Autonomous Systems in the Internet of Vehicles
by Balamurugan Balusamy, Sandeep Kumar Mathivanan, Prabhu Jayagopal, S. K. B. Sangeetha, Ali Kashif Bashir
6A Hybrid Neurosymbolic Decision-Making Approach with Multimodal Sensor Fusion for Autonomous Vehicles
Devi A.1, Rose Bindu Joseph P.2* and Meram Munirathnam3
1School of Computer Science and Applications, REVA University, Bangalore, India
2Department of Mathematics, Dayananda Sagar College of Engineering, Bangalore, India
3Department of Mathematics, Rajiv Gandhi University of Knowledge Technologies, R.K. Valley, Kadapa District, Idupulapaya, Andhra Pradesh, India
Abstract
Advanced decision models are required for autonomous vehicles (AVs) to drive in sophisticated environments in efficient and safe manners. The existing methods such as deep reinforcement learning (DRL) and rule-based methods have problems with interpretability as well as the lack of real-time adaptability. The study introduces a novel hybrid decision-making framework that integrates DRL and neurosymbolic reasoning for enhancing decision accuracy as well as robustness. The architecture follows a hierarchical design where the symbolic reasoning module ensures compliance with traffic and ethics regulations, and the DRL controls low-level control actions for maximization. Situational awareness is achieved via a multimodal sensor fusion paradigm that incorporates LiDAR, cameras, and V2X communication. The proposed model performs better than the conventional DRL-based approaches with a 92.4% decision rate, 18.6-ms average latency, and 96.8% collision avoidance rate when evaluated on the A*3D dataset. The study makes ...
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