Author(s)

XiangYu Chen

Affiliation(s)

College of Social Sciences, Shenzhen University, Shenzhen 518060, Guangdong, China.

Corresponding Author

XiangYu Chen

ABSTRACT

Vehicles must ensure safety and efficiency and deal with complex ethical dilemmas in autonomous driving. In order to deal with these ethical dilemmas effectively, moral decision-making models based on multi-objective reinforcement learning (MORL) provide a technical path to resolve such ethical dilemmas. Unlike traditional reinforcement learning (RL), MORL can generate more socially moral decision-making strategies in conflict scenarios by simultaneously optimizing multiple objectives. Of course, significant challenges remain in this research path. Assigning reward function weights is highly dependent on subjective judgement and cultural context; the dynamic environment is not adaptable enough, and the scarcity of ethical dilemma data limits model training. To address these issues, this paper points out that future research needs to focus on the dynamic weight adjustment mechanism, the construction of cross-cultural ethical frameworks, and large-scale real-world validation.

KEYWORDS

Autonomous driving; Moral decision-making; Ethical dilemmas; Multi-objective reinforcement learning

CITE THIS PAPER

XiangYu Chen. Moral decision-making for autonomous driving based on multi-objective reinforcement learning. Innovation and Technology Studies. 2025, 2(1): 33-40. DOI: https://doi.org/10.61784/its3009.

REFERENCES

[1] Chen L, Li Y, Huang C, et al. Milestones in autonomous driving and intelligent vehicles: Survey of surveys. IEEE Transactions on Intelligent Vehicles, 2022, 8(2): 1046–1056.

[2] Geisslinger M, Poszler F, Betz J, et al. Autonomous driving ethics: from trolley problem to ethics of risk. Philosophy & Technology, 2021, 34(4): 1033–1055.

[3] Liu Q, Li X, Yuan S, et al. Decision-making technology for autonomous vehicles: Learning-based methods, applications and future outlook. 2021 IEEE International Intelligent Transportation Systems Conference (ITSC), IEEE, 2021: 30–37.

[4] Wang H, Khajepour A, Cao D, et al. Ethical decision making in autonomous vehicles: challenges and research progress. IEEE Intelligent Transportation Systems Magazine, 2020, 14(1): 6–17.

[5] Li W, Devidze R, Mustafa W, et al. Ethics in Action: Training Reinforcement Learning Agents for Moral Decision-making in Text-based Adventure Games. International Conference on Artificial Intelligence and Statistics, 2024: 1954–1962.

[6] Ecoffet A, Lehman J. Reinforcement learning under moral uncertainty. International Conference on Machine Learning, 2021: 2926–2936.

[7] Noothigattu R, Bouneffouf D, Mattei N, et al. Teaching AI agents ethical values using reinforcement learning and policy orchestration. IBM Journal of Research and Development, 2019, 63(4/5): 2–1.

[8] Abel D, MacGlashan J, Littman ML. Reinforcement learning as a framework for ethical decision making. Workshops at the Thirtieth AAAI Conference on Artificial Intelligence, 2016.

[9] Deschamps T, Chaput R, Matignon L. Multi-objective reinforcement learning: an ethical perspective. RJCIA, 2024.

[10] Tennant E, Hailes S, Musolesi M. Modelling moral choices in social dilemmas with multi-agent reinforcement learning. Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence, IJCAI-23, 2023: 317–325.

[11] Rodriguez-Soto M, Serramia M, Lopez-Sanchez M, et al. Instilling moral value alignment by means of multi-objective reinforcement learning. Ethics and Information Technology, 2022, 24(1): 9.

[12] Peschl M, Zgonnikov A, Oliehoek FA, et al. MORAL: Aligning AI with human norms through multi-objective reinforced active learning. arXiv preprint arXiv:2201.00012, 2021.

[13] Bruers S, Braeckman J. A review and systematization of the trolley problem. Philosophia, 2014, 42: 251–269.

[14] Thomson JJ. The trolley problem. Yale LJ, 1984, 94: 1395.

[15] Thomson JJ. Killing, letting die, and the trolley problem. The Monist, 1976: 204–217.

[16] Paulo N. The Trolley Problem in the Ethics of Autonomous Vehicles. The Philosophical Quarterly, 2023, 73(4): 1046–1066.

[17] Wu SS. Autonomous vehicles, trolley problems, and the law. Ethics and Information Technology, 2020, 22(1): 1–13.

[18] Wiseman Y, Grinberg I. The Trolley Problem Version of Autonomous Vehicles. The Open Transportation Journal, 2018, 12(1).

[19] Himmelreich J. Never mind the trolley: The ethics of autonomous vehicles in mundane situations. Ethical Theory and Moral Practice, 2018, 21(3): 669–684.

[20] Moolayil AK. The modern trolley problem: ethical and economically sound liability schemes for autonomous vehicles. Case W. Res. JL Tech. & Internet, 2018, 9: 1.

[21] Etienne H. When AI ethics goes astray: a case study of autonomous vehicles. Social Science Computer Review, 2022, 40(1): 236–246.

[22] Tolmeijer S, Christen M, Kandul S, et al. Comparing AI and human expert collaboration in ethical decision making. Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems, 2022: 1–17.

