Consider the problem of real-time motion generation for a mobile manipulator called to execute aggressive motions. For this problem, we developed a method that is based on Nonlinear Model Predictive Control (NMPC). The robot balance is maintained through an appropriate constraint that limits the robot motions to those generating non-negative moments around the edges of the support polygon.
In this video, we show 5 scenarios of increasing difficulty in which we validate our approach and we compare it with two methods from the literature.
The simulations show that our method can effectively handle end-effector tasks that require aggressive motions maintaining the robot balance even in cases where the compared methods fail.
