A Force-Based Formation Synthesis Approach for the Cooperative Transportation of Objects

Abstract

Increasing demands in automation entail that some tasks either cannot be accomplished by a single robotic manipulator or it is economically not meaningful to indiscriminately increase the robot’s size or sophistication. At the same time, ongoing technological progress has paved the way for an increased usage of cost-effective robotic agents. Hence, the question whether the cooperation of multiple simple robotic agents is expedient to accomplish automation tasks is raised. Possible benefits are manifold, e.g., an increased versatility, resilience regarding failure, and the ability to solve more complex and larger tasks. A meaningful benchmark problem in distributed robotics is the cooperative object transportation where mobile robots organize around the object to move it by pushing. Thus, this paper deals with the question of how to allocate robotic agents around the object. Novelly, the problem is treated using second-order dynamics, explicitly including and limiting the exerted force of the agents. We derive a description of the manipulation space in terms of a zonotope which is useful for robotic manipulation beyond the scope of this benchmark problem. The proposed scheme’s versatility and functionality is demonstrated by experimental results.

This work is supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Grant 433183605 and through Germany’s Excellence Strategy (Project PN4-4 Theoretical Guarantees for Predictive Control in Adaptive Multi-Agent Scenarios) under Grant EXC2075-390740016.