Warped Gaussian Processes Occupancy Mapping with Uncertain Inputs

TL;DR

This paper extends Gaussian Process occupancy mapping by incorporating pose uncertainty and non-Gaussian perception noise using Warped Gaussian Processes, resulting in improved map quality despite increased uncertainty.

Contribution

It introduces methods to handle pose uncertainty and non-Gaussian noise in occupancy mapping using expected kernels and Warped Gaussian Processes, enhancing map accuracy.

Findings

Modeling pose uncertainty increases map uncertainty.

Warped Gaussian Processes improve map quality.

Methods can be applied separately or together.

Abstract

In this paper, we study extensions to the Gaussian Processes (GPs) continuous occupancy mapping problem. There are two classes of occupancy mapping problems that we particularly investigate. The first problem is related to mapping under pose uncertainty and how to propagate pose estimation uncertainty into the map inference. We develop expected kernel and expected sub-map notions to deal with uncertain inputs. In the second problem, we account for the complication of the robot's perception noise using Warped Gaussian Processes (WGPs). This approach allows for non-Gaussian noise in the observation space and captures the possible nonlinearity in that space better than standard GPs. The developed techniques can be applied separately or concurrently to a standard GP occupancy mapping problem. According to our experimental results, although taking into account pose uncertainty leads, as expected, to more uncertain maps, by modeling the nonlinearities present in the observation space WGPs improve the map quality.