Papers for

robot control developers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Unified alignment improves robot perception across multiple sensor types

UniOMA: Unified Optimal-Transport Multi-Modal Structural Alignment for Robot Perception

Abstract: Achieving generalizable and well-aligned multimodal representations remains a core challenge in robot learning. While recent approaches have attempted to align modalities by modeling conditional or higher-order statistical dependencies, they do not explicitly consider the structural coherence across modalities. In this work, we propose a novel multimodal alignment method that augments existing contrastive losses with a geometry-aware Gromov--Wasserstein (GW) distance-based regularization. To this end, we encode intra-modal geometry for each modality and reduce cross-modal structural discrepancies through a dynamically learned GW barycenter, thereby encouraging structural consistency in addition to instance-level contrastive alignment. We apply this optimal-transport-based alignment strategy to robot perception tasks involving underexplored modalities such as force and tactile signals, where modality data often exhibit varying sample densities. Extensive experiments on five robotic benchmarks show that our method improves structural alignment and downstream performance over state-of-the-art multimodal contrastive baselines across a wide range of downstream tasks, including robot and environment state prediction. Moreover, our GW-based regularization can be seamlessly integrated into most InfoNCE-like objectives.

Mon 9 NovRoboticsArtificial Intelligence
The gist
Robots often need to understand information that comes from very different sensors, like touch or force sensors, but making sense of all these signals together is tricky. The authors designed a new way to better align these different types of sensor data by looking not just at individual matching parts, but at the overall structure and relationships in the data. Their method helps robots make sense of complex environments more reliably by combining data in a more organized way. They tested their approach on several robot tasks and found it works better than previous methods.
Open → 2611.12351v1

Off-policy reinforcement learning improves robot bimanual task performance

Off-Policy RL with Online Inference Tuning: My Solution to the CoRL 2026 RoboEval Challenge

Abstract: The CoRL 2026 RoboEval challenge has eight bimanual tasks, and it ranks a policy not only on whether it succeeds but on how: every stage of the task, done safely, efficiently and with both arms in coordination, scored against human demonstrations. My solution has two parts. The generic part, carried over from my LeHome challenge solution, is off-policy RL on a $π_{0.5}$-based policy that is also its own value function. It is trained with AWR and RECAP on advantages of the exact official score, in a loop where every rollout is at the same time a Thompson-sampling trial of about 20 inference parameters. Out of the box, it took the policy to 0.85 RDS on held-out seeds in two days (and it was still very far from convergence). The competition-specific part adds data that local exploration cannot find: generated demonstrations that reach the full reward, and recovery episodes that start from the policy's own failures. The submitted policy reaches 0.990 RDS on held-out seeds and 0.987 on the official seeds.

Sun 8 NovRoboticsComputer Vision and Pattern RecognitionMachine Learning
The gist
Teaching robots to do tasks with both hands needs precise and safe coordination. The researchers improved robot learning by tuning how the robot guesses its best moves while learning from past experiences, even reusing its own mistakes to get better. Their method helped the robot complete complex tasks almost as well as human demonstrations. The approach combines a general learning method with specific corrections to reach high scores on challenging robot tasks.
Open → 2611.11464v1

Humanoid locomotion improved by separating aesthetics from physical smoothness

PhySA: Decoupling Aesthetic and Physical Rewards for Humanoid Locomotion

Abstract: Humanoid locomotion policies trained without an explicit smoothness objective tend to produce jerky, high-energy gaits that stress actuators. Vision-language models (VLMs) offer an appealing reward source: given a rendered rollout, a VLM returns an aesthetic quality score on [0,10] without task-specific feature engineering. We find, however, that aesthetic preference and physical smoothness point in opposite directions: across four production VLMs, clips with higher joint acceleration receive higher aesthetic scores (mean Pearson r=+0.66). We study PhySA (Physical Smoothness and Aesthetics), which keeps the two on separate, independently removable reward channels: a sparse, asynchronous VLM score and a phase-triggered correction channel that adjusts physical penalty weights from simulator state. Ablations attribute the smoothness gain to the correction channel. On MuJoCo Humanoid-v5 at 2.6M steps, PhySA lowers joint acceleration by 26 to 32% relative to vanilla PPO, Text2Reward, and Eureka (three seeds), at the cost of shorter episodes. In a single extended run it lowers joint acceleration by 59.5% relative to Prompt2Policy, while a three-seed replication shows a smaller, non-significant reduction (7.2%). The gain does not transfer to IsaacLab or the Unitree G1, even after rescaling the smoothness term.

