Tanniru Abhinav Siddharth supervised by Dr. Deepak Gangadharan received his  Master of Science –  Dual Degree in Electronics and Communication Engineering  (ECD). Here’s a summary of his research work on Efficient Computation Offloading in Vehicular Edge Computing: Scheduling and Reinforcement Learning-based Approaches:

With the growing interest in Autonomous Vehicles (AVs) and Connected Vehicles (CVs), modern transportation systems are undergoing a rapid transformation. These vehicles are equipped with advanced sensors (LiDAR, radar, cameras), real-time control algorithms, and artificial intelligence (AI) systems that generate and process massive volumes of data—up to 20 TB per day. However, the onboard computational resources of vehicles are often insufficient to handle these workloads, especially under tight latency constraints necessary for safety-critical operations such as collision avoidance and path planning. Vehicular Edge Computing (VEC) addresses this challenge by enabling vehicles to offload computation-intensive tasks to nearby Roadside Units (RSUs) equipped with edge servers. By doing so, latency is reduced, and resource usage becomes more efficient. However, the dynamic nature of vehicular networks, characterized by rapid vehicle movement and fluctuating edge server availability, introduces significant challenges in offloading and scheduling computation tasks. This thesis explores novel, mobility-aware, and adaptive computation offloading strategies that improve task completion rates and reduce latency in VEC environments. The proposed techniques integrate classical scheduling and modern learning-based algorithms, targeting real-time, scalable, and practical solutions for intelligent transportation systems.

July 2025