Oranits: Mission Assignment and Task Offloading in Open RAN-based ITS using Metaheuristic and Deep Reinforcement Learning

The paper introduces Oranits, a system model for mission assignment and task offloading in Open RAN-based intelligent transportation systems (ITS), focusing on the integration of mobile edge computing. It presents a twofold optimization approach utilizing a metaheuristic algorithm, the Chaotic Gaussian-based Global ARO (CGG-ARO), and an enhanced deep reinforcement learning framework, the Multi-agent Double Deep Q-Network (MA-DDQN). Simulation results indicate that CGG-ARO improves mission completion and overall benefits by approximately 7.1% and 7.7%, respectively, while MA-DDQN achieves increases of 11.0% and 12.5%, underscoring its potential for enhancing task processing efficiency in dynamic ITS environments, which is critical for practitioners developing autonomous vehicle systems.