Thu 5 NovRobotics
The gist
Humanoid robots often move jerkily when trained without care for smoothness, which can wear out their joints. This paper finds that AI models judging movement aesthetics tend to prefer less smooth, high-energy motions. The authors introduce PhySA, a method that separates aesthetic rewards from physical smoothness penalties, enabling smoother robot walking. Their approach reduces joint jerkiness significantly in simulation, although results vary across different robot types.
Open → 2611.07679v1

Quadruped robot leg design analyzed for walking and flying modes

Kinematic analysis of a double-closed-loop four-bar chain-based leg for a multimodal quadruped-pentarotor robot

Abstract: This paper presents the kinematic modeling and singularity analysis of a three-degree-of-freedom, double four-bar closed-loop quadruped leg mechanism. This leg is essential for transitioning between the two operational modes; therefore, it is examined in both joint and task spaces, leading to the development of its Jacobian matrix, which is vital for designing the gaits required for terrestrial mode and mid-air aerial reconfiguration (for aerial mode). Various gaits are devised for the system's motion in the MATLAB simulation, and their practical implementation on the single-leg test rig is demonstrated.

Thu 5 NovRobotics
The gist
Robots that can both walk on land and fly need special legs that work well in both situations. This paper looks at a specific robot leg with multiple moving parts, explaining how it moves and the tricky positions where it might get stuck. The researchers made formulas and models to understand these movements and tested different walking patterns on a computer and a physical leg model. This helps the robot switch smoothly between walking and flying.
Open → 2611.07470v1

Astrobench benchmark measures robotic capture of floating objects in space

AstroBench: Benchmarking Free-Floating Capture and Space Robotic Manipulation

Abstract: Capturing a loose object inside a spacecraft requires a robot to approach a target that can translate and rotate, and whose motion changes when contact is made. These conditions motivate evaluation beyond manipulation of supported objects. We introduce AstroBench, an open benchmark for free-floating capture and space robotic manipulation with a fixed-base arm. The task suite combines four initial-motion tracks with sample-storage operations based on MELFI hardware and cap and captive-screw servicing based on the Robotic Refueling Mission. The two workcells provide nine task definitions and 29 task-target combinations. A metadata-filtered Objaverse inventory supplies 8,414 object instances for asset expansion, alongside five cabin visual contexts and versioned task instructions. We report two acquisition-focused experiments using existing checkpoints. A frozen $π_{0.5}$ DROID joint-position policy captures 2 of 200 static targets and 4 of 200 moving targets across twenty object identities. A cube-specific ACT policy captures 8 of 10 static targets and 6 of 10 moving targets in a separate experiment. Each experiment retains its recorded observation, action and capture configuration. The results document difficult direct transfer for the tested generalist checkpoint and successful task-specific capture behavior, while identifying concrete settings for studying motion-aware perception and contact-sensitive control. Code, task definitions and editable workcell assets are available at https://github.com/AstroUpLab/AstroBench.

Thu 22 OctRobotics
The gist
Catching loose objects in space is hard because the objects float freely and move unpredictably when touched. The authors created AstroBench, a set of robot tasks and tools to test how well robots can capture and handle these free-floating objects using a fixed arm. They tested different robot control methods and found that general-purpose approaches struggle, but specialized ones work better. Their work helps identify challenges in sensing motion and controlling contact with objects in space.
Open → 2610.31064v1

