Robot Policy Learning

Ask about this topic

Overview

Robot policy learning studies how robots can turn camera, depth, tactile, and proprioceptive inputs into actions for tasks such as manipulation, locomotion, and navigation. Common approaches include behavior cloning from demonstrations, reinforcement learning in simulation or on hardware, and diffusion, flow-matching, or vision-language-action policies that generate action sequences; world models and hierarchical planners can add foresight and long-horizon structure. Researchers evaluate these methods in simulation suites such as LIBERO, CALVIN, RoboTwin, and navigation benchmarks, as well as on physical robots, with growing attention to memory, contact, safety, and recovery from failure. A central tension is how to make policies generalize across tasks, robots, and changing conditions while remaining data-efficient, reactive, reliable, and fast enough to deploy.

Latest papers

CardsList

377 papers85 papers in the last four weeks, up 47% on the four weeks before. 0.5% of all new papers.

Apr 15, 2026

ESCAPE: Episodic Spatial Memory and Adaptive Execution Policy for Long-Horizon Mobile Manipulation

Coordinating navigation and manipulation with robust performance is essential for embodied AI in complex indoor environments. However, as tasks extend over long horizons, existing methods often struggle due to catastrophic forgetting, spatial inconsistency, and rigid execution. To address these issues, we propose ESCAPE (Episodic Spatial Memory Coupled with an Adaptive Policy for Execution), operating through a tightly coupled perception-grounding-execution workflow. For robust perception, ESCAPE features a Spatio-Temporal Fusion Mapping module to autoregressively construct a depth-free, persistent 3D spatial memory, alongside a Memory-Driven Target Grounding module for precise interaction mask generation. To achieve flexible action, our Adaptive Execution Policy dynamically orchestrates proactive global navigation and reactive local manipulation to seize opportunistic targets. ESCAPE achieves state-of-the-art performance on the ALFRED benchmark, reaching 65.09% and 60.79% success rates in test seen and unseen environments with step-by-step instructions. By reducing redundant exploration, our ESCAPE attains substantial improvements in path-length-weighted metrics and maintains robust performance (61.24% / 56.04%) even without detailed guidance for long-horizon tasks.

Jingjing Qian, Zeyuan He, Chen Shi +2

The Chinese University of Hong Kong, Shenzhen · LiAuto Inc.

Robotic Spatial MemoryLong-Horizon Robotic ManipulationMobile ManipulationRoboticsRobot Policy Learning
AskPDF
Apr 13, 2026

ComSim: Building Scalable Real-World Robot Data Generation via Compositional Simulation

Recent advancements in foundational models, such as large language models and world models, have greatly enhanced the capabilities of robotics, enabling robots to autonomously perform complex tasks. However, acquiring large-scale, high-quality training data for robotics remains a challenge, as it often requires substantial manual effort and is limited in its coverage of diverse real-world environments. To address this, we propose a novel hybrid approach called Compositional Simulation, which combines classical simulation and neural simulation to generate accurate action-video pairs while maintaining real-world consistency. Our approach utilizes a closed-loop real-sim-real data augmentation pipeline, leveraging a small amount of real-world data to generate diverse, large-scale training datasets that cover a broader spectrum of real-world scenarios. We train a neural simulator to transform classical simulation videos into real-world representations, improving the accuracy of policy models trained in real-world environments. Through extensive experiments, we demonstrate that our method significantly reduces the sim2real domain gap, resulting in higher success rates in real-world policy model training. Our approach offers a scalable solution for generating robust training data and bridging the gap between simulated and real-world robotics.

Yiran Qin, Jiahua Ma, Li Kang +11

CUHK-Shenzhen · University of Oxford · Sun Yat-sen University +4

Synthetic Data Generation for RoboticsRobotics SimulationSynthetic Data AugmentationSim-to-Real TransferRobot Policy Learning
AskPDF
Apr 7, 2026

Rectified Schrödinger Bridge Matching for Few-Step Visual Navigation

Visual navigation is a core challenge in Embodied AI, requiring autonomous agents to translate high-dimensional sensory observations into continuous, long-horizon action trajectories. While generative policies based on diffusion models and Schrödinger Bridges (SB) effectively capture multimodal action distributions, they require dozens of integration steps due to high-variance stochastic transport, posing a critical barrier for real-time robotic control. We propose Rectified Schrödinger Bridge Matching (RSBM), a framework that exploits a shared velocity-field structure between standard Schrödinger Bridges (ε=1\varepsilon=1, maximum-entropy transport) and deterministic Optimal Transport (ε→0\varepsilon\to 0, as in Conditional Flow Matching), controlled by a single entropic regularization parameter ε\varepsilon. We prove two key results: (1) the conditional velocity field's functional form is invariant across the entire ε\varepsilon-spectrum (Velocity Structure Invariance), enabling a single network to serve all regularization strengths; and (2) reducing ε\varepsilon linearly decreases the conditional velocity variance, enabling more stable coarse-step ODE integration. Anchored to a learned conditional prior that shortens transport distance, RSBM operates at an intermediate ε\varepsilon that balances multimodal coverage and path straightness. Empirically, while standard bridges require ≥10\geq 10 steps to converge, RSBM achieves over 94% cosine similarity and 92% success rate in merely 3 integration steps -- without distillation or multi-stage training -- substantially narrowing the gap between high-fidelity generative policies and the low-latency demands of Embodied AI.

