Jardón Huete, AlbertoGonzález Víctores, Juan CarlosMartínez de la Casa Díaz, SantiagoBalaguer Bernaldo de Quirós, Carlos2014-05-272014-05-272012-05Automation in Construction (2012), vol. 23, 33-46.0926-5805https://hdl.handle.net/10016/18913This paper describes an innovative modeling, interfacing and training framework and application for microtunneling machines under heterogeneous gravel and sand soils, based on a machine simulator. It is initialized using a selective collection of skilled pilots' know-how during the performance of a pipe jacking microtunneling machine operation, generating a rule-based system based on grouped rules and states that replicate the machine's performance. The adjustment of these states and associated rules allows the creation, setup and analysis of a realistic functional model for tunneling machines. The developed system integrates a friendly Human-Machine Interface (HMI) that closely resembles real machines' pilot cabinets and allows natural interaction with the implemented inference engine through the simulated control panel. Additionally, the framework allows the training of tunneling machine operators by simulation and subsequent gathered data analysis, obviating or reducing mechanical movement times if desired. The virtual pilot's desk allows global training time and cost reductions, and increases safety for future operatorsand machinery. The HMI is divided into two screens, which replicate the data and the command panels of a real machine's control desk. The presented framework has allowed the first implementation of a jack piping microtunneling machine simulator by means of the developed pilots' steering know-how capture methodology.10 p.application/pdfeng© 2011 Elsevier B.V.Tunneling machineRobotic automationTunnel execution modelingSimulatorPipe jacking trainingresearch articleRobótica e Informática Industrial10.1016/j.autcon.2011.12.002open access33046Automation in Construction23AR/0000009903