arXiv_RO
-
Evaluating the Impact of Adaptive Extended Reality on Human-Robot Interaction Across the Reality-Virtuality Continuum
2026-09-25 11:27:38Carl Tornberg, Alicia Torck, Lotfi El Hafi, Tadahiro TaniguchiAbstract
As populations in developed countries age and labor shortages intensify, Cybernetic Avatars (CAs) are proposed to extend human capabilities through robotic embodiments, requiring effective Human-Robot Interaction (HRI) frameworks. Extended Reality (XR), an umbrella term for Augmented Reality (AR), Augmented Virtuality (AV), and Virtual Reality (VR), offers such interfaces, but prior research typically fixes the XR modality without evaluating its effect on task outcomes. This study examines whether the XR modality impacts HRI performance and whether an adaptive interface adjusting the level of virtuality along the Reality-Virtuality Continuum (RVC) at runtime improves it. A custom XR application interfaced with a mobile manipulator supports immersive control and runtime modality switching. In a within-participant multi-room pick-and-place experiment comparing fixed AR, AV, and VR with dynamic RVC through task metrics, the NASA-TLX, and the System Usability Scale (SUS), this study demonstrates that 1) the fixed reality modality affects HRI results, and 2) dynamically changing the modality along the RVC improves them. AR yielded significantly lower mental demand, effort, and frustration than AV and VR, while the dynamic RVC condition achieved the highest throughput and lowest workload, highlighting the value of adaptive XR interfaces for human-robot symbiosis. The implementation is available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2609.31138
PDF
-
INTERACT: Interactive Planning for Autonomous Driving via Anchor-Conditioned Prediction and Trust-Region Refinement
2026-09-25 11:27:27Aron Distelzweig, Andreas Look, Faris Janjo\v{s}, Steffen Hagedorn, Luigi Palmieri, Joschka BoedeckerAbstract
Driving in dense urban traffic is interactive: whether a merge or an unprotected turn succeeds depends on how surrounding agents respond to the ego vehicle. Conventional planners predict first and plan second and, therefore, cannot account for this dependency. Methods that integrate prediction and planning either train both jointly, which introduces task interference, or keep them separate and are restricted to a predefined set of proposals. We present INTERACT: Interactive Planning for Autonomous Driving via Anchor-Conditioned Prediction and Trust-Region Refinement. Our key insight is that surrounding agents react to the intent a trajectory expresses rather than to its exact realization, so a single reactive prediction stays valid across an entire family of plans. INTERACT therefore decomposes interactive planning into prediction across driving intents and optimization within each intent. We derive a small set of diverse intents, which we call anchors, from map geometry, query a dedicated ego-conditioned prediction model once per anchor, and refine every anchor with the Cross-Entropy Method under a trust-region penalty that keeps the refined plan close enough to its anchor for the conditioned reaction to still apply. Prediction thus remains a separate model, avoiding task interference, while conditioning on anchors preserves the dependency. Because each anchor is refined continuously, the final plan is not restricted to the anchor set, yet INTERACT requires only one predictor query per anchor rather than one per candidate plan, with all anchors processed in parallel. On the nuPlan and interPlan closed-loop benchmarks, INTERACT sets a new state of the art, with the largest gains precisely in the interactive scenarios that motivate the method. The code will be released upon acceptance.
