Lift Path Planning without Prior Picking/Placing Configurations: Using Crane Location Regions

Abstract

The authors present a probabilistically complete planning approach for handling the lift path planning problem without prior picking/placing configurations. Firstly, this research introduces the concept of crane location regions (CLRs) to represent the constraints that the pose of the lifted object imposes on the crane’s pose at the picking or placing moment. And then the authors develop a rapidly-exploring random trees (RRT)-based planner called the bidirectional multiple RRTs (BiMRRTs), which simultaneously gradually grows multiple trees from the picking pose-constrained space and the placing pose-constrained space instead of just growing a single tree from the picking configuration and the placing configuration. Finally, three simulation experiments are conducted to demonstrate the effectiveness of the proposed approach. The results show that the proposed approach can efficiently generate a collision-free lift path when the picking configuration and the placing configuration are unknown.