Robotic Control

Follow this topic Ask about this topic

Momentum

78 papers in the last four weeks, up 388% on the four weeks before. 0.8% of all new papers.

Jul 13Week of Sep 28

Latest papers 367

Jun 11, 2026cs.RO

Learning Dynamic Swing-Up of an Inverted Pendulum using Remote Magnetic Actuation

Electromagnetic Navigation Systems (eMNS) have gained considerable attention for minimally invasive surgery and targeted drug delivery. While most of the literature relies on quasi-static control of these systems, recent work has demonstrated the benefits of dynamic approaches. However, trajectory tracking far from equilibrium states remains largely unaddressed. We close this gap by demonstrating the first swing-up of a magnetically actuated inverted pendulum using the clinically-ready Navion eMNS. Although the inverted pendulum is not clinically relevant in itself, the proposed method utilizes torques and forces as control objectives, making it applicable to other magnetically actuated devices such as catheters and guidewires. Our approach combines trajectory optimization that accounts for internal eMNS dynamics with time-varying Linear Quadratic Regulator (LQR) state feedback and Iterative Learning Control (ILC), which leverages previous trial data and the system's dynamic model to progressively refine the feedforward command. While LQR alone fails due to the complex phenomena of magnetic actuation, ILC enables successful swing-up within six iterations. Furthermore, post-experimental analysis reveals that the learned ILC correction closely matches the torque discrepancy predicted by high-fidelity magnetic field model calibration, suggesting learning and adaptation as a promising tool to deal with uncertainties in electromagnetic actuation arising, e.g., from patient-specific physiological motion patterns and field model calibration inaccuracies.

Viacheslav Sydora, Jasan Zughaibi, Denis von Arx +2

Learning and Dynamical Systems, Max Planck Institute for Intelligent Systems, Tübingen, Germany · Multi-Scale Robotics Lab, Institute of Robotics and Intelligent Systems, D-MAVT, ETH Zürich, Zürich, Switzerland · Medical Robotics Lab, Institute of Robotics and Intelligent Systems, D-MAVT, ETH Zürich, Zürich, Switzerland

Robotic Control Adaptive Control

Jun 11, 2026cs.RO

PhysVLA: Towards Physically-Grounded VLA for Embodied Robotic Manipulation

Vision-Language-Action (VLA) models excel at mapping visual inputs and natural language instructions directly to robotic control policies. However, because they are trained primarily to fit behavioural demonstration data, they do not explicitly enforce fundamental physical principles such as rigid-body dynamics or contact constraints. This exposes a critical physics gap: standard temporal smoothing applied on top of single-step or chunked VLAs trades trajectory quality for added failures that short-term memory cannot resolve. To bridge this gap, we introduce PhysVLA (Physics-VLA), a plug-and-play, inference-time framework designed to wrap any frozen VLA backbone without retraining, fine-tuning, or weight access, with less than 1 ms of overhead per control step. PhysVLA intercepts the predicted control action, captures only the simulator or system state, and applies a dual-layered correction: (i) a phase-aware finite-state machine that structures discrete task segments (approach, grasp, transport, and place), and (ii) a selective Euler-Lagrange gate that activates only when a dynamics oracle detects kinodynamic inconsistency. Evaluated across OpenVLA, OpenVLA-OFT, Force-VLA, and Generalist-VLA on LIBERO-Spatial with a 7-DoF Franka Panda, the framework delivers absolute success rate increases of up to 17% and stability increases of up to 19% with no per-task regressions, improves trajectory efficiency by up to 15% across all four backbones, and shows up to a 10x improvement in trajectory jerk robustness on a Robosuite Lift cross-simulator sweep. We further validate the framework on a real Agilex Piper arm with a pick-and-place task, confirming that PhysVLA transfers to physical hardware without retraining, with success-rate improvements of up to 50%, establishing physical awareness as a composable, backbone-agnostic runtime module.

Namai Chandra, Shriram Damodaran, Lin Wang

Electronic Systems IIT Madras, India · EmPACT Lab Nanyang Technological University Singapore

Robotic Control Robotic Manipulation Vision-Language-Action Models

Jun 11, 2026eess.SY

Interaction Dynamics MPC for Knee Rehabilitation Exoskeletons: A Closed-Loop SEA Outer-Loop Study

Safe rehabilitation is an interaction-dynamics problem: the controller must regulate a prescribed motion while absorbing involuntary spasm, voluntary effort, actuator compliance, and model mismatch as disturbances. This paper instantiates the predictive interaction-dynamics framework of the base pHRI formulation on a SEA knee joint. SEA feedforward reduces the gravity-compensated knee to the same scalar double integrator as the base framework, while a dynamic-residual measurement from spring deflection supplies an interaction-disturbance observation. A steady-state target converts the estimated disturbance into a cancelling input, and a finite-horizon quadratic program regulates deviations from that target under range-of-motion, torque, and velocity constraints. The evaluation matches stiffness and damping across controllers so gains cannot be attributed to higher impedance. Under a motion-opposing

15\unit{Nm}

step, classical impedance and MPC without estimation produce about

500\unit{mrad}

steady-state error, whereas Kalman-augmented interaction MPC reduces this to

1.17\unit{mrad}

at 100~~Hz and $0.70\unit{mrad}$ at 500~~Hz; the 500Hz peak is $7.27\unit{mrad}$ . In 30 randomized trials, the 95th-percentile peak is $21.57\unit{mrad}$ . Bounded Assist-as-Needed scheduling, a corrective-channel energy tank, constrained OSQP stress cases, direct MuJoCo execution, and a posture-clamped MyoSuite knee slice are implemented. The framework holds on a single-mass, closed-inner-loop SEA approximation; an explicit two-mass plant with a finite-bandwidth, pole-placed inner torque loop (SectionVIII) confirms this for nominal tracking but shows delivered torque can overshoot the commanded bound by 21.7% near saturation. Scope excludes clinical intent recognition, full-system passivity, safety certification, hardware trials, and multi-joint validation.

