RL for Robotics

RL: Reinforcement Learning

Follow this topic Ask about this topic

Reinforcement learning for robotics studies how robots can learn actions that achieve goals while coping with contact, uncertainty, and changing environments. Common methods include PPO, SAC, and TD3, often trained in parallel simulation with domain randomization; imitation learning, motion priors, and pretrained vision-language-action models increasingly provide useful starting points. Work spans humanoid locomotion, aerial and mobile navigation, and manipulation that uses tactile feedback, with safety filters and learned world models helping address risky or partially observed tasks. A central challenge is making learning more sample-efficient while ensuring that simulated skills and learned rewards remain reliable on real robots and unfamiliar situations.

Latest papers

253 papers45 papers in the last four weeks, up 45% on the four weeks before. 0.3% of all new papers.

Mar 16, 2026

HALO:Closing Sim-to-Real Gap for Heavy-loaded Humanoid Agile Motion Skills via Differentiable Simulation

Humanoid robots deployed in real-world scenarios often need to carry unknown payloads, which introduce significant mismatch and degrade the effectiveness of simulation-to-reality reinforcement learning methods. To address this challenge, we propose a two-stage gradient-based system identification framework built on the differentiable simulator MuJoCo XLA. The first stage calibrates the nominal robot model using real-world data to reduce intrinsic sim-to-real discrepancies, while the second stage further identifies the mass distribution of the unknown payload. By explicitly reducing structured model bias prior to policy training, our approach enables zero-shot transfer of reinforcement learning policies to hardware under heavy-load conditions. Extensive simulation and real-world experiments demonstrate more precise parameter identification, improved motion tracking accuracy, and substantially enhanced agility and robustness compared to existing baselines. Project Page: https://mwondering.github.io/halo-humanoid/

Xingyi Wang, Chenyun Zhang, Weiji Xie +4

Zhejiang University · Institute of Artificial Intelligence (TeleAI), China Telecom · Shanghai Jiao Tong University +1

Sim-to-Real Transfer RL for Robotics Robot Dynamics Identification Robot Locomotion

Mar 16, 2026

CycleRL: Sim-to-Real Deep Reinforcement Learning for Robust Autonomous Bicycle Control

Autonomous bicycles offer a promising agile solution for urban mobility and last-mile logistics. However, conventional control strategies often struggle with underactuated nonlinear dynamics, suffering from sensitivity to model mismatches and limited adaptability to real-world uncertainties. To address this, we develop CycleRL, a comprehensive sim-to-real framework for robust autonomous bicycle control. Our approach establishes a direct perception-to-action mapping within the high-fidelity NVIDIA Isaac Sim environment, leveraging Proximal Policy Optimization (PPO) to optimize the control policy. The framework features a composite reward function tailored for concurrent balance maintenance, velocity tracking, and steering control. Crucially, systematic domain randomization is employed to reduce the reliance on precise system modeling, bridge the simulation-to-reality gap and facilitate direct transfer. In simulation, CycleRL achieves promising performance, including a 99.90% balance success rate, a heading tracking error of 1.15°, and a velocity tracking error of 0.18 m/s. These quantitative results, coupled with successful hardware deployment, validate DRL as an effective paradigm for autonomous bicycle control, offering superior adaptability over traditional methods. Video demonstrations are available at https://cpnt-lab.github.io/CycleRL/.

Gelu Liu, Teng Wang, Zhijie Wu +3

School of Electronics and Communication Engineering and Shenzhen Key Laboratory of Navigation and Communication Integration, Sun Yat-sen University, Shenzhen 518107, China

Robotic Control RL for Robotics Sim-to-Real Transfer Robust RL

Mar 13, 2026

Learning Energy-Efficient Air--Ground Actuation for Hybrid Robots on Stair-Like Terrain

Hybrid aerial--ground robots can use thrust to cross obstacles that impede wheel-driven motion, but deciding how much thrust to apply during contact remains challenging. We present an energy-aware reinforcement learning framework that jointly commands wheels, tilt servos and propellers through a single continuous policy, without prescribing locomotion modes. Hardware-calibrated power models penalise estimated electrical energy, while a terrain curriculum and a terminal reward for upright, settled arrivals support learning of thrust-assisted climbing. In simulation, continuous thrust allocation improves single-step clearance over fixed-thrust and mode-switching baselines as steps become taller. At the wheel-radius step height, it draws approximately 27 percent less mean power than the best fixed allocation. An energy-weight ablation shows the accompanying trade-off between efficiency and reliability. On a physical DoubleBee prototype, the unchanged network with a deployment interface clears two 6 centimetre steps in 8 of 10 trials. Additional hardware tests show both the potential for transfer to different terrain geometries and the remaining limitations in heading control, contact robustness and recovery.

Jiaxing Li, Ishaan Bhimwal, Wen Tian +5

Carnegie Mellon University, US · Technische Hochschule Würzburg-Schweinfurt, Germany · Nanyang Technological University, Singapore +1

Robot Locomotion Energy-Efficient Robot Locomotion Terrain-Aware Robot Locomotion RL for Robotics

Mar 12, 2026

NFPO: Stabilized Policy Optimization of Normalizing Flow for Robotic Policy Learning

Deep Reinforcement Learning (DRL) has experienced significant advancements in recent years and has been widely used in many fields. In DRL-based robotic policy learning, however, current de facto policy parameterization is still multivariate Gaussian (with diagonal covariance matrix), which lacks the ability to model multi-modal distribution. In this work, we explore the adoption of a modern network architecture, i.e. Normalizing Flow (NF) as the policy parameterization for its ability of multi-modal modeling, closed form of log probability and low computation and memory overhead. However, naively training NF in online Reinforcement Learning (RL) usually leads to training instability. We provide a detailed analysis for this phenomenon and successfully address it via simple but effective technique. With extensive experiments in multiple simulation environments, we show our method, NFPO could obtain robust and strong performance in widely used robotic learning tasks and successfully transfer into real-world robots.

