Machine Learning for Autonomous Vehicular Optical Camera Communications (OCC)

Optical Camera Communication (OCC) has emerged as a key technology for enabling the seamless operation of future autonomous vehicles. By leveraging the high performance of OCC, we can meet the stringent requirements of ultra-reliable and low-latency communication (uRLLC) in vehicular systems. In this project, we introduce a rate maximization approach for vehicular OCC that aims to optimize vehicle speed, channel code rate, and modulation order while adhering to uRLLC requirements. We model reliability by targeting a specific bit error rate (BER) and latency by considering transmission delays. To enhance transmission rates and ensure reliability, we employ low-density parity-check (LDPC) codes and adaptive modulation techniques. First, we formulate the rate maximization problem as an optimization task aimed at determining the optimal vehicle speed, channel code rates, and modulation order under reliability and latency constraints. This problem is NP-hard, even with a small set of modulation orders. To address this challenge, we model the optimization problem as a Markov Decision Process (MDP) and adopt multi-agent deep reinforcement learning (DRL) to solve it in a distributed manner.