AcTExplore: Active Tactile Exploration on Unknown Objects

Tactile exploration plays a crucial role in understanding object structures for fundamental robotics tasks such as grasping and manipulation. However, efficiently exploring such objects using tactile sensors is challenging, primarily due to the large-scale unknown environments and limited sensing coverage of these sensors. To this end, we present AcTExplore, an active tactile exploration method driven by reinforcement learning for object reconstruction at scales that automatically explores the object surfaces in a limited number of steps. Through sufficient exploration, our algorithm incrementally collects tactile data and reconstructs 3D shapes of the objects as well, which can serve as a representation for higher-level downstream tasks. Our method achieves an average of 95. 97% IoU coverage on unseen YCB objects while just being trained on primitive shapes.

Authors

• Amir-Hossein Shahidzadeh
• Seong Jong Yoo
• Pavan Mantripragada
• Chahat Deep Singh
• Cornelia Fermüller
• Yiannis Aloimonos

Keywords

• Surface reconstruction
• Three-dimensional displays
• Shape
• Tactile sensors
• Reinforcement learning
• Grasping
• Sensors
• Unknown Objects
• Active Tactile Exploration
• Number Of Steps
• Intersection Over Union
• Tactile Sensor
• Unknown Environment
• Efficient Exploration
• Unseen Objects
• Deep Learning
• Short-term Memory
• Long Short-term Memory
• Workspace
• Local Optimum
• Temporal Information
• State Representation
• Reward Function
• Deep Reinforcement Learning
• Markov Decision Process
• Tactile Information
• Intrinsic Rewards
• Proximal Policy Optimization
• Scene Perception
• Skin Deformation
• Exploration Algorithm
• Negative Reward
• Temporal Representation