Wuyang Luan, Junhui Li, Weiguang Zhao +3

School of Artificial Intelligence, Jilin University, Changchun 130012, China · College of Computer Science, Chongqing University, Chongqing 400044, China · Department of Computer Science, University of Liverpool, Liverpool L69 3BX, UK +1

Schrödinger BridgesRobot Policy LearningVisual NavigationRobot Motion Generation
AskPDF
Apr 4, 2026

Multi-Robot Multi-Queue Control via Exhaustive Assignment Actor-Critic Learning

We study online task allocation for multi-robot, multi-queue systems with asymmetric stochastic arrivals and switching delays. We formulate the problem in discrete time: each location can host at most one robot per slot, servicing a task consumes one slot, switching between locations incurs a one-slot travel delay, and arrivals at locations are independent Bernoulli processes with heterogeneous rates. Building on our previous structural result that optimal policies are of exhaustive type, we formulate a discounted-cost Markov decision process and develop an exhaustive-assignment actor-critic policy architecture that enforces exhaustive service by construction and learns only the next-queue allocation for idle robots. Unlike the exhaustive-serve-longest (ESL) queue rule, whose optimality is known only under symmetry, the proposed policy adapts to asymmetry in arrival rates. Across different server-location ratios, loads, and asymmetric arrival profiles, the proposed policy consistently achieves lower discounted holding cost and smaller mean queue length than the ESL baseline, while remaining near-optimal on instances where an optimal benchmark is available. These results show that structure-aware actor-critic methods provide an effective approach for real-time multi-robot scheduling.

Mohammad Merati, H. M. Sabbir Ahmad, Wenchao Li +1

Divisions of Systems Engineering, Boston University, 8 St Mary’s St, Boston, MA 02215, United States. · Department of Electrical and Computer Engineering, Boston University, 8 St Mary’s St, Boston, MA 02215, United States.

Multi-Robot Task AllocationRobot Policy LearningQueueing TheoryActor-Critic MethodsMarkov Decision Processes
AskPDF
Apr 2, 2026

Tune to Learn: How Controller Gains Shape Robot Policy Learning

Position controllers have become the dominant interface for executing learned manipulation policies. Yet a critical design decision remains understudied: how should we choose controller gains for policy learning? The conventional wisdom is to select gains based on desired task compliance or stiffness. However, this logic breaks down when controllers are paired with state-conditioned policies: effective stiffness emerges from the interplay between learned reactions and control dynamics, not from gains alone. We argue that gain selection should instead be guided by learnability: how amenable different gain settings are to the learning algorithm in use. In this work, we systematically investigate how position controller gains affect three core components of modern robot learning pipelines: behavior cloning, reinforcement learning from scratch, and sim-to-real transfer. Through extensive experiments across multiple tasks and robot embodiments, we find that: (1) behavior cloning benefits from compliant and overdamped gain regimes, (2) reinforcement learning can succeed across all gain regimes given compatible hyperparameter tuning, and (3) sim-to-real transfer is harmed by stiff and overdamped gain regimes. These findings reveal that optimal gain selection depends not on the desired task behavior, but on the learning paradigm employed. Project website: https://younghyopark.me/tune-to-learn

Antonia Bronars, Younghyo Park, Pulkit Agrawal

MIT · MIT Improbable AI Lab

Robot Policy LearningRobotic ControlRobotic RLSim-to-Real TransferRobot Imitation Learning
AskPDF
Mar 31, 2026

Efficient Camera Pose Augmentation for View Generalization in Robotic Policy Learning

Prevailing 2D-centric visuomotor policies exhibit a pronounced deficiency in novel view generalization, as their reliance on static observations hinders consistent action mapping across unseen views. In response, we introduce GenSplat, a feed-forward 3D Gaussian Splatting framework that facilitates view-generalized policy learning through novel view rendering. GenSplat employs a permutation-equivariant architecture to reconstruct high-fidelity 3D scenes from sparse, uncalibrated inputs in a single forward pass. To ensure structural integrity, we design a 3D-prior distillation strategy that regularizes the 3DGS optimization, preventing the geometric collapse typical of purely photometric supervision. By rendering diverse synthetic views from these stable 3D representations, we systematically augment the observational manifold during training. This augmentation forces the policy to ground its decisions in underlying 3D structures, thereby ensuring robust execution under severe spatial perturbations where baselines severely degrade.

Sen Wang, Huaiyi Dong, Jingyi Tian +7

Xi’an Jiaotong University, Noah’s Ark Lab

Robot Policy GeneralizationRobot Policy LearningVisuomotor Policy Learning3D Gaussian Splatting
AskPDF
Mar 30, 2026

Enhancing Policy Learning with World-Action Model

This paper presents the World-Action Model (WAM), an action-regularized world model that jointly reasons over future visual observations and the actions that drive state transitions. Unlike conventional world models trained solely via image prediction, WAM incorporates an inverse dynamics objective into DreamerV2 that predicts actions from latent state transitions, encouraging the learned representations to capture action-relevant structure critical for downstream control. We evaluate WAM on enhancing policy learning across eight manipulation tasks from the CALVIN benchmark. We first pretrain a diffusion policy via behavioral cloning on world model latents, then refine it with model-based PPO inside the frozen world model. Without modifying the policy architecture or training procedure, WAM improves average behavioral cloning success from 59.4% to 71.2% over DreamerV2 and DiWA baselines. After PPO fine-tuning, WAM achieves 92.8% average success versus 79.8% for the baseline, with two tasks reaching 100%, using 8.7x fewer training steps.

Yuci Han, Alper Yilmaz

Photogrammetry and Computer Vision Lab, The Ohio State University, Columbus, OH 43210, USA.

