In this work, an inverse-kinematics model, an inverse-dynamics model, an efficient and inexpensive controller design for a 2PRU-1PRS parallel manipulator, and a duly validated physical prototype of the same have been presented. The work is important because such manipulators offer wide workspaces and thus can be used for several applications including simultaneous multidirectional vibration testing. It was ensured that the dynamics model was accurate and could also be simplified such that it could be integrated with an efficient and fast-feedforward controller. Our models were duly validated in the MSC \(\hbox {ADAMS}^\text {TM}\) environment. Further, three different controllers were developed and evaluated in terms of their trajectory-tracking capability. Here, the aim was to develop a controller that can accurately track the system trajectory, require a reasonable amount of actuator effort, and are computationally fast enough to be implemented on inexpensive development boards. Out of the three controllers designed, the first one, i.e., the computed force controller (CFC), is very accurate but uses intensive computational resources as it requires real-time solutions of coupled nonlinear inverse-dynamic equations. In contrast, the second controller, i.e., the modified CFC (MCFC), is equally precise but also very fast, as it utilizes a simplified and yet accurate version of our original inverse-dynamics model. Both controllers significantly outperformed the standard PID controller. Finally, our model was prototyped and used to validate the proposed trajectory-tracking method, i.e., the MCFC. The physical prototype consisted of a moving platform in the task space with three degrees of freedom: roll, pitch, and heave. Predictions from simulations and experimental data on trajectory tracking firmly establish that the MCFC performs as per expectations and can be used for several applications.
