Haptic terrain classification for legged robots

In this paper, we are presenting a method to estimate terrain properties (such as small-scale geometry or surface friction) to improve the assessment of stability and the guiding of foot placement of legged robots in rough terrain. Haptic feedback, expressed through joint motor currents and ground contact force measurements that arises when prescribing a predefined motion was collected for a variety of ground samples (four different shapes and four different surface properties). Features were extracted from this data and used for training and classification by a multiclass AdaBoost machine learning algorithm. In a single leg testbed, the algorithm could correctly classify about 94% of the terrain shapes, and about 73% of the surface samples.

Authors

• Mark A. Höpflinger
• C. David Remy
• Marco Hutter 0001
• Luciano Spinello
• Roland Siegwart

Keywords

• Haptic interfaces
• Legged locomotion
• Rough surfaces
• Surface roughness
• Machine learning algorithms
• Computational geometry
• Friction
• Stability
• Foot
• Force feedback
• Legged Robots
• Terrain Classification
• Surface Properties
• Contact Force
• Single Leg
• Motor Current
• Center Of Mass
• Classification Performance
• Actuator
• Sensor Data
• Knee Joint
• Geometric Properties
• Friction Coefficient
• Geometric Shapes
• Test Setup
• Hemispherical
• False Alarm Rate
• Force Sensor
• Dominant Frequency
• Concave Surface
• Terrain Surface
• Shape Classification
• Dynamic Friction
• Concave Shape
• Standard Laptop
• Type Of Paper
• Haptic Information
• AdaBoost Classifier
• Leg Motion