Real-time management of kinematic and safety constraints in motion planning systems

Abstract

In industrial contexts plant efficiency and productivity represent targets which should be pursued by simultaneously accounting for the workers' safeness. Indeed, in modern plants, human workers and autonomous robotic systems share the same workspace and, consequently, safety issues may rise due to the particular working conditions. This thesis deals with two robotic systems largely used in industrial plants: autonomous guided vehicles and anthropomorphic manipulators. Both systems impose satisfying kinematic and safety constraints. Anthropomorphic manipulators, when trajectories are planned in the operational space, could incur in unfeasible situations, the so-called kinematic singularities. A trajectory planner able to manage wrist kinematic singularities in real time has been proposed in the first part of this thesis. For which concerns autonomous guided vehicles, in details laser guided ones, a novel velocity planner has been proposed. It is able to generate, in real time, velocity reference signals which are almost minimum-time and jerk-limited. Despite the heuristic nature of the proposed planning strategy, the solutions obtained are only slightly sub-optimal w.r.t. the ones obtained through a nonlinear solver.