Optimizing Parallel Task Execution for Multi-Agent Mission Planning

Multi-agent systems have received a tremendous amount of attention in many areas research and industry, especially robotics computer science. With the increased number agents missions, problem allocation tasks to arose, it is one most fundamental classes problems robotics, formally known as Multi-Robot Task Allocation (MRTA) problem. MRTA encapsulates numerous dimensions, aims at providing formulations solutions various configurations, i.e., complex multi-agent missions. One dimension has not caught much attention. In particular, configurations including Multi-Task (MT) robots been neglected. However, increase computational power, robotic systems, allowed utilization parallel task execution. This turn had benefit allowing creation more missions; however, came cost complexity. Our contribution aforementioned domain can be grouped into three categories. First, we model using two different approaches, Integer Linear Programming Constraint Programming. these models, aim filling gap literature related formal definition MT robot configuration. Second, introduce distinction between physical virtual their mutual relationship terms allows modeling wider range missions while exploiting possible Finally, provide comprehensive performance analysis both by implementing validating them CPLEX CP Optimizer on set problems. Each consists same test instances gradually increasing complexity, percentage each different. The results includes exploration scalability models solvers, effect solvers’ performance, overall solution quality.