Adaptive risk-aware reinforcement learning framework for time-constrained navigation tasks. This MSc thesis studies how an RL agent can adapt its risk attitude when mission time changes, using a hierarchical controller that switches from a risk-aware CVaR-constrained policy to a risk-seeking policy when time pressure makes switching benificial.
reinforcement-learning autonomous-navigation safe-reinforcement-learning cvar-optimization safety-gym hierarchical-rl risk-sensitive-rl policy-switching time-constrained-planning
-
Updated
Jun 7, 2026 - Python