EvoNav automates the design of reward functions for RL robot navigation by evolving LLM proposals through a three-stage cheap-to-expensive evaluation process and claims better policies than hand-crafted or prior automated rewards.
Challenges of real-world reinforcement learning: definitions, benchmarks and analysis.Machine Learning, 110:2419–2468
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
EvoNav: Evolutionary Reward Function Design for Robot Navigation with Large Language Models
EvoNav automates the design of reward functions for RL robot navigation by evolving LLM proposals through a three-stage cheap-to-expensive evaluation process and claims better policies than hand-crafted or prior automated rewards.