Humanoid control system improves body posture and position tracking

BeJEPA: A Goal-Conditioned Humanoid Behavior World Model via Unsupervised RL

Abstract: Humanoid control requires coordinating whole-body posture with position and heading in the world frame. Mainstream controllers imitate trajectories in the root frame, but struggle to regain the target world pose after displacement. We introduce BeJEPA, a goal-conditioned Behavior World Model learned through unsupervised reinforcement learning from human motion priors and robot interaction. BeJEPA learns a shared representation of body configuration and world placement through a joint-embedding predictive architecture (JEPA). A Dynamic Predictor predicts the next state embedding under an action, while a Successor Predictor summarizes a goal-conditioned policy's expected discounted future state embeddings in a single vector. During joint training, Dynamic grounds the initial representation in observed physical transitions, while Successor increasingly organizes it around long-term behavior. Their composition guides policy improvement through one physical prediction and one behavioral summary, without long imagined rollouts. In 20-second motion-tracking experiments, BeJEPA with online action refinement using Dynamic outperforms all evaluated baselines in global body-position error, reducing it on AMASS by 61.7% relative to BFM-Zero with world-pose inputs and 58.7% relative to SONIC. Physical deployment on a Unitree G1 demonstrates joint attainment of posture, world position, and heading goals, recovery from the ground, and closed-loop return after disturbances and goal changes. Project page: https://ylyem9x.github.io/BeJEPA-project.

Fri 16 OctRobotics
The gist
Controlling humanoid robots so they move naturally and stay oriented in the real world is hard. The paper presents BeJEPA, a new approach that learns to represent a robot’s body and position together, using a type of AI training without labeled goals. It can predict how actions affect the body now and over time, helping the robot recover from falls and reach goals better. The system beats previous methods at keeping track of where the robot is in the world and was tested both in simulation and on a real robot.
Open → 2610.23127v1

Gait stability changes with different impact timing in robots

The Effect of Gait Stability Based on Two Types of Impact Strategies for Two-Link Walking and Brachiating Robots

Abstract: In this paper, we explore the impulsive dynamics common to single-joint, two-link models of walking and brachiating gaits with respect to slope and switching time. In particular, we investigate how the stability of a gait and bifurcations encountered within a family of gaits change under time-based and state-based switching of the impulsive dynamics.

Thu 1 OctRobotics
The gist
Walking and swinging robots need to keep their movement stable to avoid falling or stopping. The authors looked at simple robot models with two connected parts to see how the timing of impacts during walking or swinging affects stability. They compared switching impact timing based on clock time versus using the robot's position or state. Their study shows how these timing strategies influence the robot’s balance and the points where its movement changes dramatically.
Open → 2610.01004v1

Soft robot octopus crawlers learn diverse adaptable arm movements

Diverse and Adaptable Arm Coordination for Octopus-Crawling via Diffusion-Based Uncertainty-Aware Optimization

Abstract: Octopus crawling motivates soft robots that exploit redundancy, yet discovering and organizing diverse coordination modes for adaptation remains challenging. To address this, we introduce a Diffusion-based Uncertainty-aware Optimization (DUO) algorithm that learns demonstration-free crawling controllers for a simulated, muscle-actuated CyberOctopus. This work represents the first application of diffusion-based control to soft multi-arm robots in contact-rich simulations. By embedding a variety of locomotion behaviors within a shared control distribution, this approach enables the simulated octopus to navigate dynamic physical constraints, demonstrating that learned coordination diversity inherently facilitates robust adaptation. The main contributions include: (i) a symmetry-structured policy representation that folds radially equivalent controllers into a canonical directional sector, (ii) an online black-box optimization strategy, the DUO algorithm, that discovers and retains diverse coordination modes, and (iii) a control editing technique that adapts existing controllers to novel actuator constraints without retraining. These results show how learned coordination diversity makes motor abundance a practical resource for adaptation in soft multi-arm robots.

Thu 17 SeptRoboticsMachine Learning
The gist
Controlling an octopus-like soft robot with many arms is hard because there are many ways to move, and the robot must adapt to obstacles or broken parts. The authors created a new method, called DUO, that teaches a simulated robot how to crawl in many different ways without needing example movements. This variety helps the robot adjust when conditions change, like when some arms don’t work properly. Their approach also lets the robot quickly update its movements without starting from scratch.
Open → 2609.21138v1