Search papers, labs, and topics across Lattice.
100 papers published across 9 labs.
Real-time trajectory adaptation in UAVs can now effectively balance safety, efficiency, and dynamic risks, setting a new standard for autonomous inspection.
Sparse graph policies not only outperform traditional image-based methods but also expose hidden dataset biases, enhancing both performance and interpretability in robot learning.
Automated architecture search can yield substantial performance gains for embodied agents, but also reveals critical limitations that challenge its efficacy in real-world applications.
Achieving a 95.8% success rate in resolving self-collisions, PoseShield transforms how we handle human pose estimation under extreme articulations.
Transforming mistakes into actionable knowledge, R²LPL achieves state-of-the-art performance in autonomous driving with minimal retraining.
Automated architecture search can yield substantial performance gains for embodied agents, but also reveals critical limitations that challenge its efficacy in real-world applications.
Achieving a 95.8% success rate in resolving self-collisions, PoseShield transforms how we handle human pose estimation under extreme articulations.
Transforming mistakes into actionable knowledge, R²LPL achieves state-of-the-art performance in autonomous driving with minimal retraining.
Near-zero forgetting in motion-language agents is achievable, but only with careful expert isolation during task transitions.
Motion artifacts no longer spell doom for heart and respiratory rate estimation—this new method disentangles vital signs from noise with remarkable precision.
A novel duty cycle predictive MPCC reduces current harmonics in EV charging by over 75%, tackling a critical grid integration challenge.
Adaptive imagination can transform limited target data into reliable rollouts, enabling robust sim-to-real transfer in visual reinforcement learning.
State-aware tokenization can double the success rate in robotic manipulation tasks by adapting actions to the robot's current state.
The Data Centre serves as the true body of AI, embodying human desires while remaining devoid of its own, complicating our understanding of intelligence in the context of Capital.
GaussDet achieves a remarkable 16.7% boost in referential grounding accuracy, redefining the capabilities of open-vocabulary 3D scene understanding.
Achieving robust pose tracking and memory efficiency, KiloGS-SLAM can handle over 10,000 frames in challenging outdoor environments without sacrificing accuracy or detail.
A novel transformer framework achieves superior assessment of rehabilitation exercises by effectively extracting and utilizing joint position features from RGBD data.
Real-world deployment of VLA systems reveals that the success lies in the meticulous management of the entire data-model-control pipeline rather than just model improvements.
ZR-0 achieves seamless cross-embodiment transfer in robotic manipulation by aligning high-level cognitive processes through innovative ECoT supervision.
OWMDrive's innovative use of a 4D Occupancy World Model allows for foresighted trajectory planning that adapts to dynamic traffic conditions, enhancing safety and reliability.
Sparse graph policies not only outperform traditional image-based methods but also expose hidden dataset biases, enhancing both performance and interpretability in robot learning.
Achieving state-of-the-art underwater image enhancement with a model that has only 4.23K parameters and processes images at over 600 FPS.
Predicting diverse human movement goals is now possible with a generative model that captures the stochastic nature of behavior in real-time environments.
CylindTrack achieves superior identity preservation in panoramic multi-object tracking by effectively modeling depth consistency and leveraging the unique topology of equirectangular images.
HUMEMBR achieves superior long-horizon reasoning about human behavior with fewer tokens, revolutionizing how robots navigate human-centered environments.
SPARK outperforms traditional code-generation agents by more than doubling their success rates, showcasing a new paradigm in robotic manipulation that prioritizes perception over costly re-querying.
The Unscented Kalman Filter outperforms traditional methods in estimating wind velocity, especially in highly nonlinear conditions, ensuring accurate data for critical applications like wildfire response.
Urban facade reconstruction can achieve superior geometric accuracy by integrating lightweight structural supervision, overcoming common pitfalls of traditional methods.
GeoEdit achieves unprecedented geometric accuracy and identity fidelity in object editing, overcoming the limitations of existing diffusion-based methods.
Agents using a structured memory framework can achieve significantly better manipulation performance, outperforming traditional methods in task completion and skill generalization.
Fine-grained contact states can be distinguished through the dynamic correlation of tactile motion, transforming how we approach contact-rich manipulation in robotics.
A novel graspability field allows robots to autonomously determine optimal object configurations for grasping, eliminating the need for predefined poses.
RoamFlow achieves efficient image-goal navigation with reduced inference latency while outperforming traditional methods in trajectory generation.
Extracting contact event sequences from real demonstrations can significantly enhance sim-to-real transfer, making robotic manipulation more reliable and effective.
Robots can now learn new tasks on-the-fly from just one demonstration, revolutionizing how we teach machines to manipulate their environments.