Diyuan Shi, Yiqi Tang, Zifeng Zhuang +1

Zhejiang University · Westlake University

RL for Robotics Robotic RL Normalizing Flows

Mar 11, 2026

AdaClearGrasp: Learning Adaptive Clearing for Zero-Shot Robust Dexterous Grasping in Densely Cluttered Environments

In densely cluttered environments, physical interference, visual occlusions, and unstable contacts often cause direct dexterous grasping to fail, while aggressive singulation strategies may compromise safety. Enabling robots to adaptively decide whether to clear surrounding objects or directly grasp the target is therefore crucial for robust manipulation. We propose AdaClearGrasp, a closed-loop decision-execution framework for adaptive clearing and zero-shot dexterous grasping in densely cluttered environments. The framework formulates manipulation as a controllable high-level decision process that determines whether to directly grasp the target or first clear surrounding objects. A pretrained vision-language model (VLM) interprets visual observations and language task descriptions to reason about grasp interference and generate a high-level planning skeleton, which invokes structured atomic skills through a unified action interface. For dexterous grasping, we train a reinforcement learning policy with a relative hand-object distance representation, enabling zero-shot generalization across diverse object geometries and physical properties. During execution, visual feedback monitors outcomes and triggers replanning upon failures, forming a closed-loop correction mechanism. To evaluate language-conditioned dexterous grasping in clutter, we introduce Clutter-Bench, the first simulation benchmark with graded clutter complexity. It includes seven target objects across three clutter levels, yielding 210 task scenarios. We further perform sim-to-real experiments on three objects under three clutter levels (18 scenarios). Results demonstrate that AdaClearGrasp significantly improves grasp success rates in densely cluttered environments. For more videos and code, please visit our project website: https://chenzixuan99.github.io/adaclear-grasp.github.io/.

Zixuan Chen, Wenquan Zhang, Jing Fang +7

Nanjing University · Nanjing University of Posts and Telecommunications · National University of Singapore

Robotic Grasping Robotics Zero-Shot Robotic Manipulation RL for Robotics Robot Manipulation Benchmarks

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 Robotics Robot Policy Learning Robot Skill Learning Data-Efficient Robot Learning Residual Policy Learning Policy Learning+2

Mar 10, 2026

Emerging Extrinsic Dexterity in Cluttered Scenes via Dynamics-aware Policy Learning

Extrinsic dexterity leverages environmental contact to overcome the limitations of prehensile manipulation. However, achieving such dexterity in cluttered scenes remains challenging and underexplored, as it requires selectively exploiting contact among multiple interacting objects with inherently coupled dynamics. Existing approaches lack explicit modeling of such complex dynamics and therefore fall short in non-prehensile manipulation in cluttered environments, which in turn limits their practical applicability in real-world environments. In this paper, we introduce a Dynamics-Aware Policy Learning (DAPL) framework that can facilitate policy learning with a learned representation of contact-induced object dynamics in cluttered environments. This representation is learned through explicit world modeling and used to condition reinforcement learning, enabling extrinsic dexterity to emerge without hand-crafted contact heuristics or complex reward shaping. We evaluate our approach in both simulation and the real world. Our method outperforms prehensile manipulation, human teleoperation, and prior representation-based policies by over 25% in success rate on unseen simulated cluttered scenes with varying densities. The real-world success rate reaches around 50% across 10 cluttered scenes, while a practical grocery deployment further demonstrates robust sim-to-real transfer and applicability.

Yixin Zheng, Jiangran Lyu, Yifan Zhang +8

Institute of Automation, Chinese Academy of Sciences · Beijing Academy of Artificial Intelligence · Galbot +2

Non-Prehensile Robotic Manipulation Robotic Manipulation RL for Robotics Robot Dynamics Learning Generalization in Robotic Manipulation

Mar 10, 2026

MO-Playground: Massively Parallelized Multi-Objective Reinforcement Learning for Robotics

Multi-objective reinforcement learning (MORL) is a powerful tool to learn Pareto-optimal policy families across conflicting objectives. However, unlike traditional RL algorithms, existing MORL algorithms do not effectively leverage large-scale parallelization to concurrently simulate thousands of environments, thus facing vastly increased computation time. Ultimately, this has limited the application of MORL towards complex multi-objective robotics problems. To address these challenges, we present 1) MORLAX, a new GPU-native, fast MORL algorithm, and 2) MO-Playground, a pip-installable playground of GPU-accelerated multi-objective environments. Together, MORLAX and MO-Playground approximate Pareto sets within minutes, offering 26-271x speed-ups compared to legacy CPU-based approaches and up to 19x speed-ups over prior GPU-based approaches whilst learning superior Pareto front hypervolumes. We demonstrate MO-Playground's versatility by implementing a custom BRUCE humanoid robot environment and learning Pareto-optimal locomotion policies across 6 practical objectives in simulation, such as smoothness, efficiency and arm swinging. Our toolbox is available at https://dynamicmobility.github.io/moplayground/.