[23] Madhav AS, Tyagi AK. Explainable Artificial Intelligence (XAI): Connecting artificial decision-making and human trust in autonomous vehicles. Proceedings of Third International Conference on Computing, Communications, and Cyber-Security: IC4S 2021, Springer Nature Singapore, 2022: 123–136.

[24] Martinho A, Herber N, Kroesen M, et al. Ethical issues in focus by the autonomous vehicles industry. Transport Reviews, 2021, 41(5): 556–577.

[25] Nassar A, Kamal M. Ethical dilemmas in AI-powered decision-making: a deep dive into big data-driven ethical considerations. International Journal of Responsible Artificial Intelligence, 2021, 11(8): 1–11.

[26] Conitzer V, Sinnott-Armstrong W, Borg JS, et al. Moral decision-making frameworks for artificial intelligence. Proceedings of the AAAI Conference on Artificial Intelligence, 2017, 31(1).

[27] Noothigattu R, Gaikwad S, Awad E, et al. A voting-based system for ethical decision-making. Proceedings of the AAAI Conference on Artificial Intelligence, 2018, 32(1).

[28] Qian L, Yin J, Huang Y, et al. The role of values and ethics in influencing consumers' intention to use autonomous vehicle hailing services. Technological Forecasting and Social Change, 2023, 188: 122267.

[29] Ernst D, Louette A. Introduction to reinforcement learning. Feuerriegel S, Hartmann J, Janiesch C, et al., eds. AI and Data Science in Business Applications, 2024: 111–126.

[30] Wiering MA, Van Otterlo M. Reinforcement learning. Adaptation, Learning, and Optimization, 2012, 12(3): 729.

[31] Sutton R S, Barto A G. Reinforcement learning. Journal of Cognitive Neuroscience, 1999, 11(1): 126–134.

[32] Barto A G. Reinforcement learning. In Neural systems for control (pp. 7–30). Academic Press, 1997.

[33] Kaelbling L P, Littman M L, Moore A W. Reinforcement learning: a survey. Journal of Artificial Intelligence Research, 1996, 4: 237–285.

[34] Kiran B R, Sobh I, Talpaert V, et al. Deep reinforcement learning for autonomous driving: a survey. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(6): 4909–4926.

[35] Kendall A, Hawke J, Janz D, et al. Learning to drive in a day. In 2019 International Conference on Robotics and Automation (ICRA) (pp. 8248–8254). IEEE, 2019.

[36] Singh J. The Ethics of Data Ownership in Autonomous Driving: Navigating Legal, Privacy, and Decision-Making Challenges in a Fully Automated Transport System. Australian Journal of Machine Learning Research & Applications, 2022, 2(1): 324–366.

[37] Titu A M, Stanciu A, Mihaescu L. Technological and ethical aspects of autonomous driving in a multicultural society. In 2020 12th International Conference on Electronics, Computers and Artificial Intelligence (ECAI) (pp. 1–6). IEEE, 2020.

[38] Wang Z, Rangaiah G P. Application and analysis of methods for selecting an optimal solution from the Pareto-optimal front obtained by multiobjective optimization. Industrial & Engineering Chemistry Research, 2017, 56(2): 560–574.

[39] Deb K, Gupta H. Searching for robust Pareto-optimal solutions in multi-objective optimization. In International Conference on Evolutionary Multi-Criterion Optimization (pp. 150–164). Berlin, Heidelberg: Springer Berlin Heidelberg, 2005.

[40] Yan Y, Du H, He D, Li W. Pareto optimal information flow topology for control of connected autonomous vehicles. IEEE Transactions on Intelligent Vehicles, 2022, 8(1): 330–343.

[41] MacLeod B P, Parlane F G, Rupnow C C, et al. A self-driving laboratory advances the Pareto front for material properties. Nature Communications, 2022, 13(1): 995.

[42] Li G, Li S, Qu X. Continuous decision-making for autonomous driving at intersections using deep deterministic policy gradient. IET Intelligent Transport Systems, 2022, 16(12): 1669–1681.

[43] Hu H, Lu Z, Wang Q, Zheng C. End-to-end automated lane-change manoeuvring considering driving style using a deep deterministic policy gradient algorithm. Sensors, 2020, 20(18): 5443.

[44] Hayes C F, R?dulescu R, Bargiacchi E, et al. A practical guide to multi-objective reinforcement learning and planning. Autonomous Agents and Multi-Agent Systems, 2022, 36(1): 26.

[45] Millard-Ball A. Pedestrians, autonomous vehicles, and cities. Journal of Planning Education and Research, 2018, 38(1): 6–12.

[46] World Health Organization. Pedestrian safety: a road safety manual for decision-makers and practitioners. World Health Organization, 2023.

[47] Muhammad K, Ullah A, Lloret J, et al. Deep learning for safe autonomous driving: current challenges and future directions. IEEE Transactions on Intelligent Transportation Systems, 2020, 22(7): 4316–4336.

[48] Ho D. Asian vs. liberal democracy: Identifying the locus of conflict in the Asian values debate. Political Science, 2023, 75(2): 165–192.

[49] Taeihagh A, Lim H S M. Governing autonomous vehicles: emerging responses for safety, liability, privacy, cybersecurity, and industry risks. Transport Reviews, 2019, 39(1): 103–128.