Robot Policy LearningWorld Action ModelsAction-Conditioned World ModelsModel-Based RLRobot Imitation Learning
AskPDF
Mar 26, 2026

Policy-Guided World Model Planning for Language-Conditioned Visual Navigation

Navigating to a visually specified goal given natural language instructions remains a fundamental challenge in embodied AI. Existing approaches either rely on reactive policies that struggle with long-horizon planning, or employ world models that suffer from poor action initialization in high-dimensional spaces. We present PiJEPA, a two-stage framework that combines the strengths of learned navigation policies with latent world model planning for instruction-conditioned visual navigation. In the first stage, we finetune an Octo-based generalist policy, augmented with a frozen pretrained vision encoder (DINOv2 or V-JEPA-2), on the CAST navigation dataset to produce an informed action distribution conditioned on the current observation and language instruction. In the second stage, we use this policy-derived distribution to warm-start Model Predictive Path Integral (MPPI) planning over a separately trained JEPA world model, which predicts future latent states in the embedding space of the same frozen encoder. By initializing the MPPI sampling distribution from the policy prior rather than from an uninformed Gaussian, our planner converges faster to high-quality action sequences that reach the goal. We systematically study the effect of the vision encoder backbone, comparing DINOv2 and V-JEPA-2, across both the policy and world model components. Experiments on real-world navigation tasks demonstrate that PiJEPA significantly outperforms both standalone policy execution and uninformed world model planning, achieving improved goal-reaching accuracy and instruction-following fidelity.

Amirhosein Chahe, Lifeng Zhou

Drexel University Philadelphia PA 19104, USA

Embodied NavigationVision-Language NavigationRobot Policy LearningWorld Model-Based Planning
AskPDF
Mar 26, 2026

Modernising Reinforcement Learning-Based Navigation for Embodied Semantic Scene Graph Generation

Semantic world models enable embodied agents to reason about objects, relations, and spatial context beyond purely geometric representations. In Organic Computing, such models are a key enabler for objective-driven self-adaptation under uncertainty and resource constraints. The core challenge is to acquire observations maximising model quality and downstream usefulness within a limited action budget. Semantic scene graphs (SSGs) provide a structured and compact representation for this purpose. However, constructing them within a finite action horizon requires exploration strategies that trade off information gain against navigation cost and decide when additional actions yield diminishing returns. This work presents a modular navigation component for Embodied Semantic Scene Graph Generation and modernises its decision-making by replacing the policy-optimisation method and revisiting the discrete action formulation. We study compact and finer-grained, larger discrete motion sets and compare a single-head policy over atomic actions with a factorised multi-head policy over action components. We evaluate curriculum learning and optional depth-based collision supervision, and assess SSG completeness, execution safety, and navigation behaviour. Results show that replacing the optimisation algorithm alone improves SSG completeness by 21% relative to the baseline under identical reward shaping. Depth mainly affects execution safety (collision-free motion), while completeness remains largely unchanged. Combining modern optimisation with a finer-grained, factorised action representation yields the strongest overall completeness--efficiency trade-off.

Roman Küble, Marco Hüller, Mrunmai Phatak +2

Organic Computing Group · University of Augsburg, Am Technologiezentrum 8, Augsburg, Germany · Machine Learning and Computer Vision Group

Robotic RLScene Graph GenerationRobot Policy LearningSemantic Mapping
AskPDF
Mar 25, 2026

Decentralized End-to-End Multi-AAV Pursuit Using Predictive Spatio-Temporal Observation via Deep Reinforcement Learning

Decentralized cooperative pursuit in cluttered environments is challenging for autonomous aerial swarms, especially under partial and noisy perception. Existing methods often rely on abstracted geometric features or privileged ground-truth states, and therefore sidestep perceptual uncertainty in real-world settings. We propose a decentralized end-to-end multi-agent reinforcement learning (MARL) framework that maps raw LiDAR observations directly to continuous control commands. Central to the framework is the Predictive Spatio-Temporal Observation (PSTO), an egocentric grid representation that aligns obstacle geometry with predictive adversarial intent and teammate motion in a unified, fixed-resolution projection. Built on PSTO, a single decentralized policy enables agents to navigate static obstacles, intercept dynamic targets, and maintain cooperative encirclement. Simulations demonstrate that the proposed method achieves superior capture efficiency and competitive success rates compared to state-of-the-art learning-based approaches relying on privileged obstacle information. Furthermore, the unified policy scales seamlessly across different team sizes without retraining. Finally, fully autonomous outdoor experiments validate the framework on a quadrotor swarm relying on only onboard sensing and computing.

Yude Li, Zhexuan Zhou, Huizhe Li +6

The authors are with the School of Intelligence Science and Engineering, the Guangdong Key Laboratory of Intelligent Morphing Mechanisms and Adaptive Robotics, and the Shenzhen Key Lab for Advanced Motion Control and Modern Automation Equipments, Harbin Institute of Technology, Shenzhen, Guangdong 518055, China

Multi-Agent SystemsMulti-Robot CoordinationRobot Policy LearningDecentralized MARLAerial Robotics
AskPDF
Mar 23, 2026

VersualRL: Closed-Loop Verbal Reinforcement Learning with Visual Execution Feedback for Task-Level Robot Planning

We introduce VersualRL, a closed-loop framework for task-level robot planning that uses visual execution feedback to iteratively refine executable Behavior Trees through structured natural-language critique. VersualRL combines a vision-language model critic, which analyzes visual observations and Behavior Tree execution traces, with a large language model actor that performs discrete and interpretable policy updates. During physical deployment, both foundation models remain frozen, while the executed Behavior Tree is updated at the symbolic level without online gradient-based policy optimization. This enables transparent reasoning, explicit causal feedback, and human-interpretable policy evolution. We validate VersualRL on a real mobile robot performing a multi-stage navigation and manipulation task under execution uncertainty. Experimental results show that the framework supports explainable policy improvements, closed-loop adaptation to execution failures, and successful deployment on physical robotic systems.