Neil Janwani, Ellen Novoseller, Vernon J. Lawhern +1

Georgia Institute of Technology, Atlanta, GA USA · DEVCOM Army Research Laboratory, MD, USA

GPU-Accelerated RL Multi-Objective Reinforcement Learning Reinforcement Learning RL for Robotics

Mar 9, 2026

MetaWorld-X: Hierarchical World Modeling via VLM-Orchestrated Experts for Humanoid Loco-Manipulation

Learning natural, stable, and compositionally generalizable whole-body control policies for humanoid robots performing simultaneous locomotion and manipulation (loco-manipulation) remains a fundamental challenge in robotics. Existing reinforcement learning approaches typically rely on a single monolithic policy to acquire multiple skills, which often leads to cross-skill gradient interference and motion pattern conflicts in high-degree-of-freedom systems. As a result, generated behaviors frequently exhibit unnatural movements, limited stability, and poor generalization to complex task compositions. To address these limitations, we propose MetaWorld-X, a hierarchical world model framework for humanoid control. Guided by a divide-and-conquer principle, our method decomposes complex control problems into a set of specialized expert policies (Specialized Expert Policies, SEP). Each expert is trained under human motion priors through imitation-constrained reinforcement learning, introducing biomechanically consistent inductive biases that ensure natural and physically plausible motion generation. Building upon this foundation, we further develop an Intelligent Routing Mechanism (IRM) supervised by a Vision-Language Model (VLM), enabling semantic-driven expert composition. The VLM-guided router dynamically integrates expert policies according to high-level task semantics, facilitating compositional generalization and adaptive execution in multi-stage loco-manipulation tasks.

Yutong Shen, Hangxu Liu, Penghui Liu +6

School of Information Science and Technology, Beijing University of Technology, Beijing, China · School of Information Science and Engineering, Fudan University, Shanghai, China · University of Alberta, Canada +1

Humanoid Loco-Manipulation Humanoid Robot Control RL for Robotics Mixture of Experts

Mar 8, 2026

Exoskeleton Control through Learning to Reduce Biological Joint Moments in Simulations

Data-driven joint-moment predictors offer a scalable alternative to laboratory-based inverse-dynamics pipelines for biomechanics estimation and exoskeleton control. Meanwhile, physics-based reinforcement learning (RL) enables simulation-trained controllers to learn dynamics-aware assistance strategies without extensive human experimentation. However, quantitative verification of simulation-trained exoskeleton torque predictors, and their impact on human joint power injection, remains limited. This paper presents (1) an RL framework to learn exoskeleton assistance policies that reduce biological joint moments, and (2) a validation pipeline that verifies the trained control networks using an open-source gait dataset through inference and comparison with biological joint moments. Simulation-trained multilayer perceptron (MLP) controllers are developed for level-ground and ramp walking, mapping short-horizon histories of bilateral hip and knee kinematics to normalized assistance torques. Results show that predicted assistance preserves task-intensity trends across speeds and inclines. Agreement is particularly strong at the hip, with cross-correlation coefficients reaching 0.94 at 1.8 m/s and 0.98 during 5° decline walking, demonstrating near-matched temporal structure. Discrepancies increase at higher speeds and steeper inclines, especially at the knee, and are more pronounced in joint power comparisons. Delay tuning biases assistance toward greater positive power injection; modest timing shifts increase positive power and improve agreement in specific gait intervals. Together, these results establish a quantitative validation framework for simulation-trained exoskeleton controllers, demonstrate strong sim-to-data consistency at the torque level, and highlight both the promise and the remaining challenges for sim-to-real transfer.

Zihang You, Xianlian Zhou

Department of Biomedical Engineering, New Jersey Institute of Technology, Newark, NJ 07102, USA

Sim-to-Real Transfer RL for Robotics

Mar 6, 2026

HybridMimic: Hybrid RL-Centroidal Control for Humanoid Motion Mimicking

Motion mimicking, i.e., encouraging the control policy to mimic human motion, facilitates the learning of complex tasks via reinforcement learning (RL) for humanoid robots. Although standard RL frameworks demonstrate impressive locomotion agility, they often bypass explicit reasoning about robot dynamics during deployment, which is a design choice that can lead to physically infeasible commands when the robot encounters out-of-distribution environments. By integrating model-based principles, hybrid approaches can improve performance; however, existing methods typically rely on predefined contact timing, limiting their versatility. This paper introduces HybridMimic, a framework in which a learned policy dynamically modulates a centroidal-model-based controller by predicting continuous contact states and desired centroidal velocities. This architecture exploits the physical grounding of centroidal dynamics to generate feedforward torques that remain feasible even under domain shift. Using physics-informed rewards, the policy is trained to efficiently utilize the centroidal controller's optimization by outputting precise control targets and reference torques. Through hardware experiments on the Booster T1 humanoid, HybridMimic reduces the average base position tracking error by 13% compared to a state-of-the-art RL baseline, demonstrating the robustness of dynamics-aware deployment.

Ludwig Chee-Ying Tay, I-Chia Chang, Yan Gu

Department of Computer Science, Purdue University, West Lafayette, IN 47907, USA. · School of Mechanical Engineering, Purdue University, West Lafayette, IN 47907, USA.