Abstract (translated)
URL
https://arxiv.org/abs/2609.31137
PDF
-
Comparative Evaluation of an XR Pen-based Control Interface for Semi-Autonomous Mobile Robot Navigation in Service Environments
2026-09-25 11:07:24Alicia Torc, Carl Tornberg, Eric Piette, Renaud Ronsse, Benoit Macq, Gustavo Alfonso Garcia Ricardez, Lotfi El Hafi, Tadahiro TaniguchiAbstract
Service robots remain difficult to deploy in domestic environments, partly because fully autonomous operation is not yet reliable in unpredictable surroundings, and partly because conventional control methods remain inaccessible to novice users. Extended Reality (XR) enables operators to visualize robot information overlaid onto the real world and to interact with augmented elements. Yet, common XR control methods, such as motion controllers and hand gestures, are still perceived as unintuitive. This paper presents a control interface that uses a commercial XR pen to command a semi-autonomous mobile robot in Augmented Reality (AR): the operator points at a position in the room, selects it, and drags an augmented arrow to set the desired orientation of the robot at this destination. Two additional interfaces, based on the XR motion controllers and hand gestures, were developed within the same framework. To assess the performance and users' perception of these interfaces, and of the XR pen in particular, a study with 10 participants compared four control methods, i.e., the XR pen, the XR motion controllers, hand gestures, and a computer-based baseline RViz, in navigation tasks performed in a home-like environment. Results show that the XR pen significantly outperforms the other methods in task selection time with the most consistent selections, and that the XR motion controllers obtain the best perceived workload and usability scores, ahead of the computer-based baseline, supporting XR-based control as an intuitive alternative for novice users. However, technical limitations in the integration of the recently released XR pen currently hold back its user experience.
Abstract (translated)
URL
https://arxiv.org/abs/2609.31117
PDF
-
DualManip: Agentic Dynamic Manipulation via Dual-Path Semantic Reasoning and Geometric Adaptation
2026-09-25 10:57:02Chengxi Li, Yan Di, Yingyue Li, Ruida Zhang, Mingyang Li, Xiangyang JiAbstract
Vision-language models (VLMs) enable open-vocabulary reasoning for robot manipulation, but their high inference latency limits responsiveness in dynamic scenes. Many scene changes, however, alter object geometry without invalidating task intent. We present DualManip, a dual-path framework that decouples infrequent semantic reasoning from responsive geometric adaptation. The semantic path decomposes the task and grounds task-relevant interactions, followed by a constraint-solving module for pose optimization. During execution, the geometric path continuously updates template-to-observation correspondences from live RGB-D observations via a shape-adaptive network. These correspondences transfer task-relevant grasp contacts across observations, enabling online grasp reconstruction under object motion and non-rigid deformation. The Information Interaction Module bridges the two paths by initializing task-relevant grasps from semantic grounding, validating geometric updates, and triggering semantic replanning upon update failures. Real-world evaluation spans six manipulation tasks covering non-rigid deformation, articulated reconfiguration, rigid motion, and high-precision assembly across three settings: static, single-change, and continuous dynamic. DualManip demonstrates superior manipulation robustness, particularly under continuous scene changes, while achieving geometric adaptation approximately 46$\times$ faster than agentic verification and semantic replanning. Our project page: this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2609.31112
PDF
-
AuthGuard-R: Safety-Compliant Mission Hijacking and Dual-Gate Defense for LLM-Controlled Robots
2026-09-25 10:56:22Saidattu Chepuri, Vikas SrivastavaAbstract
Large language models are increasingly used as high-level planners for mobile robots, robot manipulators, and autonomous vehicles. Recent studies show that these systems can be influenced through malicious text, speech, visual instructions, retrieved documents, and poisoned sensory context. Most defenses ask whether a proposed action is physically safe. This paper studies a different problem: an action may be physically safe and still violate the mission authorized by the user. An attacker may redirect a delivery robot, replace an approved object, extend a robot's operating region, activate an unnecessary sensor, or delay a mission without creating an immediate physical hazard. We call this attack \emph{safety-compliant mission hijacking}. We propose MissionPAIR, an adaptive attack framework that searches for executable plans that pass a safety gate while violating an authenticated mission. We also propose AuthGuard-R, a deterministic authorization layer that binds every executable action to a signed mission, robot identity, object and region scope, current state, time, and input provenance. AuthGuard-R operates with an independent safety gate, giving a dual-gate architecture. We formalize mission policies over robot traces, define security games, and prove authorization soundness, mission non-escalation, replay resistance, robot binding, provenance separation, threshold-approval security, audit-log tamper evidence, and trace-level composition. We report a preliminary cross-model evaluation with Claude Haiku~4.5 and the open-source Qwen2.5~7B planner. Across 240 live attack trials, the planners followed an injected mission deviation in 109 trials; AuthGuard-R rejected all 109 resulting unauthorized actions. A separate hand-constructed suite of eleven protocol- and policy-level attacks was also blocked completely.
