Behavior-based Arm Control for an Autonomous Bucket Excavator

Abstract

The paper at hand presents a novel, biologically inspired approach to numerically solve the inverse kinematics problem of arbitrary kinematic chains without instability problems and with low computational complexity. The kinematic chain of the robot arm is transferred to a network of behavior-based joint controllers, which try to deliver the best control value to solve the global problem. By means of controlling a wheeled excavator robot arm, the capabilities of this approach to work for redundant structures with the desired precision are shown.