A5034301879- Robotics and Sensor-Based Localization
- Robotic Path Planning Algorithms
- Indoor and Outdoor Localization Technologies
- Remote-Sensing Image Classification
- Modular Robots and Swarm Intelligence
- Advanced Image and Video Retrieval Techniques
- Distributed Control Multi-Agent Systems
- Remote Sensing in Agriculture
- Remote Sensing and Land Use
- Fault Detection and Control Systems
- Advanced Optical Sensing Technologies
- Geographic Information Systems Studies
- Oil Spill Detection and Mitigation
- Advanced Sensor and Control Systems
- Seismic Imaging and Inversion Techniques
- Handwritten Text Recognition Techniques
- Laser Design and Applications
- Ammonia Synthesis and Nitrogen Reduction
- Structural Health Monitoring Techniques
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Seismic Waves and Analysis
- Biometric Identification and Security
- Seismology and Earthquake Studies
- Tactile and Sensory Interactions
- Graphene research and applications
University of Chinese Academy of Sciences
2022-2024
Chinese Academy of Sciences
2005-2024
Xi’an University of Posts and Telecommunications
2024
Aerospace Information Research Institute
2022-2024
Inner Mongolia Electric Power Survey & Design Institute (China)
2024
Korea Advanced Institute of Science and Technology
2022
Institute of Microelectronics
2021
Space Engineering University
2019
Ningbo Institute of Industrial Technology
2016
Purdue University West Lafayette
2004-2006
Deep learning approaches have contributed to the rapid development of remote sensing (RS) image interpretation. The most widely used training paradigm is use ImageNet pretrained models process RS data for specified tasks. However, there are issues such as domain gap between natural and scenes poor generalization capacity models. It makes sense develop a foundation model with general feature representation. Since large amount unlabeled available, self-supervised method has more significance...
Abstract Micro/nano robots have exhibited excellent application potential in the biomedical field, such as drug delivery, minimally invasive surgery, and bio-sensing. Furthermore, order to achieve practical application, it is essential for swimming micro/nano navigate towards specific targets or adjust their speed morphology complete environments. The navigation of with temporal spatial precision critical fulfilling demand applications. Here, we introduced a fully integrated wearable control...
Compared with binary change detection (BCD), semantic (SCD) further provides the category information of bitemporal changed regions which is significant for practical application Earth Observation. Although recently proposed triple-branch structures including one BCD branch and two classification branches can effectively achieve task balance, they still need to employ carefully designed difference extraction module interactions capture correlations, increases complexity utilization. In this...
The artificial potential field method is used in mobile robot path planning extensively because of its simpleness, high efficiency and smooth path, but it also has disadvantages. To overcome the shortcomings traditional planning, this paper analyzes reasons that lead to failure puts forward an improved method, which attractive repulsive optimized, we propose a strategy filling escape GNRON local minima problems. At last, introduce regression search optimize path. As result, can find better...
In this paper, we investigate energy-time-efficient dispatching methods for ant-like robots to cover an unmapped region effectively. These have very limited energy and sensor ability, making them practical inexpensive build. Our model was based on bio-inspired algorithms from ant colony system. We assumed that all the start their home starting point, nest, is composed of floor tiles can be modelled as vertices in a graph. system, leave pheromone use information region. developed analyzed two...
We developed an automatic seismic wave and phase detection software based on PhaseNet, efficient highly generalized deep learning neural network for P- S-wave picking. The organically combines multiple modules including application terminal interface, docker container, data visualization, SSH protocol transmission other auxiliary modules. Characterized by a series of technologically powerful functions, the is convenient all users. To obtain picks, one only needs to prepare three-component as...
Simultaneous localization and map building (SLAM) is a fundamental complex problem in mobile robot research. In SLAM, Kalman-filter-like implementations are widely adopted to localize build simultaneously incrementally. However, this approach requires extensive computations of order O(N/sup 2/), where N the total number landmarks. To make more manageable, we propose logarithmic partitioning algorithm that partitions global into one local region several sub-maps. The size each sub-map based...
Traditionally, the SLAM problem solves localization and mapping in explored sensed regions. This paper presents a prediction-based algorithm (called P-SLAM), which has an environmental structure predictor to predict inside unexplored region (i.e., look-ahead mapping). The prediction process is based on observation of surroundings comparing it with built map If similar matched regions, hypothesis generated indicate that been before. environment repeated structures, mobile robot can utilize...
For a conventional Corner Matching Method (CMM), corner is identified only when two edges of the are sensed (scanned). Utilizing geometric features sharp corners, novel and general CMM method proposed in this paper. The major advantage approach over that with one edge can be through comparing scan data between subsequent points same scan, as will cause adjacent an abrupt change. In order to identify point scans, effective matching algorithm formulated based on distance matrix constructed...
FastSLAM is well known as a Rao-Blackwellised particle filter formulation of simultaneous localization and mapping. The accuracy conventional degenerates over time due to the depletion in resampling phase. In this paper, an improved with based on Particle Swarm Optimization (PSO) proposed address problem. Firstly, instead rejection replication, PSO employed process for pose convergence set. Then, order update proposal distribution map, special framework adopted method. Finally, result...
Recently, the transformer has achieved notable success in remote sensing (RS) change detection (CD). Its outstanding long-distance modeling ability can effectively recognize of interest (CoI). However, order to obtain precise pixel-level regions, many methods directly integrate stacked blocks into UNet-style structure, which causes high computation costs. Besides, existing generally consider bitemporal or differential features separately, makes utilization ground semantic information still...
Abstract Positron annihilation lifetime spectroscopy (PALS) is a popular method for studying defects in materials. With the development of fast digital oscilloscopes and digitizers, conventional analog positron (PAL) spectrometer being replaced by PAL spectrometer. Recently, newly conducted research has developed using silicon photomultiplier (SiPM) instead tube (PMT). Timing resolution needs to be improved identify lifetimes positrons. Many parameters affect timing performance SiPM-based...
The indoor mobile robots have been widely employed in manufacturing and service industries recently years. However, the conventional simultaneous localization mapping (SLAM) method is mainly meant for with traditional drives (differential drive Ackerman steering) that coupled translation rotation motions. Hence, error will affect heading estimations of robot, accuracy FastSLAM algorithm be reduced over time. In addition, are not suitable narrow, crowded environment lots obstacles. this...
Laser micro thruster based on semiconductor laser has great advantages in volume. However, the design of optical collimating and focusing system is very challenging under highly integrated conditions. The operating distance from exit to working ablation target needs be tens millimeters. Based ray tracing method, a convenient fast method proposed small distance. verified by an example. conditions are as follows: wavelength 940 nm, fiber core diameter 105 microns, numerical aperture 0.22; spot...