Robust robot grasping of heterogeneous objects

In the last decade, due to several factors such as the increasing cost of human labour, the spread of automation and the decreasing cost of robotic systems, impressive advancements have been made in grasping devices. Therewith, facilitating further automation of daily tasks in diverse application domains, ranging from industrial production processes, logistics, agriculture, medical, end-of-life up to household and healthcare applications. While initial approaches focused on the development of hard grippers, often tuned to the robot’s freedom of movement and offering large force exertion and high precision, the research focus evolved to the development of robotic grippers endowed with soft deformable structures. Because of higher degrees of freedom and structural compliance, these soft grippers show their potential to operate in the emerging uncertain and dynamic environments, respectively handling unknown (e.g. advanced random bin picking) or non-rigid (e.g. food, textile) objects and human-robot collaboration. However, the main obstacles for fast industrial implementation are the inherent lack of repeatability, precision, lower grasping forces, relatively slow operating speeds, high costs and often complex gripper control. Therefore, this PhD research targets to combine the advantages of both (hard and soft) gripper types in hybrid stiffness-adaptable grippers, by finding the most appropriate region within the continuous spectrum between soft and hard grippers. For this purpose, magnetorheological (MR) fluids, whose aggregation state can be continuously controlled from a free-flowing liquid to a viscoelastic solid by a magnetic field, show promising potential.