How can I know and apply underlying pose graph information for lidarSLAM

When we apply "lidarSLAM" in the MATLAB, the description says that:

The SLAM algorithm takes in lidar scans and attaches them to a node in an underlying pose graph.

So, it seems that we need to know "scan data" and underlying pose praph for the application of "lidarSLAM."

To my understanding, "scan data" is composed of angels and ranges with respect to the robot coordinates.

And, what about the "underlying pose graph"? how do we know this information? and, if we know the odometer of the robot; then, we

have the information of the pose and how can we input to the "lidarSLAM"?

So, how do we know the information of the underlying pose graph and how do we input the pose graph in to "lidarSLAM"?

Thank you

Answer 1

[Amrtanshu Raj]

Hi,

A posegraph is an underlying graph that the LidarSlam uses to store all the scanned data and compute state estimate and optimize the graph. There is no need to input a pose graph to the lidarSlam, a new posegraph is created and associated with the LidarSlam when a new LidarSlam object is created.

You can view a posegraph by this code:

a  = lidarSLAM; 
Show(a.PoseGraph);  
 

The LidarSlam uses the scans to compute a state estimate of the vehicle by scan matching, you need not input the state estimate from the odometer. However, to fasten the process of scan matching, you can give a relative pose estimate from the odometer in the addScan function.