Advanced 3D Solid Modeling, Virtual Animation and Control of Smart Cranes

This project proposes the high precision intelligent sensor based robotic control system to enhance the performance accuracy of a robotics system. The proposed design concept: (1) optimally combines several classes of sensors (e.g. touch, force, proximity, etc.) to improve the path planning and trajectory evaluation (i.e., controlled trajectory enhancement); and (2) innovatively employs multi-level intelligent controllers, which enable robots to intelligently decide for themselves the actions to be taken in new situations or totally new operating environments. The proposed concept employs the synergistic characteristics of robust control and intelligent control to improve the robot closed-loop servo control system. Its mechanism possesses three integrated levels of control. The robust control realizes the lowest (inner loop) level stabilization and slewing control. The learning control in the middle level is used as a turbocharger to enhance the precision of the servo control loop. The fuzzy neural network at the highest level merges the knowledge of human experts and input-output data from a multi-sensor platform o provide adaptive operation at capabilities for the robot. The proposed scheme provides a power technique in supporting the in-process stabilization motion control, decision making, and fault detection and reconfiguration subject to system failures or stressing operating conditions. A high precision redundant robotic manipulator which has been designed and fabricated at LLNL will be utilized as a testbed for the advanced intelligent robotic control methodologies to be developed in the proposed Phase I work.