Yongyan Cao, Jinshan Tang

Voryx Robotics · George Mason University, Dept. of Health Administration and Policy

Robotic Control Model Predictive Control Assistive Robotics Continuous Control Human-Robot Interaction

Jun 11, 2026cs.RO

Real-Time Execution with Autoregressive Policies

Real-time execution, enabled by asynchronous inference that ensures both smooth action trajectories and fast reactivity, is critical for realistic deployments of large-scale Vision-Language-Action models. However, recent work on real-time execution primarily focuses on variants of diffusion policies, even though it is more critical for autoregressive policies given their slower rollout speed in synchronous inference. In contrast, we demonstrate that autoregressive policies can achieve real-time execution by adjusting the tokenization horizon and applying constrained decoding, thereby guaranteeing strict latency bounds that enable multi-trajectory decoding to maximize performance. Across simulated and real-world environments, we find that the autoregressive policy consistently outperforms its equivalent-level flow-matching policy counterpart while achieving significantly improved task completion speeds from synchronous inference. Coupled with the inherent advantages of autoregressive policies, such as faster convergence and better generalizability in instruction-following, these results confirm that autoregressive policies can remain a competitive policy type supporting real-time execution.

Sangkyu Lee, Seohyeon Park, Tackgeun You +4

Korea Institute of Science and Technology · Seoul National University · Google Research

Robotic Control Vision-Language-Action Models Autoregressive Decoding

Jun 11, 2026cs.RO

EMG-Based Adaptation of Anisotropic Virtual Fixtures for Robot-Assisted Surgical Resection and Dissection

In this paper, we address the development of an adaptive assistance system for robot-assisted laparoscopic surgery, specifically for delicate tasks such as Resection and Dissection. Even if Virtual Fixtures offer significant advantages for guiding a surgeon's movements, conventional Virtual Fixtures are often defined by fixed geometries, lacking the flexibility to adapt to the surgical workflow or the surgeon's immediate intent. To address these limitations, we propose a novel framework for an adaptive and anisotropic virtual fixture. In addition, we introduce an intuitive control interface that modulates the fixture's geometry in real-time based on the surgeon's intent, inferred from EMG signals. This approach allows the surgeon to dynamically expand or disengage the constraint by contracting their forearm muscles, enabling seamless transitions between precise guided motion and free repositioning of the tool. Experimental results from a pilot user study, based on a standardized surgical training task, demonstrate the effectiveness of the proposed method. The system showed significant improvements in task accuracy and movement consistency, alongside a reduction in perceived cognitive load, effort, and frustration.

Dario Onfiani, Michael Dyck, Luigi Biagiotti +1

Department of Engineering ”Enzo Ferrari”, University of Modena and Reggio Emilia, Italy · Institute of Robotics and Mechatronics, German Aerospace Center (DLR), Munich, Germany

Robotic Control Human-Robot Interaction Surgical Robotics

Jun 10, 2026cs.RO

EWAM: An Enhanced World Action Model for Closed-Loop Online Adaptation in Embodied Intelligence

In this paper, we propose the Enhanced World Action Model (EWAM), a closed-loop online adaptation architecture built upon a pretrained and fully frozen Cosmos3 backbone network. Evaluated entirely under a zero-shot task protocol, EWAM is centrally focused on reducing the amount of additional deployment data required to adapt to new task layouts. Notably, no extra task-specific demonstration sets were introduced in any of the evaluations, and no fine-tuning was performed on the backbone network. Its performance gains stem entirely from an inference-time co-reasoning mechanism composed of four inserted lightweight neural layers: the Neural Experience Memory Layer located in the intermediate layers of the Diffusion Transformer (DiT) provides task-relevant execution context; the Neural Anomaly Detection Layer after the state prediction head monitors the divergence between predicted and actual states in real time; the Neural Policy Routing Layer dynamically selects direct execution, conservative replanning, or rollback recovery based on the anomaly severity; and the Neural Action Correction Layer refines the generated action chunks using execution diagnostics. Unlike naive feature fusion, the memory, anomaly detection, and correction modules are deeply integrated into the Cosmos3 forward path in a differentiable manner, with only the final routing decision being a discrete supervised one.

Xin Zhou, Cong Miao

Astronex Robotics · Nanjing University of Information Science and Technology

Robotic Control Adaptive Inference Test-Time Adaptation Memory-Augmented Agents Embodied World Models

Jun 10, 2026cs.RO

LUCID: Learning Embodiment-Agnostic Intent Models from Unstructured Human Videos for Scalable Dexterous Robot Skill Acquisition

The most widely-adopted robot learning pipelines today learn skills from robot demonstrations or structured human data, which are expensive to collect and tied to specific embodiments. In contrast, unstructured human videos provide a scalable alternative. They contain diverse manipulation demonstrations across objects, scenes, and strategies, but are not directly connected to robot action. We propose LUCID, a two-stage framework that learns task intent from unstructured human videos drawn from internet-scale datasets and learns robot control in massively-parallel simulation. The intent model predicts short-horizon intent (what should happen next in the scene) from the current observation in closed loop. An embodiment-specific sensorimotor policy converts this intent into robot actions. The intent interface is shared across controllers, so the same intent model can be applied to different embodiments, from our primary dexterous hand to a parallel-jaw gripper. We evaluate LUCID on five real-world manipulation tasks: stirring, wiping, and binning supervised by only internet video, with zero-shot transfer to novel scenes and object instances; and push-T and cable routing supervised by 1 hr each of self-collected smartphone video. Project page: https://lucid-robot.github.io/.

