A Hierarchical Approach-Based Energy Consumption Optimization of Industrial Robots

Abstract

Due to the wide application and low energy efficiency of industrial robots (IRs), energy consumption (EC) optimization techniques for them have attracted increasing attention. At present, the methods for optimizing the EC of IRs are generally simple. Energy saving for IRs through conventional methods is always achieved through optimizing the joint trajectory, which generally requires pre-planning of the joint trajectory and has limited energy efficiency improvement. Besides, it is time and labor consuming. In this paper, a hierarchical EC optimization method for IRs based on inverse kinematics solution and dynamic time-scaling method based on modified particle swarm optimization (DTS-MPSO) is proposed, with the previously established EC prediction model. For the primary EC optimization, inverse kinematics for the joint angles is performed firstly under a given initial operating posture of the IRs, then the set of joint angles with the lowest EC is selected. In the secondary EC optimization process, DTS-MPSO optimization is applied to further minimize the EC of IRs. Verification experiments with an ABB robot were conducted according to the joint angles obtained through inverse kinematics and those after secondary optimization. It is proved that EC is saved by 10.68% and 39.27% by the inverse kinematics and DTS-MPSO method respectively. The single EC for the robot to move from the initial posture to the initial operating posture was reduced by 45.76% in total when it operated under the trajectory after hierarchical optimization. The EC is effectively saved, and the trajectory for minimum EC can be automatically planned with this method.