Human-to-Robot Motion Imitation Robotic Control Humanoid Robot Control RL for Robotics Robot Imitation Learning

Mar 6, 2026

Dual-Agent Multiple-Model Reinforcement Learning for Event-Triggered Human-Robot Co-Adaptation in Decoupled Task Spaces

This paper presents a shared-control rehabilitation policy for a custom 6-degree-of-freedom (6-DoF) upper-limb robot that decomposes complex reaching tasks into decoupled spatial axes. The patient governs the primary reaching direction using binary commands, while the robot autonomously manages orthogonal corrective motions. Because traditional fixed-frequency control often induces trajectory oscillations due to variable inverse-kinematics execution times, an event-driven progression strategy is proposed. This architecture triggers subsequent control actions only when the end-effector enters an admission sphere centred on the immediate target waypoint, and was validated in a semi-virtual setup linking a physical pressure sensor to a MuJoCo simulation. To optimise human--robot co-adaptation safely and efficiently, this study introduces Dual Agent Multiple Model Reinforcement Learning (DAMMRL). This framework discretises decision characteristics: the human agent selects the admission sphere radius to reflect their inherent speed--accuracy trade-off, while the robot agent dynamically adjusts its 3D Cartesian step magnitudes to complement the user's cognitive state. Trained in simulation and deployed across mixed environments, this event-triggered DAMMRL approach effectively suppresses waypoint chatter, balances spatial precision with temporal efficiency, and significantly improves success rates in object acquisition tasks.

Yaqi Li, Zhengqi Han, Huifang Liu +1

Faculty of Engineering and IT, University of Technology Sydney, Australia · College of Medical information and Artificial intelligence, Shandong First Medical University (Shandong Academy of Medical Science) 6699 Qingdao Road, Jinan, Shandong, China

Human-Robot Interaction RL for Robotics Shared Autonomy Assistive Robotics Event-Triggered Control

Mar 6, 2026

Swooper: Learning High-Speed Aerial Grasping With a Simple Gripper

High-speed aerial grasping presents significant challenges due to the high demands on precise, responsive flight control and coordinated gripper manipulation. In this work, we propose Swooper, a deep reinforcement learning (DRL) based approach that achieves both precise flight control and active gripper control using a single lightweight neural network policy. Training such a policy directly via DRL is nontrivial due to the complexity of coordinating flight and grasping. To address this, we adopt a two-stage learning strategy: we first pre-train a flight control policy, and then fine-tune it to acquire grasping skills. With the carefully designed reward functions and training framework, the entire training process completes in under 60 minutes on a standard desktop with an Nvidia RTX 3060 GPU. To validate the trained policy in the real world, we develop a lightweight quadrotor grasping platform equipped with a simple off-the-shelf gripper, and deploy the policy in a zero-shot manner on the onboard Raspberry Pi 4B computer, where each inference takes only about 1.0 ms. In 25 real-world trials, our policy achieves an 84% grasp success rate and grasping speeds of up to 1.5 m/s without any fine-tuning. This matches the robustness and agility of state-of-the-art classical systems with sophisticated grippers, highlighting the capability of DRL for learning a robust control policy that seamlessly integrates high-speed flight and grasping. The supplementary video is available for more results. Video: https://zikenhuang.github.io/Swooper/.

Ziken Huang, Xinze Niu, Bowen Chai +2

School of Automation and Intelligent Sensing, Shanghai Jiao Tong University, Shanghai, 200240, China

Aerial Robotics Aerial Manipulation Robotic Grasping RL for Robotics Quadrotor Control

Mar 5, 2026

PRISM: Personalized Refinement of Imitation Skills for Manipulation via Human Instructions

This paper presents PRISM: an instruction-conditioned refinement method for imitation policies in robotic manipulation. This approach bridges Imitation Learning (IL) and Reinforcement Learning (RL) frameworks into a seamless pipeline, such that an imitation policy on a broad generic task, generated from a set of user-guided demonstrations, can be refined through reinforcement to generate new unseen fine-grain behaviours. The refinement process follows the Eureka paradigm, where reward functions for RL are iteratively generated from an initial natural-language task description. Presented approach, builds on top of this mechanism to adapt a refined IL policy of a generic task to new goal configurations and the introduction of constraints by adding also human feedback correction on intermediate rollouts, enabling policy reusability and therefore data efficiency. Results for a pick-and-place task in a simulated scenario show that proposed method outperforms policies without human feedback, improving robustness on deployment and reducing computational burden.

Arnau Boix-Granell, Alberto San-Miguel-Tello, Magí Dalmau-Moreno +1

Eurecat, Centre Tecnològic Catalunya, Robotics & Automation Unit, Barcelona, Spain

Robot Imitation Learning Interactive Imitation Learning RL for Robotics Language-Conditioned Robot Manipulation

Mar 5, 2026

Latent Policy Steering through One-Step Flow Policies

Offline reinforcement learning (RL) allows robots to learn from offline datasets without risky exploration. Yet, offline RL's performance often hinges on a brittle trade-off between (1) return maximization, which can push policies outside the dataset support, and (2) behavioral constraints, which typically require sensitive hyperparameter tuning. Latent steering offers a structural way to stay within the dataset support during RL, but existing offline adaptations commonly approximate action values using latent-space critics learned via indirect distillation, which can lose information and hinder convergence. We propose Latent Policy Steering (LPS), which enables high-fidelity latent policy improvement by backpropagating original-action-space Q-gradients through a differentiable one-step MeanFlow policy to update a latent-action-space actor. By eliminating proxy latent critics, LPS allows an original-action-space critic to guide end-to-end latent-space optimization, while the one-step MeanFlow policy serves as a behavior-constrained generative prior. This decoupling yields a robust method that works out-of-the-box with minimal tuning. Across OGBench and real-world robotic tasks, LPS achieves state-of-the-art performance and consistently outperforms behavioral cloning and strong latent steering baselines.

