In industrial settings, robots are typically employed to accurately track a reference force to exert on the surrounding environment to complete interaction tasks. Interaction controllers are typically used to achieve this goal. Still, they either require manual tuning, which demands a significant amount of time, or exact modeling of the environment the robot will interact with, thus possibly failing during the actual application. A significant advancement in this area would be a high-performance force controller that does not need operator calibration and is quick to be deployed in any scenario. With this aim, this paper proposes an Actor-Critic Model Predictive Force Controller (ACMPFC), which outputs the optimal setpoint to follow in order to guarantee force tracking, computed by continuously trained neural networks. This strategy is an extension of a reinforcement learning-based one, born in the context of human-robot collaboration, suitably adapted to robot-environment interaction. We validate the ACMPFC in a real-case scenario featuring a Franka Emika Panda robot. Compared with a base force controller and a learning-based approach, the proposed controller yields a reduction of the force tracking MSE, attaining fast convergence: with respect to the base force controller, ACMPFC reduces the MSE by a factor of 4.35.

In industrial settings, robotic tasks often require interaction with various objects, necessitating compliant manipulation to prevent damage while accurately tracking reference forces.To this aim, interaction controllers are typically employed, but they need either human tinkering for parameter tuning or precise environmental modeling.Both these aspects can be problematic, as the former is a time-consuming procedure, and the latter is unavoidably affected by approximations, hence being prone to failure during the actual application.Addressing these challenges, current research focuses on devising high-performance force controllers.Along this line, this work introduces ORACLE (Optimized Residual Action for Interaction Control with Learned Environments), a novel force control approach.Utilizing neural networks, ORACLE predicts robot-environment interaction forces, which are then used in an optimal residual action controller to locally correct actions from a base force controller, minimizing the expected force-tracking error.Tested on a real Franka Emika Panda robot, ORACLE demonstrates superior force tracking performance compared to state-of-the-art controllers, with a short setup time.