BROWSE
Detail
Conference Paper
A Person-Following Robot Using Siamese Network-Based RGBD Tracking
Cited - time in 
Share 
Share
- Authors
- Jiyong Oh, Joon-Goo Lee, Hea-Min Lee, Seung-Woo Nam, Man-Soo Shin, Sangjoon Choi
- Issue Date
- 2021-10
- Citation
- International Conference on Control, Automation and Systems (ICCAS) 2021, pp.1-2
- Publisher
- IEEE
- Language
- English
- Type
- Conference Paper
- Abstract
- In this paper, we present a person-following robot equipped with an RGBD camera and a 2D LiDAR. Based on the data captured by the RGBD camera, it computes the location of a target person using a Siamese network-based tracker. Together with RGB image, our Siamese network utilizes depth information of the RGBD camera as another input of the network. Our robot moves using PI control based on the relative location between the target person and the robot. In addition to the simple navigation, our robot can avoid static obstacles such as corners and pillars between the robot and the person utilizing the 2D LiDAR.
- KSP Keywords
- 2D LiDAR, Depth information, Network-based, PI control, Person-following robot, RGB image, RGB-D cameras, Relative location, Siamese network, Static obstacles