Robust Multi-Modal Policies for Industrial Assembly via Reinforcement Learning and Demonstrations: A Large-Scale Study
This work addresses the lack of industrial adoption of learning-based assembly methods, showing DRL can surpass engineered and human performance, though it is incremental in refining existing approaches for practical use.
The paper tackles the challenge of adopting deep reinforcement learning (DRL) for industrial assembly by proposing an industry-oriented paradigm and demonstrates that their DRL system consistently outperforms a professional industrial integrator in speed and reliability on the NIST benchmark, and even beats a human on a moving target insertion task.
Over the past several years there has been a considerable research investment into learning-based approaches to industrial assembly, but despite significant progress these techniques have yet to be adopted by industry. We argue that it is the prohibitively large design space for Deep Reinforcement Learning (DRL), rather than algorithmic limitations per se, that are truly responsible for this lack of adoption. Pushing these techniques into the industrial mainstream requires an industry-oriented paradigm which differs significantly from the academic mindset. In this paper we define criteria for industry-oriented DRL, and perform a thorough comparison according to these criteria of one family of learning approaches, DRL from demonstration, against a professional industrial integrator on the recently established NIST assembly benchmark. We explain the design choices, representing several years of investigation, which enabled our DRL system to consistently outperform the integrator baseline in terms of both speed and reliability. Finally, we conclude with a competition between our DRL system and a human on a challenge task of insertion into a randomly moving target. This study suggests that DRL is capable of outperforming not only established engineered approaches, but the human motor system as well, and that there remains significant room for improvement. Videos can be found on our project website: https://sites.google.com/view/shield-nist.