A reinforcement learning framework that uses formal rulebook-based safety scores as rewards can fine-tune a perception model to reduce simulated driving rule violations.
Valet parking without a valet,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Incorporating System-level Safety Requirements in Perception Models via Reinforcement Learning
A reinforcement learning framework that uses formal rulebook-based safety scores as rewards can fine-tune a perception model to reduce simulated driving rule violations.