基于对象无关掩码的通用且高效视觉目标条件强化学习
摘要
目标条件强化学习（GCRL）允许智能体使用统一的策略学习多样化的目标。然而,GCRL的成功取决于目标表示的选择。本文提出一种基于掩码的目标表示系统,提供对象无关的视觉线索给智能体,实现高效学习和优异的泛化。在 contrast, existing goal representation methods, such as target state images, 3D coordinates, and one-hot vectors, face issues of poor generalization to unseen objects, slow convergence, and the need for special cameras. Masks can be processed to generate dense rewards without requiring error-prone distance calculations. Learning with ground truth masks in simulation, we achieved 99.9% reaching accuracy on training and unseen test objects. Our proposed method can be utilized to perform pick-up tasks with high accuracy, without using any positional information of the target. Moreover, we demonstrate learning from scratch and sim-to-real transfer applications using two different physical robots, utilizing pretrained open vocabulary object detection models for mask generation.
引用
@article{arxiv.2510.06277,
title = {General and Efficient Visual Goal-Conditioned Reinforcement Learning using Object-Agnostic Masks},
author = {Fahim Shahriar and Cheryl Wang and Alireza Azimi and Gautham Vasan and Hany Hamed Elanwar and A. Rupam Mahmood and Colin Bellinger},
journal= {arXiv preprint arXiv:2510.06277},
year = {2025}
}