Hokyun Im, Andrey Kolobov, Jianlong Fu +1

Department of Artificial Intelligence, Yonsei University · Microsoft Research

RL for Robotics Latent Action Learning Offline RL

Mar 5, 2026

Decoupling Task and Behavior: A Two-Stage Reward Curriculum in Reinforcement Learning for Robotics

Deep Reinforcement Learning is a promising tool for robotic control, yet practical application is often hindered by the difficulty of designing effective reward functions. Real-world tasks typically require optimizing multiple objectives simultaneously, necessitating precise tuning of their weights to learn a policy with the desired characteristics. To address this, we propose a two-stage reward curriculum where we decouple task-specific objectives from behavioral terms. In our method, we first train the agent on a simplified task-only reward function to ensure effective exploration before introducing the full reward that includes auxiliary behavior-related terms such as energy efficiency. Further, we analyze various transition strategies and demonstrate that reusing samples between phases is critical for training stability. We validate our approach on the DeepMind Control Suite, ManiSkill3, and a mobile robot environment, modified to include auxiliary behavioral objectives. Our method proves to be simple yet effective, substantially outperforming baselines trained directly on the full reward while exhibiting higher robustness to specific reward weightings.

Kilian Freitag, Knut Åkesson, Morteza Haghir Chehreghani

Chalmers University of Technology · Gothenburg University

RL for Robotics Reinforcement Learning Reward Design for RL Curriculum RL Multi-Objective Reinforcement Learning

Mar 4, 2026

Tendon Force Modeling for Sim2Real Transfer of Reinforcement Learning Policies for Tendon-Driven Robots

Robots which make use of soft or compliant inter- actions often leverage tendon-driven actuation which enables actuators to be placed more flexibly, and compliance to be maintained. However, controlling complex tendon systems is challenging. Simulation paired with reinforcement learning (RL) could be enable more complex behaviors to be generated. Such methods rely on torque and force-based simulation roll- outs which are limited by the sim-to-real gap, stemming from the actuator and system dynamics, resulting in poor transfer of RL policies onto real robots. To address this, we propose a method to model the tendon forces produced by typical servo motors, focusing specifically on the transfer of RL policies for a tendon driven finger. Our approach extends existing data- driven techniques by leveraging contextual history and a novel data collection test-bench. This test-bench allows us to capture tendon forces undergo contact-rich interactions typical of real- world manipulation. We then utilize our force estimation model in a GPU-accelerated tendon force-driven rigid body simulation to train RL-based controllers. Our transformer-based model is capable of predicting tendon forces within 3% of the maximum motor force and is robot-agnostic. By integrating our learned model into simulation, we reduce the sim-to-real gap for test trajectories by 41%. RL-based controller trained with our model achieves a 50% improvement in fingertip pose tracking tasks on real tendon-driven robotic fingers. This approach is generalizable to different actuators and robot systems, and can enable RL policies to be used widely across tendon systems, advancing capabilities of dexterous manipulators and soft robots.

Valentin Yuryev, Josie Hughes

These authors are with with CREATE Lab, Swiss Federal Institute of Technology Lausanne (EPFL), Lausanne, Switzerland.

Sim-to-Real Transfer Tendon-Driven Actuation Robot Dynamics Learning RL for Robotics

Mar 4, 2026

UrbanHuRo: A Two-Layer Human-Robot Collaboration Framework for the Joint Optimization of Heterogeneous Urban Services

In the vision of smart cities, technologies are being developed to enhance the efficiency of urban services and improve residents' quality of life. However, most existing research focuses on optimizing individual services in isolation, without adequately considering reciprocal interactions among heterogeneous urban services that could yield higher efficiency and improved resource utilization. For example, human couriers could collect traffic and air quality data along their delivery routes, while sensing robots could assist with on-demand delivery during peak hours, enhancing both sensing coverage and delivery efficiency. However, the joint optimization of different urban services is challenging due to potentially conflicting objectives and the need for real-time coordination in dynamic environments. In this paper, we propose UrbanHuRo, a two-layer human-robot collaboration framework for joint optimization of heterogeneous urban services, demonstrated through crowdsourced delivery and urban sensing. UrbanHuRo includes two key designs: (i) a scalable distributed MapReduce-based K-submodular maximization module for efficient order dispatch, and (ii) a deep submodular reward reinforcement learning algorithm for sensing route planning. Experimental evaluations on real-world datasets from a food delivery platform demonstrate that UrbanHuRo improves sensing coverage by 29.7% and courier income by 39.2% on average in most settings, while also significantly reducing the number of overdue orders.

Tonmoy Dey, Lin Jiang, Zheng Dong +1

Department of Computer Science, Florida State University. · Department of Computer Science, Wayne State University.

Human-Robot Collaboration RL for Robotics

Mar 3, 2026

ULTRA: Unified Multimodal Control for Autonomous Humanoid Whole-Body Loco-Manipulation

Achieving autonomous and versatile whole-body loco-manipulation remains a central barrier to making humanoids practically useful. Yet existing approaches are fundamentally constrained: retargeted data are often scarce or low-quality; methods struggle to scale to large skill repertoires; and, most importantly, they rely on tracking predefined motion references rather than generating behavior from perception and high-level task specifications. To address these limitations, we propose ULTRA, a unified framework with two key components. First, we introduce a physics-driven neural retargeting algorithm that translates large-scale motion capture to humanoid embodiments while preserving physical plausibility for contact-rich interactions. Second, we learn a unified multimodal controller that supports both dense references and sparse task specifications, under sensing ranging from accurate motion-capture state to noisy egocentric visual inputs. We distill a universal tracking policy into this controller, compress motor skills into a compact latent space, and apply reinforcement learning finetuning to expand coverage and improve robustness under out-of-distribution scenarios. This enables coordinated whole-body behavior from sparse intent without test-time reference motions. We evaluate ULTRA in simulation and on a real Unitree G1 humanoid. Results show that ULTRA generalizes to autonomous, goal-conditioned whole-body loco-manipulation from egocentric perception, consistently outperforming tracking-only baselines with limited skills.

