arXiv subjects

Yusuke Tanaka

Publications and source records attributed to Yusuke Tanaka.

At least 19 recordsLinked to original sources

arXiv2026

Labeled-Data-Free Meta-Learning: Efficient Task Generation Using Pre-trained Models and Unlabeled Data

Meta-learning without labeled data is crucial for real-world applications, where obtaining labeled datasets can be expensive or restricted due to privacy concerns. Data-Free Meta-Learning (DFML) addresses this challenge by leveraging pre-trained models without access to training data. However, existing DFML methods rely on model inversion to generate training data, a process that is generally difficult and computationally expensive due to the need to generate high-dimensional data matching the original distribution. To address this limitation, we propose a novel meta-learning setting that avoids model inversion by jointly leveraging pre-trained models and unlabeled data. Our method generates meta-training tasks by assigning soft labels from pre-trained models to unlabeled data. Since the quality of these tasks can vary, we introduce a task-weighting mechanism based on task confidence and class distribution balance to ensure effective meta-learning. Extensive experiments demonstrate that our approach substantially reduces computational cost and improves generalization, achieving up to 104-fold speedup and 8.4 percent to 36.4 percent improvements in few-shot classification accuracy compared to state-of-the-art DFML methods.

cs.LG↗

arXiv2026

Rapid co-design of Buoyancy-assisted robots for Challenging Locomotion using Gaussian Evolutionary Specialists

Designing high-performance legged robots requires jointly optimizing morphology and control. Model-free Reinforcement Learning (RL) offers an alternative to model-predictive control for developing robust controllers without explicitly specifying robot dynamics. Thus, we have seen theuse of RL to train controllers and evaluate designs for robot morphology optimization. While RL has shown success inlocomotion, using it in the co-design inner loop is expensive due to repeated policy training. Universal policies conditioned on morphology offer a promising alternative, but suffer from behavioral diversity collapse, converging to a single strategy that performs sub-optimally across designs. On the other hand, end-to-end Mixture-of-Experts (MoE) architectures fail due to a collapse in its representation. We propose Gaussian Evolutionary Specialists (GES), a framework that decouples design-space partitioning from policy learning to capture diverse behaviors explicitly. GES assigns specialist policies to evolving Gaussian regions and iteratively refines them via training, probing, and territory expansion. The resulting specialists are integrated into a design sampling loop, replacing costly re-training with direct evaluation. When tested on the Buoyancy-Assisted Light Legged Unit (BALLU), GES discovers designs with 5 - 25% higher performance than naive universal policies. On hardware, a GES optimized design overcomes a 24 cm tall obstacle - 3x improvement over the baseline BALLU design. Moreover, GES curtails design optimization time by 37%.

cs.RO↗

arXiv2026

Symplectic Neural Operators for Learning Infinite Dimensional Hamiltonian Systems

The modeling and simulation of infinite-dimensional Hamiltonian systems are central problems in mathematical physics and engineering, however they pose significant computational and structural challenges for standard data-driven architectures. In this work, we introduce the Symplectic Neural Operator, a neural operator architecture designed to preserve the symplectic structure intrinsic to Hamiltonian PDEs. We provide a theoretical characterization of their symplecticity and establish a rigorous long-term stability result based on the combination of symplectic structure preservation and learning accuracy. Numerical experiments on canonical Hamiltonian PDEs corroborate this theoretical result and show that SNOs exhibit improved energy behavior compared with non-structure-preserving neural operators.

math.DS↗

arXiv2025

Postseismicity of slow-slip doublets discerned on the outermost of the Nankai Trough subduction megathrust

Despite dissimilar slip rates, slow earthquakes are faulting as ordinary earthquakes are. It is therefore physically natural that slow earthquakes also cause postseismic motions similarly to ordinary earthquakes, even though coseismic and postseismic slips remain undifferentiated for slow earthquakes. We pursue the slow-earthquake postseismicity based on the analysis of a fault slip beneath the Bungo Channel, the westernmost region of the Nankai Trough subduction zone in southwestern Japan. Its 2010 long-term slow slip event (SSE) was mispredicted by physics-based models, which concludes that the initial acceleration of this SSE was too abrupt for a slow variant of a fault rupture. We identify that a mispredicted GNSS signal evolves logarithmically in time, preceded by minor signals that evolve exponentially, lasting about two years west and about half a year east. By performing sparse inverse modeling on the GNSS, we have estimated that exponential slips occur at the same depth, bracketing a logarithmic slip that occurs beneath the channel. The regions of exponential slips match repeating slow-slip regions, and deep tremors synchronize exclusively with the logarithmic slip. This source complexity can be explained as a neighboring rupture doublet and its afterslip and aftershocks by the known mechanics of ordinary earthquakes. If slow earthquakes have a dual origin in exponentially nucleating slow rupture and logarithmically decelerating postseismic creep, it is possible to pick the slow earthquake nuclei that could accelerate into megathrust catastrophes.

