# GBPP: Grasp-Aware Base Placement Prediction for Robots via Two-Stage Learning

arXiv: https://arxiv.org/abs/2509.11594 · HTML: https://mlsanity.com/paper/2509.11594/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 19 references that are in the map.
- First posted on arXiv 15 Sep 2025.

## Builds on (most cited first, 11 of 11)

- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [Point Transformer](https://mlsanity.com/paper/2011.00931/) (2020)
- [Detecting Twenty-thousand Classes using Image-level Supervision](https://mlsanity.com/paper/2201.02605/) (2022)
- [ProcTHOR: Large-Scale Embodied AI Using Procedural Generation](https://mlsanity.com/paper/2206.06994/) (2022)
- [Integrated Task and Motion Planning](https://mlsanity.com/paper/2010.01083/) (2020)
- [ManiSkill: Generalizable Manipulation Skill Benchmark with Large-Scale Demonstrations](https://mlsanity.com/paper/2107.14483/) (2021)
- [PointNet++ Grasping: Learning An End-to-end Spatial Grasp Generation Algorithm from Sparse Point Clouds](https://mlsanity.com/paper/2003.09644/) (2020)
- [Object-Centric Task and Motion Planning in Dynamic Environments](https://mlsanity.com/paper/1911.04679/) (2019)
- [A Survey of Robotic Language Grounding: Tradeoffs between Symbols and Embeddings](https://mlsanity.com/paper/2405.13245/) (2024)
- [Reuleaux: Robot Base Placement by Reachability Analysis](https://mlsanity.com/paper/1710.01328/) (2017)
- [Planning Robot Placement for Object Grasping](https://mlsanity.com/paper/2405.16692/) (2024)

## Built on by (most cited first, 1 of 1)

- [UniWAM Technical Report: Unified Mobile Manipulation via Mixed-Stream World-Action Modeling and Manipulation Anchor Pose Supervision](https://mlsanity.com/paper/2609.39388/) (2026)

## Abstract (arXiv)

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.