Active exploration using Gaussian Random Fields and Gaussian Process Implicit Surfaces

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 demostrate how to use the online framework for object detection and terrain classification.

Authors

• Sergio Caccamo
• Yasemin Bekiroglu
• Carl Henrik Ek
• Danica Kragic

Keywords

• Surface treatment
• Gaussian processes
• Three-dimensional displays
• Robot sensing systems
• Shape
• Surface reconstruction
• Random Fields
• Gaussian Process
• Gaussian Random Field
• Gaussian Random Process
• Implicit Surface
• Point Cloud
• Robotic Arm
• Tactile Sensor
• Representation Of The Environment
• Probabilistic Framework
• Interactive
• Training Set
• Physical Interaction
• Contact Point
• Digital Elevation Model
• Mental Representations
• Surface Model
• Kriging
• Surface Reflectance
• Test Points
• Unmanned Ground Vehicles
• Delaunay Triangulation
• Amount Of Interaction
• 3D Point
• Contact Force
• Covariance Function
• Occluded Regions
• World Frame
• Occluded Objects
• Empty Regions
• Active perception
• Random field
• Tactile exploration