Cooperative Multi-Robot Task Allocation with Reinforcement Learning

This paper deals with the concept of multi-robot task allocation, referring to the assignment of multiple robots to tasks such that an objective function is maximized. The performance of existing meta-heuristic methods worsens as the number of robots or tasks increases. To tackle this problem, a novel Markov decision process formulation for multi-robot task allocation is presented for reinforcement learning. The proposed formulation sequentially allocates robots to tasks to minimize the total time taken to complete them. Additionally, we propose a deep reinforcement learning method to find the best allocation schedule for each problem. Our method adopts the cross-attention mechanism to compute the preference of robots to tasks. The experimental results show that the proposed method finds better solutions than meta-heuristic methods, especially when solving large-scale allocation problems.

6G Mobile Communications for Multi-Robot Smart Factory

Private or special-purpose wireless networks present a new technological trend for future mobile communications, while one attractive application scenario is the wireless communication in a smart factory. In addition to wireless technologies, this paper pays special attention to treat a smart factory as the integration of collaborative multi-robot systems for production robots and transportation robots. Multiple aspects of collaborative multi-robot systems enabled by wireless networking have been investigated, dynamic multi-robot task assignment for collaborative production robots and subsequent transportation robots, social learning to enhance precision and robustness of collaborative production robots, and more efficient operation of collaborative transportation robots. Consequently, the technical requirements of 6G mobile communication can be logically highlighted.

Multi-robot task allocation problem with multiple nonlinear criteria using branch and bound and genetic algorithms

The paper proposes the formulation of a single-task robot (ST), single-robot task (SR), time-extended assignment (TA), multi-robot task allocation (MRTA) problem with multiple, nonlinear criteria using discrete variables that drastically reduce the computation burden. Obtaining an allocation is addressed by a Branch and Bound (B&B) algorithm in low scale problems and by a genetic algorithm (GA) specifically developed for the proposed formulation in larger scale problems. The GA crossover and mutation strategies design ensure that the descendant allocations of each generation will maintain a certain level of feasibility, reducing greatly the range of possible descendants, and accelerating their convergence to a sub-optimal allocation. The proposed MRTA algorithms are simulated and analyzed in the context of a thermosolar power plant, for which the spatially distributed Direct Normal Irradiance (DNI) is estimated using a heterogeneous fleet composed of both aerial and ground unmanned vehicles. Three optimization criteria are simultaneously considered: distance traveled, time required to complete the task and energetic feasibility. Even though this paper uses a thermosolar power plant as a case study, the proposed algorithms can be applied to any MRTA problem that uses a multi-criteria and nonlinear cost function in an equivalent way. The performance and response of the proposed algorithms are compared for four different scenarios. The results show that the B&B algorithm can find the global optimal solution in a reasonable time for a case with four robots and six tasks. For larger problems, the genetic algorithm approaches the global optimal solution in much less computation time. Moreover, the trade-off between computation time and accuracy can be easily carried out by tuning the parameters of the genetic algorithm according to the available computational power.

Multi-Robot Preemptive Task Scheduling with Fault Recovery: A Novel Approach to Automatic Logistics of Smart Factories

This paper presents a novel approach for Multi-Robot Task Allocation (MRTA) that introduces priority policies on preemptive task scheduling and considers dependencies between tasks, and tolerates faults. The approach is referred to as Multi-Robot Preemptive Task Scheduling with Fault Recovery (MRPF). It considers the interaction between running processes and their tasks for management at each new event, prioritizing the more relevant tasks without idleness and latency. The benefit of this approach is the optimization of production in smart factories, where autonomous robots are being employed to improve efficiency and increase flexibility. The evaluation of MRPF is performed through experimentation in small-scale warehouse logistics, referred to as Augmented Reality to Enhanced Experimentation in Smart Warehouses (ARENA). An analysis of priority scheduling, task preemption, and fault recovery is presented to show the benefits of the proposed approach.