基于网络权重参数敏感度分析的终身视觉回环检测方法
Visual Loop Detection Method with Sensitivity Analysis of Network Weight Parameters
沈晔湖 1李欢 1张大庆 1苗洋 1赵冲 2蒋全胜1
作者信息
- 1. 苏州科技大学机械工程学院,苏州,215009
- 2. 厦门大学计算机科学与技术系,厦门,361005
- 折叠
摘要
在基于视觉的同时定位和地图构建系统中通过回环检测判别机器人是否到达过先前的位置,可有效消除位姿估计造成的累积误差.随着数据量的增大,现有方法会导致网络模型泛化性能降低.为实现可持续回环检测,提出一种网络权重参数敏感度分析视觉回环检测方法,将残差神经网络与广义均值池化相结合来构建轻量化特征提取网络;设计了可变相似感知区域,将可变滑动窗口和相似度矩阵相结合来提取三重样本;提出的网络权重参数敏感度分析方法降低了网络模型的灾难性遗忘.与典型方法MAC相比,所提方法在Nordland等数据集的召回率提高了 42％左右.
Abstract
In visual-based simultaneous localization and mapping,the loop-closure detection might determine whether the robot reached the previous positions,so that the accumulated errors caused by pose estimation might be effectively eliminated.With the increase of data volume,the generalization performance of network model will be reduced by the existing methods.In order to realize sustainable loopback detection,a visual loop detection method was proposed herein based on sensitivity analysis of network weight parameters.A lightweight feature extraction network was constructed by combining residual neural network and generalized mean pooling.The variable similarity sensing areas were de-signed,and the variable sliding windows were combined with the similarity matrix to extract three samples.The proposed sensitivity analysis method of network weight parameters reduced the cata-strophic forgetting of network models.Compared with the typical method MAC,the recall rate of the proposed method in Nordland and other data sets is improved by about 42％.
关键词
回环检测/终身学习/对比学习/同时定位和地图构建/网络权重参数敏感度分析Key words
loop-closure detection/lifelong learning/comparative learning/simultaneous localiza-tion and mapping/sensitivity analysis of network weight parameters引用本文复制引用
基金项目
国家自然科学基金(51975394)
江苏省自然科学基金(BK20211336)
江苏省研究生科研创新计划(092290148)
出版年
2024
国家科技期刊平台
NSTL
万方数据