Changsha, 410083
China
Central South University
Safe reinforcement learning, Human risk preferences, Constrained MDP, model predictive control, Action shielding