Self-Supervised Visual Planning with Temporal Skip Connections

TL;DR

This paper introduces a self-supervised robotic learning approach using a novel video prediction model with temporal skip-connections, enabling robots to learn complex manipulation skills through direct video prediction and planning.

Contribution

The work presents a new video prediction model with temporal skip-connections and a planning method, significantly improving robotic manipulation capabilities in complex scenarios.

Findings

Outperforms prior models in video prediction-based control

Enables manipulation of unseen objects and multiple objects

Handles occlusions and complex spatial arrangements effectively

Abstract

In order to autonomously learn wide repertoires of complex skills, robots must be able to learn from their own autonomously collected data, without human supervision. One learning signal that is always available for autonomously collected data is prediction: if a robot can learn to predict the future, it can use this predictive model to take actions to produce desired outcomes, such as moving an object to a particular location. However, in complex open-world scenarios, designing a representation for prediction is difficult. In this work, we instead aim to enable self-supervised robotic learning through direct video prediction: instead of attempting to design a good representation, we directly predict what the robot will see next, and then use this model to achieve desired goals. A key challenge in video prediction for robotic manipulation is handling complex spatial arrangements such as occlusions. To that end, we introduce a video prediction model that can keep track of objects through occlusion by incorporating temporal skip-connections. Together with a novel planning criterion and action space formulation, we demonstrate that this model substantially outperforms prior work on video prediction-based control. Our results show manipulation of objects not seen during training, handling multiple objects, and pushing objects around obstructions. These results represent a significant advance in the range and complexity of skills that can be performed entirely with self-supervised robotic learning.