Harsh Gupta, Guanya Shi, Wenzhen Yuan

University of Illinois Urbana-Champaign · Carnegie Mellon University

Robotic Control Robot Skill Learning Generalization in Robotic Manipulation Cross-Embodiment Robot Learning Robotic Manipulation Robot Imitation Learning

Jun 9, 2026cs.RO

Learning Control as Enabling Layer for Embodied Intelligence Research explored with Soft Robotic Swimming in diverse Flow Speeds

Soft robots are valuable robophysical platforms for studying body-caudal undulatory locomotion, but their compliant bodies are difficult to control precisely under changing hydrodynamic loading. Conventional proportional-integral-derivative (PID) feedback stabilizes periodic undulation in static water, but can accumulate flow-dependent tracking delay and increasing inter-trial variability when environmental flow becomes non-trivial. Here, we evaluate whether augmenting PID control with a Linear Repetitive Learning Estimation Scheme (PID-LRLES) recovers tracking accuracy and repeatability under dynamic flow. The LRLES generalizes classical integral action from constant to periodic, non-constant references, while using a stable transfer-function realization whose poles have negative real parts to avoid the long-term instability issues of classical repetitive control. Closed-loop experiments were carried out in a recirculating flow tank at five bulk flow speeds spanning 0 to 32.6 cm s^-1, using an embedded soft capacitive bending sensor at a 1 kHz control-loop rate. With controller gains tuned once in static water and then held fixed across all conditions, PID-LRLES tracked the periodic bending-envelope reference more closely than the PID baseline and significantly reduced the inter-trial spread of the per-trial RMSE (paired Wilcoxon signed-rank test, p = 1.8 x 10^-4, n = 25). Embedded soft proprioception and cycle-to-cycle learning act as complementary contributors to robustness: the sensor exposes the periodic hydrodynamic bias in body deformation, while the learning term absorbs it over recent oscillation cycles. By reducing flow-dependent control-induced variability, the approach provides an enabling layer for future robophysical studies seeking to isolate the effects of morphology, sensing, and environmental flow on aquatic locomotion.

Fabian Schwab, Federico Allione, Bingcheng Wang +5

1Engineering Sciences Department, Swiss Federal Laboratories for Materials Science and Technology, Switzerland · Institute for NeuroInformatics, University of Zurich, Switzerland · Department of Electronic Engineering, University of Rome Tor Vergata, Italy

Robotic Control Robust Control Robot Locomotion Soft Robotics

Jun 9, 2026cs.RO

IMPACT: Learning Internal-Model Predictive Control for Forceful Robotic Manipulation

Real-world robotic manipulation tasks often involve forceful interactions with the environment, such as using tools of varying weights, transporting objects with different masses, and performing contact-rich tasks like table wiping. Previous learning-based approaches typically employ imitation learning policies that output target end-effector poses tracked by low-level impedance controllers. In these systems, forceful interactions are either implicitly realized through steady-state tracking errors or explicitly commanded using wrist force/torque or tactile sensors. However, implicit approaches generalize poorly across object weights, while explicit approaches require specialized hardware and increase system complexity. In this work, we propose IMPACT, a framework that decouples these forceful tasks into task-planning and internal-model-based predictive control. Extensive simulation and real-world experiments demonstrate that the proposed framework achieves higher success rates and improved generalization to unseen object weights, as well as better safety and energy efficiency.

Jiawei Gao, Chaoqi Liu, Peilin Wu +2

Harvard University · Stanford University

Robotic Control Contact-Rich Robotic Manipulation Generalization in Robotic Manipulation Model Predictive Control Robotic Manipulation Force-Controlled Robotic Manipulation

Jun 9, 2026cs.RO

Uncovering Vulnerability of Vision-Language-Action Models under Joint-Level Physical Faults

Deploying Vision-Language-Action (VLA) models in real robotic systems requires robustness not only to semantic and perceptual variations, but also to embodiment-side faults that change how actions are physically realized. Real robots can experience joint-level changes caused by actuator degradation, hardware faults, safety limits, collision damage, or wear-induced friction. These faults are critical because they alter the action-to-motion interface of a policy, disrupting the learned closed-loop relationship between commanded actions, realized motion, and subsequent observations. In this work, we study realistic joint-level physical faults and show that VLA models are vulnerable when predicted actions are executed through a perturbed robot body. Our analysis reveals joint-dependent effects, with heterogeneous degradation in task success across affected joints. We also show that performance drops cannot be attributed solely to physical infeasibility, since feasible faults such as increased joint friction can still substantially reduce success rates and induce closed-loop execution mismatch. Motivated by these findings, we propose Joint-level Physical-fault Aware Residual Calibrator (J-PARC), a lightweight residual calibration framework built on top of a frozen VLA policy. J-PARC infers a latent joint-fault regime from recent joint dynamics and conditions a shared residual calibrator on this regime, enabling adaptive action correction across faulty joints. Experiments show that J-PARC improves robustness under joint-level faults while preserving fault-free environment performance.

Minsoo Jo, Taeju Kwon, Junha Chun +2

Graduate School of Data Science, Seoul National University

Robotic Control Vision-Language-Action Models Residual Policy Learning Robustness of VLA Models

Jun 9, 2026cs.RO

Hierarchical Policies from Verbal and Egocentric Human Signals for Natural Human-Robot Interaction

For natural human-robot interaction, a robot must understand human intent expressed not only through language but also through nonverbal signals such as gestures and gaze. However, current robot policies rely on language instructions as the sole interface for conveying intent, leaving nonverbal signals unused and placing the full burden of communication. In this work, we present EDITH, a robot framework that captures the human's nonverbal signals through continuous streams of first-person view and gaze from smart glasses, and uses them alongside language instructions as inputs to the robot policy. Our hardware system streams the human's first-person view, gaze, and speech to the robot in real time, transcribing the speech into language instructions. To handle these rich but noisy signals, we design a hierarchical policy in which a high-level policy infers the human's intent and produces a sequence of subtasks, where each subtask is represented as a fine-grained instruction paired with a keyframe that grounds the intent in the scene (e.g., the frame where the human points at the target object). A low-level policy then executes these subtasks. In our experiments on human-robot interactive tasks, EDITH enables the robot to act on the human's nonverbal signals even when intent is expressed only briefly, and significantly reduces user effort to convey intent compared to using language instructions alone. Visit our project page for source code and real-robot demo videos.

