Spatially-Distributed Missions With Heterogeneous Multi-Robot Teams

Abstract

This work is about mission planning in teams of mobile autonomous agents. We consider tasks that are spatially distributed, non atomic, and provide an utility for integral and also partial task completion. Agents are heterogeneous, therefore showing different efficiency when dealing with the tasks. The goal is to define a system-level plan that assigns tasks to agents to maximize mission performance. We define the mission planning problem through a model including multiple sub-problems that are addressed jointly: task selection and allocation, task scheduling, task routing, control of agent proximity over time. The problem is proven to be NP-hard and is formalized as a mixed integer linear program (MILP). Two solution approaches are proposed: one heuristic and one exact method. Both combine a generic MILP solver and a genetic algorithm, resulting in efficient anytime algorithms. To support performance scalability and to allow the effective use of the model when online continual replanning is required, a decentralized and fully distributed architecture is defined top-down from the MILP model. Decentralization drastically reduces computational requirements and shows good scalability at the expenses of only moderate losses in performance. Lastly, we illustrate the application of the mission planning framework in two demonstrators. These implementations show how the framework can be successfully integrated with different platforms, including mobile robots (ground and aerial), wearable computers, and smart-phone devices.

Hierarchical Planning for Heterogeneous Multi-Robot Routing Problems via Learned Subteam Performance

“…The last decade witnessed a growing interest in this field from the robotics community as well, under flavors that more closely match the application scenarios of multi-robot systems. Examples can be found in applications like search [5], monitoring [6], coverage [7], agriculture [8], package delivery [9], [10], and, more generally, the execution of spatially distributed "tasks" [11], [12]. Oberlin et al [11] and Prasad et al [12] study routing problems that are similar to the basic routing task considered in our testbed problem, but with modeling assumptions that might not be applicable to many robotic scenarios.…”

“…Oberlin et al [11] and Prasad et al [12] study routing problems that are similar to the basic routing task considered in our testbed problem, but with modeling assumptions that might not be applicable to many robotic scenarios. Solving routing problems by formulating them as MILPs is one of the most pupular solution approaches [5], [6], [9]- [11], and for this reason the one we use for our experimental campaign. We remark that the proposed planning framework is not designed around the specific application used as a testbed in this paper.…”

“…The robots are initially placed at the entrance of the area to be inspected, which is denoted by a special entry vertex (not to be inspected) s ∈ V . 5 An inspection path π r for a robot r ∈ R is defined as a sequence of inspected vertices starting at the entry vertex: π r = [v 1 = s, v 2 , . .…”

“…We start by considering routing tasks defined on areas represented as a grid graphs, an abstraction widely used in the multi-robot routing literature [5], [6], [10]. In particular, each area is represented as a 9x9 square grid graph with holes modeling untraversable locations.…”