Abstract (translated)
URL
https://arxiv.org/abs/2609.31110
PDF
-
Accuracy Evaluation of INS/ZUPT Filtering Methods Based on Different Geometric Error Definitions
Abstract
Geometric filters have recently been introduced to improve the accuracy and consistency of inertial-based integrated navigation systems. Error states were defined through specific group operations, introducing state correlations in error definition, which were lacked in the additive error used by a conventional indirect Kalman filter. The desirable consistent filtering models can be obtained based on specific geometric errors. For zero-velocity measurements expressed in the reference frame, this paper derives left-error process and measurement models from invariant filtering, two-frame-group filtering, and equivariant filtering. Importantly, a new group operation is introduced for the left tangent-group equivariant error. The analysis shows that the two-frame-group invariant extended Kalman filter (TFG-IEKF) and the tangent-group equivariant filter (TG-EqF) do not offer a significant consistency advantage over the invariant extended Kalman filter (IEKF). Experiments with an INS/ZUPT measurement system show that, under small initial attitude errors, the conventional indirect extended Kalman filter (EKF) achieves loop-closure position errors below $0.1\%$ of the traveled distance, while the three geometric filters achieve comparable positioning accuracy.
Abstract (translated)
URL
https://arxiv.org/abs/2609.31057
PDF
-
Kintsugi-VLA: Turning Failed Robot Rollouts into Recovery Data through Interventional Recoverability
2026-09-25 09:42:20Ivan Snegirev, Elizaveta Semenyakina, Dmitrii Maliukov, Miguel Altamirano Cabrera, Dzmitry TsetserukouAbstract
Simulation enables scalable training of Vision-Language-Action policies by using privileged experts to generate visual demonstrations without requiring every trajectory to be collected through manual teleoperation. However, such pipelines typically retain successful demonstrations while failed rollouts are discarded, even though they expose precisely the off-nominal states from which recovery must be learned. We introduce Kintsugi-VLA, a framework for converting failed rollouts into targeted synthetic recovery data by exploiting exact state restoration and branching in simulation. For a fixed privileged expert, we define interventional recoverability as the probability of completing the original task after the simulator is restored to a given state, estimate it using adaptive Monte Carlo continuations with pointwise Wilson intervals, and characterize its non-monotonic evolution along failed trajectories. These estimates identify an observed terminal low-recoverability frontier-the point after which measured recoverability remains below a threshold-which is then used to select informative recovery starting states. In a simulated Franka manipulation task, targeted recovery data yield aggregate SmolVLA recovery success of 34.6\% and 38.4\% under difficulty- and frame-budget matching, respectively, 5.8 and 6.7 percentage points above uniform sampling within the same recovery window. The same ordering is observed under disturbed end-to-end execution and shifted clutter and physics conditions, while clean-task success decreases from 76.8\% to 74.7\%. Kintsugi-VLA demonstrates how failed simulator rollouts can be transformed from discarded experience into structured recovery-training data through direct interventional measurement.
Abstract (translated)
URL
https://arxiv.org/abs/2609.31048
PDF
-
Compact Force Sensor for Dual-UAV Cable-Suspended Payload Transport with Tension-Aware Outer-Loop Control
2026-09-25 09:28:16Andrea Delbene, Giorgio Cannata, Giorgio Carlini, Filippo Sante, Marco BagliettoAbstract
Cooperative payload transportation using multiple \textit{Unmanned Aerial Vehicles} (UAVs) poses challenges in stability, coordination, and robustness, especially under external disturbances and unmodeled dynamics. This work proposes a dual-UAV payload transportation framework supported by a compact, custom-designed force sensor measuring the interaction force at the UAV cable anchor point. The sensor design and mathematical model are presented, and its performance is characterized through static and dynamic tests evaluating linearity, hysteresis, repeatability, and crossload. The control architecture follows a cascade structure: fast inner loops handle vehicle stabilization, while outer loops are designed to compensate for the measured forces. The approach is validated through simulations and indoor experiments under position uncertainty. Payload-drop and constrained-space tests assess the proposed sensing and control architecture against literature-based distributed references, showing improved stabilization, coordination, and disturbance rejection. A video of the experiments is available at: this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2609.31035
PDF
-
Precision at Speed: Sample-Efficient Online Model-Based Reinforcement Learning for Hydraulic Excavator Control
2026-09-25 09:22:12Claudio Canales, Fang Nan, Marco Hutter, Javier Ruiz-del-SolarAbstract
Precise, high-speed control remains challenging for robots with complex actuation dynamics. Learning directly on hardware is further constrained by the cost of real-world interaction. We present an online model-based reinforcement learning framework that learns a probabilistic dynamics ensemble model from scratch for sampling-based model predictive control. A precision-gated contouring objective conditions the progress reward on path accuracy, prioritizing precision over speed. In a data-driven excavator simulator, the framework achieves higher sample efficiency than the evaluated model-based reinforcement learning baselines. We validate the framework by learning directly on an 11.5-ton Menzi Muck M445 hydraulic excavator, without demonstrations or simulation pretraining. After 20 minutes of interaction, the controller reaches tracking accuracy comparable to prior learned controllers trained on 100-150 minutes of data. After 40 minutes, it sustains sub-centimeter mean path error at high operating speeds.
