This repository contains source codes and demo files for our paper Online Continuous Mapping using Gaussian Process Implicit Surfaces (GPIS), which is going to be presented at IEEE ICRA 2019.

The representation of the environment strongly affects how robots can move and interact with it. The paper presents an online approach for continuous mapping using Gaussian Process Implicit Surfaces (GPIS). Compared with grid-based methods, GPIS better utilizes sparse measurements to represent the world seamlessly. It provides direct access to the signed-distance function (SDF) and its derivatives which are invaluable for other robotic tasks and incorporates uncertainty in the sensor measurements. Our approach incrementally and efficiently updates GPIS by employing a regressor on observations and a spatial tree structure.

Licensed under GNU General Public License version 3.

1. Eigen

2. MATLAB

1. Clone this repository

git clone https://github.com/leebhoram/GPisMap.git

2. Cd to the mex directive in MATLAB

cd mex

3. Compile the mex functions by executing the make script.

   • Setup mex

   mex -setup
   

   • Run the make scripts

   make_GPisMap
   make_GPisMap3
   

4. Run the demo scripts

   • For 2D

   run('../matlab/demo_gpisMap.m')
   

   • For 3D

   run('../matlab/demo_gpisMap3.m')
   

5. Trouble shooting

   • If mex complains about not finding eigen, configure the eigen path appropriately in both make_GPisMap.m and make_GPisMap3.m

The major contributors of this work include Bhoram Lee, Clark Zhang and HUANG Zonghao.

Code has been tested under:

• Ubuntu 16.04 with Intel Core i7-4900MQ @ 2.90GHz