Xialin He, Sirui Xu, Xinyao Li +4

University of Illinois Urbana-Champaign

Robotic Control Humanoid Loco-Manipulation Humanoid Robot Control Motion Retargeting Robot Imitation Learning RL for Robotics

Mar 3, 2026

Rhythm: Learning Interactive Whole-Body Control for Dual Humanoids

Realizing interactive whole-body control for multi-humanoid systems is critical for unlocking complex collaborative capabilities in shared environments. Although recent advancements have significantly enhanced the agility of individual robots, bridging the gap to physically coupled multi-humanoid interaction remains challenging, primarily due to severe kinematic mismatches and complex contact dynamics. To address this, we introduce Rhythm, the first unified framework enabling real-world deployment of dual-humanoid systems for complex, physically plausible interactions. Our framework integrates three core components: (1) an Interaction-Aware Motion Retargeting (IAMR) module that generates feasible humanoid interaction references from human data; (2) an Interaction-Guided Reinforcement Learning (IGRL) policy that masters coupled dynamics via graph-based rewards; and (3) a real-world deployment system that enables robust transfer of dual-humanoid interaction. Extensive experiments on physical Unitree G1 robots demonstrate that our framework achieves robust interactive whole-body control, successfully transferring diverse behaviors such as hugging and dancing from simulation to reality.

Hongjin Chen, Wei Zhang, Pengfei Li +10

Fudan University · TARS Robotics · Tsinghua University +3

RL for Robotics Humanoid Robot Control Whole-Body Control Motion Retargeting

Feb 28, 2026

HydroShear: Hydroelastic Shear Simulation for Tactile Sim-to-Real Reinforcement Learning

In this paper, we address the problem of tactile sim-to-real policy transfer for contact-rich tasks. Existing methods primarily focus on vision-based sensors and emphasize image rendering quality while providing overly simplistic models of force and shear. Consequently, these models exhibit a large sim-to-real gap for many dexterous tasks. Here, we present HydroShear, a non-holonomic hydroelastic tactile simulator that advances the state-of-the-art by modeling: a) stick-slip transitions, b) path-dependent force and shear build up, and c) full SE(3) object-sensor interactions. HydroShear extends hydroelastic contact models using Signed Distance Functions (SDFs) to track the displacements of the on-surface points of an indenter during physical interaction with the sensor membrane. Our approach generates physics-based, computationally efficient force fields from arbitrary watertight geometries while remaining agnostic to the underlying physics engine. In experiments with GelSight Minis, HydroShear more faithfully reproduces real tactile shear compared to existing methods. This fidelity enables zero-shot sim-to-real transfer of reinforcement learning policies across four tasks: peg insertion, bin packing, book shelving for insertion, and drawer pulling for fine gripper control under slip. Our method achieves a 93% average success rate, outperforming policies trained on tactile images (34%) and alternative shear simulation methods (58%-61%).

An Dang, Jayjun Lee, Mustafa Mukadam +4

Robotics Department, University of Michigan · Amazon Industrial Robotics (AIR)

Sim-to-Real Transfer Robotic Tactile Sensing Contact-Rich Robotic Manipulation RL for Robotics

Feb 25, 2026

LessMimic: Long-Horizon Humanoid Interaction with Unified Distance Field Representations

Humanoid robots that autonomously interact with physical environments over extended horizons represent a central goal of embodied intelligence. Existing approaches rely on reference motions or task-specific rewards, tightly coupling policies to particular object geometries and precluding multi-skill generalization within a single framework. A unified interaction representation enabling reference-free inference, geometric generalization, and long-horizon skill composition within one policy remains an open challenge. Here we show that Distance Field (DF) provides such a representation: LessMimic conditions a single whole-body policy on DF-derived geometric cues--surface distances, gradients, and velocity decompositions--removing the need for motion references, with interaction latents encoded via a Variational Auto-Encoder (VAE) and post-trained using Adversarial Interaction Priors (AIP) under Reinforcement Learning (RL). Through DAgger-style distillation that aligns DF latents with egocentric depth features, LessMimic further transfers seamlessly to vision-only deployment without motion capture (MoCap) infrastructure. A single LessMimic policy achieves 80--100% success across object scales from 0.4x to 1.6x on PickUp and SitStand where baselines degrade sharply, attains 62.1% success on 5 task instances trajectories, and remains viable up to 40 sequentially composed tasks. By grounding interaction in local geometry rather than demonstrations, LessMimic offers a scalable path toward humanoid robots that generalize, compose skills, and recover from failures in unstructured environments.

