Active Exploration Using Gaussian Random Fields and Gaussian Process Implicit Surfaces
This work addresses the challenge of efficient environment exploration for robots, though it appears incremental as it builds on existing probabilistic methods for active perception.
The paper tackles the problem of actively exploring surfaces and building 3D representations for robots by proposing an online probabilistic framework that merges visual and tactile measurements using Gaussian Random Fields and Gaussian Process Implicit Surfaces, demonstrating experimental results with a PrimeSense camera, Kinova Jaco2 arm, and Optoforce sensors for object detection and terrain classification.
In this work we study the problem of exploring surfaces and building compact 3D representations of the environment surrounding a robot through active perception. We propose an online probabilistic framework that merges visual and tactile measurements using Gaussian Random Field and Gaussian Process Implicit Surfaces. The system investigates incomplete point clouds in order to find a small set of regions of interest which are then physically explored with a robotic arm equipped with tactile sensors. We show experimental results obtained using a PrimeSense camera, a Kinova Jaco2 robotic arm and Optoforce sensors on different scenarios. We then demonstrate how to use the online framework for object detection and terrain classification.