physics.geo-ph↗

arXiv2025

MOBIUS: A Multi-Modal Bipedal Robot that can Walk, Crawl, Climb, and Roll

This paper presents the MOBIUS platform, a bipedal robot capable of walking, crawling, climbing, and rolling. MOBIUS features four limbs, two 6-DoF arms with two-finger grippers for manipulation and climbing, and two 4-DoF legs for locomotion--enabling smooth transitions across diverse terrains without reconfiguration. A hybrid control architecture combines reinforcement learning for locomotion and force control for compliant contact interactions during manipulation. A high-level MIQCP planner autonomously selects locomotion modes to balance stability and energy efficiency. Hardware experiments demonstrate robust gait transitions, dynamic climbing, and full-body load support via pinch grasp. Overall, MOBIUS demonstrates the importance of tight integration between morphology, high-level planning, and control to enable mobile loco-manipulation and grasping, substantially expanding its interaction capabilities, workspace, and traversability.

cs.RO↗

arXiv2025

Mechanical Intelligence-Aware Curriculum Reinforcement Learning for Humanoids with Parallel Actuation

Reinforcement learning (RL) has enabled advances in humanoid robot locomotion, yet most learning frameworks do not account for mechanical intelligence embedded in parallel actuation mechanisms due to limitations in simulator support for closed kinematic chains. This omission can lead to inaccurate motion modeling and suboptimal policies, particularly for robots with high actuation complexity. This paper presents general formulations and simulation methods for three types of parallel mechanisms: a differential pulley, a five-bar linkage, and a four-bar linkage, and trains a parallel-mechanism aware policy through an end-to-end curriculum RL framework for BRUCE, a kid-sized humanoid robot. Unlike prior approaches that rely on simplified serial approximations, we simulate all closed-chain constraints natively using GPU-accelerated MuJoCo (MJX), preserving the hardware's mechanical nonlinear properties during training. We benchmark our RL approach against a model predictive controller (MPC), demonstrating better surface generalization and performance in real-world zero-shot deployment. This work highlights the computational approaches and performance benefits of fully simulating parallel mechanisms in end-to-end learning pipelines for legged humanoids. Project codes with parallel mechanisms: https://github.com/alvister88/og_bruce

cs.RO↗

arXiv2025

AURA: Autonomous Upskilling with Retrieval-Augmented Agents

Designing reinforcement learning curricula for agile robots traditionally requires extensive manual tuning of reward functions, environment randomizations, and training configurations. We introduce AURA (Autonomous Upskilling with Retrieval-Augmented Agents), a schema-validated curriculum reinforcement learning (RL) framework that leverages Large Language Models (LLMs) as autonomous designers of multi-stage curricula. AURA transforms user prompts into YAML workflows that encode full reward functions, domain randomization strategies, and training configurations. All files are statically validated before any GPU time is used, ensuring efficient and reliable execution. A retrieval-augmented feedback loop allows specialized LLM agents to design, execute, and refine curriculum stages based on prior training results stored in a vector database, enabling continual improvement over time. Quantitative experiments show that AURA consistently outperforms LLM-guided baselines in generation success rate, humanoid locomotion, and manipulation tasks. Ablation studies highlight the importance of schema validation and retrieval for curriculum quality. AURA successfully trains end-to-end policies directly from user prompts and deploys them zero-shot on a custom humanoid robot in multiple environments - capabilities that did not exist previously with manually designed controllers. By abstracting the complexity of curriculum design, AURA enables scalable and adaptive policy learning pipelines that would be complex to construct by hand. Project page: https://aura-research.org/

cs.RO↗

arXiv2025

Buoyant Choreographies: Harmonies of Light, Sound, and Human Connection

BALLU, the Buoyancy Assisted Lightweight Legged Unit, is a unique legged robot with a helium balloon body and articulated legs \fig{fig:fig1}. Since it is buoyant-assisted, BALLU is inherently stable, never falling over, while being able to walk, jump, and interact safely with people. The BALLU art installation builds on this playful platform to express fluidity, serendipity, and connection. It transforms robotic motion into an artistic visual and acoustic experience, merging technology and creativity into a dynamic, interactive display. This exhibition intentionally does not have a physical boundary for the robots, emphasizing the harmony of the technologies and humanity. This work significantly extends BALLU's existing permanent exhibition in the Seoul Robotics & Artificial Intelligence Museum, Seoul RAIM (https://anc.masilwide.com/2261), emphasizing the harmony of robotics and humanity through visual, acoustic, and physical expression.

