Autonomous Systems in the Internet of Vehicles
by Balamurugan Balusamy, Sandeep Kumar Mathivanan, Prabhu Jayagopal, S. K. B. Sangeetha, Ali Kashif Bashir
4Adaptive Multimodal Fusion for Robust Autonomous Driving Perception with Attention-Based Learning
Sangeetha R.
School of Computing, Department of Computer Science and Engineering, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Avadi, Chennai, India
Abstract
Intelligent transportation systems for vehicle safety allow autonomous and advanced driver-assistance systems to sense and make decisions appropriately; vehicle safety systems are essential in preventing traffic accidents. To enhance object detection and collision prediction, in this study, a novel multimodal fusion strategy is introduced, which fuses LiDAR point clouds, RGB images, and radar information using an adaptive attention-based approach. The proposed approach improves the system’s robustness under poor conditions through the dynamic adjustment of each modality’s contribution depending on external conditions. The Waymo Open Dataset, with a wide range of driving conditions across different weather and lighting conditions, is used to train and test the model. According to the experimental results, the new fusion technique is competitive with traditional early and late fusion methods with 96.3% precision for item detection, 94.7% accuracy, and 95.2% recall. This approach significantly improves vehicle safety and provides a reliable solution for real-time autonomous driving applications.
Keywords: Intelligent transportation systems, multimodal fusion, LiDAR, RGB images, radar, attention-based ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access