Optimizing Robotic Placement via Grasp-Dependent Feasibility Prediction

2025-12-21

AI Summary

This work addresses budget-constrained robotic pick-and-place tasks by proposing a low-cost feasibility prediction method that requires neither physics simulation nor real-world interaction supervision. To rank candidate pick-and-place pairs, we design a lightweight dual-output MLP that takes a grasp pose as input and jointly predicts two geometric labels: inverse-kinematics feasibility and path collision risk. We enhance geometric generalization via path-aware pose encoding and a waypoint mesh sweep template, and adopt a “rank-then-plan” strategy for efficient resource allocation. Our approach is the first to achieve cross-domain transfer under purely geometric supervision. On real hardware, it significantly reduces planner invocations, accelerates discovery of successful paths, and maintains—or even improves—task success rates. This establishes an efficient, unsupervised learning paradigm for low-budget robotic manipulation.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

In this paper, we study whether inexpensive, physics-free supervision can reliably prioritize grasp-place candidates for budget-aware pick-and-place. From an object's initial pose, target pose, and a candidate grasp, we generate two path-aware geometric labels: path-wise inverse kinematics (IK) feasibility across a fixed approach-grasp-lift waypoint template, and a transit collision flag from mesh sweeps along the same template. A compact dual-output MLP learns these signals from pose encodings, and at test time its scores rank precomputed candidates for a rank-then-plan policy under the same IK gate and planner as the baseline. Although learned from cheap labels only, the scores transfer to physics-enabled executed trajectories: at a fixed planning budget the policy finds successful paths sooner with fewer planner calls while keeping final success on par or better. This work targets a single rigid cuboid with side-face grasps and a fixed waypoint template, and we outline extensions to varied objects and richer waypoint schemes.

Problem

• Predict grasp-place feasibility using physics-free geometric labels
• Rank candidates to reduce planner calls in pick-and-place tasks
• Transfer learned scores to physics-enabled robotic execution

Innovation

• Physics-free geometric labels for grasp-place feasibility
• Compact MLP learns dual-output signals from pose encodings
• Rank-then-plan policy reduces planner calls with cheap supervision