LLM-driven self-evolving architectures can outperform expert-designed models in robotic tactile perception tasks, achieving unprecedented levels of performance and diversity.
Normalizing flow-enhanced message passing boosts multirobot localization accuracy and adaptability, outperforming traditional methods in real-world scenarios.
TACO achieves over 90% success in robustly optimizing SLAM trajectories, even in the presence of up to 50% outlier measurements.
Achieving a mean nozzle-position error below 10μm while eliminating all collision violations represents a breakthrough in collision-aware trajectory optimization for robotic additive manufacturing.
CSAR revolutionizes robotics software development by providing a robust framework that ensures dependency isolation and reproducibility in complex distributed environments.
CI-MSE dramatically improves the correlation between offline validation and real-world performance, making it a game-changer for robot policy evaluation.
Self-supervised learning can transform noisy LiDAR SLAM into a robust system by recursively refining local geometric representations.
Large-scale human motion data can now be effectively repurposed to teach diverse non-humanoid robots, unlocking new capabilities in locomotion and manipulation.
AUSLUN achieves superior localization and search efficiency in GNSS-denied environments, outperforming traditional methods by leveraging innovative scanning and guidance techniques.
Models trained on small tether configurations can accurately control larger, unseen systems, showcasing unprecedented spatial transfer capabilities.
Achieving fleet-wide localization accuracy without sharing sensitive position data could redefine privacy standards in multi-robot systems.
A novel cross-spectral VTI system achieves superior accuracy and robustness in challenging environments by dynamically balancing visual and thermal data reliance.
Users can now intuitively grasp a robot's inferred goals through its motion, reducing control effort and enhancing collaboration.
Multi-UAV formations can now navigate complex environments safely while dynamically adapting their shape to avoid obstacles, all without losing cohesion.
Lateral string stability can make or break the safety of CAV platoons, and V2V communication is the game-changer that ensures error attenuation.
STEAM redefines how robots learn from mixed-quality data, achieving up to 59% higher success rates in real-world tasks by effectively identifying reliable progress.
FalconTrack automates the generation of photorealistic labeled data, achieving 100% success in real-world tracking while traditional methods falter under pressure.
Achieving RGB-D performance with only monocular input, MyGO-Splat revolutionizes SLAM by integrating closed-loop geometric feedback for enhanced scale stability.
Achieving accurate 6-DoF pose estimation for fixed-wing UAVs without relying on CAD models could revolutionize UAV operations in complex environments.
A soft robotic arm can achieve precise control of compliance and position simultaneously, enabling it to excel in dynamic and unstructured environments.
HTT enables tactile learning across diverse sensors, achieving adaptability that was previously unattainable in contact-rich manipulation tasks.
Articulated 3D object reconstruction can now achieve high fidelity and internal structure recovery from mere text or images, thanks to a debate-driven agentic approach.
KYON's innovative design allows it to seamlessly switch between wheeled and legged locomotion, enhancing its adaptability in diverse environments.
WARP achieves zero-shot whole-body mobile manipulation from offline human demonstrations, eliminating the reliance on teleoperation data.
Chronos redefines how we leverage historical context in robot manipulation, achieving a 6.6x performance boost with a fraction of the parameters.
Digital vehicle forensics faces unique challenges that could compromise investigations, but a new framework helps prioritize evidence sources effectively.
Searching in a compressed latent space can revolutionize motion planning by enabling flexible optimization of any objective function on-the-fly.
TAPE achieves a 4.1% reduction in travel distance while ensuring tether safety, revolutionizing autonomous exploration in complex 3D environments.
Task-driven mapping with GaussLite boosts 3D reconstruction quality by over 2.7 dB while optimizing compute resources for robotic tasks.
A soft robotic tail enables quadruped robots to perform complex contamination surveys in hazardous environments without human intervention.
Grasp datasets can revolutionize robotic dexterity, enabling significant improvements in articulated tool use performance.
Feasible solutions to TWTL specifications not only guarantee satisfaction but also optimize control inputs through a novel MILP approach that adapts to task dynamics in real-time.
GROW$^2$ enables robots to creatively use open-category tools with zero-shot generalization, outperforming existing methods in both simulated and real-world tasks.
SubEdge slashes latency by nearly half during subscriber mobility, ensuring uninterrupted AI service delivery in 6G networks.
Achieving a 21.2% improvement in hypervolume, the CIMORL framework redefines how multi-robot systems can optimize competing objectives without sacrificing coordination.
Real-time trajectory adaptation in UAVs can now effectively balance safety, efficiency, and dynamic risks, setting a new standard for autonomous inspection.