cs.RO↗

arXiv2025

Learning Hamiltonian Density Using DeepONet

In recent years, deep learning for modeling physical phenomena which can be described by partial differential equations (PDEs) have received significant attention. For example, for learning Hamiltonian mechanics, methods based on deep neural networks such as Hamiltonian Neural Networks (HNNs) and their variants have achieved progress. However, existing methods typically depend on the discretization of data, and the determination of required differential operators is often necessary. Instead, in this work, we propose an operator learning approach for modeling wave equations. In particular, we present a method to compute the variational derivatives that are needed to formulate the equations using the automatic differentiation algorithm. The experiments demonstrated that the proposed method is able to learn the operator that defines the Hamiltonian density of waves from data with unspecific discretization without determination of the differential operators.

cs.LG↗

arXiv2025

Tethered Variable Inertial Attitude Control Mechanisms through a Modular Jumping Limbed Robot

This paper presents the concept of a tethered variable inertial attitude control mechanism for a modular jumping-limbed robot designed for planetary exploration in low-gravity environments. The system, named SPLITTER, comprises two sub-10 kg quadrupedal robots connected by a tether, capable of executing successive jumping gaits and stabilizing in-flight using inertial morphing technology. Through model predictive control (MPC), attitude control was demonstrated by adjusting the limbs and tether length to modulate the system's principal moments of inertia. Our results indicate that this control strategy allows the robot to stabilize during flight phases without needing traditional flywheel-based systems or relying on aerodynamics, making the approach mass-efficient and ideal for small-scale planetary robots' successive jumps. The paper outlines the dynamics, MPC formulation for inertial morphing, actuator requirements, and simulation results, illustrating the potential of agile exploration for small-scale rovers in low-gravity environments like the Moon or asteroids.

cs.RO↗

arXiv2024

Mechanisms and Computational Design of Multi-Modal End-Effector with Force Sensing using Gated Networks

In limbed robotics, end-effectors must serve dual functions, such as both feet for locomotion and grippers for grasping, which presents design challenges. This paper introduces a multi-modal end-effector capable of transitioning between flat and line foot configurations while providing grasping capabilities. MAGPIE integrates 8-axis force sensing using proposed mechanisms with hall effect sensors, enabling both contact and tactile force measurements. We present a computational design framework for our sensing mechanism that accounts for noise and interference, allowing for desired sensitivity and force ranges and generating ideal inverse models. The hardware implementation of MAGPIE is validated through experiments, demonstrating its capability as a foot and verifying the performance of the sensing mechanisms, ideal models, and gated network-based models.

cs.RO↗

arXiv2024

Cycloidal Quasi-Direct Drive Actuator Designs with Learning-based Torque Estimation for Legged Robotics

This paper presents a novel approach through the design and implementation of Cycloidal Quasi-Direct Drive actuators for legged robotics. The cycloidal gear mechanism, with its inherent high torque density and mechanical robustness, offers significant advantages over conventional designs. By integrating cycloidal gears into the Quasi-Direct Drive framework, we aim to enhance the performance of legged robots, particularly in tasks demanding high torque and dynamic loads, while still keeping them lightweight. Additionally, we develop a torque estimation framework for the actuator using an Actuator Network, which effectively reduces the sim-to-real gap introduced by the cycloidal drive's complex dynamics. This integration is crucial for capturing the complex dynamics of a cycloidal drive, which contributes to improved learning efficiency, agility, and adaptability for reinforcement learning.

cs.RO↗

arXiv2024

Photonic topological phase transition induced by material phase transition

Photonic topological insulators (PTIs) have been proposed as an analogy to topological insulators in electronic systems. In particular, two-dimensional PTIs have gained attention for the integrated circuit applications. However, controlling the topological phase after fabrication is difficult because the photonic topology requires the built-in specific structures. This study experimentally demonstrates the band inversion in two-dimensional PTI induced by the phase transition of deliberately-designed nanopatterns of a phase-change material, Ge2Sb2Te5 (GST), which indicates the first observation of the photonic topological phase transition with changes in the Chern number. This approach allows us to directly alter the topological invariants, which is achieved by symmetry-breaking perturbation through GST nanopatterns with different symmetry from original PTI. The success of our scheme is attributed to the ultrafine lithographic alignment technologies of GST nanopatterns. These results demonstrate to control photonic topological properties in a reconfigurable manner, providing an insight into new possibilities for reconfigurable photonic processing circuits.

physics.optics↗

arXiv2024

Neural Operators Meet Energy-based Theory: Operator Learning for Hamiltonian and Dissipative PDEs