Dmitrii Plotnikov, Iaroslav Kolomiets, Dmitrii Maliukov +9

The authors are with the Intelligent Space Robotics Laboratory, Skolkovo Institute of Science and Technology.

Robot Policy LearningLarge Language Model-Based Robot PlanningBehavior TreesVLMs for Robotics
AskPDF
Mar 21, 2026

Speedup Patch: Learning a Plug-and-Play Policy to Accelerate Embodied Manipulation

While current embodied policies exhibit remarkable manipulation skills, their execution remains unsatisfactorily slow as they inherit the tardy pacing of human demonstrations. Existing acceleration methods typically require policy retraining or costly online interactions, limiting their scalability for large-scale foundation models. In this paper, we propose Speedup Patch (SuP), a lightweight, policy-agnostic framework that enables plug-and-play acceleration using solely offline data. SuP introduces an external scheduler that adaptively downsamples action chunks provided by embodied policies to eliminate redundancies. Specifically, we formalize the optimization of our scheduler as a Constrained Markov Decision Process (CMDP) aimed at maximizing efficiency without compromising task performance. Since direct success evaluation is infeasible in offline settings, SuP introduces World Model based state deviation as a surrogate metric to enforce safety constraints. By leveraging a learned world model as a virtual evaluator to predict counterfactual trajectories, the scheduler can be optimized via offline reinforcement learning. Empirical results on simulation benchmarks (Libero, Bigym) and real-world tasks validate that SuP achieves an overall 1.8x execution speedup for diverse policies while maintaining their original success rates.

Zhichao Wu, Junyin Ye, Zhilong Zhang +6

National Key Laboratory for Novel Software Technology, Nanjing University, China · School of Artificial Intelligence, Nanjing University, China · Mila-Quebec AI Institute & Université de Montréal, Canada

Robot Policy LearningRobotic Manipulation
AskPDF
Mar 19, 2026

Articulated-Body Dynamics Network: Dynamics-Grounded Prior for Robot Learning

Recent work in reinforcement learning has shown that incorporating structural priors for articulated robots, such as link connectivity, into policy networks improves learning efficiency. However, dynamics properties, despite their fundamental role in determining how forces and motion propagate through the body, remain largely underexplored as an inductive bias for policy learning. To address this gap, we present the Articulated-Body Dynamics Network (ABD-Net), a novel graph neural network architecture grounded in the computational structure of forward dynamics. Specifically, we adapt the inertia propagation mechanism from the Articulated Body Algorithm, systematically aggregating inertial quantities from child to parent links in a tree-structured manner, while replacing physical quantities with learnable parameters. Embedding ABD-NET into the policy actor enables dynamics-informed representations that capture how actions propagate through the body, leading to efficient and robust policy learning. Through experiments with simulated humanoid, quadruped, and hopper robots, our approach demonstrates increased sample efficiency and generalization to dynamics shifts compared to transformer-based and GNN baselines. We further validate the learned policy on real Unitree G1 and Go2 robots, state-of-the-art humanoid and quadruped platforms, generating dynamic, versatile and robust locomotion behaviors through sim-to-real transfer with real-time inference.

Sangwoo Shin, Kunzhao Ren, Xiaobin Xiong +1

University of Wisconsin–Madison. · Shanghai Innovation Institute (SII).

Robot Policy LearningRobot LearningRL for Legged Locomotion
AskPDF
Mar 19, 2026

MemoAct: Atkinson-Shiffrin-Inspired Hierarchical Memory-Augmented Policy for Robotic Manipulation

Memory-augmented robotic policies are essential in handling memory-dependent tasks. However, existing approaches typically rely on simply extending the observation window, struggling to simultaneously achieve precise task-state tracking and robust long-horizon retention. To overcome these challenges, inspired by the Atkinson--Shiffrin memory model, we propose MemoAct, a hierarchical memory-augmented policy that leverages distinct memory tiers to tackle specific bottlenecks. Specifically, sensory memory filters immediate perceptual inputs, lossless short-term memory supports precise task-state tracking, and compressed long-term memory facilitates robust long-horizon retention. To enrich the evaluation landscape, we construct MemoryRTBench based on RoboTwin 2.0, comprising 6 manipulation tasks that systematically evaluate policy memory capabilities across three dimensions: sequential, spatial, and episodic memory. Extensive experiments across simulated and real-world scenarios demonstrate that MemoAct achieves superior performance compared to both existing Markovian baselines and history-aware policies. The project page is available at https://tlf-tlf.github.io/MemoActPage/.

Liufan Tan, Jiale Li, Gangshan Jing

Memory-Augmented AgentsRobot Policy LearningLong-Horizon Robotic ManipulationRobotic ManipulationRobot Manipulation Benchmarks
AskPDF
Mar 18, 2026

REAL: Robust Extreme Agility via Spatio-Temporal Policy Learning and Physics-Guided Filtering

Extreme legged parkour demands rapid terrain assessment and precise foot placement under highly dynamic conditions. While recent learning-based systems achieve impressive agility, they remain fundamentally fragile to perceptual degradation, where even brief visual noise or latency can cause catastrophic failure. To overcome this, we propose Robust Extreme Agility Learning (REAL), an end-to-end framework for reliable parkour under sensory corruption. Instead of relying on perfectly clean perception, REAL tightly couples vision, proprioceptive history, and temporal memory. We distill a cross-modal teacher policy into a deployable student equipped with a FiLM-modulated Mamba backbone to actively filter visual noise and build short-term terrain memory actively. Furthermore, a physics-guided Bayesian state estimator enforces rigid-body consistency during high-impact maneuvers. Validated on a Unitree Go2 quadruped, REAL successfully traverses extreme obstacles even with a 1-meter visual blind zone, while strictly satisfying real-time control constraints with a bounded 13.1 ms inference time.

