A Dynamic Neural Network with Feedback for Trajectory Generation

被引：5

作者：

Atmeh, Ghassan ^{[1
]}

Subbarao, Kamesh ^{[2
]}

机构：

[1] Univ Texas Arlington, Aerosp Syst Lab, Arlington, TX 76019 USA

[2] Univ Texas Arlington, Mech & Aerosp Engn, Arlington, TX 76019 USA

来源：

IFAC PAPERSONLINE | 2016年 / 49卷 / 01期

关键词：

Neural Networks; Dynamic Neural Networks; Recurrent Neural Networks; Trajectory planning; Path planning; BACKPROPAGATION; TIME;

D O I：

10.1016/j.ifacol.2016.03.081

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

A Dynamic Neural Network (DNN), comprising of a Recurrent Neural Network (RNN) and two Feedforward Neural Networks (FFNN), is detailed in this paper. This neurodynamic system is used to adaptively solve the problem of trajectory generation; planning a path connecting a locus of points as a function of time. The BAN is designed to exhibit desired dynamic behavior which drives the output FFNN to generate a specific trajectory. To make the system adaptive, a feedback FFNN is introduced to generate inputs to the R\N based on the outputs of the output FFNN. Simulations conducted for generation of different types of trajectories show the capabilities of the DNN with feedback in solving the trajectory generation problem. (C) 2016, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.

引用

页码：367 / 372

页数：6

共 50 条

[1] Neural network approaches to dynamic collision-free trajectory generation
Yang, SX
Meng, M
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2001, 31 (03): : 302 - 318
[2] Trajectory generation and modulation using dynamic neural networks
Zegers, P
Sundareshan, MK
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2003, 14 (03): : 520 - 533
[3] Application of a neural network to the generation of a robot arm trajectory
Imajo, Shuya
Konishi, Masami
Nishi, Tatsushi
Imai, Jun
ARTIFICIAL LIFE AND ROBOTICS, 2005, 9 (03) : 107 - 111
[4] Dynamic search trajectory methods for neural network training
Petalas, YG
Tasoulis, DK
Vrahatis, MN
ARTIFICIAL INTELLIGENCE AND SOFT COMPUTING - ICAISC 2004, 2004, 3070 : 241 - 246
[5] A neural network approach to real-time trajectory generation
Meng, M
Yang, XY
1998 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION, VOLS 1-4, 1998, : 1725 - 1730
[6] Vehicle Trajectory Prediction Based on Dynamic Graph Neural Network
Cai, Jijing
Zhu, Han
Feng, Hailin
Wen, Long
Wang, Wei
Lv, Meilei
Fang, Kai
PROCEEDINGS OF THE 2024 27 TH INTERNATIONAL CONFERENCE ON COMPUTER SUPPORTED COOPERATIVE WORK IN DESIGN, CSCWD 2024, 2024, : 67 - 72
[7] A Neural Network and Linear Feedback Based Trajectory Control Method for Robot Manipulators
Jiang, Zhao-Hui
Nie, Shiliang
Ishita, Taiki
2008 7TH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION, VOLS 1-23, 2008, : 1603 - +
[8] FEEDBACK-ERROR-LEARNING NEURAL NETWORK FOR TRAJECTORY CONTROL OF A ROBOTIC MANIPULATOR
MIYAMOTO, H
KAWATO, M
SETOYAMA, T
SUZUKI, R
NEURAL NETWORKS, 1988, 1 (03) : 251 - 265
[9] Neural network architecture for trajectory generation and control of automated car parking
Gorinevsky, D
Kapitanovsky, A
Goldenberg, A
IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY, 1996, 4 (01) : 50 - 56
[10] Fast Trajectory Generation with a Deep Neural Network for Hypersonic Entry Flight
Li, Haochen
Chen, Haibing
Tan, Chengpeng
Jiang, Zaiming
Xu, Xinyi
AEROSPACE, 2023, 10 (11)

← 1 2 3 4 5 →