This paper presents a novel approach for realtime 3D navigation in large-scale complex environments by introducing a hierarchical 3D visibility graph (V-graph) and an efficient path search method. The proposed algorithm addresses the computational challenges of V-graph construction and shortest path search on the graph simultaneously. By introducing hierarchical 3D V-graph construction with heuristic visibility update, the 3D V-graph is constructed in$O\left(K \cdot n^{2} \log n\right)$time, which guarantees real-time performance. The proposed iterative divide-and-conquer path search method can achieve near-optimal path solutions within the constraints of realtime operations. The algorithm ensures efficient 3D V-graph construction and path search. Extensive simulated and realworld environments validated that our algorithm reduces the travel time by 42%, achieves up to 24.8% higher trajectory efficiency, and runs faster than most benchmarks by orders of magnitude in complex environments. The code and developed simulator have been open-sourced to facilitate future research.

This paper investigates Path planning Among Movable Obstacles (PAMO), which seeks a minimum cost collision-free path among static obstacles from start to goal while allowing the robot to push away movable obstacles (i.e., objects) along its path when needed. To develop planners that are complete and optimal for PAMO, the planner has to search a giant state space involving both the location of the robot as well as the locations of the objects, which grows exponentially with respect to the number of objects. This paper leverages a simple yet under-explored idea that, only a small fraction of this giant state space needs to be searched during planning as guided by a heuristic, and most of the objects far away from the robot are intact, which thus leads to runtime efficient algorithms. Based on this idea, this paper introduces two PAMO formulations, i.e., bi-objective and resource constrained problems in an occupancy grid, and develops PAMO*, a planning method with completeness and solution optimality guarantees, to solve the two problems. We then further extend PAMO* to hybrid-state PAMO* to plan in continuous spaces with high-fidelity interaction between the robot and the objects. Our results show that, PAMO* can often find optimal solutions within a second in cluttered maps with up to 400 objects.

When observing objects, humans benefit from their spatial visualization and mental rotation ability to envision potential optimal viewpoints based on the current observation. This capability is crucial for enabling robots to achieve efficient and robust scene perception during operation, as optimal viewpoints provide essential and informative features for accurately representing scenes in 2D images, thereby enhancing downstream tasks.To endow robots with this human-like active viewpoint optimization capability, we propose ViewActive, a modernized machine learning approach drawing inspiration from aspect graph, which provides viewpoint optimization guidance based solely on the current 2D image input. Specifically, we introduce the 3D Viewpoint Quality Field (VQF), a compact and consistent representation for viewpoint quality distribution similar to an aspect graph, composed of three general-purpose viewpoint quality metrics: self-occlusion ratio, occupancy-aware surface normal entropy, and visual entropy. We utilize pre-trained image encoders to extract robust visual and semantic features, which are then decoded into the 3D VQF, allowing our model to generalize effectively across diverse objects, including unseen categories. The lightweight ViewActive network (72 FPS on a single GPU) significantly enhances the performance of state-of-the-art object recognition pipelines and can be integrated into real-time motion planning for robotic applications. Our code and dataset are available here https://github.com/jiayi-wu-umd/ViewActive.

Landslides, as a significant geological hazard widely distributed globally, require precise risk assessment to ensure the utmost protection of human lives and property. Existing studies typically rely on historical landslide inventories and consider factors such as topography, geology, geomorphology, and hydrology for landslide susceptibility evaluation. However, these inventories often need more timely updates and seldom account for the impact of human engineering activities. Therefore, this study focused on Yongping County in Yunnan Province, China, a region prone to frequent landslides. It updated and supplemented the landslide inventory using geolocalization methods. Additionally, the study incorporated various human engineering factors in the evaluation process and employed a convolutional neural network (CNN) model to assess landslide susceptibility. The results align more closely with field survey realities, demonstrating the reliability of this method. This method can potentially be applied to the susceptibility assessment of other geological disasters in the future.

One of the core activities of an active observer involves moving to secure a "better" view of the scene, where the definition of "better" is task-dependent. This paper focuses on the task of human pose estimation from videos capturing a person’s activity. Self-occlusions within the scene can complicate or even prevent accurate human pose estimation. To address this, relocating the camera to a new vantage point is necessary to clarify the view, thereby improving 2D human pose estimation. This paper formalizes the process of achieving an improved viewpoint. Our proposed solution to this challenge comprises three main components: a NeRF-based Drone-View Data Generation Framework, an On-Drone Network for Camera View Error Estimation, and a Combined Planner for devising a feasible motion plan to reposition the camera based on the predicted errors for camera views. The Data Generation Framework utilizes NeRF-based methods to generate a comprehensive dataset of human poses and activities, enhancing the drone’s adaptability in various scenarios. The Camera View Error Estimation Network is designed to evaluate the current human pose and identify the most promising next viewing angles for the drone, ensuring a reliable and precise pose estimation from those angles. Finally, the combined planner incorporates these angles while considering the drone’s physical and environmental limitations, employing efficient algorithms to navigate safe and effective flight paths. This system represents a significant advancement in active 2D human pose estimation for an autonomous UAV agent, offering substantial potential for applications in aerial cinematography by improving the performance of autonomous human pose estimation and maintaining the operational safety and efficiency of UAVs.