Adapting pretrained policies with just a modest multisensory dataset can enhance robot manipulation performance across diverse tasks without sacrificing prior knowledge.
ActiveVital achieves vital signs monitoring accuracy comparable to static methods by actively controlling radar alignment, reducing respiration interval error from 0.87 s to 0.14 s.
OpenSPM achieves 85.6% task success with a control frequency of 1033.3 Hz, revolutionizing high-frequency robotic manipulation with minimal computational demands.
Discrete quadrotor models paired with UKF can achieve superior wind velocity estimation accuracy, even with low-cost sensors.
Quadrotors can now navigate to specific objects in images by intelligently selecting viewpoints, significantly enhancing their operational capabilities in complex environments.
SWAM achieves superior navigation performance by seamlessly integrating observation and action generation, significantly enhancing efficiency and accuracy in embodied tasks.
ReactiveBFM enables humanoids to achieve zero-shot moving target reaching with unprecedented agility and real-time adaptability.
AERIS enables real-time, role-driven intelligence for aerial robots, adapting dynamically to resource constraints while maintaining high-performance navigation capabilities.
REPAIR-Bench reveals that understanding user recovery strategies can dramatically improve robot interaction reliability and adaptability.
Sphere-VIO achieves unprecedented accuracy and efficiency in visual-inertial odometry for heterogeneous multi-camera systems, setting a new standard for real-time performance.
Synthesizing 48,000 interaction trajectories without human input enables a humanoid robot to learn complex loco-manipulation tasks effectively.
It is concluded that the future of DRL in autonomous systems lies in the development of sample-efficient, verifiably safe, and generalizable algorithms, guiding future work toward closing the gap between simulated success and real-world deployment.
LLM-driven robots provide facilitation-oriented explanations for their interventions, revealing nuanced roles that shape group interactions.
Ghosting artifacts can be significantly reduced by treating dynamic objects as persistent identities rather than fleeting appearances in monocular mapping.
SPACE achieves four to seventeen times fewer inter-robot collisions compared to traditional greedy planners, redefining efficiency in large-scale swarm exploration.
The first detailed closed-form representation of the covariance matrix for the kinematic bicycle model reveals critical insights into vehicle pose estimation under uncertainty.
Visual goal prototypes can boost robot manipulation success rates by up to 17% compared to text-based instructions, revealing the power of leveraging action-free demonstrations.
Robust fusion of low-cost MEMS sensors can rival tactical-grade inertial navigation systems, transforming land navigation capabilities.
A broader class of degenerate trajectories can obscure the identifiability of delay and initial navigation states, complicating aided inertial navigation systems.
Causal Spectral Policy achieves superior performance in precision-sensitive tasks by effectively separating motion intent from execution details, revealing a new paradigm in hierarchical policy learning.
LAMP can tackle complex multi-robot manipulation tasks in cluttered spaces that previous methods fail to solve, showcasing a new frontier in robotic collaboration.
Achieving decimeter-level positioning accuracy with low-cost single-frequency GNSS receivers could revolutionize autonomous navigation by eliminating the need for expensive infrastructure.
MTD-Map achieves robust dynamic object removal and change detection in a single framework, cutting computational costs without sacrificing performance.
PL-LIT achieves state-of-the-art performance in thermal SLAM, even in challenging environments where traditional methods falter.
Event-VLA achieves robust robotic manipulation even in near-dark conditions by effectively integrating motion-sensitive event streams with traditional RGB inputs.
AnyBody enables humanoid robots to seamlessly control movements from any chosen subset of body keypoints, revolutionizing how we approach humanoid locomotion and manipulation.
Achieving zero-shot sim-to-real transfer, the CORE Planner reduces travel distance by up to 48% compared to existing learning-based methods, revolutionizing robot navigation in unknown environments.
Accurate force estimation in continuum robots can now be achieved even in complex multi-contact scenarios, outperforming traditional methods.
Touch is not just an add-on; it fundamentally enhances object representation, leading to dramatic improvements in physical property estimation and manipulation tasks.
Action-conditioned world modeling can yield reusable dynamics priors that enhance robot learning across both simulation and real-world applications.
Bridging human and robot manipulation through wrist translation leads to a dramatic improvement in skill transfer efficiency, outpacing traditional methods reliant on noisy hand poses.
Policies trained in SimFoundry's automated environments achieve up to 40% higher success rates in real-world tasks by leveraging affordance-preserving scene variations.
Physically aligned video models can boost robotic manipulation success rates by over 50% compared to traditional methods.
Language components in VLA models are often redundant, allowing for significant performance gains by reducing their size without sacrificing control quality.