Yutang Lin, Jieming Cui, Yixuan Li +3

Humanoid Robot Control Long-Horizon Robotic Manipulation Robot Policy Generalization RL for Robotics Robot Imitation Learning

Feb 24, 2026

Task-oriented grasping for dexterous robots using postural synergies and reinforcement learning

In this paper, we address the problem of task-oriented grasping for humanoid robots, emphasizing the need to align with human social norms and task-specific objectives. Existing methods, employ a variety of open-loop and closed-loop approaches but lack an end-to-end solution that can grasp several objects while taking into account the downstream task's constraints. Our proposed approach employs reinforcement learning to enhance task-oriented grasping, prioritizing the post-grasp intention of the agent. We extract human grasp preferences from the ContactPose dataset, and train a hand synergy model based on the Variational Autoencoder (VAE) to imitate the participant's grasping actions. Based on this data, we train an agent able to grasp multiple objects while taking into account distinct post-grasp intentions that are task-specific. By combining data-driven insights from human grasping behavior with learning by exploration provided by reinforcement learning, we can develop humanoid robots capable of context-aware manipulation actions, facilitating collaboration in human-centered environments.

Dimitrios Dimou, José Santos-Victor, Plinio Moreno

Institute for Systems and Robotics Instituto Superior Tecnico Lisboa, Portugal

Robotic Grasping RL for Robotics

Feb 23, 2026

What Matters for Simulation to Online Reinforcement Learning on Real Robots

We investigate what specific design choices enable successful online reinforcement learning (RL) on physical robots. Across 100 real-world training runs on three distinct robotic platforms, we systematically ablate algorithmic, systems, and experimental decisions that are typically left implicit in prior work. We find that some widely used defaults can be harmful, while a set of robust, readily adopted design choices within standard RL practice yield stable learning across tasks and hardware. These results provide the first large-sample empirical study of such design choices, enabling practitioners to deploy online RL with lower engineering effort.

Yarden As, Dhruva Tirumala, René Zurbrügg +4

ETH Zurich · Google DeepMind

RL for Robotics Robotic RL

Feb 23, 2026

Sample-Efficient Learning with Online Expert Correction for Autonomous Catheter Steering in Endovascular Bifurcation Navigation

Robot-assisted endovascular intervention offers a safe and effective solution for remote catheter manipulation, reducing radiation exposure while enabling precise navigation. Reinforcement learning (RL) has recently emerged as a promising approach for autonomous catheter steering; however, conventional methods suffer from sparse reward design and reliance on static vascular models, limiting their sample efficiency and generalization to intraoperative variations. To overcome these challenges, this paper introduces a sample-efficient RL framework with online expert correction for autonomous catheter steering in endovascular bifurcation navigation. The proposed framework integrates three key components: (1) A segmentation-based pose estimation module for accurate real-time state feedback, (2) A fuzzy controller for bifurcation-aware orientation adjustment, and (3) A structured reward generator incorporating expert priors to guide policy learning. By leveraging online expert correction, the framework reduces exploration inefficiency and enhances policy robustness in complex vascular structures. Experimental validation on a robotic platform using a transparent vascular phantom demonstrates that the proposed approach achieves convergence in 123 training episodes -- a 25.9% reduction compared to the baseline Soft Actor-Critic (SAC) algorithm -- while reducing average positional error to 83.8% of the baseline. These results indicate that combining sample-efficient RL with online expert correction enables reliable and accurate catheter steering, particularly in anatomically challenging bifurcation scenarios critical for endovascular navigation.

Hao Wang, Tianliang Yao, Bo Lu +4

Department of Control Science and Engineering, College of Electronics and Information Engineering, and Shanghai Institute of Intelligent Science and Technology, Tongji University, Shanghai 200092, China · Department of Electronic Engineering, Faculty of Engineering, The Chinese University of Hong Kong, Hong Kong SAR 999077, China · Robotics and Microsystems Center, School of Mechanical and Electric Engineering, Soochow University, Suzhou, Jiangsu 215131, China +3

Robotic Endovascular Intervention RL for Robotics Data-Efficient Robot Learning RL from Human Feedback

Feb 20, 2026

Cross-Embodiment Offline Reinforcement Learning for Heterogeneous Robot Datasets

Scalable robot policy pre-training has been hindered by the high cost of collecting high-quality demonstrations for each platform. In this study, we address this issue by uniting offline reinforcement learning (offline RL) with cross-embodiment learning. Offline RL leverages both expert and abundant suboptimal data, and cross-embodiment learning aggregates heterogeneous robot trajectories across diverse morphologies to acquire universal control priors. We perform a systematic analysis of this offline RL and cross-embodiment paradigm, providing a principled understanding of its strengths and limitations. To evaluate this offline RL and cross-embodiment paradigm, we construct a suite of locomotion datasets spanning 16 distinct robot platforms. Our experiments confirm that this combined approach excels at pre-training with datasets rich in suboptimal trajectories, outperforming pure behavior cloning. However, as the proportion of suboptimal data and the number of robot types increase, we observe that conflicting gradients across morphologies begin to impede learning. To mitigate this, we introduce an embodiment-based grouping strategy in which robots are clustered by morphological similarity and the model is updated with a group gradient. This simple, static grouping substantially reduces inter-robot conflicts and outperforms existing conflict-resolution methods.