Abstract (translated)
URL
https://arxiv.org/abs/2609.31025
PDF
-
Co-design of trajectory and morphology for a vertical jump-climbing robot
Abstract
Animals such as squirrels and even bears have adapted to rapidly climb up trees and other complex vertical terrain, but achieving comparable agility has been a challenge for climbing robots. Existing robots often use walking gaits and move conservatively to stay in contact with the surface, which limits the range of dynamic maneuvers. In this paper we present a 290 g robot that, to our knowledge, is the first to climb vertically by bounding (with an aerial phase). We leverage a co-design workflow, in which the morphology and trajectory are jointly optimized for fast locomotion, subject to adhesion force limitations seen in spined grippers. The resulting trajectory includes a rapid maneuver that launches the robot vertically, and an aerial reorientation that brings the front grippers back to the surface using the rear leg as an inertial tail. We evaluate the resulting jump forces in 2D force space and demonstrate that the optimized morphology is capable of continuous climbing at a speed of 0.375 m/s (1.97 body lengths/s), and can also achieve ground locomotion and transition to a vertical surface. Our work proposes design insights for the jump-climbing maneuver and serves as an important step toward creating climbing robots with agility on par with that of animals.
Abstract (translated)
URL
https://arxiv.org/abs/2609.31014
PDF
-
Quadruped Obstacle Avoidance and Footstep Planning with Distributed Low-cost Time-of-Flight Sensors
2026-09-25 08:52:16Giammarco Caroleo, Timoth\'ee Mahamoodally, Matteo Manzardo, Jin Jin, Marco Pontin, Matias Mattamala, Renato Vidoni, Perla Maiolino, Maurice FallonAbstract
Quadruped robots typically rely on depth cameras and LiDAR sensors to map their local environment. However, these sensors have limited close-range coverage, are relatively expensive, and consume significant power. This study investigates whether distributed Time-of-Flight (ToF) sensors can serve as a low-cost alternative to depth cameras for near-field terrain mapping for locomotion and local navigation. We designed a distributed ToF sensing architecture for the ANYbotics ANYmal quadruped, assessed its environment reconstruction accuracy, and benchmarked it against depth cameras for terrain mapping and obstacle avoidance. Distributing these sensors around the robot can also avoid the blind spots of traditional sensors. Our results show that, despite their low resolution and higher measurement noise, distributed ToF sensors can support reliable perceptual locomotion with centimeter-level local mapping accuracy. The proposed sensing strategy provides sufficient geometric information for near-field obstacle avoidance and footstep planning, at substantially lower cost, energy consumption, and system complexity than depth cameras.
