Adversarial Attacks, Smart Mobility.},
        month = {April},
        }