Biblio

For the task with complicated manipulation in unstructured environments, traditional hand-coded methods are ineffective, while reinforcement learning can provide more general and useful policy. Although the reinforcement learning is able to obtain impressive results, its stability and reliability is hard to guarantee, which would cause the potential safety threats. Besides, the transfer from simulation to real-world also will lead in unpredictable situations. To enhance the safety and reliability of robots, we introduce the force and haptic perception into reinforcement learning. Force and tactual sensation play key roles in robotic dynamic control and human-robot interaction. We demonstrate that the force-based reinforcement learning method can be more adaptive to environment, especially in sim-to-real transfer. Experimental results show in object pushing task, our strategy is safer and more efficient in both simulation and real world, thus it holds prospects for a wide variety of robotic applications.

To decouple the multi-axis motion in the 6 degrees of freedom magnetically levitated actuators (MLAs), this paper introduces a numerical method to model the force and torque distribution. Taking advantage of the Gaussian quadrature, the concept of coil node is developed to simplify the Lorentz integral into the summation of the interaction between each magnetic node in the remanence region and each coil node in the coil region. Utilizing the coordinate transformation in the numerical method, the computation burden is independent of the position and the rotation angle of the moving part. Finally, the experimental results prove that the force and torque predicted by the numerical model are rigidly consistent with the measurement, and the force and torque in all directions are decoupled properly based on the numerical solution. Compared with the harmonic model, the numerical wrench model is more suitable for the MLAs undertaking both the translational and rotational displacements.