Abstract (translated)
URL
https://arxiv.org/abs/2609.31008
PDF
-
TACTIC: Understanding Tactile Encoders and Conditioning for Contact-rich Robot Manipulation Policies
2026-09-25 08:17:23Seongjin Bien, D\'ebora Oliveira Makowski, Carlo Kneissl, Reihaneh Mirjalili, Pankhuri Vanjani, Rudolf Lioutikov, Gitta Kutyniok, Florian Walter, Wolfram BurgardAbstract
Tactile information is essential for contact-rich manipulation tasks in robotics. Vision-based tactile sensors make it particularly easy to design end-to-end manipulation policies with tactile sensing, as they enable the use of existing encoders from computer vision. However, this has led to a huge variety of architectures, training datasets, and evaluation protocols, making it difficult to determine which design choices best encode touch. In this work, we address this gap and present a comprehensive study of tactile encoders and fusion strategies across various contact-rich manipulation tasks in real-world experiments. To enable a controlled comparison, we train and evaluate all models under the same pipeline and experimental setup, comprising more than 2000 real-world rollouts. Our results go beyond other studies that only compare simulation performance, which does not necessarily translate to real-world settings, where large-scale evaluations are needed to obtain reliable statistics. Our key finding is that there is no universally optimal representation or fusion strategy for encoding visual-tactile. Instead, the best encoder backbone and fusion scheme depend strongly on the task.
Abstract (translated)
URL
https://arxiv.org/abs/2609.30969
PDF
-
FRAM: Trajectory-Guided Visual Feature Selection for Compact Language-Conditioned Robot Manipulation
2026-09-25 08:16:16Hiroshi Ito, Hyogo Hiruma, Yoshiki Kanai, Takahiro Yoshida, Akira Kanazawa, Hiroki YamadaAbstract
Vision-Language-Action models achieve strong performance in robot manipulation, but often require large numbers of parameters. In this work, we propose the Future Representation Action Model (FRAM), a small policy that explicitly links the future end-effector trajectory to the current visual input. FRAM uses the image coordinates of the predicted trajectory as spatial pointers and reads local visual features related to the motion from the current image. This organizes the information for action generation into the reference position (Where), the visual state (What), and the future motion (Future). Trajectory labels are generated automatically from demonstrations and camera geometry, so no manual annotation is needed. With 138.7M parameters, including a frozen language encoder, FRAM reaches an average success rate of 92.2% over the four standard LIBERO suites, close to the 94.2% of $\pi_0$ with 3.3B parameters. Without extra training, it also reaches an average of 67.3% on LIBERO-Plus. Ablations confirm that both the future trajectory and the local visual features improve performance and robustness. On a real dual-arm UR5e, FRAM stacks cups using only wrist cameras, including choosing and switching between the left and right arms. These results show that selecting visual information based on future motion is an effective way to obtain both high performance and robustness in a small robot policy.
Abstract (translated)
URL
https://arxiv.org/abs/2609.30965
PDF
-
VisTacAlign: Co-Training Dexterous Policies on Tactile Human and Robot Demonstrations
2026-09-25 08:12:04Julien Poffet, Matthew Strong, Ankush Dhawan, Baiyu Shi, Shalika Neelaveni, Yujia Yuan, Zhenan Bao, Monroe Kennedy IIIAbstract
Human demonstrations are a cheap source of data for dexterous manipulation, but co-training a robot policy on them requires closing the human--robot gap in every modality the policy consumes. We present VisTacAlign, a framework for co-training 3D-visual-tactile dexterous policies on human and robot demonstrations. Glove-tracked human hand motion is retargeted to a 17-DoF tactile robot hand with a one-time fingertip correction. The human hand is then erased from both stereo views and replaced by a posed robot-hand mesh painted with pixels from robot recordings, and a real-time stereo foundation model is re-run on the composite, so the human point clouds carry the same stereo errors and visibility as the robot ones. Finally, a capacitive tactile glove is aligned to the robot's fingertip sensors in its signal space, giving one interpretable per-finger force representation. A diffusion transformer consumes point-cloud, proprioceptive, and per-finger tactile tokens. On three real-world tasks requiring precise force -- Lego assembly, plucking strawberries of varying size, and activating and lifting a power drill -- adding aligned human demonstrations to existing robot data improves over robot-only policies, and ablations show that both tactile input and visual alignment are necessary. Project page: this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2609.30959
