AGNC-PGO - GitLite

KO | EN

gitlite — search

#ai #llm #rust #deep-learning #nodejs #windows #awesome #cli #kubernetes #awesome-list #pytorch #mcp #ios #openai

AGNC-PGO

(★ 69)

Adaptive Graduated Non-Convexity for Pose Graph Optimization

File Explorer

Download Latest Version (.zip)

- - README.md
- - CMakeLists.txt
  - README.md
- - CMakeLists.txt
  - README.md
- - animate-traj
  - comparisons.py
  - g2o-2-irl
  - gtg2o-2-values
  - make-gridworld-dataset
  - multi-plot
  - plot-timing
  - plot-traj
  - plot.py
  - precision-recall
  - README.md
  - run-batch-g2o
  - run-dataset-parallel
  - toro-2-g2o
- - CMakeLists.txt
  - dcsam
  - Kimera-RPGO
- CMakeLists.txt
- README.md
- run-experiment.cpp
- figure.png
- - GraduatedFactor.cpp
  - GraduatedKernel.cpp
  - RISAM2.cpp
- .clang-format
- CMakeLists.txt
- README.md
.gitignore
.gitmodules
CMakeLists.txt
DEPENDENCIES.md
LICENSE
NOTICE
README.md

# Installation Guide

1. Get the code

git clone https://github.com/SNU-DLLAB/AGNC-PGO

Downloads the entire project code from GitHub to your computer.

cd AGNC-PGO

Moves into the project folder you just downloaded.

2. CMake

Medium Recommended

Prerequisites

Git Needed to download the project code from GitHub.
CMake The tool used to generate build configuration.
C/C++ 컴파일러 Windows needs Visual Studio (Community edition, free), macOS needs Xcode Command Line Tools, Linux needs the gcc/g++ package.

* Configure cmake with following options: `cmake .. -DGTSAM_POSE3_EXPMAP=ON -DGTSAM_ROT3_EXPMAP=ON -DGTSAM_USE_SYSTEM_EIGEN=ON`

Type this command into your terminal and run it.

* `cmake ..`

Type this command into your terminal and run it.

✅ Check that an executable was created inside the build folder, then run it directly (e.g. ./build/app_name).

Pulled directly from this repo's README.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Adaptive Graduated Non-Convexity for Pose Graph Optimization (AGNC-PGO)

Presented in [IROS 2023 Workshop "ROBOTIC PERCEPTION AND MAPPING: FRONTIER VISION & LEARNING TECHNIQUES"](https://sites.google.com/view/ropem/).

[![arXiv](https://img.shields.io/badge/cs.RO-2308.11444-b31b1b?logo=arxiv&logoColor=red)](https://arxiv.org/abs/2308.11444)

<p align="center">
<img src="media/figure.png"
    alt="AGNC-PGO might be around 20% faster than riSAM" 
    width="70%"/>
</p>

# Structure
* `experiments`: Contains the implementations for experiments.
* `risam`: Contains the implementation of the algorithm.

## Building Instructions (Validated as of Aug 2023)
* Version Summary (tested and confirmed with the following dependency versions)
    * GTSAM: Tag=4.2a8, exact hash=9902ccc0a4f62123e91f057babe3612a95c15c20
    * KimeraRPGO: exact hash=8c5e163ba38345ff583d87403ad53bf966c0221b
    * dcsam: exact hash=b7f62295eec201fb00ee6d1d828fa551ac1f4bd7
    * GCC: 11.4.0
* These should be checked out when the git submodules are initialized, but are included here for completeness

* GTSAM 
    * Download [GTSAM version 4.2a8](https://github.com/borglab/gtsam/releases/tag/4.2a8) **! 4.2a9 not working!**
    * Setup compile time options [required by KimeraRPGO](https://github.com/MIT-SPARK/Kimera-RPGO)
    * Build and optionally install GTSAM (Scripts assume GTSAM python is installed in the active python environment)
* Clone riSAM and Submodules
    * `git clone --recursive https://github.com/SNU-DLLAB/AGNC-PGO.git`
* Build GTSAM
    * Configure cmake with following options: `cmake .. -DGTSAM_POSE3_EXPMAP=ON -DGTSAM_ROT3_EXPMAP=ON -DGTSAM_USE_SYSTEM_EIGEN=ON`
* Link GTSAM
    * If you `install` GTSAM this should be automatic
    * If you are working with a local build of GTSAM set `GTSAM_DIR` and `GTSAM_INCLUDE_DIR` to the appropriate directories.
* Build AGNC-PGO with riSAM
    * `cd AGNC-PGO`
    * `mkdir build`
    * `cd build`
    * `cmake ..`
    * `make`

## Acknowlodgement
The original code is from "Robot Perception Lab - Carnegie Mellon University". link: https://github.com/rpl-cmu/risam