The field of autonomous systems is moving towards a greater emphasis on ethical decision-making, with a focus on developing frameworks and methods that can handle complex, real-world scenarios. Researchers are exploring ways to integrate human values and preferences into the decision-making process, and to ensure that autonomous systems behave in ways that are aligned with human moral principles. This includes the development of new approaches to reinforcement learning, such as preference-based multi-objective reinforcement learning, and the integration of reason-based moral decision-making into reinforcement learning architectures. Additionally, there is a growing recognition of the need to consider the impact of autonomous systems on non-human species, and to design systems that can coexist with animals in a safe and respectful way. Notable papers in this area include:

• Principles and Reasons Behind Automated Vehicle Decisions in Ethically Ambiguous Everyday Scenarios, which proposes a principled conceptual framework for automated vehicle decision-making in routine, ethically ambiguous scenarios.
• Integrating Reason-Based Moral Decision-Making in the Reinforcement Learning Architecture, which presents a first implementation of a reason-based artificial moral agent and demonstrates its potential in initial experiments.