A safety-constrained perception-control framework with lightweight convolutions and Lagrangian safe PPO yields higher success rates, safety, and efficiency in UAV navigation experiments.
Addressing function approxi- mation error in actor-critic methods,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Lightweight Safe Reinforcement Learning for End-to-End UAV Navigation
A safety-constrained perception-control framework with lightweight convolutions and Lagrangian safe PPO yields higher success rates, safety, and efficiency in UAV navigation experiments.