Towards Transferring Tactile-based Continuous Force Control Policies from Simulation to Robot
This work addresses the challenge of safe object manipulation in robotics by enabling sim-to-real transfer for force control, though it is incremental as it builds on existing tactile sensor and reinforcement learning methods.
The paper tackles the problem of transferring tactile-based continuous force control policies from simulation to a real robot without fine-tuning, using a model-free deep reinforcement learning approach, and shows it outperforms a hand-modeled baseline in evaluation.
The advent of tactile sensors in robotics has sparked many ideas on how robots can leverage direct contact measurements of their environment interactions to improve manipulation tasks. An important line of research in this regard is that of grasp force control, which aims to manipulate objects safely by limiting the amount of force exerted on the object. While prior works have either hand-modeled their force controllers, employed model-based approaches, or have not shown sim-to-real transfer, we propose a model-free deep reinforcement learning approach trained in simulation and then transferred to the robot without further fine-tuning. We therefore present a simulation environment that produces realistic normal forces, which we use to train continuous force control policies. An evaluation in which we compare against a baseline and perform an ablation study shows that our approach outperforms the hand-modeled baseline and that our proposed inductive bias and domain randomization facilitate sim-to-real transfer. Code, models, and supplementary videos are available on https://sites.google.com/view/rl-force-ctrl