Improving Robotic Manipulation: Techniques for Object Pose Estimation, Accommodating Positional Uncertainty, and Disassembly Tasks from Examples

TL;DR

This paper advances robotic manipulation by integrating tactile sensing and reinforcement learning to improve object pose estimation, handle positional uncertainty, and perform disassembly tasks efficiently from human examples.

Contribution

It introduces novel techniques combining tactile sensing and reinforcement learning for better manipulation in unstructured environments, addressing pose estimation and task execution from examples.

Findings

Tactile sensing improves object pose accuracy under occlusion.

Reinforcement learning reduces grasp attempts amid positional uncertainty.

Learning from human examples accelerates disassembly task training.

Abstract

To use robots in more unstructured environments, we have to accommodate for more complexities. Robotic systems need more awareness of the environment to adapt to uncertainty and variability. Although cameras have been predominantly used in robotic tasks, the limitations that come with them, such as occlusion, visibility and breadth of information, have diverted some focus to tactile sensing. In this thesis, we explore the use of tactile sensing to determine the pose of the object using the temporal features. We then use reinforcement learning with tactile collisions to reduce the number of attempts required to grasp an object resulting from positional uncertainty from camera estimates. Finally, we use information provided by these tactile sensors to a reinforcement learning agent to determine the trajectory to take to remove an object from a restricted passage while reducing training time by pertaining from human examples.