Market-based Coordination of Recharging Robots

The effectiveness of mobile robots has always been affected by the amount of time they can spend in the field. Most mobile robots today are equipped with rechargeable batteries which allow a robot to recharge in the field without extended interruption of its work. Mobile recharging units are crucial for a team of robots to effectively perform their assigned tasks over long periods of time. Mobile recharging stations allow robots the option of performing the work-recharge cycle without returning to the base station. This extends the amount of time the robots can spend in the field and consequently increases the productivity of the group as a whole. When groups of worker robots are sharing recharger robots it is critical that these robots coordinate amongst themselves to optimize the team’s work-recharge cycles. The first hypothesis of this thesis is that coordinating the efforts of the rechargers and workers will increase the amount of work completed by the team of robots. Market-based approaches have risen in popularity over the last few years for tackling the problem of multi-robot coordination. Three features that propel the popularity of market-based approaches are their simplicity, their distributed nature, and their fault tolerance. In short, market-based coordination utilizes a simulated economy where robots buy and sell tasks according to their estimated costs for completion. In this way, the robots effectively coordinate amongst themselves to allocate the necessary tasks according to the best information they have at the time. The second hypothesis explored in this thesis is that marketbased coordination techniques can be used effectively to allocate recharging tasks to maximize the productivity and efficiency of the team. The metric I will use to evaluate the effectiveness of my proposed solution is the total quantity and duration of work performed by the team of robots, both rechargers and workers, engaged in autonomous exploration and mapping of unknown environments.