PDF
-
Bundled Contact Gradients: Stabilizing Differentiable Simulation for Deployable Dynamic Tasks
2026-09-25 08:02:29Dyuman Aditya, Jin Cheng, Clemens Schwarke, Quan Nguyen, Gaurav Sukhatme, Stelian Coros, Gabriele FadiniAbstract
Differentiable simulation provides analytic gradients of robot dynamics, enabling fast and sample-efficient first-order policy optimization. However, obtaining smooth and informative gradients through rigid-body contact typically requires softened contact models, often at the expense of physical fidelity and thereby limiting learned policies largely to simulation. This trade-off becomes particularly consequential for dynamic humanoid motions, where accurate contact dynamics are critical for transferring policies to the real world. Increasing contact stiffness in rigid-body simulation improves the fidelity of interactions, but also makes the dynamics increasingly sensitive to small state perturbations, producing high-variance gradients that can destabilize first-order policy learning. To address this, we propose \emph{Bundled Contact Gradients (BCG)}, a contact-local randomized smoothing framework for differentiable policy learning. When stiff contact is detected, our method evaluates a local bundle of randomized perturbation rollouts around the stiff contact configuration and aggregates their gradient signal thereby reducing gradient variance. We demonstrate the effectiveness of our method by successfully training and transferring dynamic motions zero-shot onto a real-world Unitree G1 humanoid platform. Videos and supplementary information can be found at this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2609.30951
PDF
-
Cloak: Zero-Shot Cross-Embodiment Manipulation by Masking the End-Effector from the VLA
2026-09-25 07:48:37Michael Piseno, Guy Tevet, C. Karen LiuAbstract
We present Cloak, a training recipe that endows a Vision-Language-Action (VLA) model with zero-shot cross-embodiment transfer by cloaking the end-effector from its own wrist camera. The end-effector occupies a large and consistent region of the wrist view and masking it allows for embodiment-agnostic visual reasoning. Cloak renders a mask in simulation from the robot's known geometry, accurately and in real time, with no segmentation or generative models. During training, we augment the mask so the model generalizes to embodiments unseen at training time. We demonstrate the recipe with Cloak-VLA, a VLA trained with Cloak on a single parallel-jaw gripper dataset. No data of new embodiments is ever collected. Cloak-VLA transfers zero-shot to various unseen embodiments, including another gripper, another arm, and a five-fingered hand, while preserving the source embodiment's performance. By decoupling the wrist view from its own embodiment, Cloak allows data to outlive the hardware it was collected on.
Abstract (translated)
URL
https://arxiv.org/abs/2606.22836
PDF
-
Causeway: Restoring Task Accessibility for Instruction Switching in VLA Policies
2026-09-25 07:25:19Qingzi Wang, Kaixi Feng, Guangyao Shi, Xiyang Wu, Ang Li, Dinesh ManochaAbstract
Vision-language-action (VLA) policies can execute many tasks from standard initial states, yet a new instruction may fail after another task has altered the robot's physical state. We study instruction switching, where a new task is issued during or after the execution of a different one. We observe that a target task that is reliably completed from its standard initial states can become inaccessible from states produced by a preceding task. We call such states task islands. We propose Causeway, a training-free inference-time intervention. Given the current state and a re-entry pose for the target task, Causeway back-propagates through the frozen decoding computation and applies a state-directed write within the action-stream representation. The VLA decodes the return motion itself, without parameter updates, a new action head, or external action generation. Across 71 cross-object pairs, three switch timings, and three VLA architectures on LIBERO-Goal, Causeway raises bare-switch success from 3-26% to 47-65% and increases the rate of reaching the handoff neighborhood by 42-72 percentage points across models. Additional experiments on LIBERO-Object and a real xArm platform show that the recovery extends beyond the main LIBERO-Goal setting, both in simulation and on a robot.
