http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Deep Learning based-State Estimation for Holonomic Mobile Robots Using Intrinsic Sensors
Dinh Van Nam,Kim Gon-Woo 제어로봇시스템학회 2021 제어로봇시스템학회 국제학술대회 논문집 Vol.2021 No.10
State estimation is a fundamental component of the navigation system of autonomous mobile robots. Generally, the robot setup is equipped with intrinsic and extrinsic sensors. The state estimators have relied almost on intrinsic sensors such as wheel encoders and inertial measurement units in textureless and structureless environments. This paper will analyze and propose the learning state estimation frameworks for the dead-reckoning of autonomous holonomic vehicles based only on intrinsic sensors. First, we review and categories the intrinsic-only estimation problem. Second, we describe the problem formulation using learning-based techniques. Next, the learning inertial-only estimation is presented with several strategies using the deep learning technique. The initial experiment results are analyzed and deployed using a holonomic mobile robot in real-world environments.
Fuzzy Logic and Deep Steering Control based Recommendation System for Self-Driving Car
Nam Dinh Van,Gon-Woo Kim 제어로봇시스템학회 2018 제어로봇시스템학회 국제학술대회 논문집 Vol.2018 No.10
Automatic cruise control system such as steering and velocity control are complex tasks that work as fundamental to all automatic systems in self-driving cars. The challenging task is to deal with steering signals at high speeds and to manage the vehicle’s velocity with significant changes in steering signals. In order to reach a destination with optimized time and stable trajectory, an autonomous vehicle system should utilize steering control and velocity signal in constraints. In this article, we propose a recommendation system based on Fuzzy logic and Deep Steering Neural Networks. Convolution neural networks(CNNs) works as a front-end stage for predicting steering control, and fuzzy logic in a back-end stage works as natural inferences for recommending velocity and adapting new steering control. The front-end stage uses CNNs with the input from raw sensory data, and the output extracts steering control prediction that feeds forward to the back-end stage. Key functions of back-end stage combine a number of dynamic vehicle information including, steering prediction to extract more effective steering control, and velocity for the autonomous car. The convolution neural network was trained on the Udacity driving datasheet with a number of hours of training and testing data, and the all system is built on MATLAB with GPU coder.
Hoang Nam Nhat,Dinh Van Chau,Dinh Van Thuong,Nguyen Thi Hang 한국인터넷방송통신학회 2015 International Journal of Internet, Broadcasting an Vol.7 No.1
This paper presents the application of the bond valence method to estimate the valence charge distribution in several perovskite systems: La1xPbxMnO3 (x=0.1-0.5), La0.6Sr0. - xTixMnO3 (x=0.0-0.25) and La1xSrxCoO3 (x=0.1-0.5); the reviewing of their crystal structures is also incorporated. The results showed the failure of the elastic bonding mechanism in all studied systems and revealed the general deficit of the valence charge in their unit cells. This valence deficit was not associated with the structural defects and was not equally localized in all coordination spheres. As the content of substitution increased, the charge deficit declined systematically from balanced level, signifying the transfer of valence charge from the O6 to O12 spheres. This transfer depended on the valence deviation of spheres and the average reached near 2 electron per unit cell. The possible impact of the limitted accuracy of the available structural data on the bond valence results has also been considered.
Nhat, Hoang Nam,Chau, Dinh Van,Thuong, Dinh Van,Hang, Nguyen Thi The Institute of Internet 2015 International Journal of Internet, Broadcasting an Vol.7 No.1
This paper presents the application of the bond valence method to estimate the valence charge distribution in several perovskite systems: $La_{{\tilde{1}}x}Pb_xMnO_3$ (x=0.1-0.5), $La_{0.6}Sr_{0.{\tilde{4}}x}Ti_xMnO_3$ (x=0.0-0.25) and $La_{{\tilde{1}}x}Sr_xCoO_3$ (x=0.1-0.5); the reviewing of their crystal structures is also incorporated. The results showed the failure of the elastic bonding mechanism in all studied systems and revealed the general deficit of the valence charge in their unit cells. This valence deficit was not associated with the structural defects and was not equally localized in all coordination spheres. As the content of substitution increased, the charge deficit declined systematically from balanced level, signifying the transfer of valence charge from the ${\tilde{B}}O_6$ to ${\tilde{A}}O_{12}$ spheres. This transfer depended on the valence deviation of spheres and the average reached near 2 electron per unit cell. The possible impact of the limitted accuracy of the available structural data on the bond valence results has also been considered.
Nguyen, Van-Dinh,Tran, Le-Nam,Duong, Trung Q.,Shin, Oh-Soon,Farrell, Ronan IEEE 2017 IEEE Transactions on Vehicular Technology VT Vol.66 No.5
<P>We consider a linear precoder design for an underlay cognitive radio multiple-input multiple-output (MIMO) broadcast channel, where the secondary system consisting of a secondary base station (BS) and a group of secondary users is allowed to share the same spectrum with the primary system. All the transceivers are equipped with multiple antennas, each of which has its own maximum power constraint. Assuming zero-forcing (ZF) method to eliminate the multiuser interference, we study the sum rate maximization problem for the secondary system subject to both per-antenna power constraints at the secondary BS and the interference power constraints at the primary users. The problem of interest differs from the ones studied previously that often assumed a sum power constraint and/or single antenna employed at either both the primary and secondary receivers or the primary receivers. To develop an efficient numerical algorithm, we first invoke the rank relaxation method to transform the considered problem into a convex–concave problem based on a downlink-uplink result. We then propose a barrier interior-point method to solve the resulting saddle point problem. In particular, in each iteration of the proposed method we find the Newton step by solving a system of discrete-time Sylvester equations, which help reduce the complexity significantly, compared to the conventional method. Simulation results are provided to demonstrate fast convergence and effectiveness of the proposed algorithm.</P>