Self-Supervised Visual Planning with Temporal Skip Connections

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.

accepted at the Conference on Robot Learning (CoRL) 2017

References in corpus (3)

• Generating Videos with Scene Dynamics
• Uncertainty-Aware Reinforcement Learning for Collision Avoidance
• Parallel Multiscale Autoregressive Density Estimation

Cited by in corpus (10)

• Keypoints into the Future: Self-Supervised Correspondence in Model-Based Reinforcement Learning
• Learning Robotic Manipulation through Visual Planning and Acting
• RoboNet: Large-Scale Multi-Robot Learning
• Experience-Embedded Visual Foresight
• Learning Semantic Embedding Spaces for Slicing Vegetables
• Reasoning and Generalization in RL: A Tool Use Perspective
• Morphology-Agnostic Visual Robotic Control
• Cubic LSTMs for Video Prediction
• Cross-Identity Motion Transfer for Arbitrary Objects through Pose-Attentive Video Reassembling
• Depth by Poking: Learning to Estimate Depth from Self-Supervised Grasping