Towards a Development of Robotics Tower Crane System

摘要

This paper describes the outcomes of a research study of a Robotic Tower Crane (RTC) which is part of a Digital Production Inventory Logistic Management System (DPILMS), that can be viewed as a complex system as it is required to deal with various inputs and perform different tasks in an unpredictable environment such as the construction field. The system has to work day and night, rain or shine under all-weather conditions. The RTC must sense the environment using multiple sensors. The proposed solution consists of various sensors capable of localizing the RTC in real-time, detect and track objects within the surroundings, and subsequently make decisions about how to react correctly within the appropriate time in the perceived environment, and several real-time subsystems. To achieve autonomous hoisting in an unpredictable environment, the real-time subsystems must interoperate with sensor processing, perception, localization, planning and control, processing an enormous amount of sensor data via a high complexity computation pipeline.