Dongjun Lee, Juheon Choi, Dong Kyu Shin +2

1KAIST · 2Seoul National University · 3Config

Robotic Control Language-Conditioned Robot Control Human-Robot Interaction

Jun 8, 2026cs.RO

Flow Control: Steering Vision-Language-Action Models with Simple Real-Time Inputs

We introduce flow control of vision-language-action (VLA) models, a simple and effective way to steer VLA actions in real-time through generic inputs, such as a keyboard. This method can be used out-of-the-box and does not require retraining or fine-tuning VLAs. It enables relatively crude user inputs to steer a VLA to align with user intent. The VLA transforms these inputs into action samples drawn from the VLA expert action distribution learned during training, so that the generated actions are high quality (conformity to the action expert distribution) and high fidelity (reflecting the user's intent). We demonstrate that flow control has many desirable properties: (1) flow control accurately and responsively steers robot actions with user inputs, (2) it is robust to suboptimal user inputs, (3) it enables users to steer VLAs to achieve significantly higher success rates and faster task completion, and (4) fine-tuning a VLA on flow control trajectories improves the autonomous policy. Together, these results provide a simple and intuitive way for users to help steer VLA actions, increasing task performance.

Jonathan C. Kao, Jason Chan, Andy Wang

Departments of Electrical and Computer Engineering, Computer Science, and Neurobiology, University of California, Los Angeles · Department of Computer Science, University of California, Los Angeles

Robotic Control Vision-Language-Action Models Vision-Based Robot Control VLMs for Robotics

Jun 8, 2026cs.RO

Physics-Aware Sparse Learning and Selective Online Adaptation for Euler-Lagrange Robot Dynamics

Accurate dynamics models are essential for model-based robotic control, yet nominal Euler--Lagrange models often become inaccurate in the presence of payload variation, unmodeled coupling, friction, aerodynamic effects, and changing operating conditions. Most learning-based correction methods improve prediction accuracy by introducing a single additive residual, but do not preserve the internal mechanical structure of Euler--Lagrange systems. This leads to models that do not preserve symmetry, positive-definiteness, or the coupling between inertia and velocity-dependent terms, which can result in physically inconsistent predictions and reduced reliability when embedded in model-based controllers. We propose a structure-preserving residual learning framework that decomposes model mismatch into an inertia correction, the corresponding induced Coriolis term, and a generalized-force residual. The mechanical component is learned under physical constraints, while the disturbance-sensitive component is represented through a sparse history-dependent latent interaction model and adapted online using Bayesian linear regression. This separation preserves key mechanical structure while restricting adaptation to the part of the dynamics most affected by changing conditions. Experiments across multiple robotic platforms, including mobile, aerial, and manipulator systems, show that the proposed method improves dynamics prediction and trajectory tracking under coupled and time-varying dynamics. These results highlight the value of combining structured residual modeling, compact latent interaction selection, and selective online adaptation for real-world model-based control.

Rishabh Dev Yadav, Samaksh Ujjawal, Sihao Sun +2

Department of Computer Science, The University of Manchaster, U.K. · Department of Cognitive Robotics, Delft University of Technology, Netherlands · International Institute of Information Technology Hyderabad. +1

Robotic Control Robot Dynamics Identification Online System Identification Residual Learning Residual Dynamics Learning Robot Dynamics Learning+2

Jun 8, 2026cs.RO

CT-VAM: A Cerebello-Thalamic-Inspired Vision-Action Model for Efficient Visuomotor Control

Vision-language-action models have shown strong promise for robot manipulation, yet raw language is primarily needed to specify task intent rather than to be repeatedly processed during high-frequency low-level execution. Motivated by this separation, we propose a cerebello-thalamic-inspired vision-action model (CT-VAM) for efficient task-conditioned visuomotor control. CT-VAM acts as a compact local execution policy that predicts action chunks from dualview visual observations, proprioception, and a lightweight task condition, potentially enabling a practical cloud-edge paradigm in which high-level semantic reasoning can be handled by large models while fast closed-loop control runs on local hardware. To fuse heterogeneous inputs effectively, CT-VAM introduces TARS (Thalamic Action Routing Stream), a stream-separated conditional attention decoder that independently routes action, visual and task streams, preventing dense sensory tokens from overwhelming compact task-relevant conditions. With only 68M parameters, CT-VAM achieves LIBERO success rates competitive with substantially larger VLA models, while reducing inference latency. Together with flow-consistent inpainting for asynchronous chunk execution, CT-VAM supports high-frequency control and demonstrates robust realworld deployment on resource-constrained robotic platforms.

Jiacheng Li, Yize Guo, Jiabin Guo +2

AIRLab, Department of Automation University of Science and Technology of China Hefei, China

Robotic Control Visuomotor Policy Learning Vision-Language-Action Models Vision-Based Robot Control

Jun 8, 2026cs.RO

Harness Engineering for Physical AI: Robot Middleware Is the Harness Layer

Robot middleware faces a new role in the era of Physical AI. Learned policies, planners, and vision-language-action (VLA) models now enter deployed robots as causal participants on the control path, but the layer that integrates them with timing, scheduling, and network has not been named. Recent language-agent work names this layer the harness, the external system that mediates tools, manages state, bounds resources, and records execution. The robotics community has not yet adopted this framing, and we propose that robot middleware is that harness. A Physical AI harness differs from a software harness in where it intervenes. A software harness mediates at tool-call boundaries. A Physical AI harness must mediate at control, computing, and communication simultaneously, because a learned policy's output crosses all three: its commands shift the trajectory, its inference time shifts the schedule, and its payload shifts the bandwidth. Robot middleware is the lowest robot-stack layer with mediating abstractions over all three, so it is best positioned to compose their enforcement. It already provides most of what a harness needs but lacks the enforcement for an AI model. We name this missing enforcement as three functions: Projection gates each output at emission, Isolation bounds the model's execution and transmission slot, and Transfer falls back to a verified baseline when checks fail. Each appears today as hand-built application code in deployed robot systems, built on surfaces robot middleware already provides. Robot middleware should host them not as the best single-axis enforcer but as the layer that composes all three. We sketch this as a ROS 2 Harness Profile, a deployment artifact that carries an AI model's declared output region, inference budget, and operating regime while the middleware enforces them across ROS 2, DDS, and Zenoh.