Abstract (translated)
URL
https://arxiv.org/abs/2609.30913
PDF
-
PHASE: Compliance-Enabled Tactile Phase Retrieval for Few-Shot Insertion Learning
2026-09-25 06:53:56Jeremy Siburian, Cristian C. Beltran-Hernandez, Tatsuya Matsushima, Yusuke Iwasawa, Masashi Hamaya, Mai NishimuraAbstract
Contact-rich assembly tasks such as peg-in-hole insertion remain difficult to learn from limited demonstrations. While retrieval-augmented imitation learning, which augments target demonstrations with relevant prior data, offers a promising direction, its applicability to contact-rich manipulation remains largely unexplored. Contact-rich insertion unfolds over multiple phases from search to insert, and retrieving phase-specific experience from prior data in principled ways remains an open question. Our key insight is that a compliant wrist enables the robot to sustain contact throughout execution, producing rich tactile and force signals that naturally reveal the phase structure of insertion and inform what should be retrieved. Based on this insight, we present PHASE (PHase-Aware Segmentation and REtrieval), a framework for compliance-enabled tactile phase retrieval that integrates multimodal contact-aware representation learning, variable-length phase segmentation from tactile signals, and phase-consistent retrieval for policy learning. We evaluate PHASE on real-world peg-in-hole insertion across five peg geometries, comparing against retrieval strategies drawn from state-of-the-art methods under a shared policy architecture. PHASE improves the overall success rate by 13 percentage points over the strongest non-phase-aware baseline, and improves performance under unseen initial positions by 30 percentage points. These results demonstrate that aligning retrieval with interaction-defined contact phases substantially improves robustness in few-shot insertion learning.
Abstract (translated)
URL
https://arxiv.org/abs/2609.30889
PDF
-
A Second Torque Port for Series Elastic Actuators: Parallel-Integrated Design and Time-Scale Torque Allocation
2026-09-25 06:30:31Donghao Jia, Xiubo Xia, Junhang Liu, Zeyu Zhu, Xiaoyu Geng, Jian SunAbstract
A series elastic actuator has a single torque port and pays for it twice: the geared motor must swing its own reflected inertia through the spring, so the amplitude it delivers collapses as $\omega^{-2}$ in the command frequency $\omega$ once it saturates, while commands below the transmission's breakaway friction never arrive at all. This letter opens a second torque port on the load side, placing a frameless direct-drive micro motor in parallel with a fixed-stiffness spring -- a parallel-integrated SEA, or Pi-SEA, whose delivered torque is read from spring deflection and micro current without a sensor -- and dividing the commanded torque between the two channels by time scale rather than by filter design. The micro torque loop is the fast subsystem, which makes the closed loop singularly perturbed and turns the separation the channels need into a bound to check rather than a crossover to tune; a leaky mid-ranging integrator returns the steady load to the spring; and the amplitude ceiling, read backwards, becomes a closed-form sizing rule that matches spring, geared motor and micro motor to the amplitudes and frequencies an application asks for. Against SEAs, the Pi-SEA widens the tracked band at small amplitudes and lowers the residual the joint imposes on its environment, each by an order of magnitude.
Abstract (translated)
URL
https://arxiv.org/abs/2609.30873
PDF
-
VLaRL: Augmenting Vision-Language-Action Models with Simulation-Trained Latent-Conditioned Residual RL
2026-09-25 06:18:40Namiko Saito, Kinam Kim, Heecheol Kim, Katsushi Ikeuchi, Yasuyuki MatsushitaAbstract
Vision-language-action (VLA) models provide broad, instruction-conditioned manipulation behaviors, but their physical execution can remain imprecise during contact-rich interaction. Residual reinforcement learning (RL) can correct such errors while keeping the VLA frozen, but real-robot RL is costly and safety-critical. We propose VLA Latent-Conditioned RL (VLaRL), which enables residual RL for frozen VLAs to be trained in simulation and deployed on real robots without real-world RL or online adaptation. The key challenge is transferring the learned residual policy despite the visual gap between simulation and reality. Rather than requiring pixel-level visual correspondence, VLaRL uses the VLA's internal vision-language latent representation to condition residual control and as the sim-to-real transfer interface, and learns a lightweight mapper that transforms simulation-derived latents toward the real latent distribution. Across four contact-rich manipulation tasks and two VLA backbones, VLaRL improves real-world success in all task-backbone combinations, while controlled ablations demonstrate the importance of both latent conditioning and latent alignment for transferring simulation-trained residual control.
Abstract (translated)
URL
https://arxiv.org/abs/2609.30868
PDF
