Multi-Modal Anomaly Detection for Unstructured and Uncertain Environments
This work provides a method for improving robot autonomy by enabling better failure identification in complex, real-world environments, which is crucial for reducing human supervision.
This paper addresses the challenge of anomaly detection in robots operating in unstructured environments by proposing a deep learning neural network called Supervised Variational Autoencoder (SVAE). The SVAE model effectively fuses high-dimensional, multi-modal sensor data to identify failures, demonstrating superior performance compared to baseline methods on real field robot data.
To achieve high-levels of autonomy, modern robots require the ability to detect and recover from anomalies and failures with minimal human supervision. Multi-modal sensor signals could provide more information for such anomaly detection tasks; however, the fusion of high-dimensional and heterogeneous sensor modalities remains a challenging problem. We propose a deep learning neural network: supervised variational autoencoder (SVAE), for failure identification in unstructured and uncertain environments. Our model leverages the representational power of VAE to extract robust features from high-dimensional inputs for supervised learning tasks. The training objective unifies the generative model and the discriminative model, thus making the learning a one-stage procedure. Our experiments on real field robot data demonstrate superior failure identification performance than baseline methods, and that our model learns interpretable representations. Videos of our results are available on our website: https://sites.google.com/illinois.edu/supervised-vae .