Run Your Visual-Inertial Odometry on NVIDIA Jetson: Benchmark Tests on a Micro Aerial Vehicle

TL;DR

This paper benchmarks various visual and visual-inertial odometry algorithms on NVIDIA Jetson platforms, evaluating their accuracy and resource usage for UAV applications, and introduces a new challenging dataset for testing.

Contribution

It provides the first comprehensive comparison of VO/VIO algorithms on Jetson boards and releases a new dataset for UAV odometry testing.

Findings

VINS-Mono and VINS-Fusion showed high accuracy on Jetson platforms.

Jetson Xavier NX offers a good balance of performance and resource usage.

The new KAIST VIO dataset includes challenging geometric trajectories for robust testing.

Abstract

This paper presents benchmark tests of various visual(-inertial) odometry algorithms on NVIDIA Jetson platforms. The compared algorithms include mono and stereo, covering Visual Odometry (VO) and Visual-Inertial Odometry (VIO): VINS-Mono, VINS-Fusion, Kimera, ALVIO, Stereo-MSCKF, ORB-SLAM2 stereo, and ROVIO. As these methods are mainly used for unmanned aerial vehicles (UAVs), they must perform well in situations where the size of the processing board and weight is limited. Jetson boards released by NVIDIA satisfy these constraints as they have a sufficiently powerful central processing unit (CPU) and graphics processing unit (GPU) for image processing. However, in existing studies, the performance of Jetson boards as a processing platform for executing VO/VIO has not been compared extensively in terms of the usage of computing resources and accuracy. Therefore, this study compares representative VO/VIO algorithms on several NVIDIA Jetson platforms, namely NVIDIA Jetson TX2, Xavier NX, and AGX Xavier, and introduces a novel dataset 'KAIST VIO dataset' for UAVs. Including pure rotations, the dataset has several geometric trajectories that are harsh to visual(-inertial) state estimation. The evaluation is performed in terms of the accuracy of estimated odometry, CPU usage, and memory usage on various Jetson boards, algorithms, and trajectories. We present the {results of the} comprehensive benchmark test and release the dataset for the computer vision and robotics applications.