Devices
- Desktop (passmark 1639, 130W) https://www.cpubenchmark.net/cpu.php?cpu=Intel+Xeon+E5-2680+%40+2.70GHz&id=1221
- Laptop (passmark 2140, 15W) https://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i7-8550U+%40+1.80GHz&id=3064
Benchmark
- Gazebo simulation [http://gazebosim.org/tutorials?tut=ros_overview]
VO/VSLAM/VINS Systems Assessed
- GF-ORB-SLAM Stereo + Inertial Fusion [https://github.com/ivalab/gf_orb_slam2]
- ORB-SLAM Stereo + Inertial Fusion [same as above]
- SVO Stereo + Inertial Fusion https://github.com/YipuZhao/rpg_svo
- VINS-Fusion https://github.com/YipuZhao/VINS-Fusion
- MSCKF Stereo https://github.com/YipuZhao/msckf_vio
Compared to the official repos, the baseline methods in my repo are with explicit logging on pose tracking and time cost.
References
@article{zhao2020closedLoop,
title={Closed-Loop Benchmarking of Stereo Visual-Inertial SLAM Systems: Understanding the Impact of Drift and Latency on Tracking Accuracy},
author={Zhao, Yipu and Smith, Justin S. and Karumanchi, Sambhu H. and Vela, Patricio A.},
journal={IEEE Conference on Robotics and Automation},
year={2020}
}
Contact information
- Yipu Zhao yipu.zhao@gatech.edu
- Justin S. Smith jssmith@gatech.edu
- Patricio A. Vela pvela@gatech.edu