xiaojake / tinygrapekit

!wheel/visual/gps MSCKF融合里程计:A bunch of state estimation algorithms

TinyGrapeKit

A bunch of state estimation algorithms.

This repo is divided into two parts, one is the basic algorithm, in the library folder. The other is the actual multi-sensor fusion algorithm (e.g. SLAM), in the app folder.

Detailed derivations can be found in: https://www.zhihu.com/column/slamTech

Install

Prerequisites

opencv, ceres, Eigen

Build

chmod +x build.sh
./build.sh

Applications in the app folder

FilterFusion : Filter-Based Sensor Fusion.

Fuse wheel, visual, and GNSS in an Extended Kalman Filter.

For visual-wheel fusion, please refer to: https://zhuanlan.zhihu.com/p/270670373

For fusing of GNSS data, please refer to: https://zhuanlan.zhihu.com/p/330880853

You can select the sensors to participate in the fusion through the configuration file.

sys_config.enable_plane_update: 1
sys_config.enable_gps_update: 1

Dataset

We used the KAIST dataset to test our method. https://irap.kaist.ac.kr/dataset/

Example

For examples, please refer to the Example folder.

./RunKAISTData ${REPO_PATH}/TinyGrapeKit/app/FilterFusion/params/KAIST.yaml ${KAIST_PATH}

Contact us

For any issues, please feel free to contact Dongsheng Yang: [email protected], [email protected]