The recent advancements in autonomous systems research are significantly enhancing the efficiency and safety of robotic operations, particularly in challenging environments such as high-traffic waters and constrained missions. A notable trend is the integration of active learning and risk-sensitive control frameworks, which optimize decision-making processes by considering both the cost of actions and the potential risks involved. This approach not only improves the accuracy of data collection but also ensures safer navigation and more efficient resource utilization. Additionally, the development of novel reinforcement learning techniques, such as risk-constrained and trust region methods, is addressing the critical issue of safety in autonomous systems, ensuring that policies remain safe throughout the training process. These innovations are paving the way for more robust and reliable autonomous systems capable of operating in complex and dynamic environments, such as inland waterway transport and autonomous shipping. Notably, the incorporation of topological modeling for obstacle avoidance and the use of Rényi divergence in risk-sensitive control are particularly groundbreaking, offering new methodologies for enhancing the performance and safety of autonomous systems.