通过不确定性自适应融合传感器，提升自动驾驶中视觉惯性里程计的精度与鲁棒性。
Uncertainty-Aware Adaptive Sensor Fusion for Autonomous Navigation
- 结合ViT与MCNN，分别提取IMU时间依赖性和视觉运动特征。
- 在KITTI上实现0.129的ATE和0.005的RPE，优于基线方法。
- 适合部署在算力受限的自动驾驶系统，支持155 FPS实时处理。
本文提出一种融合深度学习与无迹卡尔曼滤波（UKF）的混合方法，用于提升自主导航中视觉惯性里程计（VIO）的位姿估计精度。模型采用视觉变压器（ViT）捕捉惯性测量单元（IMU）数据中的时序依赖性，利用多尺度卷积神经网络（MCNN）从视觉数据中学习基于光流的运动线索。一个自适应传感器融合模块根据估计的不确定性动态加权IMU与视觉特征，增强了在复杂环境下的鲁棒性。此外，提出一种新型不确定性感知损失函数，将预测不确定性显式融入学习过程，使系统在噪声大、数据缺失或传感器不可靠的情况下仍能实现稳定准确导航。在KITTI数据集上的全面评估表明，该方法显著优于基线模型，绝对轨迹误差（ATE）达0.129，相对位姿误差（RPE）为0.005。轻量级设计在NVIDIA A100 GPU上实现155 FPS的实时处理，适用于资源受限的自主系统部署。
原文摘要 · Abstract (English)
This work introduces a hybrid deep learning approach integrated with an Unscented Kalman Filter (UKF) to enhance pose estimation accuracy in Visual-Inertial Odometry (VIO) for autonomous navigation. The proposed model employs a Vision Transformer (ViT) network to effectively capture temporal dependencies from inertial measurement unit (IMU) data and utilizes a Multiscale Convolutional Neural Network (MCNN) to learn optical flow-based motion cues from visual data. An adaptive sensor fusion module dynamically weights IMU and visual features by leveraging estimated uncertainty, thus improving robustness in diverse and challenging environmental conditions. Additionally, a novel uncertainty-aware loss function is proposed to explicitly incorporate prediction uncertainty into the learning process, enabling robust and accurate navigation under noisy, incomplete, or unreliable sensor inputs. Comprehensive evaluations of the KITTI dataset demonstrate that the proposed method significantly outperforms baseline approaches, achieving superior performance in terms of Absolute Trajectory Error (ATE) and Relative Pose Error (RPE). The lightweight and computationally efficient model processes data at 155 FPS on an NVIDIA A100 GPU, making it highly suitable for deployment in resource-constrained autonomous systems.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。