Intelligent Machine Operator Identification to Develop Damage-Reducing Operating Strategies for Mobile Machines

Mobile machines are exposed to a multitude of influencing factors, such as the working task, the operator and the environmental conditions. This leads to a broad spectrum of load collectives for the machine components. In many cases it is difficult to influence the working task and the environmental conditions under the objective function of achieving the required work goals optimally while at the same time minimizing the component load. The operation of the machine offers a more evident degree of freedom to minimize the component damage. With control systems adapted to the operator, the external environmental conditions and the working task, which instructs the operator to a less damaging operating behaviour or override the damage-initiating control signals, the loads and damage can be reduced. An explicit operator identification is the basis for such control approaches. This paper presents a method for machine operator identification (MOI) based on Hidden Markov Models (HMM). Through a parameter influence analysis and a combination with operation state recognition (OSR), a machine operator can be successfully identified among others. To create and validate the method, measurement data from 150 work cycles of seven different operators are analysed. Based on the MOI, a method for an operator-specific damage reduction using adaptive control strategies is developed. The results and limits of this strategy are presented and discussed by means of a complete machine simulation, considering the traction drive and function drives (working hydraulics), a multi-body simulation and the driving dynamics. The conclusion is made by considering predictive operating strategies to avoid damage-intensive operating points.