Open Source
FlowDrive: moderated flow matching with data balancing for trajectory planning
arXiv cs.RO0 views
arXiv:2509.21961v4 Announce Type: replace Abstract: Learning-based planners are sensitive to the long-tailed distribution of driving data. Common maneuvers dominate datasets, while dangerous or rare scenarios are sparse. This imbalance can bias models toward the frequent cases and degrade performance on critical scenarios. To tackle this problem, we compare balancing strategies for sampling training data and find reweighting by trajectory pattern an effective approach. We then present FlowDrive, a flow-matching trajectory planner that learns a conditional rectified flow to map noise directly to trajectory distributions with few flow-matching steps. We further introduce moderated, in-the-loop guidance that injects small perturbation between flow steps to systematically increase trajectory diversity while remaining scene-consistent. On nuPlan and the interaction-focused interPlan benchmarks, FlowDrive achieves state-of-the-art results among learning-based planners and approaches methods with rule-based refinements. After adding moderated guidance and light post-processing (FlowDrive*), it achieves overall state-of-the-art performance across nearly all benchmark splits. Our code is available at https://github.com/einsteinguang/flow_drive_planner.
Source:https://arxiv.org/abs/2509.21961PreviousImproving Imitation Policies for Multi-Phase Tasks through Residual Learning AllocationNextLCLA: Language-Conditioned Latent Alignment for Vision-Language Navigation
Related posts
Show HN: Voxstead – Live captions and bilingual call assistance for macOS
0Hacker News Show
Show HN: Food Sound · White Noise ShiYin
0Hacker News Show
Can You Use an Email Alias for GitHub?
0dev.to GitHub
Close ZIP-area rent differences are not precise rankings
0dev.to Data
Notion vs Obsidian vs Logseq 2026 (2026 Guide)
0dev.to Programming
Turn a Bambu Lab P1S 3D printer into a pen plotter
0Hacker News New