A5034700638- Robotics and Sensor-Based Localization
- Advanced Image and Video Retrieval Techniques
- Visual Attention and Saliency Detection
- Advanced Vision and Imaging
- Robotic Path Planning Algorithms
- Video Surveillance and Tracking Methods
- Water Quality Monitoring Technologies
- Soft Robotics and Applications
- Image Processing Techniques and Applications
- Vehicle Dynamics and Control Systems
- Anatomy and Medical Technology
- Control and Dynamics of Mobile Robots
- Surgical Simulation and Training
- Transportation Safety and Impact Analysis
- 3D Surveying and Cultural Heritage
- Dynamics and Control of Mechanical Systems
- Video Coding and Compression Technologies
- Electromagnetic Launch and Propulsion Technology
- Advanced Neural Network Applications
- Image Enhancement Techniques
Chongqing University
2020-2024
Beijing Jishuitan Hospital
2021
Peking University
2021
Visual odometry (VO) is a prevalent way to deal with the relative localization problem, which becoming increasingly mature and accurate, but it tends be fragile under challenging environments. Comparing classical geometry-based methods, deep learning-based methods can automatically learn effective robust representations, such as depth, optical flow, feature, ego-motion, etc., from data without explicit computation. Nevertheless, there still lacks thorough review of recent advances VO (Deep...
Most visual simultaneous localization and mapping (SLAM) algorithms rely on geometric features, such as points, lines, planes, these treated all features equally important in the optimization process, ignoring semantic information of features. However, robust rich should play a more role, which is similar to mechanism saliency or attention. Therefore, this article aims mimic SLAM framework by using prediction model. To solve center bias generic salient dataset, we proposed method for...
Precise and drift-free motion estimation is an essential technology for autonomous driving. Single-sensor methods such as laser-based or vision-based have proven to be inadequate. To solve the problem, we proposed optimization-based fusion approach that incorporates information from complementary sensors achieve high accuracy global drift-free. The core idea construct a globally unified pose graph through dual-layer optimization strategy. local layer obtains relative LiDAR odometry...
For manually driven rubber trams to track, virtual tracks can easily cause driver fatigue. Therefore, based on visual navigation, an automatic steering and trajectory following method are proposed. First, the vehicle kinematic dynamic model of Delight Tram is Then, methods introduced, which prediction control Ackermann theory, respectively. Finally, effectiveness proposed has been evaluated via both multi-body simulations road tests under various working conditions. The results show that...
Recently, the philosophy of visual saliency and attention has started to gain popularity in robotics community. Therefore, this paper aims mimic mechanism SLAM framework by using prediction model. Comparing with traditional that treated all feature points as equal important optimization process, we think salient should play more role process. proposed a model predict map, which can capture both scene semantic geometric information. Then, Salient Bundle Adjustment value map weight approach....
Abstract Objective To explore navigation-related factors interfering with accuracy of robot-assisted surgery. Methods We made a measurement model to test the TianJi Robot system when performing stimulated screw placement procedure. The three-coordinate machine was used measure deviation between actual position and planned position. designed corresponding experiments effects different on accuracy. deviations were measured at distance (ranging from 1.2 m 2.2 m) navigation optical stereo camera...