Searcharxiv⌕ Search

arXiv subjects

Yina Jian

Publications and source records attributed to Yina Jian.

3 recordsLinked to original sources

Seeing What Execution Needs: Benchmarking Task-Conditioned Actionable Spatial Belief for Manipulation

When should a robot look at the cup, and when at the portafilter? A view that is informative for one task can be useless for another, even at the same workstation. We introduce Seeing What Execution Needs, a benchmark for task-conditioned actionable spatial belief in simulated tabletop manipulation. Its aligned three-domain track contains 1,080 frozen cases (360 each for coffee workstation, drawer/cabinet, and microwave/appliance). The primary coffee protocols -- closed-loop actionability and same-object functional out-of-distribution generalization -- produce over 115,000 evaluator episodes across evaluated methods, budgets, and inference repeats. Each case pairs candidate RGB views with a manipulation command, shared scripted skills, and hidden postconditions, directly measuring whether selected observations support execution. A metadata-assisted diagnostic reference pipeline maintains spatial-functional beliefs and ranks views by missing evidence. At primary budget settings, it reaches 98.8% closed-loop and 98.9% functional-OOD success; removing functional fields or command binding reduces functional-OOD success to 0.0% and 19.7%, respectively. Under a separate full-preview contract, seven vision-language models reach 26-43%. Replacing coverage metadata with frozen RGB-derived scores retains 98.5%/98.9% success; shuffling these scores reduces success to 15%/11%, supporting dependence on image content. Separately, the aligned three-domain evaluation at H=3 supports portability of the benchmark interface under frozen domain-specific schemas. These results suggest that spatial belief for manipulation should be evaluated by whether acquired evidence meets the current command's execution requirements.

cs.RO↗

II-NVM: Enhancing Map Accuracy and Consistency with Normal Vector-Assisted Mapping

SLAM technology plays a crucial role in indoor mapping and localization. A common challenge in indoor environments is the "double-sided mapping issue", where closely positioned walls, doors, and other surfaces are mistakenly identified as a single plane, significantly hindering map accuracy and consistency. To address this issue this paper introduces a SLAM approach that ensures accurate mapping using normal vector consistency. We enhance the voxel map structure to store both point cloud data and normal vector information, enabling the system to evaluate consistency during nearest neighbor searches and map updates. This process distinguishes between the front and back sides of surfaces, preventing incorrect point-to-plane constraints. Moreover, we implement an adaptive radius KD-tree search method that dynamically adjusts the search radius based on the local density of the point cloud, thereby enhancing the accuracy of normal vector calculations. To further improve realtime performance and storage efficiency, we incorporate a Least Recently Used (LRU) cache strategy, which facilitates efficient incremental updates of the voxel map. The code is released as open-source and validated in both simulated environments and real indoor scenarios. Experimental results demonstrate that this approach effectively resolves the "double-sided mapping issue" and significantly improves mapping precision. Additionally, we have developed and open-sourced the first simulation and real world dataset specifically tailored for the "double-sided mapping issue".

cs.RO↗

Regularized Maximum Likelihood Image Synthesis and Validation for ALMA Continuum Observations of Protoplanetary Disks

Regularized Maximum Likelihood (RML) techniques are a class of image synthesis methods that achieve better angular resolution and image fidelity than traditional methods like CLEAN for sub-mm interferometric observations. To identify best practices for RML imaging, we used the GPU-accelerated open source Python package MPoL, a machine learning-based RML approach, to explore the influence of common RML regularizers (maximum entropy, sparsity, total variation, and total squared variation) on images reconstructed from real and synthetic ALMA continuum observations of protoplanetary disks. We tested two different cross-validation (CV) procedures to characterize their performance and determine optimal prior strengths, and found that CV over a coarse grid of regularization strengths easily identifies a range of models with comparably strong predictive power. To evaluate the performance of RML techniques against a ground truth image, we used MPoL on a synthetic protoplanetary disk dataset and found that RML methods successfully resolve structures at fine spatial scales present in the original simulation. We used ALMA DSHARP observations of the protoplanetary disk around HD 143006 to compare the performance of MPoL and CLEAN, finding that RML imaging improved the spatial resolution of the image by up to a factor of 3 without sacrificing sensitivity. We provide general recommendations for building an RML workflow for image synthesis of ALMA protoplanetary disk observations, including effective use of CV. Using these techniques to improve the imaging resolution of protoplanetary disk observations will enable new science, including the detection of protoplanets embedded in disks.

astro-ph.EP↗