汇报题目：参加CIE48国际会议总结报告会
汇报时间：2018年12月21日(星期五) 9:00
汇报地点：曲江校区西五楼南楼405会议室
汇报人：李润泽
会议名称：The 48th International Conference on Computers and Industrial Engineering(CIE48)
会议时间：2-5，DEC，2018
会议地点：Auckland, New Zealand
会议简介：The 48th International Conference on Computers and Industrial Engineering (CIE 48) was held in Auckland, New Zealand from the 2nd to the 5th of December 2018. The conference aims to disseminate recent theoretical and methodological developments, significant technical applications, case studies and survey results in areas such as Computers, Industrial Engineering, Manufacturing and Management. The special sessions of the conference are as follows:
-New technologies for designing, operating and maintaining of the intelligent workshop
-Data Science and Analytics for Smart Manufacturing
-Industrial Blockchain
-Next Generation Logistics and Supply Chain Management
-Stakeholder-oriented Design Innovation of Smart Product-Service Systems
-Efficiency & Productivity
-Sustainability & Innovation
-Digital Twin: Theories, challenges, methodologies and applications
-Cloud-enabled Cyber-Physical Production Systems (C2P2S)
参会论文信息：
Title: Deep Reinforcement Learning Based Tracking Control of Ball Screw Feed Drive
Authors: Runze Li, Dun Lyu*, Wanhua Zhao
Abstract: A deep reinforcement learning based intelligent controller is proposed to solve the problem of tracking control of ball screw feed drive. The deep reinforcement learning based intelligent controller is composed of two neural networks. One network called actor network generates control signal to drive ball screw to feed and the other network called critic network evaluates whether the generated control signal is accurate. The two neural networks is modified through a deep deterministic policy gradient(DDPG). The intelligent controller and a traditional PID controller whose parameters have been tuned optimal are compared on an accurate simulation model of a ball screw feed drive. Based on simulation results, the intelligent controller outperforms the PID controller in tracking control of ball screw feed drive, encouraging future research.
