Home // International Journal On Advances in Networks and Services, volume 17, numbers 1 and 2, 2024 // View article

People Detection and Tracking Using Distributed LiDAR Network

Authors:
Ryota Imai
Marino Matsuba
Masafumi Hashimoto
Kazuhiko Takahashi

Keywords: LiDAR network; people detection and tracking; one-dimensional convolutional neural network; distributed interacting multimodel estimator

Abstract:
This paper presents a people detection and tracking method using Light Detection And Ranging sensors (LiDARs) set in an environment. Each LiDAR detects people from its own LiDAR measurements, exchanges information of people positions by communicating with its neighboring LiDARs, and then fuses the information of people positions. Thereafter, each LiDAR estimates people’s poses from the information of people’s positions, and the estimates are fused by exchanging information among neigboring LiDARs. A one-dimensional convolutional neural network is applied to accurately detect people from LiDAR measurements in an environment where various objects, such as people, two-wheelers, and cars, coexist. A distributed interacting multimodel estimator is applied to accurately estimate the poses of people under various motion modes, such as stopping, walking, and suddenly running and stopping, in a distributed manner without a central server. Simulation results of eight people tracked by four Velodyne 32-layer LiDARs in an intersection environment where people and cars coexist show the performance of the proposed method.

Pages: 11 to 21

Copyright: Copyright (c) to authors, 2024. Used with permission.

Publication date: June 30, 2024

Published in: journal

ISSN: 1942-2644