Intelligent Energy Management for Mobile Manipulators Using Machine Learning

Abstract

Integrated robotic systems combining manipulators with mobile robots provide outstanding improvement opportunities for semi-automatic assembly processes leveraged by Industry 4.0. Factory operations are released from the rigid layout constraints imposed by conventional fixed robots. Thus, they introduce new challenges in managing the recharge cycles as the energy consumption of mobile manipulators is not simply related to the travelled distance but to the overall tasks executed. Its estimation requires a systemic approach. In the proposed solution, an intelligent monitoring system is implemented on board. Data gathered online, and Key Performance Indicators (KPIs) calculated during the working tasks are exploited by Machine Learning (ML) to optimize energy recharging cycles. Although the development of an intelligent monitoring framework for a mobile manipulator was the original objective of the research, the monitoring system is exploited here for energy management only, leaving space for other future applications.