Efficient Robotic Policy Learning via Latent Space Backward Planning

• URL: http://arxiv.org/abs/2505.06861v2
• Date: Tue, 27 May 2025 10:30:16 GMT
• Title: Efficient Robotic Policy Learning via Latent Space Backward Planning
• Authors: Dongxiu Liu, Haoyi Niu, Zhihao Wang, Jinliang Zheng, Yinan Zheng, Zhonghong Ou, Jianming Hu, Jianxiong Li, Xianyuan Zhan,
• Abstract summary: Current robotic planning methods often rely on predicting multi-frame images with full pixel details.<n>We propose a Latent Space Backward Planning scheme (LBP), which begins by grounding the task into final latent goals.<n>We show that LBP outperforms existing fine-grained and forward planning methods, achieving SOTA performance.
• Score: 17.770562202624962
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Current robotic planning methods often rely on predicting multi-frame images with full pixel details. While this fine-grained approach can serve as a generic world model, it introduces two significant challenges for downstream policy learning: substantial computational costs that hinder real-time deployment, and accumulated inaccuracies that can mislead action extraction. Planning with coarse-grained subgoals partially alleviates efficiency issues. However, their forward planning schemes can still result in off-task predictions due to accumulation errors, leading to misalignment with long-term goals. This raises a critical question: Can robotic planning be both efficient and accurate enough for real-time control in long-horizon, multi-stage tasks? To address this, we propose a Latent Space Backward Planning scheme (LBP), which begins by grounding the task into final latent goals, followed by recursively predicting intermediate subgoals closer to the current state. The grounded final goal enables backward subgoal planning to always remain aware of task completion, facilitating on-task prediction along the entire planning horizon. The subgoal-conditioned policy incorporates a learnable token to summarize the subgoal sequences and determines how each subgoal guides action extraction. Through extensive simulation and real-robot long-horizon experiments, we show that LBP outperforms existing fine-grained and forward planning methods, achieving SOTA performance. Project Page: https://lbp-authors.github.io

Related papers

• Hindsight Planner: A Closed-Loop Few-Shot Planner for Embodied Instruction Following [62.10809033451526]
  This work focuses on building a task planner for Embodied Instruction Following (EIF) using Large Language Models (LLMs)<n>We frame the task as a Partially Observable Markov Decision Process (POMDP) and aim to develop a robust planner under a few-shot assumption.<n>Our experiments on the ALFRED dataset indicate that our planner achieves competitive performance under a few-shot assumption.
  arXiv  Detail & Related papers  (2024-12-27T10:05:45Z)
• Closed-Loop Long-Horizon Robotic Planning via Equilibrium Sequence Modeling [21.45039811922009]
  We advocate a self-refining scheme that iteratively refines a draft plan until an equilibrium is reached.<n>A nested equilibrium sequence modeling procedure is devised for efficient closed-loop planning.<n>Our method is evaluated on the VirtualHome-Env benchmark, showing advanced performance with improved scaling w.r.t. inference-time computation.
  arXiv  Detail & Related papers  (2024-10-02T11:42:49Z)
• Propose, Assess, Search: Harnessing LLMs for Goal-Oriented Planning in Instructional Videos [48.15438373870542]
  VidAssist is an integrated framework designed for zero/few-shot goal-oriented planning in instructional videos. It employs a breadth-first search algorithm for optimal plan generation. Experiments demonstrate that VidAssist offers a unified framework for different goal-oriented planning setups.
  arXiv  Detail & Related papers  (2024-09-30T17:57:28Z)
• Learning to Reason over Scene Graphs: A Case Study of Finetuning GPT-2 into a Robot Language Model for Grounded Task Planning [45.51792981370957]
  We investigate the applicability of a smaller class of large language models (LLMs) in robotic task planning by learning to decompose tasks into subgoal specifications for a planner to execute sequentially. Our method grounds the input of the LLM on the domain that is represented as a scene graph, enabling it to translate human requests into executable robot plans. Our findings suggest that the knowledge stored in an LLM can be effectively grounded to perform long-horizon task planning, demonstrating the promising potential for the future application of neuro-symbolic planning methods in robotics.
  arXiv  Detail & Related papers  (2023-05-12T18:14:32Z)
• Imitating Graph-Based Planning with Goal-Conditioned Policies [72.61631088613048]
  We present a self-imitation scheme which distills a subgoal-conditioned policy into the target-goal-conditioned policy. We empirically show that our method can significantly boost the sample-efficiency of the existing goal-conditioned RL methods.
  arXiv  Detail & Related papers  (2023-03-20T14:51:10Z)
• A Framework for Neurosymbolic Robot Action Planning using Large Language Models [3.0501524254444767]
  We present a framework aimed at bridging the gap between symbolic task planning and machine learning approaches. The rationale is training Large Language Models (LLMs) into a neurosymbolic task planner compatible with the Planning Domain Definition Language (PDDL) Preliminary results in selected domains show that our method can: (i) solve 95.5% of problems in a test data set of 1,000 samples; (ii) produce plans up to 13.5% shorter than a traditional symbolic planner; (iii) reduce average overall waiting times for a plan availability by up to 61.4%.
  arXiv  Detail & Related papers  (2023-03-01T11:54:22Z)
• Planning to Practice: Efficient Online Fine-Tuning by Composing Goals in Latent Space [76.46113138484947]
  General-purpose robots require diverse repertoires of behaviors to complete challenging tasks in real-world unstructured environments. To address this issue, goal-conditioned reinforcement learning aims to acquire policies that can reach goals for a wide range of tasks on command. We propose Planning to Practice, a method that makes it practical to train goal-conditioned policies for long-horizon tasks.
  arXiv  Detail & Related papers  (2022-05-17T06:58:17Z)
• Active Learning of Abstract Plan Feasibility [17.689758291966502]
  We present an active learning approach to efficiently acquire an APF predictor through task-independent, curious exploration on a robot. We leverage an infeasible subsequence property to prune candidate plans in the active learning strategy, allowing our system to learn from less data. In a stacking domain where objects have non-uniform mass distributions, we show that our system permits real robot learning of an APF model in four hundred self-supervised interactions.
  arXiv  Detail & Related papers  (2021-07-01T18:17:01Z)
• Long-Horizon Visual Planning with Goal-Conditioned Hierarchical Predictors [124.30562402952319]
  The ability to predict and plan into the future is fundamental for agents acting in the world. Current learning approaches for visual prediction and planning fail on long-horizon tasks. We propose a framework for visual prediction and planning that is able to overcome both of these limitations.
  arXiv  Detail & Related papers  (2020-06-23T17:58:56Z)
• Divide-and-Conquer Monte Carlo Tree Search For Goal-Directed Planning [78.65083326918351]
  We consider alternatives to an implicit sequential planning assumption. We propose Divide-and-Conquer Monte Carlo Tree Search (DC-MCTS) for approximating the optimal plan. We show that this algorithmic flexibility over planning order leads to improved results in navigation tasks in grid-worlds.
  arXiv  Detail & Related papers  (2020-04-23T18:08:58Z)
• iCORPP: Interleaved Commonsense Reasoning and Probabilistic Planning on Robots [46.13039152809055]
  We present a novel algorithm, called iCORPP, to simultaneously estimate the current world state, reason about world dynamics, and construct task-oriented controllers. Results show significant improvements in scalability, efficiency, and adaptiveness, compared to competitive baselines.
  arXiv  Detail & Related papers  (2020-04-18T17:46:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.