Improved SLAM Method Based on Dynamic Objects Detecting with the Fusion of Vision and Lidar

2021 ◽  
pp. 347-356
Author(s):  
Ren Zhang ◽  
Yitian Li ◽  
Lu Lou
Keyword(s):  
Metrologiya ◽  
2020 ◽  
pp. 25-42
Author(s):  
Dmitrii V. Khablov

This paper describes a promising method for non-contact vibration diagnostics based on the use of Doppler microwave sensors. In this case, active irradiation of the object with electromagnetic waves and the allocation of phase changes using two-channel quadrature processing of the received reflected signal are used. The modes of further fine analysis of the resulting signal using spectral or wavelet transformations depending on the nature of the active vibration are considered. The advantages of this non-contact and remote vibration analysis method for the study of complex dynamic objects are described. The convenience of the method for machine learning and use in intelligent systems of non-destructive continuous monitoring of the state of these objects by vibration is noted.


Author(s):  
Мадина Усенбай ◽  
Акмарал Иманбаева

Конфиденциальность является одним из важных параметров для повышения безопасности в сети, цель которого - сохранить секретную информацию. Рассмотрена модель доверия, состоящая из текущих и прошлых оценок на основе репутации объекта в сети. В модели используется параметр времени для защиты конфиденциальности пользователя для статических и динамических объектов, например, в IoT или облачной технологии. Confidentiality is one of the important parameters for increasing security on the network, the coal of which is to keep secret information. A trust model consisting of current and past assessments based on the object reputation in the network is considered. The model uses a time parameter to protect user privacy for static and dynamic objects, for example, in IoT or cloud technology.


Author(s):  
Jiahui Huang ◽  
Sheng Yang ◽  
Zishuo Zhao ◽  
Yu-Kun Lai ◽  
Shi-Min Hu

AbstractWe present a practical backend for stereo visual SLAM which can simultaneously discover individual rigid bodies and compute their motions in dynamic environments. While recent factor graph based state optimization algorithms have shown their ability to robustly solve SLAM problems by treating dynamic objects as outliers, their dynamic motions are rarely considered. In this paper, we exploit the consensus of 3D motions for landmarks extracted from the same rigid body for clustering, and to identify static and dynamic objects in a unified manner. Specifically, our algorithm builds a noise-aware motion affinity matrix from landmarks, and uses agglomerative clustering to distinguish rigid bodies. Using decoupled factor graph optimization to revise their shapes and trajectories, we obtain an iterative scheme to update both cluster assignments and motion estimation reciprocally. Evaluations on both synthetic scenes and KITTI demonstrate the capability of our approach, and further experiments considering online efficiency also show the effectiveness of our method for simultaneously tracking ego-motion and multiple objects.


2021 ◽  
Vol 88 (6) ◽  
pp. 337
Author(s):  
S. F. Sergeev ◽  
A. V. Khomyakov
Keyword(s):  

Sign in / Sign up

Export Citation Format

Share Document