Sanghoon Lee, Jiyeong Chae, Kyung-Joon Park

Daegu Gyeongbuk Institute of Science and Technology (DGIST) · Daegu, Republic of Korea

Robotic Control Robotics Runtime Enforcement for AI Agents Runtime Assurance Embodied AI

Jun 7, 2026cs.RO

FAWAM: Force-Aware World Action Models for Closed-Loop Contact-Rich Manipulation

Force signals provide critical interaction cues for contact-rich robotic manipulation. However, existing methods mostly use force as an additional observation modality, without fully exploiting its role in modeling future interaction dynamics or guiding execution-time feedback correction. In this paper, we propose FAWAM, a force-aware world action model that incorporates force information at three levels: perception, prediction, and closed-loop execution. FAWAM first encodes historical 6-axis force/torque signals to modulate action generation, then jointly predicts future actions and end-effector wrenches to explicitly model contact evolution. It further introduces a residual correction module that uses the predicted wrench trajectory as an execution-time reference to refine actions online based on real-time force feedback. Real-world experiments across multiple contact-rich tasks show that FAWAM improves the average success rate by 36.25% over vision-only baselines and 21.25% over existing force-aware baselines, demonstrating the effectiveness of our force-aware framework for robust contact-rich manipulation.

Haotian He, Zeyu Yan, Qipeng Liu +2

School of Mathematical Sciences, Peking University · School of Artificial Intelligence, Shanghai Jiao Tong University.

Robotic Control World Models for Robotics Contact-Rich Robotic Manipulation Robotic Manipulation Force-Controlled Robotic Manipulation

Jun 7, 2026cs.AI

Self-Evolving Scientific Agent Designs Physically Reasoned White-Box Fluid Control

While neural networks excel in autonomous control, their black-box nature makes control decisions difficult to interpret and diagnose in dynamic fluids. Here, we show how self-evolving scientific agents can design explicit, neural-network-free white-box controllers by iteratively interpreting simulation evidence, accumulating control knowledge and refining controller code. We demonstrate this approach on an underactuated two-joint swimmer navigating unsteady flows via joint angular accelerations. Starting from a target-blind propulsive controller, the agent gradually constructs key mechanisms, including travelling-wave propulsion, body-frame guidance, phase-selective steering, redirect bursts and adaptive relief. The resulting controllers reach targets and generalize across changes in target position, wake geometry, cylinder count, and inflow speed without revision. Moreover, 2D control priors transfer successfully to accelerate 3D adaptation. Our work demonstrates that self-evolving agents can autonomously design physically reasoned and generalizable white-box fluid control, showing a promising paradigm beyond traditional reinforcement learning and black-box neural network control.

Boai Sun, Wenjin Guo, Zongmin Yu +1

National University of Singapore, Singapore.

Robotic Control Continuous Control Autonomous Scientific Discovery Self-Evolving Agents

Jun 6, 2026cs.RO

Reinforcement learning in linear embedding space unlocks generalizable control across soft robot configurations

Soft-bodied organisms such as octopuses and elephant trunks exhibit remarkable morphological adaptability, dynamically reconfiguring body shape and stiffness, and flexibly adjusting their control strategies to enable versatile behaviors. Inspired by these biological systems, various soft robots have emerged in recent decades, featuring diverse materials, stiffnesses, and morphologies tailored to specific tasks. Despite substantial advances in the materials and structural designs of soft robots, developing a generalizable control framework capable of rapid adaptation across diverse configurations remains a long-standing challenge. Existing controllers are limited to fixed configurations, demanding laborious configuration-specific remodelling and policy redesign for new configurations. Here, we introduce a generalizable control system that enables rapid adaptation across diverse soft robot configurations via reinforcement learning in a shared linear Koopman embedding space. By encoding robot dynamics into this embedding space, our method decouples control policies from specific morphologies, allowing real-time, model-free policy adaptation across diverse configurations without retraining from scratch. We validate our system across 33 distinct robot configurations. Our system achieves a 75 times reduction in transfer samples across configurations, while sustaining robust performance under high-speed motion, heavy payloads, and multiactuator faults, and achieving real-world skills previously unattainable in soft robotics. This work establishes a unified and adaptable control paradigm for diverse soft robot configurations, bridging mechanical reconfigurability with control flexibility, and may offer broader insights for generalizable control in complex physical systems.

Xinglong Zhang, Cong Li, Hangjie Mo +13

College of Intelligence Science and Technology, National University of Defense Technology, Deya Road, Changsha, 410003, Hunan, China. · School of Management, Hefei University of Technology, Nanyihuan Road, Hefei, 230002,2026 Anhui, China. · State Key NanjingLaboratoryUniversityfor Novel(SuzhouSoftwareCampus),TechnologyTaihuandRoad,the Suzhou,School of215163,ScienceChina.and Technology, 4School of Computation, Information and Technology, Technical University of Munich, Boltzmann Strasse, Munich, 85748, Germany. +5

Robotic Control RL for Robotics Cross-Embodiment Robot Learning Robot Policy Generalization Soft Robotics

Jun 4, 2026cs.RO

Multi-Robot Planning and Control from CCTV Camera Networks in a Real Warehouse

Off-board control of mobile robots from cameras embedded in the environment offers a practical path to scalable autonomy, moving sensing and compute off the robots. We extend this idea from the single-robot case to coordinated fleets in a real warehouse, driving multiple robots with only a distributed CCTV network and edge compute. The system operates entirely in image space over an uncalibrated, pixel-wise topological camera graph, enabling wide-area operation with flexible camera placement. A hierarchical planner selects a camera sequence per robot and plans its image-space motion through each view, coordinating robots with a prioritised-then-joint strategy and treating overlapping camera regions as shared resources held by one robot at a time to prevent collisions and deadlocks. We validate the approach in a real warehouse with four robots and 30 cameras across six 27 m aisles, reporting mission times and coordination statistics. To our knowledge, this is the first field demonstration of multi-robot planning and coordination using only an external camera network and off-board compute, with robots carrying no task-specific navigation hardware.

