Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Networks Presented by: Parsa Yousefi Supervisors: Dr. M. Jamshidi, Dr. P. Benavidez June 23 rd , 2017 The University of Texas at San Antonio – Department of Electrical and Computer Engineering
Outline • Data Analytics • Introduction • Clustering • Neural Networks • Long Short-Term Memory • Data prediction • Latency • Reconstructing Data Using LSTM • Fault Detection • Training Initial Model Using LSTM • Future Works • Acknowledgments • References The University of Texas at San Antonio – Department of Electrical and Computer Engineering 2 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Analytics • Introduction • Definition: Data analytics refers to qualitative and quantitative techniques and processes used to enhance productivity and business gain. Data is extracted and categorized to identify and analyze behavioral data and patterns, and techniques vary according to organizational requirements. • Big Data https://www.neuralt.com/uploads/assets/BigData%20.png The University of Texas at San Antonio – Department of Electrical and Computer Engineering 3 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Analytics • A Summary of Data Science Process: • Data Collection • Processing • Cleaning Data • Product Data • Communication https://upload.wikimedia.org/wikipedia/commons/b/ba/Data_visualization_process_v1.png The University of Texas at San Antonio – Department of Electrical and Computer Engineering 4 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Analytics • Processing Methods • Clustering • Neural Networks http://i0.wp.com/planningtank.com/wp-content/uploads/2014/01/Distribution-Process.jpg The University of Texas at San Antonio – Department of Electrical and Computer Engineering 5 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Processing Methods • Data Clustering • Definition • Centroid-based Clustering • Density-based Clustering https://upload.wikimedia.org/wikipedia/commons/c/c8/Cluster-2.svg The University of Texas at San Antonio – Department of Electrical and Computer Engineering 6 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Clustering • Centroid-based Clustering • Initiating centroids • Finding nearest members to centroids • Calculating New Centroids • Repeating the method until convergence • Advantage • Convergence Speed • Disadvantage • Number of Centroids as an input http://www.mdpi.com/sensors/sensors-15-29056/article_deploy/html/images/sensors-15-29056-g004-1024.png The University of Texas at San Antonio – Department of Electrical and Computer Engineering 7 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Clustering • Density-based Clustering • Defining clusters as areas of higher density • Advantage • No need to set the number of clusters as input • Disadvantage • Not applicable for datasets with large differences https://upload.wikimedia.org/wikipedia/commons/thumb/a/af/ DBSCAN-Illustration.svg/400px-DBSCAN-Illustration.svg.png http://web.cse.ohio-state.edu/~belkin.8/clustering_workshop_SDM2010/clusters.jpg The University of Texas at San Antonio – Department of Electrical and Computer Engineering 8 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Processing Methods • Neural Networks • What is a Neural Network? • Definition of Dr. Hecht- Nielsen: • “A computing system made up of a number of simple, highly interconnected processing elements, which process information by their dynamic state response to external inputs.” • The Basics • Applications • Long Short-Term Memory http://www.ndt.net/article/v05n07/spanner2/fig2.gif The University of Texas at San Antonio – Department of Electrical and Computer Engineering 9 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Neural Networks • Long Short-Term Memory • A type of Recurrent Neural Networks • Introduced by S. Hochreiter and J. Schmidhuber in 1997 • The Core Idea • Why LSTM? https://tc.sinaimg.cn/maxwidth.800/tc.service.weibo.com/cdn_images_1_medium_com/58ad765e09ea cb5116c9dfc5897c7296.png The University of Texas at San Antonio – Department of Electrical and Computer Engineering 10 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Prediction Using LSTM • Thrust 1, Sub-thrust 1-1 of TECHLAV • Problem definition? • The latency in sending and receiving data by UAVs and UGVs in the presence of heavy computation • The latency is an effect of limitation in communication speed • Solution? • Prediction of future data based on the current data with high accuracy and reconstructing it The University of Texas at San Antonio – Department of Electrical and Computer Engineering 11 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Prediction Using LSTM Feeding Data to UAVs Cloud Latency Processor Computation (Delay) Unit Feeding Data to UGVs The University of Texas at San Antonio – Department of Electrical and Computer Engineering 12 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network Long short-term memory
Data Prediction Using LSTM • Our Methodology • Receiving data from Reference [1], [2], and Creating Dataset • 12,000 samples of angular error • Simulating Latency • Use 70% of Dataset as the current data • Feed the current data to the Neural Network (LSTM) for Training • Finding the pattern of data • Forecasting next 30% of dataset by NN • Reconstructing Data by Adding 70% of the Original Data and the 30% forecasted one • Evaluating the predicted data comparing with original dataset The University of Texas at San Antonio – Department of Electrical and Computer Engineering 13 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Prediction Using LSTM • The Objective 70% of Original Original Dataset Original Dataset Latency Dataset Reconstructed Reconstructed Time Series Data Data Forecasting Predictive Forecasted 30% Model Reconstruction Method The University of Texas at San Antonio – Department of Electrical and Computer Engineering 14 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Prediction Using LSTM • The Structure of LSTM • One input layer Output Gate • One hidden layer with o(t) four LSTM neurons Input Gate • Input gate i(t) • Output gate x(t) sig c(t) sig h(t) • Forget gate • Current Condition of the network • One output layer f(t) Forget Gate • Sigmoid Function The University of Texas at San Antonio – Department of Electrical and Computer Engineering 15 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Prediction Using LSTM • The Structure of LSTM • Sigmoid Function: Used as an activation function for all LSTM Blocks The University of Texas at San Antonio – Department of Electrical and Computer Engineering 16 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Prediction Using LSTM • Results • Root mean square error for train: 0.01856 • Root mean square error for test: 0.02324 • 98.15% accuracy in training • 97.68% accuracy in testing The University of Texas at San Antonio – Department of Electrical and Computer Engineering 17 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Prediction Using LSTM • Results The University of Texas at San Antonio – Department of Electrical and Computer Engineering 18 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Prediction Using LSTM • Results The University of Texas at San Antonio – Department of Electrical and Computer Engineering 19 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Prediction Using LSTM • Results The University of Texas at San Antonio – Department of Electrical and Computer Engineering 20 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Data Prediction Using LSTM • Results The University of Texas at San Antonio – Department of Electrical and Computer Engineering 21 Data Analysis, Estimation, and Fault detection of Large-Scale Autonomous System of Vehicles Using Neural Network
Recommend
More recommend