This paper introduces a real-time algorithm for navigating complex unknown environments cluttered with movable obstacles. Our algorithm achieves fast, adaptable routing by actively attempting to manipulate obstacles during path planning and adjusting the global plan from sensor feedback. The main contributions include an improved dynamic Directed Visibility Graph (DV-graph) for rapid global path searching, a real-time interaction planning method that adapts online from new sensory perceptions, and a comprehensive framework designed for interactive navigation in complex unknown or partially known environments. Our algorithm is capable of replanning the global path in several milliseconds. It can also attempt to move obstacles, update their affordances, and adapt strategies accordingly. Extensive experiments validate that our algorithm reduces the travel time by 33%, achieves up to 49% higher path efficiency, and runs faster than traditional methods by orders of magnitude in complex environments. It has been demonstrated to be the most efficient solution in terms of speed and efficiency for interactive navigation in environments of such complexity. We also open-source our code in the docker demo1to facilitate future research.

Methane CH4) is the second largest greenhouse gas in the world after CO2. CH4emissions contribute 16% of the world's greenhouse gases. Although global CH4 emissions are much lower than global carbon dioxide emissions, their global warming potential (GWP) over 100 years is 28-36 times greater than that of carbon dioxide. Using satellites to observe methane is an effective means. This paper is aimed at Hebei Province, using TROPOMI onboard Sentinel-5P. The main research content: 1). The concentration of XCH4in Hebei Province has obvious seasonal trend (autumn > winter > summer > spring). 2). Due to the existence of some stable emission sources in the southwest of Hebei Province, there are obvious anomalous areas of stable high value of methane in these areas.

TanSat is crucial for detecting global CO2concentration, solar-induced chlorophyll fluorescence, and CO2flux as China's first atmospheric CO2concentration monitoring satellite. This manuscript presents a preliminary attempt to estimate anthropogenic CO2emissions from large sources with TanSat. We identified XCO2(the column-average dry air-mole fraction of CO2) anomalies and quantified CO2emissions from two TanSat observations in Northeast China. The emission rate estimations of the two XCO2plumes are 11.46 kt CO2/h and 10.09 kt CO2/h respectively, and the emission rates of the Multi-resolution Emission Inventory for China (MEIC) are 6.86 kt CO2/h and 3.10 kt CO2/h respectively. The result shows that TanSat has the ability to quantify anthropogenic CO2emissions.

Aerosol vertical structure (AVS) plays an important role in the Earth's climate system. The spectral radiance matching (SRM) method can effectively reconstruct the aerosol three-dimensional structure, but it is affected by the donor-recipient matching accuracy. To overcome the problem, in this paper, the SRM algorithm is improved by using regional spectral radiation matching instead of single-pixel spectral matching. First, the cost function is constructed by the multispectral radiometric difference between the off-nadir recipient pixel neighborhood and the potential donor neighborhood within the MODIS scan. The best donor-recipient match is selected by minimizing the cost function. Then, the corresponding donor pixel aerosol profile information from CALIPSO observations is assigned to the recipient pixel, which fills the gaps between CALIPSO track and achieves a true AVS global estimate. Finally, the accuracy of the algorithm was verified by using AD-NET station data. The experimental results show that the reconstruction results of the region-based method are consistent with the ground-based lidar measurements and superior to the pixel-based SRM method.

The air quality model can reproduce the processes of atmospheric transport, chemical reaction and removal, and analyze the spatio-temporal evolution law, internal mechanism and origin of air pollution. WRF-Chem model is the third generation of air quality model. Its biggest advantage is that the meteorological model and chemical transmission model are fully coupled in time and spatial resolution to achieve true online feedback, which is the main development direction of the future model. Therefore, this paper selected Xuzhou City, a heavy industry city with serious air pollution, as the research area, and took December when air quality was poor as the research time. WRF-Chem model was used to simulate and analyze meteorological parameters such as wind speed, temperature, relative humidity and concentration of air pollutants including PM2.5and O3in Xuzhou in December 2020. The simulation results were compared with the site data, and the simulation results under different parameter Settings of WRF-Chem mode were studied and compared.