Luke Robinson, Benjamin Ramtoula, Anas Izaaryene +2

Oxford Robotics Institute, University of Oxford, UK · Robot Systems Group, Technical University of Munich, Germany

Robotic Control Multi-Robot Motion Planning Multi-Robot Coordination Vision-Based Robot Control

Jun 4, 2026cs.RO

AxisGuide: Grounding Robot Action Coordinate System in RGB Observations for Robust Visuomotor Manipulation

Visuomotor manipulation policies trained via large-scale behavior cloning have achieved strong semantic scene understanding, yet often fail to reliably execute correct low-level actions under distribution shifts. For example, even in a simple pickup task with identical scene layouts, camera viewpoints, and illumination, performance can degrade substantially when the object is placed at unseen locations. We argue that this gap arises from insufficient action understanding, namely the inability to interpret the robot's base-frame action coordinate system in image space. To address this issue, we introduce AxisGuide, a lightweight guidance method that bridges semantic scene understanding and action-coordinate interpretation. Using camera parameters and end-effector poses, AxisGuide renders the robot base-frame axes in each camera view and augments RGB observations with a small set of cue channels that explicitly visualize the meaning of the +x, +y, and +z motions in image space. Extensive evaluations in both the LIBERO simulation and real-world environments demonstrate that AxisGuide yields substantial performance gains and improved generalization, highlighting the effectiveness of explicit action-coordinate cues for learning reliable and transferable generalist visuomotor policies.

Jiyun Jang, Yujin Sung, Woosung Joung +5

1Korea University · University of Michigan · 3KT R&D Center +1

Robotic Control Generalization in Robotic Manipulation Visuomotor Policy Learning Robotic Manipulation Robot Imitation Learning

Jun 4, 2026cs.RO

TempoVLA: Learning Speed-Controllable Vision-Language-Action Policies

Robot manipulation alternates between low-risk transit phases that call for fast execution and high-risk contact stages that demand slow, precise motion. Yet existing Vision-Language-Action models (VLAs) only inherit a single fixed speed from training demonstrations. Prior efforts to accelerate VLAs through model compression, KV-cache reuse, or reinforcement learning only shift the policy from one fixed speed to another, and leave deceleration almost unexplored. We observe that the magnitude of each predicted action already governs how fast the robot moves, opening a direct route to controllable execution speed. We turn this observation into TempoVLA, a single VLA whose execution speed is controlled by an explicit condition. TempoVLA combines two coupled components. (1) A data-side Variable-Speed Trajectory Augmentation (VSTA) that re-times demonstration to any target speed by merging or splitting actions while preserving its motion semantics. (2) A model-side conditioning mechanism that feeds the speed to the policy. Statistics show that VSTA reaches the requested speed with negligible motion error. Experiments in simulation and on real-world tasks demonstrate that TempoVLA achieves flexible speed control in both directions, while VSTA additionally boosts the default

1\times

performance via better data utilization. Furthermore, by cooperating with a large multimodal model, TempoVLA realizes dynamic speed control, accelerating through low-risk phases and decelerating for high-risk ones.

Dong Jing, Jingchen Nie, Tianqi Zhang +4

1RUC · 2FDU · 3UNC

Robotic Control Vision-Language-Action Models Robot Learning

Jun 4, 2026cs.RO

Ensuring Interaction Safety in Multitask Exoskeleton Control: A Simulation-Trained Variable Impedance Framework

Wearable exoskeletons can augment human phys ical capabilities during complex activities. However, ensuring adaptation across diverse tasks while guaranteeing interaction safety remains a critical challenge. To address this, a simulation trained variable impedance control approach with stability guarantees is proposed. First, a simulation-based human exoskeleton motion data generation pipeline is established, utilizing Proximal Policy Optimization (PPO) to synthesize human muscle activations while the exoskeleton provides direct compensation for human biological joint torques. Subsequently, the generated dataset is used to train a dual modality policy that fuses semantic instructions with proprioceptive history, enabling the prediction of reference trajectories and variable impedance gains for nine different motion tasks. To guarantee safety, the network outputs are constrained by a stability criterion derived from Lyapunov stability theory, which bounds stiffness variations to ensure the asymptotic stability of the coupled system. Experimental results indicate that the proposed framework reduces metabolic cost in real-world scenarios com pared with standard baseline methods. These findings suggest the feasibility of the proposed framework for safe, multitask exoskeleton control.

Muyuan Ma, Houcheng Li, Haotian Zhai +4

Robotic Control Impedance Control Ankle Exoskeleton Control Language-Conditioned Robot Control

Jun 4, 2026cs.RO

TAM: Torque Adaptation Module for Robust Motion Transfer in Manipulation

A policy tuned for one robot often behaves differently on another, whether due to the sim-to-real gap, unknown payloads, or the differing dynamics of two instances of the same robot. In contact-rich, dynamic manipulation, even small motion discrepancies can result in failure to track reference motion, since they disrupt the timing and modes of contact. Common remedies, such as domain randomization or system identification, either produce overly conservative task policies or require data that must be recollected for each robot or payload. We introduce the Torque Adaptation Module (TAM), a learned module that adapts the torque commands sent to the robot to match the behavior of an ideal robot. TAM operates between the low-level controller that tracks the policy's actions and the robot's torque interface. It includes a history encoder that embeds proprioceptive history into a latent state and a torque adaptor that computes residual torque corrections. Because TAM depends only on proprioceptive history and not on policy observations, or the action space, the same TAM weights can be reused to adapt policies with different action spaces (joint targets, end-effector targets, or direct torques). The policies themselves do not need to be trained with domain randomization of robot parameters. Instead, we offload the need for domain randomization to TAM by training it entirely in randomized simulation, using multi-robot pretraining followed by a robot-specific fine-tuning step that still requires no real-robot data. We evaluate TAM zero-shot on a real Franka Panda robot across dynamic manipulation tasks that include a vision-based box pushing policy (from RL), a flip policy (from BC), and an MPC ball-on-plate balancing. Our experiments show that TAM improves zero-shot real-robot execution compared to online system identification and RMA baselines and enables robust dynamic manipulation performance.

