Experimental evaluation of energy optimized trajectories for industrial robots

Abstract

This thesis presents a method reducing the energy consumption in industrial robots up to more than 30%.The reduction is possible by the use of a smart optimization algorithm tuning the individual robot motions and improving the coordination of multiple robots.To increase the sustainability of the automated manufacturing industry, the large energy consumption in industrial robots must be reduced.Various approaches, such as new control systems, lower robot weight, DC-based architectures and smart sleep mode, are being evaluated today.However, with over 1.3 million industrial robots operating worldwide, a solution for already existing robots is also desirable.The challenges are to develop an energy-optimized solution and to integrate it in existing robots.The experiments and evaluations in this thesis have been conducted on real industrial robots available in the robotic laboratory at Chalmers University of Technology.Based on the current robot behavior, various approaches for reducing the energy consumption of predefined trajectories have been evaluated.The results show that the use of an optimization algorithm for minimizing the squared acceleration can reduce the energy consumption in a robot significantly.When coordinating multiple robots the reduction can be increased even further, by the use of a smart zone-booking system reducing waiting time.Furthermore, tools for measuring robot performance and for implementing and executing optimized trajectories in the robots have also been developed.The latter is an important contribution towards integrating the final optimization strategy in existing robots.The resulting energy reduction of over 30% indicates, along with the execution method, that a fully integrated solution is feasible and likely to provide satisfying results.