Jialong Liu, Dehan Shen, Yanbo Wen +2

PEAK-Lab, The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, 511453, China.

Robot Policy LearningRobot LocomotionLegged RoboticsVisuomotor Policy LearningTerrain-Aware Robot Locomotion
AskPDF
Mar 15, 2026

SmallSatSim: A GPU-Accelerated Microgravity Robotics Toolkit for Planning, Control, and Policy Learning

Microgravity rendezvous and close proximity operations (RPO) is a growing area of interest for applications spanning in-space assembly and manufacturing (ISAM), orbital debris remediation, and small body exploration. Developing autonomy for these operations requires integrating dynamics simulation with task definition, control and learning algorithms, robustness testing, and evaluation. We present SmallSatSim, an open-source toolkit for developing autonomy algorithms for robots operating in microgravity environments. Built on MuJoCo, SmallSatSim provides a common experiment and task abstraction for spacecraft models, planners, controllers and policies, actuator effects, disturbances, and evaluation, allowing model-based and learning-based approaches to operate on the same problem definitions. The framework combines conventional MuJoCo execution for model-based control with vectorized JAX/MJX execution for massively parallel policy learning. We demonstrate SmallSatSim through Monte Carlo experiments on model-based control under actuator perturbations, five-seed PPO and SAC training under nominal and randomized dynamics with out-of-distribution evaluation, contact-rich rendezvous and docking, and GPU scaling experiments.

David Schwartz, Alexander Hansson, Sabrina Bodmer +3

ETH Zürich, Rämistrasse 101, 8092 Zürich, Switzerland · Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA 91011, USA · University of Michigan, 500 South State Street, Ann Arbor, MI 48109, USA +1

Robotics SimulationSpace RoboticsRobotic ControlRobot Policy LearningSpacecraft Rendezvous and Proximity Operations
AskPDF
Mar 14, 2026

Pixel-level Scene Understanding in One Token: Visual States Need What-is-Where Composition

For robotic agents operating in dynamic environments, learning visual state representations from streaming video observations is essential for sequential decision making. Recent self-supervised learning methods have shown strong transferability across vision tasks, but they do not explicitly address what a good visual state should encode. We argue that effective visual states must capture what-is-where by jointly encoding the semantic identities of scene elements and their spatial locations, enabling reliable detection of subtle dynamics across observations. To this end, we propose CroBo, a visual state representation learning framework based on a global-to-local reconstruction objective. Given a reference observation compressed into a compact bottleneck token, CroBo learns to reconstruct heavily masked patches in a local target crop from sparse visible cues, using the global bottleneck token as context. This learning objective encourages the bottleneck token to encode a fine-grained representation of scene-wide semantic entities, including their identities, spatial locations, and configurations. As a result, the learned visual states reveal how scene elements move and interact over time, supporting sequential decision making. We evaluate CroBo on diverse vision-based robot policy learning benchmarks, where it achieves state-of-the-art performance. Reconstruction analyses and perceptual straightness experiments further show that the learned representations preserve pixel-level scene composition and encode what-moves-where across observations. Project page available at: https://seokminlee-chris.github.io/CroBo-ProjectPage.

Seokmin Lee, Yunghee Lee, Byeonghyun Pak +1

Agency for Defense Development

State Representation LearningSelf-Supervised Visual Representation LearningObject-Centric Representation LearningRobot Policy Learning
AskPDF
Mar 13, 2026

Prepare to Turn: Learning Anticipatory Maneuvering for Fast Sequential Quadrupedal Navigation

Fast quadrupedal navigation requires preparing for the next turn while approaching the current goal. Policies trained to reach and remain at a single target can instead produce stop-and-go motion when applied to sequential goals, compromising its agility potential. We introduce PrepTurn, a reinforcement learning framework for anticipatory maneuvering in sequential local navigation. PrepTurn maps the current and next pose goals directly to joint actions, using the upcoming goal to inform how the robot should approaches the current one. Its persistent-credit reward encourages timely completion of the whole sequence by continuing to reward reached goals while moving on and preparing subsequent maneuvers, and an automatic curriculum that gradually increases goal difficulty is critical for learning stable turning behaviors. Together, these components enable the policy to coordinate approach speed and reorientation across goal transitions without prescribed trajectories or arrival times. Simulation and onboard Unitree Go2 experiments demonstrate fast, efficient sequential traversal with high completion and reduced stop-and-go motion. We further show that when the goal-reaching tolerances are more relaxed, the policy learns to exploit the freedom allowed by the task with emergent behaviors, such as turning to the next goal in advance and cutting corners within the permitted regions behaviors, to traverse goal sequences more efficiently.

Zunzhi You, Yunke Wang, Haolan Guo +1

All authors are with School of Computer Science, University of Sydney.

