SLAM with Lidar Scans

This repository demonstrates how to implement the Simultaneous Localization And Mapping (SLAM) algorithm using a series of lidar scans. The primary goal is to build an accurate map of an environment and retrieve the trajectory of a mobile robot.

Table of Contents

• Introduction
• Installation
• Usage
• Features
• Configuration
• Contributing
• License
• Contact

Introduction

This example showcases the implementation of the SLAM algorithm with pose graph optimization. The SLAM algorithm incrementally processes lidar scans to build a pose graph linking these scans. The robot recognizes previously-visited places through scan matching and establishes loop closures along its path. The SLAM algorithm uses loop closure information to update the map and adjust the estimated robot trajectory.

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/slam-lidar.git
   cd slam-lidar

2. Ensure you have MATLAB installed, as this example relies on MATLAB functions and data structures.

3. Load the provided dataset:

   load('offlineSlamData.mat');

Usage

Load Laser Scan Data from File

Load a down-sampled dataset consisting of laser scans collected from a mobile robot in an indoor environment. The dataset is stored in offlineSlamData.mat and contains the scans variable, which holds all the laser scans used in this example.

load('offlineSlamData.mat');

Run SLAM Algorithm, Construct Optimized Map, and Plot Trajectory

1. Create a lidarSLAM object and configure the parameters:

   maxLidarRange = 8;
   mapResolution = 20;
   slamAlg = lidarSLAM(mapResolution, maxLidarRange);
   
   slamAlg.LoopClosureThreshold = 210;  
   slamAlg.LoopClosureSearchRadius = 8;

2. Incrementally add scans to the slamAlg object:

   for i = 1:10
       [isScanAccepted, loopClosureInfo, optimizationInfo] = addScan(slamAlg, scans{i});
       if isScanAccepted
           fprintf('Added scan %d \n', i);
       end
   end
   
   figure;
   show(slamAlg);
   title({'Map of the Environment', 'Pose Graph for Initial 10 Scans'});

3. Continue to add scans and observe the loop closure process:

   firstTimeLCDetected = false;
   
   figure;
   for i = 10:length(scans)
       [isScanAccepted, loopClosureInfo, optimizationInfo] = addScan(slamAlg, scans{i});
       if ~isScanAccepted
           continue;
       end
   
       if optimizationInfo.IsPerformed && ~firstTimeLCDetected
           show(slamAlg, 'Poses', 'off');
           hold on;
           show(slamAlg.PoseGraph); 
           hold off;
           firstTimeLCDetected = true;
           drawnow
       end
   end
   title('First loop closure');

4. Plot the final built map and the robot's trajectory:

   figure;
   show(slamAlg);
   title({'Final Built Map of the Environment', 'Trajectory of the Robot'});

Build Occupancy Grid Map

Generate an occupancy grid map representing the environment as a probabilistic occupancy grid:

[scans, optimizedPoses] = scansAndPoses(slamAlg);
map = buildMap(scans, optimizedPoses, mapResolution, maxLidarRange);

figure;
show(map);
hold on;
show(slamAlg.PoseGraph, 'IDs', 'off');
hold off;
title('Occupancy Grid Map Built Using Lidar SLAM');

Run Exploringscansvariable.m

This was created to compare every Lidar scan output and find pattern, for basic analysis. A video is created(in the all 71 plot video folder) for better analysis. Note that this is done because difference between 2 scans cannot be drastic and so, are comparable.

Features

• Incremental SLAM with pose graph optimization
• Loop closure detection and map correction
• Visualization of the mapping process and trajectory
• Generation of an occupancy grid map

Configuration

• Max Lidar Range: 8 meters (configurable)
• Map Resolution: 20 cells per meter (configurable)
• Loop Closure Threshold: 210 (configurable)
• Loop Closure Search Radius: 8 meters (configurable)

Contributing

Contributions are welcome! Please follow these steps:

1. Fork the repository.
2. Create a new branch.
3. Make your changes and commit them.
4. Push your branch to your forked repository.
5. Open a pull request to the main repository.

License

This project is licensed under the Apache-2.0 License. See the LICENSE file for details.

Contact

For any questions or inquiries, please contact:

• Mail: [email protected]
• GitHub: ChiragMV

---

References:

• MATLAB
• Research papers(present in the repository)
• https://www.researchgate.net/figure/Multi-line-LiDAR-based-SLAM-system-structure-diagram_fig1_369950490
• https://www.researchgate.net/figure/The-LiDAR-data-preprocessing-includes-three-parts-the-keyframe-selection-a-the_fig3_351731857
• https://ieeexplore.ieee.org/document/10137963

Thank you for using my SLAM with Lidar Scans project! I hope it helps you in your robotics and mapping endeavors.