Research map / Multi-agent systems and games / Human-AI and goal-conditioned
OccDriver: Future Occupancy Guided Dual-branch Trajectory Planner in Autonomous Driving
Zhao Huang, Bowen Zhang, Zhongzhu Li, Di Lin
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add marginal occupancy prediction (ablation) | OccDriver · nuPlan Val14 (ablation) | NR-S: 0.859 -> 0.863 | Improved |
| add trajectory-occupancy collision loss (ablation) | OccDriver · nuPlan Val14 (ablation) | Collisions: Collision from 0.943 to 0.960 and TTC from 0.914 to 0.931 | Improved |
| add trajectory-occupancy alignment loss (ablation) | OccDriver · nuPlan Val14 (ablation) | Progress: 0.882 -> 0.892 | Improved |
| add contingency planning (ablation) | OccDriver · nuPlan Val14 (ablation) | NR-S: 0.885 -> 0.896 | Improved |
| add contingency planning (ablation) | OccDriver · nuPlan Val14 (ablation) | Progress: 0.892 -> 0.885 | Hurt |
| add occupancy interference loss (ablation) | OccDriver · nuPlan Val14 (ablation) | Collisions: 0.938 -> 0.943 | Improved |
| use fine instead of coarse trajectory (ablation) | OccDriver · nuPlan Val14 (20% train subset) | NR-S: 0.836 -> 0.845 | Improved |
| add occupancy branch to vector-only planner (ablation) | OccDriver · nuPlan Val14 (20% train subset) | NR-S: 0.836 -> 0.859 | Improved |
| lengthen joint occupancy horizon in guidance loss from 2s to 6s (ablation) | OccDriver · nuPlan Val14 | NR-score: 0.828 -> 0.845 | Improved |
| extend joint occupancy horizon from 6s to 8s (ablation) | OccDriver · nuPlan Val14 | NR-score: 0.845 -> 0.828 | Hurt |
| raise occupancy threshold zeta from 0.5 to 0.7 (ablation) | OccDriver · nuPlan Val14 | NR-score: 0.837 -> 0.845 | Improved |
| raise collision loss weight w2 from 9 to 12 (ablation) | OccDriver · nuPlan Val14 | NR-score: 0.845 -> 0.826 | Hurt |
Closest papers by meaning
- Driving in the Occupancy World: Vision-Centric 4D Occupancy Forecasting and Planning via World Models for Autonomous Driving (AAAI 2025 · similarity 0.92)
- Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution (NeurIPS 2025 · similarity 0.89)
- OccProphet: Pushing the Efficiency Frontier of Camera-Only 4D Occupancy Forecasting with an Observer-Forecaster-Refiner Framework (ICLR 2025 · similarity 0.88)
- COME: Adding Scene-Centric Forecasting Control to Occupancy World Model (NeurIPS 2025 · similarity 0.88)
- UniMotion: A Unified Motion Framework for Simulation, Prediction and Planning (NeurIPS 2025 · similarity 0.87)
- Semi-Supervised Vision-Centric 3D Occupancy World Model for Autonomous Driving (ICLR 2025 · similarity 0.87)
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.