Dongwon Son, Florian Shkurti, Jason Lee +3

1KAIST · *Work done during an internship at the Allen Institute for AI. · 2Allen Institute for AI +2

Robotic Control Synthetic-to-Real Domain Adaptation Robot Policy Adaptation Contact-Rich Robotic Manipulation Generalization in Robotic Manipulation Cross-Embodiment Robot Learning+1

Jun 4, 2026cs.RO

Visuotactile and Explicitly Force-Controlled Robotic Ultrasound for Abdominal Volumetric Reconstruction

In this paper, we present a robotic ultrasound acquisition system that integrates stereo vision, touch-based feedback, and expert-informed strategies to perform autonomous and adaptive abdominal scans. The system records freehand motion and force data from expert radiologists, creating a framework to capture transducer motion, applied forces, and anatomical scanning strategies. This expert data is replayed to replicate characteristic scans with the robot, forming a foundation for further autonomous capabilities. Using stereo vision, the system generates three-dimensional topography maps of the patient's abdomen, which are refined through stiffness measurements at key points to delineate the rib cage boundary. These combined techniques enable the robot to execute two distinct scanning paths: an upward-angled sweep beneath the rib cage to visualize structures near the upper abdomen and a perpendicular sweep across soft tissue regions. A compliant, torque-controlled seven degree-of-freedom robotic manipulator is controlled to maintain consistent probe contact through closed-loop force control over the varied anatomical surfaces. Physical experiments demonstrate that the system achieves high-quality imaging comparable to expert scans while dynamically adapting to patient-specific topographies. Furthermore, the robotic system surpasses expert capabilities by enabling three-dimensional volume acquisition, which enhances diagnostic potential and provides volumetric data for advanced analyses. This work highlights the integration of expert knowledge into autonomous robotic systems and underscores the potential of combining perception-based autonomy with physical reasoning for enhanced diagnostic performance.

Adrian Piedra, R Brooke Jeffrey, Oussama Khatib

Stanford Robotics Laboratory, Computer Science Department, Stanford University, Stanford, CA 94305 USA · Department of Radiology, School of Medicine, Stanford University, Stanford, CA 94305 USA

Robotic Control 3D Medical Imaging Ultrasound Imaging Robotic Ultrasound

Jun 4, 2026eess.SY

Amortized Nonlinear Model Predictive Control

Nonlinear Model Predictive Control requires solving a constrained nonlinear program (NLP) in real-time at every sampling instant, a computational bottleneck that limits deployment on resource-constrained hardware or at high sampling rates. We address this challenge for the broad class of input-affine nonlinear systems to show that the optimal control move can be approximated by a state-dependent quadratic program (QP) whose cost parameters depend on the current state and reference. We propose a single-network residual-corrector architecture: a state-dependent analytic baseline provides initial QP parameters, and the network learns only the corrections needed to match the full NLP solution; the QP is solved by a differentiable interior-point layer, guaranteeing constraint satisfaction for the first control action. The network is trained offline on data generated by an NLP solver using a hybrid loss that combines supervised imitation and KKT-residual penalties. We validate the approach on a three-link planar robotic arm with Cartesian end-effector tracking, demonstrating orders-of-magnitude speedup over the NLP solver while maintaining comparable tracking performance.

Francesco Pillitteri, Alberto Bemporad

IMT School for Advanced Studies, Lucca, Italy

Robotic Control Nonlinear MPC Learning to Optimize Model Predictive Control Optimal Control

Jun 3, 2026cs.RO

Efficient Computation of Distance Functions for Navigation Vector Fields in Lie Groups

Vector-field-based methods are widely used for robot control and are often applied to the path-tracking problem. Some vector field approaches require repeatedly computing the distance between the robot configuration and the curve, as well as the corresponding closest point. Recently, vector fields have been extended to Lie Groups. In this case, this computation can be expensive, especially when performed at high control frequencies on embedded platforms. This paper proposes a method for efficiently computing the distance between a point and a curve represented as what is called a G-polynomial curve, which is a curve representation that generalizes polynomial curves to matrix Lie groups. The proposed approach exploits the structure of these curves to reduce the problem to a small number of polynomial root-finding computations. Simulation results show that the method significantly reduces computation time while maintaining accuracy compared to existing optimization-based approaches. Practical formulas are also provided for the case of the group SE(3), and the method is validated experimentally on a robotic manipulator. The methodology is implemented in a computational package, available online.

Vinicius M. Gonçalves, João Baião, Felipe Bartelt +4

Robotic Control Robotics Lie Group Methods

Jun 3, 2026cs.RO

Real-World Deployment of a 5G-Connected Edge-Controlled Aerial Robot in Industrial Subterranean Mines

This article presents the first real-world autonomous flight of a 5G-connected aerial robot controlled by an edge-offloaded controller, and aims to bridge the gap between controlled and factual setups. The robot operates within an active industrial subterranean mine, while the high-level controller is deployed in a nearby Kubernetes-based edge cluster. Communication between the robot and the edge is enabled via a 5G New Radio (NR) Standalone (SA) network. The chosen controller is a Model Predictive Controller (MPC), which generates control actions to allow the robot to navigate seamlessly through the mining environment. A human operator selects waypoints for the aerial robot, and the MPC generates smooth, collision-free paths for autonomous executions. The proposed 5G edge-based closed-loop system is evaluated in a real industrial setting and demonstrates the potential of edge-controlled robotic systems toward time-critical, safe and efficient future deployments.

