Session: Modeling and Simulation

Paper Number: 112957

112957 - Towards Energy-Efficient Semi-Autonomous Operation of Hydraulic Mobile Cranes 

This paper describes a method for an automatic control of hydraulic mobile crane movement that decreases required energy and allows avoiding obstacles. The method was originally developed for mobile log cranes and provides automatic transportation of the grappled logs to the desired position. The trajectory is calculated taking into account hydraulic energy needed for the movement. The method can also be applied to other machines, such as loader cranes. An input to the method is the positions of the endpoints and a map of the working space that defines the obstacles. To calculate an energy-efficient trajectory, the method uses a modified version of the A* search algorithm. The method includes two steps. First, a set of simulation experiments must be performed with a multi-physics model of the crane. The model should describe mechanical and hydraulic parts of the system and provide the value of hydraulic energy spent in each experiment. The first step represents a learning phase of the algorithm and must be executed once. A set of energy efficiency matrices is generated at this phase, which is used in the second step of the method to calculate a heuristic for the modified A* algorithm. The first step requires several hours of calculations depending on the model complexity and available computational resources. The second step can be executed faster than in a second, which makes it suitable for the real-time operation of the machine. To move the crane by the generated energy-efficient trajectory, a control algorithm was developed by combining feed-forward and PID control.

The paper presents the simulation results obtained with a real-time model of a log truck created using commercial software. The results demonstrate the median value of energy savings of 5% compared to the shortest path for the work task of loading and unloading the truck using common endpoint positions of logs. Validation of the method with a real device is planned for future research.

Presenting Author: Heikki Handroos LUT University

Presenting Author Biography: HEIKKI HANDROOS has been a Professor of machine automation at LUT University, since 1992. He has also been a Visiting Professor at the University of Minnesota, Minneapolis, MN, USA; Peter the Great St. Petersburg Polytechnic University, Saint Petersburg, Russia; and the National Defense Academy, Japan. He has published about 250 international scientic papers and supervised more than 20 D.Sc. (Tech.) theses. He has led several important domestic and international research projects. His research interests include modeling, design, and control of mechatronic transmissions to robotics and virtual engineering. He has been an Associate Editor of the Journal of Dynamic Systems, Measurement, and Control, since 2014.