Distributed Optimization Framework for Industry 4.0 Automated Warehouses

Ajay Kattepur, Hemant Rath, Arijit Mukherjee, Anantha Simha
2018 EAI Endorsed Transactions on Industrial Networks and Intelligent Systems  
Robotic automation is being increasingly proselytized in the industrial and manufacturing sectors to increase production efficiency. Typically, complex industrial tasks cannot be satisfied by individual robots, rather coordination and information sharing is required. Centralized robotic control and coordination is ill-advised in such settings, due to high failure probabilities, inefficient overheads and lack of scalability. In this paper, we model the interactions among robotic units using
more » ... ligent agent based interactions. As such agents behave autonomously, coordinating task/resource allocation is performed via distributed algorithms. We use the motivating example of warehouse inventory automation to optimally allocate and distribute delivery tasks among multiple robotic agents. The optimization is decomposed using primal and dual decomposition techniques to operate in minimal latency, minimal battery usage or maximal utilization scenarios. These techniques may be applied to a variety of deployments involving coordination and task allocation between autonomous agents. Decentralized Decisions: The ability of such systems to make autonomous decisions; only critical cases will involve human intervention. Warehouse and factory floor automation [3][4] has been a principal area of interest with respect to these requirements. Automated Guided Vehicles (AGVs) are employed in the warehousing environment to move products from one place to another [5][6]. AGVs follow fixed routes (using wires or markers) that are preprogrammed on them. As they have limited on board computational intelligence, Networked Robotics [7] have been proposed, where robotic AGVs may link to an internet based infrastructure to seamlessly exchange data. This data may be autonomously exchanged or coordinated via a central control station. This has been extended to the Cloud Robotics [8] framework, where robots make use of the cloud to coordinate or offload computational tasks. In warehouses that may have hundreds of robots on the shop floor, complex problem domains (scheduling, optimization, planning) require modular and scalable solutions [4]. A number of functional, modular components (agents) may be deployed to solve specialized problem aspects. Decomposing large problems allows 1
doi:10.4108/eai.27-6-2018.155237 fatcat:g5pkl7w4jzb3ni3jg75s4ig55i