Contrastive Learning-Based Attribute Extraction Method for Enhanced Terrain Classification

The outdoor environment has many uneven surfaces that put the robot at risk of sinking or tipping over. Recognizing the type of terrain can help robot avoid risks and choose an appropriate gait. One of the critical problems is how to extract the terrain-related knowledge from sensor data collected as the robot traversed the ground. Many existing vision-based approaches are limited in directly perceiving the intrinsic properties of various terrains. The intuitive approach entails directly analyzing data recorded by the robot’s proprioceptive sensors. However, it faces challenges in being specific to certain robot leg configurations or in the lack of interpretability of the extracted features. In this paper, a terrain attribute extraction algorithm is proposed based on contrastive learning. It leverages the haptic data generated from the interaction between the robot’s legs and terrain to automatically extract terrain attributes. The results demonstrate that the attributes extracted using this method strongly correlate with the actual softness of the terrain. Furthermore, these attributes played an important role in achieving high accuracy in terrain classification tasks.

Authors

• Xiao Liu
• Hongjin Chen
• Haoyao Chen

Keywords

• Legged locomotion
• Dimensionality reduction
• Accuracy
• Propioception
• Contrastive learning
• Feature extraction
• Robot sensing systems
• Terrain Classification
• Softening
• Sensor Data
• Self-supervised Learning
• Terrain Types
• Legged Robots
• Normal Distribution
• Loss Function
• Time Series
• Superior Performance
• Time Series Data
• Feature Representation
• Linear Discriminant Analysis
• Temporal Dimension
• Unsupervised Learning
• K-nearest Neighbor
• Gaussian Mixture Model
• Feature Extraction Methods
• Empirical Measures
• Contrastive Loss
• High-dimensional Representation
• Single-linkage Clustering
• Contextual Representation
• Silhouette Coefficient
• Comparison Of Experimental Results
• Clustering Quality
• Root Mean Square Error
• Exteroceptive
• Quantitative Metrics