MetaBot: Automated and dynamically schedulable robotic behaviors in retail environments

The ever-increasing popularity of online stores is reshaping traditional commerce models. In particular, brick-and-mortar stores are presently facing the challenge of reinventing themselves and their business models to offer attractive yet low-cost alternatives to e-commerce. Other industries have already introduced new concepts to fight inefficiency (i.e., “Just-in-Time” inventory management in Automotive), retail stores face a more challenging environment which these models cannot accommodate. Stores remain heavily vested in battling the overhead costs of personnel management when, instead, a robotic automation scheme with retail-oriented behaviors could reduce the detection latency of out-of-stock and compliance error phenomena throughout the store. These behaviors must be automated, multi-purpose, and schedulable; they must also ensure that the robot coordinates store nuances to adapt its functionality appropriately. In this paper, we present our architecture that defines retail robot behaviors as a collection of reusable activities, which, when permuted various ways, allows for various high-level and application-specific tasks to be accomplished effectively. We evaluate this system on our robotic platform by scrutinizing the integrity of navigation and machine vision tasks, which we perform concurrently in an experimental store setup. Results show the feasibility and efficiency of our proposed architecture.