PLANNING OF DYNAMICALLY FEASIBLE TRAJECTORIES FOR TRANSLATIONAL AND PLANAR CABLE-SUSPENDED ROBOTS

Cable-robots are relatively simple robotic manipulators formed by attaching multiple cables to an end-effector. Cable-robots have several desirable advantages over conventional robots. Primarily, they can be designed to have a very large workspace, a very high load capacity, or to generate very high speed motions. Additionally, their simple design makes them inexpensive, modular, transportable and easily reconfigurable. Finally, their minimal moving mass makes them very energy efficient. All these advantages are being promoting the deployment of cable-robots in several real-world applications. A major requirement that has to be met in cable-robots is ensuring that during operation all cables are under tension, and that such a tension is below the maximum permissible value related to the torque limits of the actuators or to tensile force limits of the cables. Assuring feasible tensions in all cables is particularly difficult in underconstrained or cable-suspended robots which use an external force, typically gravity, to maintain their cables in tension. A successful approach to prevent cable slackness and excessive tensions may consist in planning dynamically feasible trajectories by making use of a dynamic model of the cable-suspended robot to translate the cable tension bilateral bounds (i.e. positive and bounded tensile cable forces) into limits on the velocity and acceleration of the robot end-effector along the assigned path. The lecture starts with an introduction to cable-robots, providing some essential definitions and showing successful examples of application of these robots. Subsequently the main open research issues in cable robotics are presented. Then, proceeding from general to particular, the focus is posed on an hybrid, translational and planar cable-suspended robot, proposed as a representative example of cable-suspended robot for which the planning of dynamically feasible trajectories is particularly challenging. The dynamic model of the studied robot is then presented as well as the robot workspace. Afterwards, the attention is focused on the model-based trajectory planning method developed to ensure dynamically feasible trajectories. It is proved that the method leads to kinematic limits that can be incorporated into any trajectory planning algorithm. What is more important, the low computational complexity of the method proposed makes it suitable for implementation in real-time systems. Finally, the validity of the method is proved by experimental results obtained by referring to two paths of industrial interest.