Haruki Abe, Takayuki Osa, Yusuke Mukuta +1

The University of Tokyo, Tokyo, Japan · RIKEN Center for Advanced Intelligence Project, Tokyo, Japan

RL for Robotics Cross-Embodiment Robot Learning Data-Efficient Robot Learning Robot Locomotion

Feb 19, 2026

Reinforcement-Learning-Based Assistance Reduces Squat Effort with a Modular Hip--Knee Exoskeleton

Squatting is one of the most demanding lower-limb movements, requiring substantial muscular effort and coordination. Reducing the physical demands of this task through intelligent and personalized assistance has significant implications, particularly in industries involving repetitive low-level assembly activities. In this study, we evaluated the effectiveness of a neural network controller for a modular Hip-Knee exoskeleton designed to assist squatting tasks. The neural network controller was trained via reinforcement learning (RL) in a physics-based, human-exoskeleton interaction simulation environment. The controller generated real-time hip and knee assistance torques based on recent joint-angle and velocity histories. Five healthy adults performed three-minute metronome-guided squats under three conditions: (1) no exoskeleton (No-Exo), (2) exoskeleton with Zero-Torque, and (3) exoskeleton with active assistance (Assistance). Physiological effort was assessed using indirect calorimetry and heart rate monitoring, alongside concurrent kinematic data collection. Results show that the RL-based controller adapts to individuals by producing torque profiles tailored to each subject's kinematics and timing. Compared with the Zero-Torque and No-Exo condition, active assistance reduced the net metabolic rate by approximately 10%, with minor reductions observed in heart rate. However, assisted trials also exhibited reduced squat depth, reflected by smaller hip and knee flexion. These preliminary findings suggest that the proposed controller can effectively lower physiological effort during repetitive squatting, motivating further improvements in both hardware design and control strategies.

Neethan Ratnakumar, Mariya Huzaifa Tohfafarosh, Saanya Jauhri +1

Department of Biomedical Engineering, New Jersey Institute of Technology, Newark, NJ, USA.

Upper-Limb Exoskeletons Assistive Robotics Human-Robot Interaction RL for Robotics

Feb 16, 2026

AdaptManip: Learning Adaptive Whole-Body Object Lifting and Delivery with Online Recurrent State Estimation

This paper presents Adaptive Whole-body Loco-Manipulation, AdaptManip, a fully autonomous framework for humanoid robots to perform integrated navigation, object lifting, and delivery. Unlike prior imitation learning-based approaches that rely on human demonstrations and are often brittle to disturbances, AdaptManip aims to train a robust loco-manipulation policy via reinforcement learning without human demonstrations or teleoperation data. The proposed framework consists of three coupled components: (1) a recurrent object state estimator that tracks the manipulated object in real time under limited field-of-view and occlusions; (2) a whole-body base policy for robust locomotion with residual manipulation control for stable object lifting and delivery; and (3) a LiDAR-based robot global position estimator that provides drift-robust localization. All components are trained in simulation using reinforcement learning and deployed on real hardware in a zero-shot manner. Experimental results show that AdaptManip significantly outperforms baseline methods, including imitation learning-based approaches, in adaptability and overall success rate, while the learned estimator keeps tracking the object when visual observations are intermittent. We further demonstrate fully autonomous real-world navigation, object lifting, and delivery on a humanoid robot.

Morgan Byrd, Donghoon Baek, Kartik Garg +5

Georgia Institute of Technology, Atlanta, GA, 30308, USA · Georgia Tech Research Institute, Atlanta, GA, 30308, USA

RL for Robotics Robotic Manipulation Loco-Manipulation

Feb 9, 2026

High-Speed Vision-Based Flight in Clutter with Safety-Shielded Reinforcement Learning

Quadrotor unmanned aerial vehicles (UAVs) are increasingly deployed in complex missions that demand reliable autonomous navigation and robust obstacle avoidance. However, traditional modular pipelines often incur cumulative latency, whereas purely reinforcement learning (RL) approaches typically provide limited formal safety guarantees. To bridge this gap, we propose an end-to-end RL framework augmented with model-based safety mechanisms. We incorporate physical priors in both training and deployment. During training, we design a physics-informed reward structure that provides global navigational guidance. During deployment, we integrate a real-time safety filter that projects the policy outputs onto a provably safe set to enforce strict collision-avoidance constraints. This hybrid architecture reconciles high-speed flight with robust safety assurances. Benchmark evaluations demonstrate that our method outperforms both traditional planners and recent end-to-end obstacle avoidance approaches based on differentiable physics. Extensive experiments demonstrate strong generalization, enabling reliable high-speed navigation in dense clutter and challenging outdoor forest environments at velocities up to 7.5 m/s}.

Jiarui Zhang, Chengyong Lei, Chengjiang Dai +4

Institute of Cyber-Systems and Control, College of Control Science and Engineering, Zhejiang University, Hangzhou 310027, China · Differential Robotics, Hangzhou 311121, China.

RL for Robotics UAV Navigation Safe RL Vision-Based Robot Control Robot Collision Avoidance

Feb 2, 2026

Flow Policy Gradients for Robot Control

Likelihood-based policy gradient methods are the dominant approach for training robot control policies from rewards. These methods rely on differentiable action likelihoods, which constrain policy outputs to simple distributions like Gaussians. In this work, we show how flow matching policy gradients, a recent framework that bypasses likelihood computation, can be made effective for training and finetuning more expressive policies in challenging robot control settings. We introduce an improved objective that enables success in legged locomotion, humanoid motion tracking, and manipulation tasks, as well as robust sim-to-real transfer on two humanoid robots. We then present ablations and analysis on training dynamics. Results show how policies can exploit the flow representation for exploration when training from scratch, as well as improved finetuning robustness over baselines.

Brent Yi, Hongsuk Choi, Himanshu Gaurav Singh +9

Amazon FAR · UC Berkeley · Stanford +2

Robotic Control RL for Robotics Flow-Based Policy Learning RL for Legged Locomotion