Robot Policy LearningRobot NavigationRL for Legged LocomotionQuadruped LocomotionQuadruped Robot Navigation
AskPDF
Mar 13, 2026

FLUX: Accelerating Cross-Embodiment Generative Navigation Policies via Rectified Flow and Static-to-Dynamic Learning

Autonomous navigation requires a broad spectrum of skills, from static goal-reaching to dynamic social traversal, yet evaluation remains fragmented across disparate protocols. We introduce DynBench, a dynamic navigation benchmark featuring physically valid crowd simulation. Combined with existing static protocols, it supports comprehensive evaluation across six fundamental navigation tasks. Within this framework, we propose FLUX, the first flow-based unified navigation policy. By linearizing probability flow, FLUX replaces iterative denoising with straight-line trajectories, improving per-step inference efficiency by 47% over prior flow-based methods and 29% over diffusion-based ones. Following a static-to-dynamic curriculum, FLUX initially establishes geometric priors and is subsequently refined through reinforcement learning in dynamic social environments. This regime not only strengthens socially-aware navigation but also enhances static task robustness by capturing recovery behaviors through stochastic action distributions. FLUX achieves state-of-the-art performance across all tasks and demonstrates zero-shot sim-to-real transfer on wheeled, quadrupedal, and humanoid platforms without any fine-tuning.

Zeying Gong, Yangyi Zhong, Yiyi Ding +6

The Hong Kong University of Science and Technology (Guangzhou) · The Hong Kong University of Science and Technology · National University of Singapore

Robot Navigation BenchmarksRobot Policy LearningRobot NavigationCross-Embodiment Robot LearningSocial Robot NavigationRectified Flow
AskPDF
Mar 12, 2026

Ada3Drift: Adaptive Training-Time Drifting for One-Step 3D Visuomotor Robotic Manipulation

Diffusion-based visuomotor policies model complex action distributions through iterative denoising, but repeated inference adds latency to robotic control. One-step generators reduce this cost, motivating training objectives that retain useful action structure with few demonstrations. We present Ada3Drift, a point-cloud-conditioned policy that builds on Drifting Models to perform distribution refinement during training and generate action chunks in one forward pass. Our central design is a regression-to-drifting curriculum: paired action regression first emphasizes observation--action correspondence, while a sigmoid schedule progressively increases a batch-level action-distribution regularizer. The drifting term combines attraction to demonstrated actions and repulsion among generated samples using inherited multi-temperature aggregation. A timestep-free generator preserves single-step inference throughout. On Adroit, Meta-World, RoboTwin, and five real-world tasks, Ada3Drift achieves the highest reported average success rates among the evaluated baselines with 1 NFE, compared with 10 NFE for the diffusion baselines. Controlled ablations favor sigmoid scheduling at matched demonstration budgets. We will release our code and pretrained model weights.

Chongyang Xu, Yixian Zou, Tianyu Yang +4

College of Computer Science, Sichuan University, Chengdu, China · School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu, China

Robot Policy LearningRobot Skill LearningOne-Step Generative ModelingRobotic ManipulationRobot Imitation LearningDrifting Models
AskPDF
Mar 10, 2026

From Prior to Pro: Efficient Skill Mastery via Distribution Contractive RL Finetuning

We introduce Distribution Contractive Reinforcement Learning (DICE-RL), a framework that uses reinforcement learning (RL) as a "distribution contraction" operator to refine pretrained generative robot policies. DICE-RL turns a pretrained behavior prior into a high-performing "pro" policy by amplifying high-success behaviors from online feedback. We pretrain a diffusion- or flow-based policy for broad behavioral coverage, then finetune it with a stable, sample-efficient residual off-policy RL framework that combines selective behavior regularization with value-guided action selection. Extensive experiments and analyses show that DICE-RL reliably improves performance with strong stability and sample efficiency. It enables mastery of complex long-horizon manipulation skills directly from high-dimensional pixel inputs, both in simulation and on a real robot. Project website: https://zhanyisun.github.io/dice.rl.2026/.

Zhanyi Sun, Shuran Song

1Stanford University.

RL for RoboticsRobot Policy LearningRobot Skill LearningData-Efficient Robot LearningResidual Policy LearningPolicy Learning+2
AskPDF
Mar 10, 2026

Quality over Quantity: Demonstration Curation via Influence Functions for Data-Centric Robot Learning

Learning from demonstrations has emerged as a promising paradigm for end-to-end robot control, particularly when scaled to diverse and large datasets. However, the quality of demonstration data, often collected through human teleoperation, remains a critical bottleneck for effective data-driven robot learning. Human errors, operational constraints, and teleoperator variability introduce noise and suboptimal behaviors, making data curation essential yet largely manual and heuristic-driven. In this work, we propose Quality over Quantity (QoQ), a grounded and systematic approach to identifying high-quality data by defining data quality as the contribution of each training sample to reducing loss on validation demonstrations. To efficiently estimate this contribution, we leverage influence functions, which quantify the impact of individual training samples on model performance. We further introduce two key techniques to adapt influence functions for robot demonstrations: (i) using maximum influence across validation samples to capture the most relevant state-action pairs, and (ii) aggregating influence scores of state-action pairs within the same trajectory to reduce noise and improve data coverage. Experiments in both simulated and real-world settings show that QoQ consistently improves policy performances over prior data selection methods.

