SLAM

2-D and 3-D simultaneous localization and mapping

Simultaneous localization and mapping (SLAM) uses both Mapping and Localization and Pose Estimation algorithms to build a map and localize your vehicle in that map at the same time. Use lidarSLAM to tune your own SLAM algorithm that processes lidar scans and odometry pose estimates to iteratively build a map. Use buildMap to take logged and filtered data to create a map using SLAM. The SLAM Map Builder app lets you manually modify relative poses and align scans to improve the accuracy of your map.

Apps

SLAM Map Builder

Build 2-D grid maps using lidar-based SLAM

Functions

expand all

ekfSLAM

`ekfSLAM`	Perform simultaneous localization and mapping using extended Kalman filter
`correct`	Correct state and state error covariance
`landmarkInfo`	Retrieve landmark information
`poseHistory`	Retrieve corrected and predicted pose history
`predict`	Predict state and state error covariance
`removeLandmark`	Remove landmark from state vector
`reset`	Reset state and state estimation error covariance

lidarSLAM

`lidarSLAM`	Perform localization and mapping using lidar scans
`addScan`	Add scan to lidar SLAM map
`buildMap`	Build occupancy map from lidar scans
`removeLoopClosures`	Remove loop closures from pose graph
`scansAndPoses`	Extract scans and corresponding poses
`show`	Plot scans and robot poses

Pose Graphs

`poseGraph`	Create 2-D pose graph
`poseGraph3D`	Create 3-D pose graph
`poseplot`	3-D pose plot
`addPointLandmark`	Add landmark point node to pose graph
`addRelativePose`	Add relative pose to pose graph
`edgeNodePairs`	Edge node pairs in pose graph
`edgeConstraints`	Edge constraints in pose graph
`edgeResidualErrors`	Compute pose graph edge residual errors
`findEdgeID`	Find edge ID of edge
`nodeEstimates`	Poses of nodes in pose graph
`optimizePoseGraph`	Optimize nodes in pose graph
`removeEdges`	Remove loop closure edges from graph
`show`	Plot pose graph
`trimLoopClosures`	Optimize pose graph and remove bad loop closures

Factor Graph

`factorGraph`	Bipartite graph of factors and nodes
`factorGraphSolverOptions`	Solver options for factor graph
`importFactorGraph`	Import factor graph from g2o log file
`factorIMU`	Convert IMU readings to factor
`factorIMUParameters`	Factor IMU parameters
`factorGPS`	Factor for GPS measurement
`factorTwoPoseSE2`	Factor relating two SE(2) poses
`factorTwoPoseSE3`	Factor relating two SE(3) poses
`factorPoseSE2AndPointXY`	Factor relating SE(2) position and 2-D point
`factorPoseSE3AndPointXYZ`	Factor relating SE(3) position and 3-D point
`factorIMUBiasPrior`	Prior factor for IMU bias
`factorVelocity3Prior`	Prior factor for 3-D velocity
`factorPoseSE3Prior`	Full-state prior factor for SE(3) pose
`factorCameraSE3AndPointXYZ`	Factor relating SE(3) camera pose and 3-D point
`estimateGravityDirection`	Estimate gravity rotation using IMU measurements and factor graph optimization
`estimateGravityDirectionAndPoseScale`	Estimate gravity rotation and pose scale using IMU measurements and factor graph optimization

Featured Examples

EKF-Based Landmark SLAM

Use ekfSLAM for a reliable implementation of landmark Simultaneous Localization and Mapping (SLAM) using the Extended Kalman Filter (EKF) algorithm and maximum likelihood algorithm for data association. In this example, you create a landmark map of the immediate surroundings of a vehicle and simultaneously track the path of the vehicle. Generate a trajectory by moving the vehicle using the noisy control commands, and form the map using the landmarks it encounters along the path. Correct the vehicle trajectory and landmark estimates by observing the landmarks again.

Open Live Script

Landmark SLAM Using AprilTag Markers

Combine robot odometry data and observed fiducial markers called AprilTags to better estimate the robot trajectory and the landmark positions in the environment. The example uses a pose graph approach and a factor graph approach, and compares the two graphs.

Open Live Script

Perform SLAM Using 3-D Lidar Point Clouds

Demonstrates how to implement the simultaneous localization and mapping (SLAM) algorithm on collected 3-D lidar sensor data using point cloud processing algorithms and pose graph optimization. The goal of this example is to estimate the trajectory of the robot and create a 3-D occupancy map of the environment from the 3-D lidar point clouds and estimated trajectory.

Open Live Script

Implement Simultaneous Localization And Mapping (SLAM) with Lidar Scans

Demonstrates how to implement the Simultaneous Localization And Mapping (SLAM) algorithm on a collected series of lidar scans using pose graph optimization. The goal of this example is to build a map of the environment using the lidar scans and retrieve the trajectory of the robot.

Open Live Script

Implement Online Simultaneous Localization And Mapping (SLAM) with Lidar Scans

Demonstrates how to implement the Simultaneous Localization And Mapping (SLAM) algorithm on lidar scans obtained from simulated environment using pose graph optimization. This example requires Simulink® 3D Animation™ and Navigation Toolbox™.

Open Live Script

Map Indoor Area Using Lidar SLAM and Factor Graph

Implement offline lidar SLAM on collected lidar scans to map an indoor area using a factor graph and lidar scan map.

Open Live Script