An end-to-end system for crowdsourced 3D maps for autonomous vehicles: The mapping component

Abstract

Autonomous vehicles rely on precise high definition (HD) 3d maps for navigation. This paper presents the mapping component of an end-to-end system for crowdsourcing precise 3d maps with semantically meaningful landmarks such as traffic signs (6 dof pose, shape and size) and traffic lanes (3d splines). The system uses consumer grade parts, and in particular, relies on a single front facing camera and a consumer grade GPS. Using real-time sign and lane triangulation ondevice in the vehicle, with offline sign/lane clustering across multiple journeys and offline Bundle Adjustment across multiple journeys in the backend, we construct maps with mean absolute accuracy at sign corners of less than 20 cm from 25 journeys. To the best of our knowledge, this is the first end-to-end HD mapping pipeline in global coordinates in the automotive context using cost effective sensors.

Simultaneous Traffic Sign Detection and Boundary Estimation Using Convolutional Neural Network

“…We refer to this 'San Diego Traffic Sign' dataset as 'SDTS', and the number of signs for each shape is reported at Table I. 1 The dataset is split into two disjoint sets; 33,360 training images and 1 There are several benchmark dataset for traffic sign detection such as GTSDB dataset [8] for German traffic sign and LISA dataset [42] for U.S. traffic sign, but we can not use these datasets since they only have bounding box annotations. 3,719 test images which were captured at different paths.…”

“…Permission from IEEE must be obtained for all other users, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works. DOI: 10 (1,0) (1, 1) (0, 1) (0, 0) Fig. 1.…”

“…To reconstruct detected traffic signs to a 3D map, one should have point-wise correspondences of boundary corners of the signs across multiple frames, and then compute the 3D coordinates of the boundary corners by triangulation using the camera pose and the internal parameters of the camera (see [1] for details of the 3D mapping procedure). For accurate triangulation of 3D positions, it is required to estimate the boundary of signs with pixel-level accuracy.…”

An Efficient and Accurate Method for Different Configurations Railway Extraction Based on Mobile Laser Scanning

“…Moreover, it can identify the track networks. of urban objects [3][4][5]. MLS was applied to the railway surveying by Tulloch Engineering in Toronto in 2015, which used to collect data at night, taking only a small part of the time required to measure with conventional ground crew and reducing the safety risk associated with on track field surveys [6].…”

“…As shown in the Figure 7, the two groups that are adjacent endpoints of P i (xi, yi, zi) are selected as blue and green lines, respectively, and a new endpoint P i (xi , yi , zi ) is fitted by circular curve on plane expressed in the Formula (5). Bring the plane coordinates of P i−2 , P i−1 , and P i+1 into the Formula (5) to get a set of parameters of the circular curve.…”