Haeone Lee, Taywon Min, Junsu Kim +4

KAIST · University of California, Berkeley · New York University

Training Data CurationData-Efficient Robot LearningLearning from DemonstrationRobot Policy LearningInfluence Functions
AskPDF
Mar 9, 2026

OmniGuide: Universal Guidance Fields for Enhancing Generalist Robot Policies

Vision-language-action(VLA) models have shown great promise as generalist policies for a large range of relatively simple tasks. However, they demonstrate limited performance on more complex tasks, such as those requiring complex spatial or semantic understanding, manipulation in clutter, or precise manipulation. We propose OMNIGUIDE, a flexible framework that improves VLA performance on such tasks by leveraging arbitrary sources of guidance, such as 3D foundation models, semantic-reasoning VLMs, and human pose models. We show how many kinds of guidance can be naturally expressed as differentiable energy functions with task-specific attractors and repellers located in 3D space, that influence the sampling of VLA actions. In this way, OMNIGUIDE enables guidance sources with complementary task-relevant strengths to improve a VLA model's performance on challenging tasks. Extensive experiments in both simulation and real-world environments, across diverse sources of guidance, demonstrate that OMNIGUIDE enhances the performance of state-of-the-art generalist policies (e.g., π0.5π_{0.5}, GR00T N1.6) significantly across success and safety rates. Critically, our unified framework matches or surpasses the performance of prior methods designed to incorporate specific sources of guidance into VLA policies. Project Page: this ; url

Yunzhou Song, Long Le, Yong-Hyun Park +7

University of Pennsylvania

Robot Policy LearningVision-Language-Action ModelsRobotic Manipulation
AskPDF
Mar 9, 2026

3PoinTr: From Human Videos to Robot Policies with 3D Point-Track Plans

Human videos could reduce the reliance on expensive robot demonstrations, but extracting task-relevant features remains challenging. We introduce 3PoinTr, a method that provides robot policies with a global 3D point-track plan predicted from human videos. Given an initial point cloud, 3PoinTr predicts dense 3D point tracks spanning the full task and compresses them into a small set of tokens that remain fixed throughout execution. At each policy query, the policy combines these plan tokens with the current observation to predict robot actions. This formulation enables the policy to extract useful motion cues from human videos. We find this formulation transfers exceptionally well, even when human demonstration motions are infeasible for robots. In controlled simulation and real-world evaluations, 3PoinTr achieves an average success rate 25.0 percentage points higher than the strongest behavior-cloning and video-pretraining baselines on real-world tasks, and an improvement of 29.6 percentage points in simulation. Ablations establish the benefits of our method, including predicting non-embodiment 3D tracks once as a global plan, Perceiver-IO compression, and a residual cross-attention mechanism. Our point-track prediction network also outperforms a strong 3D-flow baseline by retaining supervision for temporarily occluded points. Project page: https://adamhung60.github.io/3PoinTr/

Adam Hung, Bardienus Pieter Duisterhof, Jeffrey Ichnowski

Carnegie Mellon University

Human-to-Robot Motion ImitationRobot Policy LearningRobot Learning from VideoRobot Imitation LearningCross-Embodiment Robot Learning3D Point Tracking+1
AskPDF
Mar 9, 2026

VORL-EXPLORE: A Hybrid Learning Planning Approach to Multi-Robot Exploration in Dynamic Environments

Hierarchical multi-robot exploration decouples frontier allocation from local navigation, making systems brittle in dynamic environments. Without awareness of execution difficulty, allocators may send robots toward frontiers whose routes are congested or unsafe, causing bottlenecks, oscillatory replanning, and redundant coverage. We propose VORL-EXPLORE, a hybrid learning and planning framework that makes exploration execution-aware through execution fidelity, an estimate of local navigability and execution reliability. This fidelity score is used to penalize risky or crowded frontiers in a fidelity-coupled Voronoi allocation objective with inter-robot repulsion, and to arbitrate between global A∗A^* guidance and a reactive reinforcement learning policy during navigation. As a result, robots follow efficient global plans in reliable regions and rely more on safer local reactions in crowded or changing areas. The fidelity model is recalibrated using self-supervised pseudo-labels derived from recent progress and safety outcomes. Quantitative experiments on randomized dynamic grids and an ablation under severe traffic show higher success rates, shorter exploration lengths, and lower inter-robot coverage overlap. A Gazebo factory scenario is included as a deployment illustration. The source code is available at https://github.com/21ning/VORL-EXPLORE.

Ning Liu, Sen Shen, Zheng Li +4

The University of Western Australia · The Chinese University of Hong Kong · Karlsruhe Institute of Technology +1

Multi-Robot ExplorationMulti-Robot Motion PlanningMulti-Robot Task AllocationRobot Policy Learning
AskPDF
Mar 9, 2026

RoboRouter: Training-Free Policy Routing for Robotic Manipulation

Research on robotic manipulation has developed a diverse set of policy paradigms, including vision-language-action (VLA) models, vision-action (VA) policies, and code-based compositional approaches. Concrete policies typically attain high success rates on specific task distributions, but limited generalization beyond it. Rather than proposing another monolithic policy, we propose to leverage the complementary strengths of existing approaches through intelligent policy routing. We introduce RoboRouter, a training-free framework that maintains a pool of heterogeneous policies and learns to select the best-performing policy for each task through accumulated execution experience. Given a new task, RoboRouter constructs a semantic task representation, retrieves historical records of similar tasks, predicts the optimal policy choice without requiring trial-and-error, and incorporates structured feedback to refine subsequent routing decisions. Integrating a new policy into the system requires only a lightweight evaluation and does not incur training overhead. Across simulation benchmark and real-world evaluations, RoboRouter consistently outperforms individual policies, improving the average success rate by more than 3% in simulation and 13% in real-world settings, while preserving execution efficiency. Our results demonstrate that intelligent routing across heterogeneous, off-the-shelf policies provides a practical and scalable pathway toward building more capable robotic systems.

