contact event detection

To address insufficient pose estimation robustness in tight-tolerance robotic assembly caused by point-cloud noise and contact uncertainty, this paper proposes a multimodal perception method integrating global geometry and local contact information. Our approach features three key contributions: (1) a novel force/torque inversion algorithm based on rejection sampling, enabling high-precision estimation of end-effector contact location; (2) embedding contact observations into the Stochastic Poisson Surface Reconstruction (SPSR) framework to construct an online-updatable Stochastic Poisson Surface Map (SPSMap) with explicit uncertainty modeling; and (3) tightly coupled fusion of point-cloud and six-axis force/torque sensor data. Simulation results demonstrate significant improvement in contact localization accuracy. Real-world peg-in-hole experiments show a 42% reduction in hole pose estimation error and a 76% decrease in insertion failure rate.

Accurately detecting the precise temporal moments of hand–object contact in first-person videos is challenging due to subtle motions and frequent occlusions. This work proposes the Hand-informed Context Enhanced (HiCE) module, which integrates spatiotemporal features from both hand regions and their surrounding context, augmented with a cross-attention mechanism to model latent contact patterns. To further refine temporal discrimination, the authors introduce a grasp-aware loss function and a soft-label training strategy. Evaluated on TouchMoment—a newly curated large-scale dataset comprising 8,456 annotated contact moments—the proposed method achieves a 16.91% improvement in average precision over existing event localization approaches under a two-frame tolerance metric, substantially advancing the accuracy of contact moment detection.

Real-time 6-DoF pose tracking of unknown dynamic objects in frequent-contact scenarios—such as dexterous in-hand manipulation—is challenged by severe occlusion, motion blur, and transient impact forces. Method: We propose a bidirectional Real2Sim/Sim2Real framework that jointly integrates visual observations with differentiable contact-aware physical modeling. First, an initial visual reconstruction is refined online via contact dynamics optimization to estimate geometry and physical properties. Subsequently, the learned contact physics model adaptively fuses visual tracking outputs, enhancing physical consistency and robustness. Contributions/Results: The method incorporates a GPU-accelerated physics engine, a multimodal adaptive fusion architecture, online collision geometry updating, and an end-to-end differentiable visual feature extraction network. Evaluated on drop-impact and dexterous manipulation tasks, it achieves >20 Hz real-time tracking, outperforming pure vision-based and conventional filtering approaches in both accuracy and robustness.

Existing evaluation metrics for visual object tracking lack a comparable, continuous-time measure for the trajectory function of time (FoT), relying instead on discrete-frame assessments that fail to characterize arbitrary-time states or disentangle distinct error types (e.g., localization, false positives, missed detections). Method: We propose Star-ID—the first spatiotemporally aligned trajectory integral distance—defining a rigorous, comparable FoT metric over continuous spacetime. Star-ID strictly distinguishes temporally aligned versus misaligned trajectory segments and analytically decouples detection and localization errors. It introduces time-averaged metrics and a theoretical error decomposition model, supported by a multi-object numerical validation framework. Contribution/Results: We provide formal theoretical analysis and demonstrate—via both single- and multi-object simulations—that Star-ID significantly enhances physical interpretability and fine-grained discriminative power in tracking evaluation, enabling precise, continuous-time performance assessment.

High-precision online estimation algorithms for robotics are highly sensitive to sensor timestamp accuracy; however, existing synchronization solutions struggle to simultaneously achieve real-time operation, low cost, and high temporal precision. To address this, we propose a real-time, trigger-based time synchronization system built on commodity hardware. Our approach employs a hardware-triggered mechanism to jointly schedule heterogeneous sensors operating at different frequencies, and integrates an enhanced clock synchronization protocol with nanosecond-resolution timestamping to ensure precise coordination between sensors and the onboard computer. Crucially, the system eliminates reliance on expensive dedicated timing hardware, thereby substantially mitigating the impact of timing errors on online estimation. Experimental evaluation on a physical robot platform demonstrates sub-microsecond synchronization accuracy, along with significant improvements in both estimation robustness and real-time performance.

Accurately localizing the first point of contact (FPOC) between players and dummies in untrimmed American football training videos is highly challenging due to camera motion, scene clutter, visual similarity among actors, and rapid pose changes. This work proposes a zero-shot, training-free method that decouples interaction candidate discovery from pixel-level contact verification for the first time. It leverages Grounding DINO to generate interaction candidates, employs motion-aware temporal reasoning to filter potential contact frames, and introduces SAM2 for explicit pixel-level contact confirmation. The approach operates without reliance on detection confidence scores or annotated data, achieving successful FPOC estimation in 97.4% of 738 videos, with 77.5% of predictions within ±10 frames and 82.7% within ±20 frames of the ground truth, substantially improving both robustness and precision.

This work addresses the challenge of reconstructing tool-tip contact trajectories when tactile sensors are located far from the point of interaction, making direct perception infeasible. Framing the problem as a geometric reasoning task under a single-point contact assumption, the method fuses tactile and proprioceptive data to first estimate the global contact location via a calibration phase and then reconstructs the full trajectory using sequential tactile marker observations. To the best of our knowledge, this is the first approach to achieve stable and robust tool-tip trajectory estimation using only grasp-level tactile information, demonstrating adaptability across diverse tool types, wrist poses, and grasp configurations. Experimental results show a trajectory root-mean-square error (RMSE) of 8.59 ± 2.41 mm and a shape RMSE of 5.96 ± 1.16 mm in world coordinates, with an online processing rate of 14.00 ± 4.11 Hz.

This work addresses the challenge of robots failing in fine manipulation tasks due to their inability to perceive transient, weak external contacts. The authors propose TECDAR, a method that integrates a miniature 6D inertial measurement unit at the gripper tip to enable high-speed dynamic tactile sensing. By fusing this tactile data with robot pose information, TECDAR achieves real-time contact detection and localization. Leveraging an ultra-compact 6D tactile sensor operating at 7 kHz, the approach realizes sub-millisecond transient contact detection and ranging while requiring only low data bandwidth, enabling millisecond-scale trajectory correction and millimeter-level positioning accuracy. Experiments demonstrate that TECDAR achieves an average localization accuracy of 7 mm within 180 ms for both point and line contact tasks, supporting precise, purely tactile-driven manipulation and environmental perception.

This work addresses the challenge of performing precise physical interaction tasks—such as sitting on a chair or pushing furniture—with humanoid robots, which cannot be reliably achieved through keypoint tracking alone. The authors propose a novel approach that decouples geometric pose from contact behavior, enabling explicit control over contact initiation or suppression via part-level binary contact commands jointly optimized with keypoint trajectories. The method incorporates a contact-following reward and trajectory augmentation strategy, facilitating diverse physical interactions without requiring task-specific reward engineering and supporting effective sim-to-real transfer. In simulation, the approach outperforms pure keypoint-based methods across ten distinct tasks, and real-world experiments successfully demonstrate controllable contact execution in five of these tasks.

This work proposes a low-latency, high-accuracy method for contact angle estimation using the event-based tactile sensor NeuroTac to meet the demands of real-time tactile perception in complex robotic interactions. By constructing and systematically evaluating static, dynamic, and fused spatial contour representations, the study demonstrates that the static representation achieves significantly superior accuracy and robustness. The approach attains mean absolute errors of 0.160° and 0.251° during continuous rolling and stationary phases, respectively, and exhibits strong adaptability to variations in sliding velocity and indentation depth. Furthermore, the end-to-end processing latency remains below 10 ms at the 99th percentile (P99), making it well-suited for high-speed robotic manipulation under varying operational conditions.