Imported air pollution has a significant impact on urban air quality. Many urban air pollution events are not caused by local emissions, but by the transport of air pollutants from surrounding areas. Therefore, it is very necessary to prevent and control imported air pollution. However, the existing supervision of urban air quality mostly relies on ground monitoring stations, which is extremely limited in time and space. In this paper, MERRA-2 data is used to grasp urban air quality from a more macroscopic perspective, and combined with ground monitoring station data and meteorological data, the transmission route of air pollution is reconstructed. This paper takes Xuzhou City, Jiangsu Province as an example. It is proved that this method is highly feasible and can provide scientific data support for efficient prevention and control of imported air pollution.

Accurately estimating crop carbon sinks at large regional scales from the perspective of remote sensing is of great significance for carbon neutrality research, crop yield estimation, and scientific agriculture. In this study, an improved Carnegie–Ames–Stanford approach (CASA) model was coupled with time-series satellite remote sensing images to estimate Net primary productivity. The NEP is then further calculated by coupling the soil respiration model to represent the carbon sink at a regional scale. The main research contents: (1) The month-by-month net primary productivity of crops in Jiangsu Province in 2021. (2) The net ecosystem productivity of crops in Jiangsu Province month by month in 2021.

In this paper, surface ozone in China were estimated by the Geographically and Temporally Weighted Regression model with its precursors and AOD data. Based on the GTWR model, the surface ozone was estimated by time, space, ozone precursors and AOD. Taking August 1, 2022 as an example, it was analyzed and confirmed about the feasibility of ozone precursors and AOD data in estimating surface ozone mass concentrations.

Graph convolutional networks have been widely applied in skeleton-based gait recognition. A key challenge in this task is to distinguish the individual walking styles of different subjects across various views. Existing state-of-the-art methods employ uniform convolutions to extract features from diverse sequences and ignore the effects of viewpoint changes. To overcome these limitations, we propose a condition-adaptive graph (CAG) convolution network that can dynamically adapt to the specific attributes of each skeleton sequence and the corresponding view angle. In contrast to using fixed weights for all joints and sequences, we introduce a joint-specific filter learning (JSFL) module in the CAG method, which produces sequence-adaptive filters at the joint level. The adaptive filters capture fine-grained patterns that are unique to each joint, enabling the extraction of diverse spatial-temporal information about body parts. Additionally, we design a view-adaptive topology learning (VATL) module that generates adaptive graph topologies. These graph topologies are used to correlate the joints adaptively according to the specific view conditions. Thus, CAG can simultaneously adjust to various walking styles and viewpoints. Experiments on the two most widely used datasets (i.e., CASIA-B and OU-MVLP) show that CAG surpasses all previous skeleton-based methods. Moreover, the recognition performance can be enhanced by simply combining CAG with appearance-based methods, demonstrating the ability of CAG to provide useful complementary information.

Methane is the world's second largest greenhouse gas after carbon dioxide, accounting for about one-sixth of the total greenhouse gas emissions. Due to the development of animal husbandry, carbon dioxide and methane produced in the process of animal husbandry breeding have gradually become one of the important sources of the greenhouse effect. The research of methane gas emissions from animal husbandry is of great significance to sorting out the sources of greenhouse gases. This paper is aimed at Hebei Province, using TROPOMI sensors onboard Sentinel-5P, TANSO-FTS onboard GOSAT observation data, and the related ground station data. The main research content: (1) The temporal and spatial distribution characteristics of methane in Hebei Province from 2009 to 2020; (2) The proportion of methane emissions from ruminants in Hebei Province Estimate using ruminant methane emission formula. The average proportion is 6.10% from 2009 to 2018.

This paper calculates the Air Quality Index (AQI) in Xuzhou from 2018 to 2020 and analyzes its annual and monthly changes in air quality to determine the month of severe pollution. Through AQI calculation and interpolation analysis of different locations in Xuzhou, the air quality conditions in different regions are determined. Through the HYSPLIT backward trajectory model, cluster analysis of the air masses in December 2020 is carried out. Combined with the PM2.5 concentration data released every hour by the Xuzhou State Control Station, PSCF and CWT methods are used to determine potential sources of PM2.5 pollution in different regions. This paper also uses the hourly PM2.5 data in Xuzhou area obtained from the retrieval of the Himawari-8/AHI AOD through the IGTWR model for the traceability analysis of pollutants. The results show that there are large potential pollution sources in Henan, northern Anhui, northern Hubei and other places. At the same time, pollutant emissions in parts of northern Jiangsu, southern Shanxi and Shandong province also contribute to PM2.5 in Xuzhou. This paper also traced the source of the PM2.5 heavy pollution days in Xuzhou in December 2020 and determined the location of the specific pollution source based on the actual situation.

Dust as a kind of atmospheric pollutants, its harm are gradually being paid attention to. However, most of the existing supervision of dust relies on ground monitoring sites, which has strong spatial limitations. In this paper, high-resolution remote sensing images are used to identify regional dust sources, which can identify dust at a more macroscopic level from “plane”. At the same time, regional particulate emission inventory was established based on multi-source data, and then the dust emission is quantified. This paper takes Quanshan District, Xuzhou City, Jiangsu Province as an example, and the experimental results show that the method can reflect the actual dust situation well, and provide scientific data support for the supervision of urban dust.