Yiteng Chen, Zhe Cao, Hongjia Ren +9

South China University of Technology · Nanjing University · University of New South Wales +5

Robot Policy LearningRobotic ManipulationRobot Policy Generalization
AskPDF
Mar 8, 2026

PanoDP: Learning Collision-Free Navigation with Panoramic Depth and Differentiable Physics

Autonomous collision-free navigation in cluttered environments requires safe decision-making under partial observability with both static structure and dynamic obstacles. We present PanoDP, a communication-free learning framework that combines four-view panoramic depth perception with differentiable-physics-based training signals. PanoDP encodes panoramic depth using a lightweight CNN and optimizes policies with dense differentiable collision and motion-feasibility terms, improving training stability beyond sparse terminal collisions. We evaluate PanoDP on a controlled ring-to-center benchmark with systematic sweeps over agent count, obstacle density/layout, and dynamic behaviors, and further test out-of-distribution generalization in an external simulator (e.g., AirSim). Across settings, PanoDP increases collision-free and completion rates over single-view and non-physics-guided baselines under matched training budgets, and ablations (view masking, rotation augmentation) confirm the policy leverages 360-degree information. Code will be open source upon acceptance.

Hao Zhong, Pei Chi, Jiang Zhao +4

School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore. · Beihang University, Beijing, China. · The University of Queensland, Australia.

Panoramic Depth EstimationSafe Robot NavigationRobot Collision AvoidanceRobot Policy Learning
AskPDF
Mar 6, 2026

Don't Freeze, Don't Crash: Extending the Safe Operating Range of Neural Navigation in Dense Crowds

Navigating safely through dense crowds requires collision avoidance that generalizes beyond the densities seen during training. Learning-based crowd navigation can break under out-of-distribution crowd sizes due to density-sensitive observation normalization and social-cost scaling, while analytical solvers often remain safe but freeze in tight interactions. We propose a reinforcement learning approach for dense, variable-density navigation that attains zero-shot density generalization using a density-invariant observation encoding with density-randomized training and physics-informed proxemic reward shaping with density-adaptive scaling. The encoding represents the distance-sorted KK nearest pedestrians plus bounded crowd summaries, keeping input statistics stable as crowd size grows. Trained with N ⁣∈ ⁣[11,16]N\!\in\![11,16] pedestrians in a 3m×3m3\mathrm{m}\times3\mathrm{m} arena and evaluated up to N ⁣= ⁣21N\!=\!21 pedestrians (1.3×1.3\times denser), our policy reaches the goal in >99%>99\% of episodes and achieves 86%86\% collision-free success in random crowds, with markedly less freezing than analytical methods and a > ⁣60>\!60-point collision-free margin over learning-based benchmark methods. Codes are available at https://github.com/jznmsl/PSS-Social.

Jiefu Zhang, Yang Xu, Vaneet Aggarwal

Purdue University, USA

Safe Robot NavigationRobot NavigationRobot Collision AvoidanceRobot Policy LearningRobot Policy Generalization
AskPDF
Mar 5, 2026

Online Self-Training for Co-Adaptation in Hierarchical Diffusion Policies

Hierarchical policies decompose language-conditioned long-horizon robotic manipulation into a high-level planner and a low-level controller. However, effective coordination between HL and LL requires that both components operate on compatible subgoal distributions. We propose ORCHID, a self-training framework that enables stable online improvement of hierarchical diffusion policies by aligning planning and control through iterative refinement. By filtering policy samples via environment feedback, ORCHID identifies trajectories where the planner and controller are jointly successful and distills them back into both modules via supervised learning. This process induces a bidirectional co-adaptation: the planner grounds its subgoals in the actual reaching capabilities of the controller, while the controller specializes in the trajectory structures the planner produces. By relying on supervised distillation of filtered on-policy samples, ORCHID avoids the instability typical of online hierarchical gradient-based RL training with diffusion models. On the CALVIN benchmark, ORCHID allows a lightweight, initially weak model to outperform pure offline methods, including a Vision-Language-Action model twice its size.

Clemence Grislain, Mathilde Kappel, Olivier Sigaud +1

ISIR, Sorbonne Université, CNRS, Paris, France

Robot Policy LearningRobot Policy AdaptationDiffusion PolicyRobotic Manipulation
AskPDF
Mar 5, 2026

VPWEM: Non-Markovian Visuomotor Policy with Working and Episodic Memory

Imitation learning from human demonstrations has achieved significant success in robotic control, yet most visuomotor policies still condition on single-step observations or short-context histories, making them struggle with non-Markovian tasks that require long-term memory. Simply enlarging the context window incurs substantial computational and memory costs and encourages overfitting to spurious correlations, leading to catastrophic failures under distribution shift and violating real-time constraints in robotic systems. By contrast, humans can compress important past experiences into long-term memories and exploit them to solve tasks throughout their lifetime. In this paper, we propose VPWEM, a non-Markovian visuomotor policy equipped with working and episodic memories. VPWEM retains a sliding window of recent observation embeddings as short-term working memory, and introduces a Transformer-based contextual memory compressor that recursively converts out-of-window observations into a fixed number of episodic memory embeddings. The compressor uses self-attention over a cache of past summary embeddings and cross-attention over a cache of historical observations, and is trained jointly with the policy. We instantiate VPWEM on diffusion policies to exploit both short-term and episode-wide information for action generation with nearly constant memory and computation per step. Experiments demonstrate that VPWEM outperforms state-of-the-art baselines including diffusion policies and vision-language-action (VLA) models by more than 20% on the memory-intensive manipulation tasks in MIKASA and achieves an average 5% improvement on the mobile manipulation benchmark MoMaRT. Code is available at https://github.com/HarryLui98/code_vpwem.

Yuheng Lei, Zhixuan Liang, Hongyuan Zhang +1

School of Computing and Data Science, University of Hong Kong

Robot Policy LearningVisuomotor Policy LearningRobot Imitation LearningEpisodic MemoryDiffusion Policy
AskPDF