Minimizing cycle time and energy consumption for a multi-degree serial manipulator using teaching-learning-based optimization

Multiple inverse kinematic solutions are obtained for a multi-degree serial manipulator. Each solution provides a different cycle time and energy consumed to perform a given task in a particular sequence. Improper selection of inverse kinematic solution in a robotic assembly line may lead to an increase in not only overall cycle time but also energy consumption. Therefore, it is necessary to optimize the robot's performance with respect to cycle time and energy consumption simultaneously. These two objectives usually conflict due to joint motors' different operational speeds and energy consumption rates. Hence, in this work, an attempt is made to simultaneously optimize both cycle time and energy consumption by minimizing the differences in joint angles of the robot manipulator while moving from one position to another in applications involving multi-point movement of the robot end effector. The proposed approach is demonstrated by considering an example of a commonly used six-degree-of-freedom Kuka KR5-A robot manipulator. It can be extended efficiently for any other robot manipulator. A relatively new algorithm known as teaching-learning-based optimization (TLBO) algorithm is employed to solve this problem. It is observed that the results obtained by TLBO show a significant reduction of about 20% and 10% for cycle time, while 3% and 15% for energy consumption over the results obtained by genetic algorithm (GA) and sequential quadratic programming (SQP), respectively.