The operator learning has received significant attention in recent years, with the aim of learning a mapping between function spaces. Prior works have proposed deep neural networks (DNNs) for learning such a mapping, enabling the learning of solution operators of partial differential equations (PDEs). However, these works still struggle to learn dynamics that obeys the laws of physics. This paper proposes Energy-consistent Neural Operators (ENOs), a general framework for learning solution operators of PDEs that follows the energy conservation or dissipation law from observed solution trajectories. We introduce a novel penalty function inspired by the energy-based theory of physics for training, in which the energy functional is modeled by another DNN, allowing one to bias the outputs of the DNN-based solution operators to ensure energetic consistency without explicit PDEs. Experiments on multiple physical systems show that ENO outperforms existing DNN models in predicting solutions from data, especially in super-resolution settings.

stat.ML↗

arXiv2024

OptiState: State Estimation of Legged Robots using Gated Networks with Transformer-based Vision and Kalman Filtering

State estimation for legged robots is challenging due to their highly dynamic motion and limitations imposed by sensor accuracy. By integrating Kalman filtering, optimization, and learning-based modalities, we propose a hybrid solution that combines proprioception and exteroceptive information for estimating the state of the robot's trunk. Leveraging joint encoder and IMU measurements, our Kalman filter is enhanced through a single-rigid body model that incorporates ground reaction force control outputs from convex Model Predictive Control optimization. The estimation is further refined through Gated Recurrent Units, which also considers semantic insights and robot height from a Vision Transformer autoencoder applied on depth images. This framework not only furnishes accurate robot state estimates, including uncertainty evaluations, but can minimize the nonlinear errors that arise from sensor measurements and model simplifications through learning. The proposed methodology is evaluated in hardware using a quadruped robot on various terrains, yielding a 65% improvement on the Root Mean Squared Error compared to our VIO SLAM baseline. Code example: https://github.com/AlexS28/OptiState

cs.RO↗

arXiv2024

Meta-Learning for Neural Network-based Temporal Point Processes

Human activities generate various event sequences such as taxi trip records, bike-sharing pick-ups, crime occurrence, and infectious disease transmission. The point process is widely used in many applications to predict such events related to human activities. However, point processes present two problems in predicting events related to human activities. First, recent high-performance point process models require the input of sufficient numbers of events collected over a long period (i.e., long sequences) for training, which are often unavailable in realistic situations. Second, the long-term predictions required in real-world applications are difficult. To tackle these problems, we propose a novel meta-learning approach for periodicity-aware prediction of future events given short sequences. The proposed method first embeds short sequences into hidden representations (i.e., task representations) via recurrent neural networks for creating predictions from short sequences. It then models the intensity of the point process by monotonic neural networks (MNNs), with the input being the task representations. We transfer the prior knowledge learned from related tasks and can improve event prediction given short sequences of target tasks. We design the MNNs to explicitly take temporal periodic patterns into account, contributing to improved long-term prediction performance. Experiments on multiple real-world datasets demonstrate that the proposed method has higher prediction performance than existing alternatives.

cs.LG↗

arXiv2023

SCALER: Versatile Multi-Limbed Robot for Free-Climbing in Extreme Terrains

This paper presents SCALER, a versatile free-climbing multi-limbed robot that is designed to achieve tightly coupled simultaneous locomotion and dexterous grasping. While existing quadrupedal-limbed robots have demonstrated impressive dexterous capabilities, achieving a balance between power-demanding locomotion and precise grasping remains a critical challenge. We design a torso mechanism and a parallel-serial limb to meet the conflicting requirements that pose unique challenges in hardware design. SCALER employs underactuated two-fingered GOAT grippers that can mechanically adapt and offer seven modes of grasping, enabling SCALER to traverse extreme terrains with multi-modal grasping strategies. We study the whole-body approach, where SCALER utilizes its body and limbs to generate additional forces for stable grasping in various environments, thereby further enhancing its versatility. Furthermore, we improve the GOAT gripper actuation speed to realize more dynamic climbing in a closed-loop control fashion. With these proposed technologies, SCALER can traverse vertical, overhanging, upside-down, slippery terrains and bouldering walls with non-convex-shaped climbing holds under the Earth's gravity.

cs.RO↗

arXiv2023

Meta-learning of Physics-informed Neural Networks for Efficiently Solving Newly Given PDEs

We propose a neural network-based meta-learning method to efficiently solve partial differential equation (PDE) problems. The proposed method is designed to meta-learn how to solve a wide variety of PDE problems, and uses the knowledge for solving newly given PDE problems. We encode a PDE problem into a problem representation using neural networks, where governing equations are represented by coefficients of a polynomial function of partial derivatives, and boundary conditions are represented by a set of point-condition pairs. We use the problem representation as an input of a neural network for predicting solutions, which enables us to efficiently predict problem-specific solutions by the forwarding process of the neural network without updating model parameters. To train our model, we minimize the expected error when adapted to a PDE problem based on the physics-informed neural network framework, by which we can evaluate the error even when solutions are unknown. We demonstrate that our proposed method outperforms existing methods in predicting solutions of PDE problems.

stat.ML↗