In the context of autonomous driving, learning-based methods have been promising for the development of planning modules. During the training process of planning modules, directly minimizing the discrepancy between expert-driving logs and planning output is widely deployed. In general, driving logs consist of suddenly appearing obstacles or swiftly changing traffic signals, which typically necessitate swift and nuanced adjustments in driving maneuvers. Concurrently, future trajectories of the vehicles exhibit their long-term decisions, such as adhering to a reference lane or circumventing stationary obstacles. Due to the unpredictable influence of future events in driving logs, reasoning bias could be naturally introduced to learning based planning modules, which leads to a possible degradation of driving performance. To address this issue, we identify the decisions and their corresponding time horizons, and characterize a so-called decision scope by retaining decisions within derivable horizons only, to mitigate the effect of irrational behaviors caused by unpredictable events. Several viable implementations have been proposed, among which batch normalization along the temporal dimension is particularly effective and achieves superior performance. It consistently outperforms baseline methods in terms of driving scores, as demonstrated through closed-loop evaluations on the nuPlan dataset. Essentially, this approach accommodates an appealing plug-and-play feature to enhance the closed-loop performance of other learning-based planning models.
View on arXiv@article{xin2025_2411.00476,
title={ PlanScope: Learning to Plan Within Decision Scope Does Matter },
author={ Ren Xin and Jie Cheng and Hongji Liu and Jun Ma },
journal={arXiv preprint arXiv:2411.00476},
year={ 2025 }
}