SLAMDATMO in roadway and sidewalk environments by helmet-mounted LiDAR and IMU

Abstract

This paper presents a method for simultaneous localization, mapping, detection, and tracking of moving objects (SLAMDATMO) using LiDAR and IMU set on a helmet of micro-mobility rider. An environmental obstacle map built by SLAMDATMO represents stationary objects, road boundaries and road surface conditions in roadway and sidewalk environments. This map can be used to active safety for micro-mobility, such as bicycles, e-bikes, and e-scooters. Distortion in the LiDAR scan data is corrected by estimating the helmet’s pose (three-dimensional position and attitude angle) based on the information from Normal Distributions Transform (NDT)-based SLAM and IMU. The road surface, boundary, stationary and moving-object scan data are classified from the corrected LiDAR scan data. The moving scan data are used for DATMO, whereas the stationary scan data are used for point-cloud mapping. The boundary scan data are used to detect obstacles, such as curbs, gutters, and steps. The road surface scan data from LiDAR are in conjunction with IMU, and potholes and humps, such as manhole cover and cracks, are detected to reduce the falling risk of micro-mobility. The experimental results in a road environment of our university campus show the effectiveness of the proposed SLAMDATMO method.