PALoc

<div align="center"> <h1>PALoc: Advancing SLAM Benchmarking with Prior-Assisted 6-DoF Trajectory Generation and Uncertainty Estimation</h1>

<a href="https://ieeexplore.ieee.org/document/10480308"><img src='https://img.shields.io/badge/TMECH 2024- PALoc -red' alt='Paper PDF'></a><a href="https://www.youtube.com/watch?v=_6a2gWYHeUk"> <img alt="Youtube" src="https://img.shields.io/badge/Video-Youtube-red"/></a><a ><img alt="PRs-Welcome" src="https://img.shields.io/badge/PRs-Welcome-white" /></a><a href="https://github.com/JokerJohn/PALoc/network/members"> <img alt="FORK" src="https://img.shields.io/github/forks/JokerJohn/PALoc?color=white" /> </a>

</div>

Introduction

PALoc presents a novel approach for generating high-fidelity, dense 6-DoF ground truth (GT) trajectories, enhancing the evaluation of Simultaneous Localization and Mapping (SLAM) under diverse environmental conditions. This framework leverages prior maps to improve the accuracy of indoor and outdoor SLAM datasets. Key features include:

• Robustness in Degenerate Conditions: Exceptionally handles scenarios frequently encountered in SLAM datasets.
• Advanced Uncertainty Analysis: Detailed covariance derivation within factor graphs, enabling precise uncertainty propagation and pose analysis.
• Open-Source Toolbox: An open-source toolbox is provided for map evaluation, indirectly assessing trajectory precision.

<div align="center">  </div>

News

• 2026/01/20: Add the generated gt trajectories of FP dataset in fp_gt folder.
• 2025/01/22: Add new data bag redbird_03 in MS-Dataset.
• 2025/01/16: A real-time map-based localization system with new features is on the way, see this repo LTLoc!
• 2024/12/26: Formally Published by IEEE/ASME TMECH.
• 2024/12/06: Adapt for real-time map-based localization, see fp_os128_corridor_loc.launch and instructions.
• 2024/11/15: Docker support!
• 2024/08/15: Support newer college dataset!
• 2024/08/15: Support FusionPortable dataset and MS-dataset
• 2024/08/14: Release codes and data.
• 2024/03/26: Early access by IEEE/ASME TMECH.
• 2024/02/01: Preparing codes for release.
• 2024/01/29: Accepted by 2024 IEEE/ASME TMECH.
• 2023/12/08: Resubmitted.
• 2023/08/22: Reject and resubmitted.
• 2023/05/13: Submitted to IEEE/ASME TRANSACTIONS ON MECHATRONICS (TMECH).
• 2023/05/08: Accepted by ICRA 2023 Workshop on Future of Construction.

Dataset

GEODE Dataset

This data was provided by Zhiqiang Chen and Prof.Yuhua Qi from SYSU and HKU.

Stairs scenes with different types of lidar and glass noise. This is very challenging due to narrow stairs , you need to tune some parameters of ICP. The prior map and raw map can be downloaded.

| Prior map without glass noise | Raw prior map | | ------------------------------------------------------------ | ------------------------------------------------- |

| Sensor setup | Download link | | ---------------------- | -------------------------------- | | Velodyne16+ xsense IMU | http://gofile.me/4jm56/yCBxjdEXA | | Ouster64 + xsense IMU | http://gofile.me/4jm56/2EoKPDfKi |

| | | | ------------------------------------------------------------ | ------------------------------------------------------------ |

FusionPortable Dataset

Our algorithms were tested on the Fusion Portable Dataset.

| Sequence | GT Map | Scene | | ------------------------------------------------------------ | ------------------------------------------------------------ | -------------------------------------------------- | | 20220216_corridor_day | corridor with x degeneracy. | | | 20220216_canteen_day | The prior map only covers a portion of the scene. | | | 20220219_MCR_normal_01 | Performance on a quadruped robot platform. | | | 20220216_escalator_day | Performance in an open stairwell scenario. | | | 20220216_garden_day | Smaller scenario, similar to an indoor environment. | | | 20220225_building_day | Three loops of indoor hallway scanning with a handheld device, taking a relatively long time. | |

Newer College Dataset

• Multicam Vision Lidar IMU dataset : Ouster 128 + Integrated IMU
• Stereo Vision Lidar IMU dataset: Ouster 64 + Integrated IMU

This dataset include 2 different maps: parkland and math-institute.

| Parkland | Math-institute | | ------------------------------------------------------------ | ------------------------------------------------------------ | | | |

Self-collected Dataset

<div align="center">

| Sequence | parkinglot_01 | redbird_02 | redbird_03 | | ---------------------------------------------------- | ------------------------------------------------------------ | ------------------------------------------------------------ | ------------------------------------------------------------ | | Scenes | | | | | Ground Truth Map | Ground Truth Trajectory | Ground Truth Trajectory | Ground Truth Trajectory |

</div>

Getting Started

Docker Support

It is recommended to run the code in the container while visualize it in the host machine, e.g.:

Pull the Docker image:

docker pull ulterzlw/paloc

Run the container with host network access:

docker run -it --network host ulterzlw/paloc bash

launch the application (inside the container):

roslaunch paloc geode_beta_os64.launch

Start RViz (on the host machine):

rviz -d ${PATH_TO_PALOC}/config/rviz/ouster_indoors.rviz

Install

• Ubuntu 20.04 / ROS Noetic
• Open3d ( >= 0.17.0) (fixed by @ljy-zju)
• PCL
• GTSAM 4.2.0 (fixed by @WangWenda98)

Recommend to use system eigen when install GTSAM.

cmake -DGTSAM_USE_SYSTEM_EIGEN=ON -DG