AI Paper Atlas

Research map / Reinforcement learning / Imitation learning

NavOL: Navigation Policy with Online Imitation Learning

Xiaofei Wei, Chun Gu, Li Zhang

ICML 2026 · Reinforcement learning › Imitation learning · approach: Offline imitation from imperfect data · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add online imitation learning with expert-in-the-loop rollouts on top of NavDPNavDP · NavOL benchmark (8 held-out scenes)mSR: 42.2 -> 69.0Improved
add online imitation learning on top of NavDPNavDP · NavDP benchmarkmSR: 77.8 -> 80.4Improved
add online imitation learning on top of NavDPNavDP · NavDP benchmark Intern-HomeSR: 60.0 -> 58.8Hurt
execute expert actions with probability 0.2 during rollout (ablation)NavOL · NavOL benchmarkSR: 43.9 -> 69.0Improved
initialize from pretrained NavDP weights (ablation)NavOL · NavOL benchmarkSR: 48.3 -> 69.0Improved
train on more scenes (1 -> 10 -> 50) (ablation)NavOL · NavOL benchmarkSR: 57.8 -> 64.8 -> 69.0Improved
vary rollout length per iteration (8, 32 vs default 128 steps) (ablation)NavOL · NavOL benchmarkSR: 67.9 (8 steps), 68.5 (32 steps) vs 69.0No clear effect
add path post-processing (local line search for clearance) in the expert planner (ablation)NavOL · NavOL benchmarkSR: 64.2 -> 69.0Improved
add camera height and pitch randomization (ablation)NavOL · NavOL benchmarkSR: 66.0 -> 69.0Improved
add critic for test-time trajectory selection (ablation)NavOL · NavOL benchmarkSR: 67.1 -> 69.0Improved
deploy on real Unitree Go2 zero-shot instead of NavDPNavOL · Real-world Office/Gym/Corridorsuccess rate: Office 4/10 -> 8/10; Gym 2/10 -> 7/10; Corridor 2/10 -> 10/10Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy