eduzhai > Applied Sciences > Engineering >

Anomaly Detection using Deep Reconstruction and Forecasting for Autonomous Systems

  • Save

... pages left unread,continue reading

Document pages: 7 pages

Abstract: We propose self-supervised deep algorithms to detect anomalies inheterogeneous autonomous systems using frontal camera video and IMU readings.Given that the video and IMU data are not synchronized, each of them areanalyzed separately. The vision-based system, which utilizes a conditional GAN,analyzes immediate-past three frames and attempts to predict the next frame.The frame is classified as either an anomalous case or a normal case based onthe degree of difference estimated using the prediction error and a threshold.The IMU-based system utilizes two approaches to classify the timestamps; thefirst being an LSTM autoencoder which reconstructs three consecutive IMUvectors and the second being an LSTM forecaster which is utilized to predictthe next vector using the previous three IMU vectors. Based on thereconstruction error, the prediction error, and a threshold, the timestamp isclassified as either an anomalous case or a normal case. The composition ofalgorithms won runners up at the IEEE Signal Processing Cup anomaly detectionchallenge 2020. In the competition dataset of camera frames consisting of bothnormal and anomalous cases, we achieve a test accuracy of 94 and an F1-scoreof 0.95. Furthermore, we achieve an accuracy of 100 on a test set containingnormal IMU data, and an F1-score of 0.98 on the test set of abnormal IMU data.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×