Episodic Memory Model for Learning Robotic Manipulation Tasks

TL;DR

This paper introduces an episodic memory model enabling robots to learn complex manipulation tasks from a single demonstration by decomposing tasks into sub-tasks and recognizing state changes.

Contribution

The proposed model allows robots to understand and perform complex tasks by learning from a single demonstration and recognizing state changes, improving task decomposition.

Findings

Effective in learning complex manipulation tasks from one demonstration

Enables task decomposition into sub-tasks based on state changes

Improves robot performance in assembly and machine tending tasks

Abstract

Machine learning, artificial intelligence and especially deep learning based approaches are often used to simplify or eliminate the burden of programming industrial robots. Using these approaches robots inherently learn a skill instead of being programmed using strict and tedious programming instructions. While deep learning is effective in making robots learn skills, it does not offer a practical route for teaching a complete task, such as assembly or machine tending, where a complex logic must be understood and related sub-tasks need to be performed. We present a model similar to an episodic memory that allows robots to comprehend sequences of actions using single demonstration and perform them properly and accurately. The algorithm identifies and recognizes the changes in the states of the system and memorizes how to execute the necessary tasks in order to make those changes. This allows the robot to decompose the tasks into smaller sub-tasks, retain the essential steps, and remember how they have been performed.