GBPP: Grasp-Aware Base Placement Prediction for Robots via Two-Stage Learning
This provides a practical, data-efficient solution for geometry-aware base placement in robotics, though it is incremental in improving existing methods.
The paper tackles the problem of selecting robot base poses for grasping from a single RGB-D snapshot, achieving safer and more reachable stances that outperform proximity and geometry-only baselines in simulation and on a real mobile manipulator.
GBPP is a fast learning based scorer that selects a robot base pose for grasping from a single RGB-D snapshot. The method uses a two stage curriculum: (1) a simple distance-visibility rule auto-labels a large dataset at low cost; and (2) a smaller set of high fidelity simulation trials refines the model to match true grasp outcomes. A PointNet++ style point cloud encoder with an MLP scores dense grids of candidate poses, enabling rapid online selection without full task-and-motion optimization. In simulation and on a real mobile manipulator, GBPP outperforms proximity and geometry only baselines, choosing safer and more reachable stances and degrading gracefully when wrong. The results offer a practical recipe for data efficient, geometry aware base placement: use inexpensive heuristics for coverage, then calibrate with targeted simulation.