Orientation Planning For Multi-Agents With Discrete-Step Locomotion And Multiple Goals

The paper addresses the coordinated path planning of mobile agents with multiple goal positions and orientations in a plane. The targeted multi-robot system uses discrete locomotion ensuring uncertainty free localization and mapping as well as simple and robust control. It is suitable for material-handling, reconfigurable-fixturing, and mobile-manipulation tasks in a flexible-manufacturing environment. Using its three leg, and matching pin-socket couplings with the base surface, each agent either stands fixed or strides along via "Swing and Dock" (SaD) locomotion. Each mounting pin can serve both as a connecting-locking device and as a pivot of a planar rotation. Previous work offered planning solutions only for the agents' positions. In reality, the orientation in which the agent arrives at the goal is very important because neither robot workspaces nor work cell geometries have axial symmetry. Herein, we provide for the required orientational planning by labelling the agent's legs to keep track of its rotation. Integer Linear Programming (ILP) is used to model the path planning problem in the so augmented configuration space. The mathematical formulations are implemented and tested using a GUROBI solver. Computational results display the effectiveness of the approach.