Task planning under uncertainty is essential for home-service robots operating in the real world. Tasks involve ambiguous human instructions, hidden or unknown object locations, and open-vocabulary object types, leading to significant open-ended uncertainty and a boundlessly large planning space. To address these challenges, we propose Tru-POMDP, a planner that combines structured belief generation using Large Language Models (LLMs) with principled POMDP planning. Tru-POMDP introduces a hierarchical Tree of Hypotheses (TOH), which systematically queries an LLM to construct high-quality particle beliefs over possible world states and human goals. We further formulate an open-ended POMDP model that enables rigorous Bayesian belief tracking and efficient belief-space planning over these LLM-generated hypotheses. Experiments on complex object rearrangement tasks across diverse kitchen environments show that Tru-POMDP significantly outperforms state-of-the-art LLM-based and LLM-tree-search hybrid planners, achieving higher success rates with significantly better plans, stronger robustness to ambiguity and occlusion, and greater planning efficiency.

Object detection is a crucial topic in computer vision. Mask Region‐Convolution Neural Network (R‐CNN) based methods, wherein a large intersection over union (IoU) threshold is chosen for high quality samples, have often been employed for object detection. However, the detection performance of such methods deteriorates when samples are reduced. To address this, the authors propose an improved Mask R‐CNN‐based method: the ResNet Group Cascade (RGC) Mask R‐CNN. First, they compared ResNet with different layers, finding that ResNeXt‐101‐64 × 4d is superior to other backbone networks. Secondly, during the training of the test model, the performance of Mask R‐CNN suffered from a small batch processing scale, resulting in inaccurately calculated mean and variance; thus, group normalisation was added to the backbone, feature pyramid network neck and bounding box head of the network. Finally, the higher the intersection of Mask R‐CNN than the threshold, the easier it is to obtain high‐quality samples. However, blindly selecting a high threshold leads to sample reduction and overfitting. Thus, a proposed cascade network configuration with three IoU thresholds was utilised in the process of model training. The model was trained and tested on the COCO and PASCAL VOC07 datasets. Their proposed algorithm demonstrated superior performance compared to that of the Mask R‐CNN.

In order to solve the problem of data shortages resulting from long revisit period and cloud contamination in high resolution imagery, this paper proposed a fusion model for generating high spatial and temporal resolution images based on semi-physical method and improved ESTARFM(enhanced spatial and temporal adaptive reflectance fusion model). The result shows that the new model can produce more precise reflectance than ESTARFM and take full use of high spatial resolution data.

RapidEye with its short revisit period and high spatial resolution provides a potential data source for monitoring vegetation fraction in arid regions. In this paper, we try to estimate the sparse vegetation fraction by using VI and MESMA (multiple end-member spectral mixture analysis methods, based on RapidEye data, field-measured spectral signatures and vegetation fraction data. The result shows that: 1) Compared with most commonly used NDVI, red-edge vegetation index is better for estimating the green vegetation fraction in arid regions; 2) MESMA performs much better than vegetation indices, based on which green and senescent vegetation fraction were acquired respectively, subsequently, the total vegetation fraction (green+senescent) has high consistency with field measured total vegetation abundance. In conclusion, Rapideye data with MESMA method provide a good choice for estimating sparse vegetation fraction in arid regions.