The Rise of Artificial Intelligence in Autonomous Systems: Exploring the Boundaries of Machine Autonomy and Human Control in Modern Technology

This work aimed at reviewing the role of AI in the self-driving vehicles with regard to shared autonomy. Through the comparison of numerous algorithms and their effectiveness, which is the main focus of our work, it is intended to reveal the impact of the advanced AI tools on system improvement and control issues solving. We employed four key algorithms: DQN, CNNs, RL, GA, and so on. The outcome of experiments indicated that DQN obtained 92% in terms of decision making, CNN 89% in classification efficiency, RL 75% in optimizing the control task and, GA's 80% enhancement in the solution compared to the conventional techniques. The comparative analysis shows that although AI improves operations, it creates new challenges for human supervision. Thus, the work underlines the need for achieving a middle ground when it comes to AI application in autonomous systems, so both ethical and practical perspectives are taken into account.