HybridGen: VLM-Guided Hybrid Planning for Scalable Data Generation of Imitation Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Wensheng, Tan, Ning
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916654519156736
author Wang, Wensheng
Tan, Ning
author_facet Wang, Wensheng
Tan, Ning
contents The acquisition of large-scale and diverse demonstration data are essential for improving robotic imitation learning generalization. However, generating such data for complex manipulations is challenging in real-world settings. We introduce HybridGen, an automated framework that integrates Vision-Language Model (VLM) and hybrid planning. HybridGen uses a two-stage pipeline: first, VLM to parse expert demonstrations, decomposing tasks into expert-dependent (object-centric pose transformations for precise control) and plannable segments (synthesizing diverse trajectories via path planning); second, pose transformations substantially expand the first-stage data. Crucially, HybridGen generates a large volume of training data without requiring specific data formats, making it broadly applicable to a wide range of imitation learning algorithms, a characteristic which we also demonstrate empirically across multiple algorithms. Evaluations across seven tasks and their variants demonstrate that agents trained with HybridGen achieve substantial performance and generalization gains, averaging a 5% improvement over state-of-the-art methods. Notably, in the most challenging task variants, HybridGen achieves significant improvement, reaching a 59.7% average success rate, significantly outperforming Mimicgen's 49.5%. These results demonstrating its effectiveness and practicality.
format Preprint
id arxiv_https___arxiv_org_abs_2503_13171
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle HybridGen: VLM-Guided Hybrid Planning for Scalable Data Generation of Imitation Learning
Wang, Wensheng
Tan, Ning
Robotics
Artificial Intelligence
The acquisition of large-scale and diverse demonstration data are essential for improving robotic imitation learning generalization. However, generating such data for complex manipulations is challenging in real-world settings. We introduce HybridGen, an automated framework that integrates Vision-Language Model (VLM) and hybrid planning. HybridGen uses a two-stage pipeline: first, VLM to parse expert demonstrations, decomposing tasks into expert-dependent (object-centric pose transformations for precise control) and plannable segments (synthesizing diverse trajectories via path planning); second, pose transformations substantially expand the first-stage data. Crucially, HybridGen generates a large volume of training data without requiring specific data formats, making it broadly applicable to a wide range of imitation learning algorithms, a characteristic which we also demonstrate empirically across multiple algorithms. Evaluations across seven tasks and their variants demonstrate that agents trained with HybridGen achieve substantial performance and generalization gains, averaging a 5% improvement over state-of-the-art methods. Notably, in the most challenging task variants, HybridGen achieves significant improvement, reaching a 59.7% average success rate, significantly outperforming Mimicgen's 49.5%. These results demonstrating its effectiveness and practicality.
title HybridGen: VLM-Guided Hybrid Planning for Scalable Data Generation of Imitation Learning
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2503.13171