GPAS: Gaussian Process Adaptive Sampling

This is currently under construction.
Please contact Lauren Miller (laurenm[@]berkeley[dot]edu) for questions.

This is demonstration code for Tumor Localization using Automated Palpation withGaussian Process Adaptive Sampling. The paper will be presented at IEEE CASE 2016.

Dependencies

In addition to numpy, scipy, matplotlib, this code depends on:

• gPy
• shapely

Running Demo Files

To run a single simulated experiment, run runPhase2.py

To run a batch of simulated experiments, varying noise and bias levels, run scaffoldPhase2.py

To run an experiment on the robot, edit configs in runPhase2.py and run (needs ROS)