dense tactile simulation

This study investigates how the type, layout, and resolution of tactile sensors influence performance and cost in dexterous manipulation. To this end, we develop a GPU-parallel simulation platform supporting multimodal tactile sensing—including contact, force/torque, elastomer displacement, audio, and temperature—augmented with realistic noise models. We introduce a voxelized temperature field to unify multimodal interfaces and enable efficient training of over 20,000 environments and 1,000 taxels on a single GPU. Through teacher-student policy distillation and real-to-sim transfer validation, we find that sensor coverage is far more critical than resolution: full-hand coverage significantly outperforms fingertip-only sensing, with force/torque information per taxel proving most practically valuable. The learned policies successfully transfer to the physical XHand1 platform.

High-fidelity tactile modeling conflicts with real-time performance, and existing methods rely heavily on scarce real-world labeled data. Method: This paper introduces the first hydroelastic contact mechanics–based tactile sensor simulation framework, enabling continuous pressure-field modeling and sensor signal synthesis for soft–soft and soft–hard non-convex surface interactions. Implemented as an efficient, plugin-based extension in MuJoCo, it ensures physical fidelity via pressure-surface discretized integration while maintaining computational efficiency. Contribution/Results: The method achieves zero-shot sim-to-real transfer—using only synthetic data, it significantly improves real-sensor performance in object state estimation. It overcomes the long-standing accuracy–speed trade-off inherent in point-contact and finite-element approaches. The code is open-sourced and integrated into the MuJoCo ecosystem.

This study addresses the challenge of high-fidelity tactile rendering of 3D geometric shapes in virtual reality, a task hindered by limited spatiotemporal resolution in existing vibrotactile approaches. Inspired by the natural deformation of the fingertip during object contact, this work proposes a novel parametric tactile rendering method that integrates fingerpad deformation modeling into electro-tactile feedback design. Leveraging a finger-worn electro-tactile interface, the system dynamically generates spatiotemporal tactile patterns based on interaction states—such as proximity, contact, and sliding—and geometric context, including shape features and surface textures. A user study (N=24) demonstrates that the proposed approach significantly outperforms baseline systems in both texture discrimination and geometric feature recognition, offering a new paradigm and practical guidance for the development of high-fidelity tactile interfaces.

Robotic hands struggle with human-like adaptive manipulation in dynamic environments due to insufficient tactile feedback. Method: We propose F-TAC, a biomimetic robotic hand that—while preserving full joint degrees of freedom—integrates a high-density, flexible tactile sensor array covering 70% of the hand surface with 0.1 mm spatial resolution. Combining biomechanics-informed structural optimization and generative hand-configuration synthesis, we establish an embodied tactile closed-loop control framework. Results: In 600 real-world dynamic grasping trials, F-TAC significantly outperforms a non-tactile baseline (p < 0.0001), demonstrating robust adaptive adjustment. This work reveals the critical role of high-fidelity embodied tactile sensing in enhancing robotic intelligent behavior and establishes a scalable tactile-motor co-design paradigm for embodied intelligence.

The field of tactile surface modeling and data representation has long lacked a systematic treatment from a signal processing perspective, resulting in technical fragmentation across modeling, acquisition, rendering, and perception. Method: This paper presents the first unified taxonomy, comparative analysis, and critical review of tactile surface modeling methods—organized along core signal processing dimensions: signal sampling, feature extraction, compression coding, geometry–physics hybrid modeling, and cross-modal representation. Contribution/Results: We construct an end-to-end technical roadmap, identifying six fundamental modeling paradigms and three critical representation bottlenecks: (1) resolution–bandwidth trade-offs, (2) insufficient physical fidelity, and (3) challenges in cross-modal alignment. Our framework provides a scalable theoretical foundation and concrete technical pathways toward standardized tactile modeling, real-time rendering, and closed-loop perception.

This work addresses the limitations of conventional host-based visual-tactile processing—namely high power consumption, substantial transmission latency, and non-deterministic scheduling—which hinder robotic perception and response performance. The authors propose a near-sensor computing framework that enables deterministic, low-latency tactile reconstruction directly at the sensor without relying on data-dependent branching or iterative convergence. The architecture integrates a fully pipelined hardware design, a spectral Poisson solver, fixed-point arithmetic, and on-chip decision logic, operating at 166 MHz. Experimental results demonstrate a first-pixel latency of only 0.211 ms per frame, a reconstruction error of 0.17% relative to peak depth, a protective reflex closed-loop latency of 28.3 ± 4.9 ms, and a power consumption of merely 347 mW.

This work addresses the challenge of achieving physically accurate simulation of optical tactile sensors, which exhibit high deformability and complex optical responses that are difficult to model. The authors propose DOT-Sim, the first framework to employ differentiable Material Point Method (MPM) for elastodynamic modeling of soft tactile sensors, enabling accurate simulation of large-scale nonlinear deformations. By calibrating the optical response with minimal real-world data and learning residual images to efficiently approximate visual outputs, DOT-Sim achieves real-to-sim alignment within minutes. In zero-shot transfer settings, the method attains 85% accuracy in complex object classification and 90% accuracy in tumor-type detection, while maintaining a mean trajectory tracking error below 0.9 mm.

This work addresses the challenge faced by blind and low-vision students in accessing statistical graphics, a barrier exacerbated by the inefficiency and specialized CAD expertise required by conventional 3D printing approaches, which hinder classroom-scale deployment. To overcome this, the authors propose a reusable, three-tier software pipeline that, for the first time, automatically integrates haptic perceptual parameters into the generation workflow. Leveraging a multimodal large language model to parse chart structures directly from images, the system supports automated tactile rendering of scatter plots, bar charts, histograms, line graphs, and box plots. Implemented in JavaScript with a modular architecture, the pipeline produces print-ready STL files in under 250 milliseconds, substantially lowering production barriers and enabling educators to rapidly review and deploy accessible data representations, thereby enhancing data accessibility in inclusive education.

This work addresses the limitations of existing social touch sensing designs, which often rely on predefined configurations and lack empirical grounding for coverage of diverse touch behaviors and spatial layout. The authors propose a需求-driven design paradigm, leveraging a virtual reality platform with haptic feedback to collect full-body social touch data across multiple scenarios. Through high-resolution contact analysis and user studies, they identify nine common social touch gestures and release an open-source dataset comprising 5,520 interactions. Building on this empirical foundation, the study provides the first quantitative guidelines for tactile skin coverage and sensor density required on humanoid robots, establishing transferable design benchmarks applicable to robots of varying morphologies.

This work addresses the challenge of tightly coupling high-fidelity physical simulation with photorealistic real-time rendering in contact-rich robotic systems, particularly in modeling deformation and tactile perception. The authors present the first deep integration of GPU-accelerated Incremental Potential Contact (IPC) into the IsaacSim/Isaac Lab platform and introduce the Geometric Mortar Contact Potential (GMCP) to more accurately capture contact pressure distributions on tactile surfaces. By establishing a deformation mapping mechanism between simulation and visual meshes, the approach enables synchronized physics simulation and rendering in scenarios involving rigid–soft interactions. Experiments demonstrate the method’s effectiveness across multiple contact benchmarks and its successful application to high-fidelity, real-time simulation and data generation for quadrupedal robots, dexterous hands, and UMI grippers.