• virtual reality
• teleoperation
• tele-impedence
• motion capture

Nowadays robots are effective in repetitive tasks, they are able to perform independent and fast actions, their usage in unstructured environments is still limited, that because critical features like adaptivity, robustness and efficiency are still far from being acceptable.
One of the causes of these limitations is the rigid structure of conventional robots. Recent developments in robotics research brought to the soft robots generation. Soft robot can adjust not only the position but also the impedance, just as humans do. Soft robots have raised new challenges in planning of impedance. This problem can be treated via optimal control techniques but they are prone to model errors and feasible just for small scale problems. An alternative approach is teleoperation.
These advances have led to new techniques such as the teleimpedance: teleoperation in which position and impedance references are sent to the robot.
The aim of this thesis is the study and the integration teleoperation techniques with the purpose to be actually usable in a bimanual framework in real contexts.

The framework developed involve a wearable device equipped with 8 s-EMG and 1 IMU 9- axis (MYO Armband), used for motion capture and management impedance. As a test-bench of testing activities, we used a bimanual platform consisting of two manipulators Kuka LWR (Active Variable Impedance), equipped with a Soft Hand as End-Effector (EE).
In order to guarantee a visual and immersive feedback, it has been used a system of Virtual reality: the Oculus Rift, VR headset equipped with a 9-axis IMU and an infrared system. A robotic head follows the orientation of the VR headset, the stereo camera mounted on the robotic head provides the images to the Oculus.

The teleoperation platform has been experimentally validated performing several tasks that highlight the different strengths of the proposed method: bimanual operations, stiffness control and assessing of the use of the Oculus.
The teleoperation platform developed in this work has been integrated in the Walkman robot for performing evaluation and object recovery task. The experimental activity, in collaboration with the Protezione Civile took place in the city of Amatrice, where an important earthquake recently happened.