A5029928323- Robotics and Sensor-Based Localization
- 3D Surveying and Cultural Heritage
- Indoor and Outdoor Localization Technologies
- Market Dynamics and Volatility
- Inertial Sensor and Navigation
- Advanced Sensor and Control Systems
- Adaptive Control of Nonlinear Systems
- Advanced Vision and Imaging
- Advanced Image and Video Retrieval Techniques
- Underwater Vehicles and Communication Systems
- Industrial Technology and Control Systems
- Robotic Path Planning Algorithms
- Remote Sensing and LiDAR Applications
- Advanced Control Systems Optimization
- Soft Robotics and Applications
- 3D Shape Modeling and Analysis
- Advanced Measurement and Metrology Techniques
- Autonomous Vehicle Technology and Safety
- Control and Dynamics of Mobile Robots
- Robotic Locomotion and Control
- Target Tracking and Data Fusion in Sensor Networks
- Robotic Mechanisms and Dynamics
- Advanced optical system design
- Oceanographic and Atmospheric Processes
- Climate variability and models
China University of Geosciences (Beijing)
2023-2024
Shanghai University
2024
Southeast University
2021-2024
Shanghai Power Equipment Research Institute
2024
Ministry of Natural Resources
2023
Virginia Wesleyan College
2020
Cytoskeleton (United States)
2020
Chienkuo Technology University
2016
China Post (China)
2016
A passive navigation method of autonomous underwater vehicles (AUVs) without transmitting signals is stressed in this article, which has the advantages good concealment and high security. An inertial system (INS) a kind navigation, can be used for positioning AUVs. However, divergence INS error limits its application long endurance long-range navigation. Thus, single source-aided integrated scheme proposed article. fixed sound source emits acoustic signal periodically,...
This study aims to employ a causal network model based on transfer entropy for the early warning of systemic risk in commodity markets. We analyzed dynamic relationships prices 25 commodities related China (including futures and spot energy, industrial metals, precious agricultural products), validating effect structure among markets risk. Our research results identified categories playing significant roles, revealing that industry metal possess stronger market information transmission...
Ultra-short baseline (USBL) is a mature navigation method for underwater applications. Acoustic outliers and the slant distance calculation error are main challenges improving positioning accuracy. The latter, in particular, often overlooked. In this paper, errors due to motion effect acoustic will be mainly investigated. To address problem, influence of on localization dynamic environments analyzed first. Aiming at two challenges, new measurement model based round trip time delay bearing...
Abstract In mainstream visual inertial odometry (VIO) systems, the method of positional solution by feature point extraction and matching in image is widely used. However, tracking accuracy features dependent on texture richness environment. Although many existing algorithms introduce line front end to improve system’s environmental adaptability, most them sacrifice system real-time exchange for higher positioning accuracy. The often require more time, thus failing meet requirements...
Robotics is an interdisciplinary science and should be treated as a part of System Integration. Automotive industry with its assembly line requirements continues to the main robot consumer. Besides automotive industry, other sectors like for instance metal machining, health semiconductor are interwoven robotics. Robot high value added product so it worth development. Therefore, goal this paper demonstrate how design fully operational rotary delta robot. Firstly, we elaborate kinematic model...
Vehicles have become one of the most popular transportation tools. More and more vehicles occur with different AI-based systems. In order to make this mainstream comfortable, efficient, safer for people use, artificial intelligence technologies are applied in vehicles, leading much smarter intelligent. paper, three main applications AI including Vehicle-to-Everything(V2X), vehicle control system, diagnostic device will be shown. The current technical issues when applying V2X challenges all...
The integrated navigation of the visual and inertial measurement is becoming a research hotspot in field autonomous driving intelligent navigation. fusion heterogeneous sensors can effectively compensate for deficiency single sensor. Therefore, developing visual-inertial calibration algorithm with good real-time performance, high accuracy, strong robustness an urgent issue. analytical solution-based avoid locally optimal solutions during process significantly increase system but achieves low...
In this article, an innovative simultaneous localization and mapping system for monocular vision in dynamic scenes is proposed, which eliminates the unreliable features via point-state segmentation strategy. A neighbor-based random sample consensus method image-sequences-based selection are presented to ensure a stable precise initial state of system. Based on accurate information, probability model orientation consistency proposed degrade disturbance environment factors process feature...
Abstract In the high-dose radiation environment of primary circuit nuclear power plants, there are problems such as low grinding efficiency and automation level in repair process sealing surface stop valve seat. Using working principle manual technology experience, we extracted key parameters for control. We researched design scheme technologies automated equipment. addition, provided a detailed description equipment’s performance, purpose, structure, function.
Mesoscale eddies play a critical role in sea navigation and route planning, yet traditional prediction methods have often overlooked their spatial relationships, relying on indirect approaches to capture distribution across extensive maps. To address this limitation, we present BiST-SA-LSTM, an end-to-end framework that combines Bidirectional Spatial Temporal LSTM Self-Attention mechanisms. Utilizing data sourced from the South China Sea its surrounding regions, which are renowned for...
Machine learning is a promising technique for many practical applications. In this perspective, we illustrate the development and application machine learning. It indicated that theories applications of method in field energy conservation indoor environment are not mature, due to difficulty determination model structure with better prediction. order significantly contribute problems, utilize ANN predict culturable fungi concentration, which achieves accuracy convenience. The proposal hybrid...
Most of the visual inertial positioning accuracy is constrained by calibration sensor, however, there are few navigation systems that take time offset and extrinsic parameters into account at same time. In this paper, an iterative-optimization-based framework for VIO (visual-inertial odometry) with limited prior conditions proposed. particular, a three-stage method presented, which includes estimation stage, intrinsic global temporal IMU parameter optimization stage. first model proposed,...
In order to improve the initialization robustness of visual inertial SLAM, complementarity optical flow method and feature-based can be used in vision data processing. The parallel is proposed, where monocular are carried out at same time. After initializations, state estimation results jointly optimized by bundle adjustment. proposed retains more mapping information, correspondingly adaptable scene. It found that map constructed features a comparable accuracy one ORB-SLAM3 mode. Since...
Exploring various environmental features is beneficial to improve the adaptability of visual-oriented localization system. However, most existing visual-based systems mainly focus on extracting more kinds landmarks from images but do not fully explore geometric correlation among different features. This article proposes a multifeatures association bundle adjustment (BA) constraints method accuracy First, this constructs an amount spatial virtual plane (VP) based point and line landmarks,...
Feature constraints in a visual oriented simultaneous localization and mapping (SLAM) are critical. Previous researches explored point line features the environment to obtain more information, however, mostly researchers focused on geometric characteristic of single but ignored association constraint between multi-features. This paper proposed an inertial-aided navigation method with point-line joint constraint. We not only use IMU measurement residual, feature re-projection error error,...