Placeit! A Framework for Learning Robot Object Placement Skills
This work addresses the fundamental problem of data acquisition for robot object placement skills, offering a versatile tool for open-environment tasks and data generation for simulation-based models.
The paper tackles the challenge of generating valid object placement positions for robots by introducing Placeit!, an evolutionary-computation framework that uses quality-diversity optimization. It significantly outperforms state-of-the-art methods and achieves a 90% success rate in real-world pick-and-place deployments.
Robotics research has made significant strides in learning, yet mastering basic skills like object placement remains a fundamental challenge. A key bottleneck is the acquisition of large-scale, high-quality data, which is often a manual and laborious process. Inspired by Graspit!, a foundational work that used simulation to automatically generate dexterous grasp poses, we introduce Placeit!, an evolutionary-computation framework for generating valid placement positions for rigid objects. Placeit! is highly versatile, supporting tasks from placing objects on tables to stacking and inserting them. Our experiments show that by leveraging quality-diversity optimization, Placeit! significantly outperforms state-of-the-art methods across all scenarios for generating diverse valid poses. A pick&place pipeline built on our framework achieved a 90% success rate over 120 real-world deployments. This work positions Placeit! as a powerful tool for open-environment pick-and-place tasks and as a valuable engine for generating the data needed to train simulation-based foundation models in robotics.