A typical robot assembly operation involves contacts with the parts of the product to be assembled and consequently requires the knowledge of not only position and orientation trajectories but also the accompanying force-torque profiles for successful performance. In this paper we proposed a complete methodology to generalize also the orientational trajectories and the accompanying force-torque profiles to compute the necessary control policy for a given condition of the assembly task. Our method is based on statistical generalization of successfully recorded executions at different task conditions, which are acquired by kinesthetic guiding.
COBISS.SI-ID: 30866727
We propose a novel human-robot cooperation scheme for bimanual robots. Within the proposed scheme, the initial task demonstration naturally evolves into a cooperative task, where the human co-worker is allowed to modify both the spatial course of motion as well as the speed of execution at any stage. To achieve this goal, speed-scaled dynamic motion primitives are applied for the underlying task representation. The main feature of the proposed adaptation scheme is that the robot adjusts its stiffness in path operational space, defend with Frenet-Serret frame. The required dynamic capabilities of the robot were obtained by decoupling the bimanual robot dynamics in operational space, which is attached to the desired trajectory.
COBISS.SI-ID: 30905127
In this study we propose a new method to enhance the performance of iterative learning control. We focus on robotic tasks dealing with adaptation to the unknown or partially known environment. We proposed a new adaptive iterative learning control scheme, where the adaptation is supervised by the reinforcement learning.
COBISS.SI-ID: 31023143
We propose a two-phase programming by demonstration framework, which enables fast deployment of complex bi-manual assembly tasks. The first phase is a pre-learning phase, where the robot observes multiple task demonstrations performed by humans. The second phase is the policy refinement with incremental learning, performed by the kinesthetic guidance of the robot. The benefit of this framework is in improved learning efficiency since the operator can concentrate only on the fine adjustment of the pre-learned trajectory.
COBISS.SI-ID: 31854119
An effective robot trajectory representation should encode all relevant aspects of the desired motion. In this paper we propose a new representation based on dynamic movement primitives, where spatial and temporal aspects of movement are well separated. We demonstrate the effectiveness of the proposed representation for statistical learning of robot skills and movement recognition. We compare the performance with standard DMPs, where temporal and spatial aspects of motion are intertwined.
COBISS.SI-ID: 29907495