In this paper, surface-level ozone in China was estimated by using the source-sink analysis and machine learning model. By analyzing the source and sink of surface ozone, it is clear that ozone mass concentration is influenced by background value, regional and local chemical generation, deposition, chemical removal and Interregional transport comprehensively. Then, the light gradient boosting machine (LGBM) model was used to integrate various corresponding satellite-based variables, numerical model-based meteorological variables and land variables to obtain the high spatial resolution surface mass concentration of ozone in China. Taking June, July, August, 2021 as example, the feasibility of the Tropospheric Monitoring Instrument (TROPOMI) data, the European Centre for Medium-Range Weather Forecasts (ECWMF) data and LGBM model in estimating surface-level ozone mass concentration was analyzed and confirmed.

Margin-based metric losses have shown great success in Person Re-identification and Face Verification. But most existing works adopt a fixed class-level margin regardless of the difference between each training sample. This paper proposes a Norm-Aware Margin Assignment (NAMA) scheme to dynamically adjust the weight of each sample during training. Combined with the existing margin-based classification losses, NAMA improves the robustness of feature embedding by assigning larger margins to more recognizable samples. NAMA is a fully trainable module that automatically models the correlation between the optimal margin and image quality during back-propagation without supervision. To stabilize the training and make the assigned margin more controllable, we introduce a margin re-balance mechanism to align the expectation of learned margins to a pre-defined value. Extensive experiments on three popular ReID benchmarks validate the effectiveness of our NAMA method. Code will be publicly available at: https://github.com/huangzongheng/NAMA.

Although gait recognition has drawn increasing research attention recently, it remains challenging to learn discriminative temporal representation since the silhouette differences are quite subtle in spatial domain. Inspired by the observation that humans can distinguish gaits of different subjects by adaptively focusing on temporal sequences with different time scales, we propose a context-sensitive temporal feature learning (CSTL) network in this paper, which aggregates temporal features in three scales to obtain motion representation according to the temporal contextual information. Specifically, CSTL introduces relation modeling among multi-scale features to evaluate feature importances, based on which network adaptively enhances more important scale and suppresses less important scale. Besides that, we propose a salient spatial feature learning (SSFL) module to tackle the misalignment problem caused by temporal operation, e.g., temporal convolution. SSFL recombines a frame of salient spatial features by extracting the most discriminative parts across the whole sequence. In this way, we achieve adaptive temporal learning and salient spatial mining simultaneously. Extensive experiments conducted on two datasets demonstrate the state-of-the-art performance. On CASIA-B dataset, we achieve rank-1 accuracies of 98.0%, 95.4% and 87.0% under normal walking, bag-carrying and coat-wearing conditions. On OU-MVLP dataset, we achieve rank-1 accuracy of 90.2%. The source code will be published at https://github.com/OliverHxh/CSTL.

Multicopters are able to perform high maneuverability yet their potential have not been fully achieved. In this work, we propose a full-body, optimization-based motion planning framework that takes shape and attitude of aerial robot into consideration such that the aggressiveness of drone maneuvering improves significantly in cluttered environment. Our method takes in a series of intersecting polyhedrons that describe a range of 3D free spaces and outputs a time-indexed trajectory in real-time with full-body collision-free guarantee. The drone is modeled as a tilted cuboid, yet we argue that our framework can be freely adjusted to fit multicopters of different shapes. Guaranteeing dynamic feasibility and safety conditions, our framework transforms the original constrained nonlinear programming problem to an unconstrained one in higher dimensions which is further solved by quasi-Newton methods. Benchmark has shown that our method improves the state-of-art with orders of magnitude in terms of computation time and memory usage. Simulations and onboard experiments are carried out as validation.

The development of aerial autonomy has enabled aerial robots to fly agilely in complex environments. However, dodging fast-moving objects in flight remains a challenge, limiting the further application of unmanned aerial vehicles (UAVs). The bottleneck of solving this problem is the accurate perception of rapid dynamic objects. Recently, event cameras have shown great potential in solving this problem. This paper presents a complete perception system including ego-motion compensation, object detection, and trajectory prediction for fast-moving dynamic objects with low latency and high precision. Firstly, we propose an accurate ego-motion compensation algorithm by considering both rotational and translational motion for more robust object detection. Then, for dynamic object detection, an event camera-based efficient regression algorithm is designed. Finally, we propose an optimization-based approach that asynchronously fuses event and depth cameras for trajectory prediction. Extensive real-world experiments and benchmarks are performed to validate our framework. Moreover, our code will be released to benefit related researches.