Safe Robot Navigation via Multi-Modal Anomaly Detection

Abstract

Navigation in natural outdoor environments requires a robust and reliable traversability classification method to handle the plethora of situations a robot can encounter. Binary classification algorithms perform well in their native domain but tend to provide overconfident predictions when presented with out-of-distribution samples, which can lead to catastrophic failure when navigating unknown environments. We propose to overcome this issue by using anomaly detection on multi-modal images for traversability classification, which is easily scalable by training in a self-supervised fashion from robot experience. In this work, we evaluate multiple anomaly detection methods with a combination of uni- and multi-modal images in their performance on data from different environmental conditions. Our results show that an approach using a feature extractor and normalizing flow with an input of RGB, depth and surface normals performs best. It achieves over 95% area under the ROC curve and is robust to out-of-distribution samples.

References in corpus (1)

• Voxblox: Incremental 3D Euclidean Signed Distance Fields for On-Board MAV Planning

Cited by in corpus (16)

• A Unifying Review of Deep and Shallow Anomaly Detection
• Learning Robust Autonomous Navigation and Locomotion for Wheeled-Legged Robots
• WayFAST: Navigation with Predictive Traversability in the Field
• Visual-Tactile Cross-Modal Data Generation using Residue-Fusion GAN with Feature-Matching and Perceptual Losses
• Learning Off-Road Terrain Traversability with Self-Supervisions Only
• ScaTE: A Scalable Framework for Self-Supervised Traversability Estimation in Unstructured Environments
• Sim-to-Lab-to-Real: Safe Reinforcement Learning with Shielding and Generalization Guarantees
• Learning Robotic Navigation from Experience: Principles, Methods, and Recent Results
• An Outlier Exposure Approach to Improve Visual Anomaly Detection Performance for Mobile Robots
• Using Visual Anomaly Detection for Task Execution Monitoring
• OneFlow: One-class flow for anomaly detection based on a minimal volume region
• Sensing Anomalies as Potential Hazards: Datasets and Benchmarks
• Do You Know the Way? Human-in-the-Loop Understanding for Fast Traversability Estimation in Mobile Robotics
• An Expert Ensemble for Detecting Anomalous Scenes, Interactions, and Behaviors in Autonomous Driving
• Out-of-Distribution Detection for Adaptive Computer Vision
• Random Similarity Isolation Forests