Deep visual foresight for planning robot motion
Citations Over TimeTop 1% of 2017 papers
Abstract
A key challenge in scaling up robot learning to many skills and environments is removing the need for human supervision, so that robots can collect their own data and improve their own performance without being limited by the cost of requesting human feedback. Model-based reinforcement learning holds the promise of enabling an agent to learn to predict the effects of its actions, which could provide flexible predictive models for a wide range of tasks and environments, without detailed human supervision. We develop a method for combining deep action-conditioned video prediction models with model-predictive control that uses entirely unlabeled training data. Our approach does not require a calibrated camera, an instrumented training set-up, nor precise sensing and actuation. Our results show that our method enables a real robot to perform nonprehensile manipulation - pushing objects - and can handle novel objects not seen during training.
Related Papers
- → Two styles of foresight: Are we predicting the future or inventing it?(1996)59 cited
- → FORESIGHT: METHODOLOGICAL AND PRACTICAL ASPECTS(2023)1 cited
- Research Status and Progress of Technology Foresight at Domestic and Abroad(2009)
- → Foresight: Country Specifics and General Patterns(2014)
- Foresight: a modern methodology of technological forecasting (Part 2)(2008)