Long-Horizon Manipulation via Trace-Conditioned VLA Planning

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Liu, Isabella, Cheng, An-Chieh, Yan, Rui, Chen, Geng, Qiu, Ri-Zhao, Zou, Xueyan, Yi, Sha, Yin, Hongxu, Wang, Xiaolong, Liu, Sifei
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910160706863104
author Liu, Isabella
Cheng, An-Chieh
Yan, Rui
Chen, Geng
Qiu, Ri-Zhao
Zou, Xueyan
Yi, Sha
Yin, Hongxu
Wang, Xiaolong
Liu, Sifei
author_facet Liu, Isabella
Cheng, An-Chieh
Yan, Rui
Chen, Geng
Qiu, Ri-Zhao
Zou, Xueyan
Yi, Sha
Yin, Hongxu
Wang, Xiaolong
Liu, Sifei
contents Long-horizon manipulation remains challenging for vision-language-action (VLA) policies: real tasks are multi-step, progress-dependent, and brittle to compounding execution errors. We present LoHo-Manip, a modular framework that scales short-horizon VLA execution to long-horizon instruction following via a dedicated task-management VLM. The manager is decoupled from the executor and is invoked in a receding-horizon manner: given the current observation, it predicts a progress-aware remaining plan that combines (i) a subtask sequence with an explicit done + remaining split as lightweight language memory, and (ii) a visual trace -- a compact 2D keypoint trajectory prompt specifying where to go and what to approach next. The executor VLA is adapted to condition on the rendered trace, thereby turning long-horizon decision-making into repeated local control by following the trace. Crucially, predicting the remaining plan at each step yields an implicit closed loop: failed steps persist in subsequent outputs, and traces update accordingly, enabling automatic continuation and replanning without hand-crafted recovery logic or brittle visual-history buffers. Extensive experiments spanning embodied planning, long-horizon reasoning, trajectory prediction, and end-to-end manipulation in simulation and on a real Franka robot demonstrate strong gains in long-horizon success, robustness, and out-of-distribution generalization. Project page: https://www.liuisabella.com/LoHoManip
format Preprint
id arxiv_https___arxiv_org_abs_2604_21924
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Long-Horizon Manipulation via Trace-Conditioned VLA Planning
Liu, Isabella
Cheng, An-Chieh
Yan, Rui
Chen, Geng
Qiu, Ri-Zhao
Zou, Xueyan
Yi, Sha
Yin, Hongxu
Wang, Xiaolong
Liu, Sifei
Robotics
Long-horizon manipulation remains challenging for vision-language-action (VLA) policies: real tasks are multi-step, progress-dependent, and brittle to compounding execution errors. We present LoHo-Manip, a modular framework that scales short-horizon VLA execution to long-horizon instruction following via a dedicated task-management VLM. The manager is decoupled from the executor and is invoked in a receding-horizon manner: given the current observation, it predicts a progress-aware remaining plan that combines (i) a subtask sequence with an explicit done + remaining split as lightweight language memory, and (ii) a visual trace -- a compact 2D keypoint trajectory prompt specifying where to go and what to approach next. The executor VLA is adapted to condition on the rendered trace, thereby turning long-horizon decision-making into repeated local control by following the trace. Crucially, predicting the remaining plan at each step yields an implicit closed loop: failed steps persist in subsequent outputs, and traces update accordingly, enabling automatic continuation and replanning without hand-crafted recovery logic or brittle visual-history buffers. Extensive experiments spanning embodied planning, long-horizon reasoning, trajectory prediction, and end-to-end manipulation in simulation and on a real Franka robot demonstrate strong gains in long-horizon success, robustness, and out-of-distribution generalization. Project page: https://www.liuisabella.com/LoHoManip
title Long-Horizon Manipulation via Trace-Conditioned VLA Planning
topic Robotics
url https://arxiv.org/abs/2604.21924