arXiv subjects

Mitsuhiro Hayashibe

Publications and source records attributed to Mitsuhiro Hayashibe.

6 recordsLinked to original sources

arXiv2026

LAC: Linear and Angular Compliance for Humanoid Whole-body Control

Real-world humanoid tasks involve physical interaction with objects and humans, yet current controllers either reject external forces as disturbances or restrict compliance to limited body links while ignoring angular effects. We present LAC, a general whole-body controller that simultaneously realizes commanded Linear and Angular Compliance for wrenches applied to the upper body. First, we synthesize whole-body compliant responses into a large-scale augmented dataset. Sampled force and couple events are imposed on contact frames extracted from human interaction data. At each contact link, the external force and a virtual torque from the passively yielding kinematic chain drive a virtual admittance under the commanded stiffness. Subsequently, teacher-student reinforcement learning trains a single policy to track the compliant motions under external wrenches. Finally, extensive simulation and real-world experiments demonstrate whole-body compliant responses to wrenches across the upper body, monotonic modulation over the full range of both stiffness commands, and applicability to teleoperated loco-manipulation tasks. Project website: https://lac-humanoid.github.io/

cs.RO↗

arXiv2026

TRANS: Terrain-aware Reinforcement Learning for Agile Navigation of Quadruped Robots under Social Interactions

This study introduces TRANS: Terrain-aware Reinforcement learning for Agile Navigation under Social interactions, a deep reinforcement learning (DRL) framework for quadrupedal social navigation over unstructured terrains. Conventional quadrupedal navigation typically separates motion planning from locomotion control, neglecting whole-body constraints and terrain awareness. On the other hand, end-to-end methods are more integrated but require high-frequency sensing, which is often noisy and computationally costly. In addition, most existing approaches assume static environments, limiting their use in human-populated settings. To address these limitations, we propose a two-stage training framework with three DRL pipelines. (1) TRANS-Loco employs an asymmetric actor-critic (AC) model for quadrupedal locomotion, enabling traversal of uneven terrains without explicit terrain or contact observations. (2) TRANS-Nav applies a symmetric AC framework for social navigation, directly mapping transformed LiDAR data to ego-agent actions under differential-drive kinematics. (3) A unified pipeline, TRANS, integrates TRANS-Loco and TRANS-Nav, supporting terrain-aware quadrupedal navigation in uneven and socially interactive environments. Comprehensive benchmarks against locomotion and social navigation baselines demonstrate the effectiveness of TRANS. Hardware experiments further confirm its potential for sim-to-real transfer.

cs.RO↗

arXiv2025

Simultaneous estimation of contact position and tool shape with high-dimensional parameters using force measurements and particle filtering

Estimating the contact state between a grasped tool and the environment is essential for performing contact tasks such as assembly and object manipulation. Force signals are valuable for estimating the contact state, as they can be utilized even when the contact location is obscured by the tool. Previous studies proposed methods for estimating contact positions using force/torque signals; however, most methods require the geometry of the tool surface to be known. Although several studies have proposed methods that do not require the tool shape, these methods require considerable time for estimation or are limited to tools with low-dimensional shape parameters. Here, we propose a method for simultaneously estimating the contact position and tool shape, where the tool shape is represented by a grid, which is high-dimensional (more than 1000 dimensional). The proposed method uses a particle filter in which each particle has individual tool shape parameters, thereby to avoid directly handling a high-dimensional parameter space. The proposed method is evaluated through simulations and experiments using tools with curved shapes on a plane. Consequently, the proposed method can estimate the shape of the tool simultaneously with the contact positions, making the contact-position estimation more accurate.

cs.RO↗

arXiv2025

Language-Guided Long Horizon Manipulation with LLM-based Planning and Visual Perception

Language-guided long-horizon manipulation of deformable objects presents significant challenges due to high degrees of freedom, complex dynamics, and the need for accurate vision-language grounding. In this work, we focus on multi-step cloth folding, a representative deformable-object manipulation task that requires both structured long-horizon planning and fine-grained visual perception. To this end, we propose a unified framework that integrates a Large Language Model (LLM)-based planner, a Vision-Language Model (VLM)-based perception system, and a task execution module. Specifically, the LLM-based planner decomposes high-level language instructions into low-level action primitives, bridging the semantic-execution gap, aligning perception with action, and enhancing generalization. The VLM-based perception module employs a SigLIP2-driven architecture with a bidirectional cross-attention fusion mechanism and weight-decomposed low-rank adaptation (DoRA) fine-tuning to achieve language-conditioned fine-grained visual grounding. Experiments in both simulation and real-world settings demonstrate the method's effectiveness. In simulation, it outperforms state-of-the-art baselines by 2.23, 1.87, and 33.3 on seen instructions, unseen instructions, and unseen tasks, respectively. On a real robot, it robustly executes multi-step folding sequences from language instructions across diverse cloth materials and configurations, demonstrating strong generalization in practical scenarios. Project page: https://language-guided.netlify.app/

cs.RO↗

arXiv2022

Sparse Identification of Lagrangian for Nonlinear Dynamical Systems via Proximal Gradient Method

Distilling physical laws autonomously from data has been of great interest in many scientific areas. The sparse identification of nonlinear dynamics (SINDy) and its variations have been developed to extract the underlying governing equations from observation data. However, SINDy faces certain difficulties when the dynamics contain rational functions. The principle of the least action governs many mechanical systems, mathematically expressed in the Lagrangian formula. Compared to the actual equation of motions, the Lagrangian is much more concise, especially for complex systems, and does not usually contain rational functions for mechanical systems. Only a few methods have been proposed to extract the Lagrangian from measurement data so far. One of such methods, Lagrangian-SINDy, can extract the true form of Lagrangian of dynamical systems from data but suffers when noises are present. In this work, we develop an extended version of Lagrangian-SINDy (xL-SINDy) to obtain the Lagrangian of dynamical systems from noisy measurement data. We incorporate the concept of SINDy and utilize the proximal gradient method to obtain sparse expressions of the Lagrangian. We demonstrated the effectiveness of xL-SINDy against different noise levels with four nonlinear dynamics: a single pendulum, a cart-pendulum, a double pendulum, and a spherical pendulum. Furthermore, we also verified the performance of xL-SINDy against SINDy-PI (parallel, implicit), a recent robust variant of SINDy that can handle implicit dynamics and rational nonlinearities. Our experiment results show that xL-SINDy is 8-20 times more robust than SINDy-PI in the presence of noise.

eess.SY↗

arXiv2019

Systematic Enhancement of Functional Connectivity in Brain-Computer Interfacing using Common Spatial Patterns and Tangent Space Mapping

Functional connectivity of cognitive tasks allows researchers to analyse the interaction mapping occurring between different regions of the brain using electroencephalography (EEG) signals. Standard practice in functional connectivity involve studying the electrode pair interactions across several trials. As the cognitive task always involves the human factor, it is inevitable to have lower quality data from the brain signals influenced by the subject concentration or other mental states which can occur anytime over the whole experimental trials. The connectivity among electrodes are heavily influenced by these low quality EEG. In this paper, we aim at enhancing the functional connectivity of mental tasks by implementing a classification step in the process to remove those incorrect EEG trials from the available set. The classification step removes the trials which were mis-classified or had a low probability of occurrence to extract only reliable EEG trials. Through our approach, we have successfully improved the separability among graph parameters for different mental tasks. We also observe an improvement in the readability of the connectivity by focusing only on a group of selected channels rather than employing all the channels.

cs.HC↗