Search papers, labs, and topics across Lattice.
Comparison to State of the Art A comprehensive comparison of BevAD to other methods on Bench, Drive. BevAD emerges from our systematic analysis of common architectural patterns, previously studied in isolation. We show that high-resolution BEV features can lead to overfitting, which we mitigate by forcing the planner to learn bottleneck. Additionally, planning with diffusion complements disentangled planning output representations, particularly excelling when scaled with data. We acknowledge several limitations. First, while compressing the BEV along its spatial dimension significantly improved closed-loop driving, our approach may not directly extend to high-speed highway scenarios, which require long-range perception. A principled, context-adaptive BEV masking strategy remains for future work. Second, our analysis of failure cases suggests potential causal confusions. Mitigating these, perhaps via incorporating world knowledge from VLMs or with reinforcement learning, requires further investigation. Acknowledgments. This work is a result of the joint research project STADT:up (
1
0
3
Seemingly beneficial architectural choices in end-to-end driving planners, like high-resolution perception, can actually hinder robust closed-loop performance, demanding a re-evaluation of design principles.
