Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Planning-Oriented Autonomous Driving
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Combine tracking and mapping with motion forecasting (ablation) | UniAD · nuScenes motion forecasting | minADE: 0.815 -> 0.736 (-9.7%) | Helped |
| Combine motion forecasting and occupancy prediction (ablation) | UniAD · nuScenes joint prediction | minADE: 0.736 -> 0.710 (-3.5%) | Helped |
| Add motion forecasting and occupancy prediction before planning (ablation) | UniAD planner · nuScenes planning | avg.L2: 1.131 -> 1.004 | Helped |
| Replace separate task heads with planning-oriented coordination | UniAD versus ID-0 multi-task baseline · nuScenes joint tasks | minADE: 0.858 -> 0.728 (-15.2%) | Helped |
| Add rotated scene-level anchors (ablation) | MotionFormer · nuScenes motion forecasting | minADE: 0.844 -> 0.768 (-15.8%) | Helped |
| Add agent-goal point interaction (ablation) | MotionFormer · nuScenes motion forecasting | minADE: 0.768 -> 0.755 | Helped |
| Add ego-vehicle query (ablation) | MotionFormer · nuScenes motion forecasting | minADE: 0.755 -> 0.747 | Helped |
| Add nonlinear target optimization (ablation) | MotionFormer · nuScenes motion forecasting | minADE: -5.0% minADE | Helped |
| Add unrestricted pixel-agent cross-attention without a mask (ablation) | OccFormer · nuScenes occupancy prediction, near | IoU: 61.2 -> 61.3 | No clear effect |
| Add occupancy-guided attention mask (ablation) | OccFormer · nuScenes occupancy prediction, near | IoU: +1.0 IoU-n.(%) | Helped |
| Reuse the mask feature for instance-level occupancy (ablation) | OccFormer · nuScenes occupancy prediction, near | IoU: 62.3 -> 62.6 | Helped |
| Add BEV attention (ablation) | Planner · nuScenes planning, 1s | Collision rate: 0.56 -> 0.35 | Helped |
Closest papers by meaning
- $AutoDrive\text{-}P^3$: Unified Chain of Perception–Prediction–Planning Thought via Reinforcement Fine-Tuning (ICLR 2026 · similarity 0.89)
- Prioritizing Perception-Guided Self-Supervision: A New Paradigm for Causal Modeling in End-to-End Autonomous Driving (NeurIPS 2025 · similarity 0.88)
- Drive-R1: Bridging Reasoning and Planning in VLMs for Autonomous Driving with Reinforcement Learning (AAAI 2026 · similarity 0.88)
- Hybrid-Driving: An Autonomous Driving Decision Framework Integrating Large Language Models, Knowledge Graphs and Driving Rules (AAAI 2025 · similarity 0.88)
- AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuning (NeurIPS 2025 · similarity 0.87)
- DriveAgent-R1: Advancing VLM-based Autonomous Driving with Active Perception and Hybrid Thinking (ICLR 2026 · 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.