Achilleas Santi Seisa, Emanuele Pagliari, Gerasimos Damigos +2

Robotic Control Aerial Robotics Robot Navigation Edge Computing Robotics

Jun 3, 2026cs.RO

COP-Q: Safety-First Reinforcement Learning for Robot Control via Cholesky-Ordered Projection

Safe robot control requires maximizing return while satisfying safety constraints. In off-policy safe reinforcement learning, reward and safety Q-values are commonly learned by separate critic ensembles, with uncertainty handled independently for each objective. This objective-wise treatment neglects inter-objective correlation and can lead to overly conservative value estimates, thereby reducing sample efficiency. To address this issue, we propose Cholesky-Ordered Projection Q-learning (COP-Q), a safety-first method that incorporates inter-objective covariance into vector-valued Q-value estimation. COP-Q constructs a generalized confidence bound in the joint Q-value space and uses Cholesky factorization to encode objective priority in a sequential form. This preserves conservatism on safety while adaptively reducing excessive conservatism on the reward objective. The resulting estimate is used in both temporal-difference target computation and actor optimization. COP-Q incurs minimal computational overhead and is readily compatible with most existing deep Q-learning frameworks. Experiments on robot locomotion in Brax and safe navigation in Safety-Gymnasium, covering both hard- and soft-safety settings, demonstrate that COP-Q achieves strong safety performance together with competitive or improved sample efficiency relative to representative baselines.

Guopeng Li, Moritz A. Zanger, Matthijs T. J. Spaan +1

Department of Cognitive Robotics, Delft University of Technology · School of Transportation, Southeast University, China · Department of Intelligent Systems, Delft University of Technology, the Netherlands

Robotic Control Multi-Objective Reinforcement Learning Deep Q-Learning Constrained RL Robot Safety Safe RL

Jun 3, 2026cs.RO

A New Quaternion-Joint Cable-Driven Redundant Manipulator Configuration and its Control Through FABRIK and Residual Reinforcement Learning

Robotic arms capable of traversing arbitrary spatial paths, especially in highly obstructed workspaces, are highly desired across several industries. Quaternion-joints have recently empowered a specific class of robotic arms -- cable-driven redundant manipulators -- beyond its prior capabilities. Specifically, quaternion-joints reduce the number of required motors per degree of freedom, paving the way for more compact solutions.An ongoing challenge is that the complexity of the kinematic model of quaternion joints challenges a priori decisions on manipulator configurations and imposes higher computational demands on the control system and its non-linearities amplify all discrepancies between design and physical artifact arising from fabrication imprecision. Here we show a that a 4-segment, 8-joint manipulator can achieve a broader workspace than extant configurations, at lower hardware cost, and that Residual Reinforcement Learning outperforms extant state-of-the-art methods -- specifically, the FABRIK algorithm -- on the control of such manipulator. Our results show that this configuration is more workspace-effective than prior designs, and that Residual Reinforcement Learning outperforms FABRIK by three orders of magnitude on positional and orientational accuracy, effecting precise control of the novel 4-segment, 8-joint manipulator. Additionally, the control implementation is simpler: we describe the complete FABRIK process for control and corresponding learning implementation. Our methodology is applicable to the design of new systems, providing designers with further tools for the development of this class of manipulators and corresponding control systems for novel configurations.

Tanapath Pornthisan, Thanapat Kemthong, Thanyapisit Kangsathien +3

International School of Engineering, Chulalongkorn University, Bangkok, Thailand

Robotic Control Robotic Manipulation Inverse Kinematics

Jun 3, 2026cs.RO

OLIVE: Online Low-Rank Incremental Learning for Efficient Adaptive Exoskeletons

Wearable exoskeleton systems hold promise for restoring mobility in individuals with physical impairments, yet most existing controllers rely on static gait policies that lack the ability to adapt to dynamic real-world environments or individual user characteristics. We present \olive (\underline{O}nline \underline{L}ow-rank \underline{I}ncremental Learning for Efficient Adapti\underline{ve} Exoskeletons), a parameter-efficient online adaptation framework that continuously personalizes exoskeleton control during deployment. \olive decomposes the adaptive component of the control policy into a low-rank residual form~

\dW = \At\Bt^\top

with rank~

r!\ll!\min(d,k)

, reducing online update cost from

\mathcal{O}(dk)

to

\mathcal{O}(r(d{+}k))

while preserving the stability of a pretrained base controller~

\Wz

. Parameters are updated via a reward-shaped policy gradient driven purely by on-body sensor feedback (EMG, IMU, vibration), eliminating dependence on offline reference trajectories. A gating mechanism modulates the strength of personalization based on contextual state, and a dynamic rank scheduler adapts the update dimensionality to terrain complexity -- allocating minimal capacity on simple flat terrain and expanding to higher-rank updates on demanding uneven surfaces -- enabling robust performance across diverse activities: flat walking, stair navigation, slopes, and uneven terrain. Experiments on the wearable platform demonstrate that \olive achieves +13, +22, and +15 percentage-point improvements in gait smoothness, effort reduction, and motion stability over the strongest baseline, converging within

\sim

1{,}800 walking steps at 7.4,ms end-to-end latency. Our code implementation is available at https://github.com/FastLM/OLIVE.

Dong Liu, Yanxuan Yu, Ben Lengerich +2

University of California, Los Angeles Los Angeles, California, USA · Columbia University New York, New York, USA · University of Wisconsin-Madison Madison, Wisconsin, USA +1

Robotic Control Low-Rank Adaptation Ankle Exoskeleton Control Assistive Robotics