arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2001.10568cs.LGeess.SPstat.ML
收藏

Landmark2Vec:一种基于无监督神经网络的地标定位方法

Landmark2Vec: An Unsupervised Neural Network-Based Landmark Positioning Method

Autonomous Transport Solutions (ATS), Scania CV AB(斯堪尼亚商用车公司自主运输解决方案部)

Alireza Razavi

摘要:

介绍了一种基于神经网络的无监督地标地图估计方法。训练网络所需的测量是由代理从地标观察/接收的信号。地标、智能体和智能体从地标获取的测量值的定义在这里相当宽泛:地标可以是视觉对象,例如,沿着道路的杆,测量值是安装在车辆上的视觉传感器(代理)中的地标的大小,或者它们可以是无线电发射机,例如,建筑物内的WiFi接入点,测量值是由人(代理)携带的移动终端从它们听到的接收信号强度(RSS)。然后,地图估计的目标是找到地标的位置,直到缩放、旋转和移位(即,地标的拓扑图)。假设存在$L$界标,则测量结果将是在该区域上收集的$L \times 1$向量。然后将训练一个浅层网络来学习地图,而不需要任何地面实况信息。

英文摘要:

A Neural Network-based method for unsupervised landmarks map estimation from measurements taken from landmarks is introduced. The measurements needed for training the network are the signals observed/received from landmarks by an agent. The definition of landmarks, agent, and the measurements taken by agent from landmarks is rather broad here: landmarks can be visual objects, e.g., poles along a road, with measurements being the size of landmark in a visual sensor mounted on a vehicle (agent), or they can be radio transmitters, e.g., WiFi access points inside a building, with measurements being the Received Signal Strength (RSS) heard from them by a mobile device carried by a person (agent). The goal of the map estimation is then to find the positions of landmarks up to a scale, rotation, and shift (i.e., the topological map of the landmarks). Assuming that there are $L$ landmarks, the measurements will be $L \times 1$ vectors collected over the area. A shallow network then will be trained to learn the map without any ground truth information.

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。

↑