Global real-time optimal control of autonomous robot and mechatronic systems.

This project focuses on a class of systems including general robot systems consisting of one or more collaborating robots, production machines ranging from machine tools to fast pick-and-place robots, and intelligent biomedical devices such as orthoses. We consider systems that operate in an uncertain environment, and that interact with only partially known objects and environments or with humans, which requires a continuous adaptation and optimization of their trajectories. The ability to reschedule and re-optimize the task of such systems on-line based on the most recent information from sensors, prognosis or prediction algorithms, is a first key element to achieve global optimal performance with respect to production output, speed, accuracy, energy consumption and interaction with humans. A second key element is the accurate realization of the (re)scheduled task. The generic approach followed in this project is to specify the tasks that have to be executed as optimal control problems, and to solve them on-line using numerical optimization algorithms that are embedded on the controller hardware of the system.

Keywords:orthoses, robotics, optimal control, FPGA, sensor-based robots, drive systems, numerical optimization, autonome systemen

Disciplines:Control systems, robotics and automation, Design theories and methods, Mechatronics and robotics, Computer theory, Manufacturing engineering, Other mechanical and manufacturing engineering, Product development