Programmable independent operation and closed-loop control of a soft robotic table for simultaneous manipulation of multiple objects

This paper presents the design, development and validation of a closed-loop control system for a soft robotic table. A soft table can simultaneously provide independent manipulation of multiple objects, which is an advantage over traditional devices such as robotic arms and conveyor belt systems. This table also allows for the inherent smooth handling of delicate objects. In order to implement the automation tasks, a vision feedback system and an operation controller are developed and integrated into the existing desk prototype to form a closed-loop control. The vision feedback system implements image processing techniques to identify the object position from captured images. The operational controller is integrated from three sets of algorithms including motion planning, stimulus selection and stimulation planning. Two sets of tests were conducted to validate the closed loop system and the soft table capability. The first set confirms the manipulative task of moving an object to a desired location with independent planning and execution. The second set confirms the table's ability to manipulate multiple objects simultaneously and independently. Videos of both tests have been prepared. Attention experts - a soft robotic table has been developed for manipulating delicate objects on a surface. This table is intended for automatic sorting tasks in an industrial environment. Multiple objects placed on the table surface can be simultaneously and independently moved and oriented on the horizontal XY plane. The deformable surface of the table allows smooth manipulation of delicate objects. These are advantages over existing manipulators such as robotic arms and conveyor belt systems. Based on previous experiments, it was found that open-loop control is inefficient for performing automation tasks due to the variability in the motion of objects with soft interactions. To solve this problem, a closed-loop control system for independent object manipulation has been developed, and its detailed design and implementation are presented in this paper. A vision feedback system and a controller are developed to form the proposed closed-loop control. Two sets of experiments involving routine manipulation tasks were performed. It has been proven that the control system is effective in controlling the uncertainty in the object movements and the automatic manipulation of the object by the soft robotic table. The performance of this control system can be improved by better predicting both the deformations on the table surface and the resulting object movements. Placing deformable sensors on the surface of the table is a potential method to achieve such an improvement, and this issue can be explored in future work.