论文arXiv 2611.217834 天前
EDL模型主导多ASV溢油监测路径规划,重建误差方差解释率达80%
本文由 AI 根据公开内容整理,可能有误,请以原文为准:Better Calibrated, Better Planned: Uncertainty-Aware Multi-ASV Informative Path Planning for Oil Spill Monitoring(arXiv 2611.21783)
核心内容
针对多自主水面艇(ASV)在时能受限下监测异构溢油污染场的问题,研究提出基于Evidential Deep Learning(EDL)的不确定性感知信息模型。在60张物理仿真地图的闭环测试中,模型选择解释了最终重建误差80%的方差,而深度模型间的路径规划器差异仅解释1%方差,确立了校准良好的不确定性模型在多机协同信息路径规划(IPP)中的核心地位。
EDL在所有五种规划器下均取得最低重建误差与最高IoU,且具备最佳不确定性校准特性,单步推理耗时仅3.9ms。相比MC Dropout和Deep Ensembles需150个采样单元,EDL仅需50个单元即可实现更优重建;后者随误差增大置信度反而提升,而EDL保持稳健。这表明信息模型对自身采集数据的鲁棒性是构建可靠不确定性感知IPP系统的必要前提。
关键看点
- 模型主导性:在60组溢油仿真闭环测试中,信息模型选择解释了80%的重建误差方差,远超规划器策略的1%影响权重
- EDL性能优势:Evidential Deep Learning在所有规划器下均获最低误差与最高IoU,单步推理仅3.9ms,校准度优于GP及集成方法
- 采样效率对比:EDL每车仅需50个采样单元即超越MC Dropout与Deep Ensembles在150单元下的重建精度,数据利用率提升3倍
- 校准鲁棒性:Deep Ensembles呈现误差增大置信度反升的过自信缺陷,EDL则保持准确校准,验证了模型自洽性对IPP的关键价值
工程与产业观察
该研究揭示了具身感知系统中“模型校准优先于规划策略”的工程范式。对于资源受限的移动机器人集群,采用EDL等显式不确定性建模方法,能以极低算力开销换取显著的采样效率与重建精度提升,为环境监测、搜索救援等场景下的在线协同探索提供了可复现的算法基准与设计准则。
英文摘要 (Abstract)
Monitoring oil spills with fleets of Autonomous Surface Vehicles (ASVs) requires mapping a heterogeneous pollutant field from sparse, noisy measurements under tight time and energy budgets. Informative Path Planning (IPP) addresses this problem by coupling a reconstruction model, which infers the field and its uncertainty, with a planner that decides where to sample next. Prior work has focused on planning strategies and has largely relied on Gaussian Processes (GPs), leaving the calibration of the model's uncertainty underexplored. We study how uncertainty calibration conditions multi-vehicle IPP with a fleet of ASVs that coordinates online through a shared uncertainty map. We introduce Evidential Deep Learning (EDL) as an uncertainty-calibrated information model. We benchmark EDL, Monte Carlo Dropout and Deep Ensembles against a GP in closed loop with five planners on physics-based oil spill simulations. Across 60 maps, the choice of model dominates the outcome, explaining 80% of the variance of the final reconstruction error, against 1% for the planner among the deep models. EDL achieves the lowest error and the highest IoU under every planner, is the best calibrated model and runs at 3.9 ms per step. With 50 cells per vehicle it reconstructs the spill better than MC Dropout and Deep Ensembles with 150. Deep Ensembles become more confident as their error grows, whereas EDL is unaffected. Robustness of the information model to the data its own planner collects is therefore a requirement of uncertainty-aware IPP.
作者:Alejandro Casado-P'erez, Marija Popovi'c, Daniel Guti'errez-Reina, Sergio L. Toral., Samuel Yanes
外部来源
本页事实对应的公开页面,以原文为准。(点击将离开本站)