Computationally efficient energy optimization of multiple robots

This paper presents convex modeling techniques for the problem of optimal velocity control of multiple robots on given intersecting paths. The optimal control problem is formulated as a nonlinear program, which generates predictive state and control trajectories that avoid collisions among the robots and minimize a certain performance index, such as operation time, energy dissipation and actuator usage. Advantages and disadvantages of two modeling approaches are discussed, involving a variable change with square of velocity, or alternatively inverse of velocity. The performance of the two approaches is evaluated in a case study with an industrial robot.