The purpose of this paper is to present a control system for a heavy duty industrial robot, including both the control structure and algorithm, which was designed and tested.
An industrial PC with TwinCAT real‐time system is chosen as the motion control unit; EtherCAT is used for command transmission. The whole system has a decoupled and centralized control structure. A novel optimal motion generation algorithm based on modified cubic spline interpolation is illustrated. The execution time and work were chosen as the objective function. The constraints are the limits of torque, velocity and jerk. The motion commands were smooth enough throughout the execution period. By using the Lagangue equation and assumed modes methods, a dynamic model of heavy duty industrial robots is built considering the elastic of both joints and links. After that a compound control algorithm based on singular perturbation theory was designed for the servo control loop.
The final experimental results showed that the control commands and algorithms could easily be calculated and transmitted in one sample unit. Both the motion generation and servo control algorithm greatly improved the control performance of the robot.
All parts of the control algorithm can be computed on‐line except the optimal motion generation part. The motion generation part is time consuming (about 2.5 seconds), which can only be performed off‐line. Hence future work will focus on improving the efficiency of this algorithm; therefore it could be performed online, increasing the robot's overall robustness and adaptability.
Aiming at the internal and external causes that limit the dynamic performance of heavy duty industrial robots, this paper proposes a realizable scheme of control system and includes both the control structure and algorithms. A novel optimal motion generation algorithm is presented.
