cuVSLAM稀疏光束平差中自适应视觉-IMU权重的智能融合
Smart Fusion for Adaptive Visual-IMU Weighting in cuVSLAM Sparse Bundle Adjustment
AI总结：
该研究针对cuVSLAM的SBA模块，提出通过训练VQE实现α的二元选择，自适应调整视觉与IMU权重，在两个数据集上分别降低27.7%和14.9%的平均RMSE APE。
摘要：
在cuVSLAM中，视觉和惯性测量单元（IMU）残差在稀疏光束平差（SBA）中融合，权重由预先校准的噪声协方差固定，没有在运行时调整视觉-惯性平衡的机制。我们引入一个标量α≥1，用于相对于IMU预积分约束放大视觉信息矩阵。在候选集{1.0,2.0,5.0,10.0,20.0}中，生成的轨迹形成两个经验性能区间：IMU-TRUST区间包含α∈{1,2}，接近校准基线；VISION-TRUST区间包含α∈{10,20}；中间值α=5被视为过渡情况。每个区间内的差异相对于区间间的差异很小，该结构促使将α选择简化为二元决策。我们训练了一个轻量视觉质量估计器（VQE），实现为具有914个可训练参数的多层感知器，它将由跟踪器已维护的统计数据组成的10维视觉质量特征向量映射到每个SBA区间的预测。该方法在EuRoC MAV和TUM-VI Room上的留一法交叉验证（LOOCV）下，分别将平均RMSE绝对轨迹误差（APE）降低了27.7%和14.9%。
英文摘要：
In cuVSLAM, visual and inertial measurement unit (IMU) residuals are fused in sparse bundle adjustment (SBA) with a weighting fixed by pre-calibrated noise covariances, leaving no mechanism to adapt the visual-inertial balance at runtime. We introduce a scalar $α\geq1$ that upscales the visual information matrix relative to the IMU preintegration constraint. Across the candidate set $\{1.0,2.0,5.0,10.0,20.0\}$, the resulting trajectories form two empirical performance regimes. The IMU-TRUST regime contains $α\in\{1,2\}$ near the calibrated baseline, while the VISION-TRUST regime contains $α\in\{10,20\}$. The intermediate value $α=5$ is treated as a transition case. Differences within each regime are small relative to those observed between the regimes. This structure motivates reducing $α$-selection to a binary decision. We train a lightweight Visual Quality Estimator (VQE), implemented as a multilayer perceptron with 914 trainable parameters. It maps a 10-dimensional visual quality feature vector assembled from statistics already maintained by the tracker to a per-SBA regime prediction. The proposed method reduces mean RMSE APE by $27.7\%$ and by $14.9\%$ under leave-one-out cross-validation (LOOCV) on EuRoC MAV and TUM-VI Room, respectively.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
