A Non-Time Based Controller for Load Swing Damping and Path-Tracking in Robotic Cranes

This paper proposes the use of the non-time based control strategy named Delayed Reference Control (DRC) to the control of industrial robotic cranes. Such a control scheme has been developed to achieve two relevant objectives in the control of autonomous operated cranes: the active damping of undesired load swing, and the accurate tracking of the planned path through space, with the preservation of the coordinated Cartesian motion of the crane. A paramount advantage of the proposed scheme over traditional ones is its ease of implementation on industrial devices: it can be implemented by just adding an outer control loop (incorporating path planning) to standard position controllers. Experimental performance assessment of the proposed control strategy is provided by applying the DRC to the control of the oscillation of a cable-suspended load moved by a parallel robot mimicking a robotic crane. In order to implement the DRC scheme on such an industrial robot it has been just necessary to manage path planning and the DRC algorithm on a separate real-time hardware computing the delay in the execution of the desired trajectory suitable to reduce load swing. Load swing has been detected by processing the images from two off-the-shelf cameras with a dedicated vision system. No customization of the robot industrial controller has been necessary.