The +1 Method: Model-Free Adaptive Repositioning Policies for Robotic Multi-Agent Systems

Abstract

Robotic multi-agent systems can efficiently handle spatially distributed tasks in dynamic environments. Problem instances of particular interest and generality are the dynamic vehicle routing problem and the dynamic traveling repairman problem. Operational policies for robotic fleets solving these two problems take decisions in an online setting with continuously arriving dynamic demands to optimize system time and efficiency. They can be classified along several lines. First, some require a model of the demand, e.g., based on historical information, while others work model-free. Second, they are designed for different operating conditions from light to heavy system load. Third, they work in a time-invariant or time-varying setting. We present a novel class of model-free operational policies for time-varying demands, with performance independent of the load factor and applicable to any number of dimensions, a combination of properties not achieved by any other operational policy in the literature. The underlying principle of the introduced policies is to send available robots to recent realizations of the stochastic process that generates service requests. In simple terms, the strategies rely on sending more than one robot for every service request arriving to the system. This leads to an advantage in scenarios where demand is non-uniformly distributed and correlated in space an time. We provide theoretical stability and performance guarantees for both the time-invariant and the time-varying cases as well as for correlated demand. We verify our theoretical results numerically. Finally, we apply our operational policy to the problem of mobility-on-demand fleet operation and demonstrate that it outperforms model-based and complex algorithms across all load ranges despite of its simplicity.

Simulation-Based Assessment of Parking Constraints for Automated Mobility on Demand: A Case Study of Zurich

“…These different changes in waiting times with different parking strategies and distributions can also be explained as follows. While typically in fleet control vehicles are rebalanced to locations of future anticipated demand [33,34,52,53] no rebalancing logic is used on top of the basic global bipartite matching in the present case. Idle and staying vehicles are merely sent to available parking spaces by the parking operating policies.…”

“…These vehicles must be repositioned to locations of future anticipated demand, a process called rebalancing or repositioning. The choice of rebalancing locations can be made in various different ways, e.g., as described in [33][34][35][36]. Furthermore, possible ride-sharing opportunities may be considered with a suitable strategy, e.g., the ones compared in Ruch et al [27] or presented by [37].…”

Analysis of the Allocation and Implementation of Tasks in the Heterogeneous Team of the Collaborative Robotic System

“…Once the overall structure of the CRS management and control system has been selected and established, and it is necessary to proceed to the choice of methods and The CRS is composed of heterogeneous participants, resulting in the need to apply collective management (control) approaches, algorithms and methods to the elements of a heterogeneous team [8,[11][12][13][14][15][16][17][18][19][20][21][22][23]. According to [8,24], two stages of the life cycle of a heterogeneous team are usually considered-its formation and functioning.…”

The Impact of Fleet Coordination on Taxi Operations

“…(2) e fluidic feedforward rebalancing policy (FFR) uses the same matching algorithm as GBM but additionally executes vehicle rebalancing according to the fluidic feedforward method presented by Pavone et al [37]. (3) e last policy (MFAR) rebalances vehicles according to the model-free adaptive repositioning policy presented by Ruch et al [38]. is results in significantly reduced wait times but does increase the empty vehicle distance.…”