R4I Aims

  1. Robot learning, autonomy and mobility. Machine learning research will ease robot adaptation to a new tasks and application environments, including robot-human cooperation. Effective and safe machine learning algorithms are an important precondition for autonomy in robotics, which has been recognized as a strategic bottleneck for smart industrial applications. A high-level reasoning in robotics needs a good representation of the environment, which continues to be a challenging research issue. We will deal with robot learning, mobility, and with the mechanical aspects essential for effective human-robot collaboration.
  2. Perception, grasping and manipulation in industrial environments. The ability of robots to perceive and understand their environment is still very limited. Reliable sensing and perception methods for mobile industrial robots constitute the precondition for the use of robots in the Industry 4.0 applications. Additional challenges are present when it comes to the combination of perception and dexterity. This workpackage will therefore address the integration of perceptual systems with dexterous manipulation in the context of cooperative robots. The perception-action cycle can be substantial simpli ed if physical laws are explored smartly. Advanced perception, calibration and hybrid sensor-fusion will be studied in conjunction with mechatronic side of the problem.
  3. Networked control systems. Strongly connected systems, which are the backbone of the Industry 4.0 concept, give rise to additional complexity due to the interaction of the subsystems. Profound understanding of the phenomena arising in the networked system is a prerequisite for the successful implementation of the Industry 4.0 paradigm. Two distinct directions will be followed. The first will deal with theoretical issues regarding `systems of systems’ from the control-theoretic and cyberphysical point of view. Two more applied areas belong here too: humans interacting with complex networks and multi-agent robotics. The second direction will research mechanisms constituting mechanical networks in two areas – controlled mechanical impedance to increase the mechanism’s stiffness and mechatronic tricks to ease grasping and manipulation.