TY  - RPRT
TI  - Collision-Aware Object-Goal Visual Navigation via Two-Stage Deep Reinforcement Learning
AU  - Hongwu Wang
AU  - Shiwei Lian
AU  - Feitian Zhang
PY  - 2026
UR  - https://arxiv.org/abs/2502.13498
ID  - 2502.13498
ER  -