This study proposes a method based on reinforcement learning (RL) for point-to-point and dynamic reference position tracking control of a planar cable-driven parallel robots, which is a multi-input multi-output system (MIMO). The method eliminates the use of a tension distribution algorithm in controlling the system’s dynamics and inherently optimizes the cable tensions based on the reward function during the learning process. The deep deterministic policy gradient algorithm is utilized for training the RL agents in point-to-point and dynamic reference tracking tasks. The performances of the two agents are tested on their specifically trained tasks. Moreover, we also implement the agent trained for point-to-point tasks on the dynamic reference tracking and vice versa. The performances of the RL agents are compared with a classical PD controller. The results show that RL can perform quite well without the requirement of designing different controllers for each task if the system’s dynamics is learned well.

Redundant actuation for the parallel kinematic machine (PKM) is a well-known technique for overcoming general drawbacks of the PKM by helping it to avoid singularity and enhance stiffness characteristics, among others. Torque distribution plays a critical role in redundant actuation because this actuation causes the PKM to consume too much energy or put a substantial amount of stress on joints and links. This paper proposes a new torque distribution method for reducing the maximum torque of the actuator of a planar PKM. Here the main idea behind the proposed method is the use of superposition of a particular solution for a non-redundant case and an optimized null-space solution for a redundant case with a constant coefficient. The optimal value of a null-space solution can be easily determined by checking only the intersection points of the profile of the actuator's torque as the coefficient varies. We consider three cases of planar PKMs—2-, 3-, and 4-RRR PKMs—and present a detailed procedure for deriving a kinematic solution for the 2-RRR PKM based on Screw theory. We compare the proposed method with the minimum-norm pseudo-inverse method and assess a limitation of the proposed method. The torque distribution algorithm can be used to determine the number of actuators in an efficient manner and to reduce energy consumption.

We recently developed the Image-Charge Solvation Model (ICSM), which is an explicit/implicit hybrid model to accurately account for long-range electrostatic forces in molecular dynamics simulations [Lin et al., J. Chem. Phys., 131,154103,2009]. The ICSM has a productive spherical volume within the simulation cell for which key physical properties of bulk water are reproduced, such as density, radial distribution function, diffusion constants and dielectric properties. Although the reaction field (RF) is essential, it typically accounts for less than 2% of the total electrostatic force on a water molecule. This observation motivates investigating further the role of the RF within the ICSM. In this report we focus on distributions of forces and torques on water molecules as a function of distance from the origin and make extensive tests over a range of model parameters where Coulomb forces are decomposed into direct interactions from waters modeled explicitly and the RF. Molecular torques due to the RF typically account for 20% of the total torque, revealing why the RF plays an important role in the dielectric properties of simulated water. Moreover, it becomes clear that the buffer layer in the ICSM is essential to mitigate artifacts caused by the discontinuous change in dielectric constants at the explicit/implicit interface.

In this paper, a novel optimal torque distribution method for a redundantly actuated parallel robot is proposed. Geometric analysis based on screw theory is performed to calculate the stiffness matrix of a redundantly actuated 3-RRR parallel robot. The analysis is performed based on statics focusing on low-speed motions. The stiffness matrix consisting of passive and active stiffness is also derived by the differentiation of Jacobian matrix. Comparing two matrices, we found that null-space vector is related to link geometry. The optimal distribution torque is determined by adapting mean value of minimum and maximum angles as direction angles of null-space vector. The resulting algorithm is validated by comparing the new method with the minimum-norm method and the weighted pseudo-inverse method for two different paths and force conditions. The proposed torque distribution algorithm shows the characteristics of minimizing the maximum torque.