Carlo Masone

I am a researcher at the Max Planck Institute for Biological Cybernetics in the group led by .

My research revolves around two main topics:

• Shared control of mobile robots: mobile-robots are platforms that can be efficiently used for many applications, for example to improve their perception of the environment, to perform complex actions or to reach locations that are not accessible by humans. In this regard, my interest is to study the problem of combining the autonomy of mobile robots with the supervision of a human operator in a shared control framework in which 1) the robot(s) autonomously control some aspect of the task (for example keeping a desired formation) and 2) the human provides high-level directives to the robot(s) (for example high-level maneuvers of a formation of robots). To make this interaction possible I am adopting a bilateral approach, namely giving to the human operator a force-feedback that provides some information of the behaviour of the robot(s).
  More informations on our group's research on this topic be found .

• Robotic platforms for motion simulation: I am also involved in the research on robotic platforms to be used as motion simulators. In this regard I have worked on the development of control and motion cueing algorithms for the CyberMotion Simulator, a motion simulator based on an anthropomorphic robotic manipulator. In comparison to other ad hoc structures, this solution offers a good tradeoff between developing/maintenance costs and motion capabilities.
  Currently I am also working on the research of a new motion simulator based on a cable driven parallel robot, together with .
  More informations on the CyberMotion Simulator can be found .

For other informations, please visit . Check also the with our videos.

CyberMotion Simulator and WABS

The represents a novel approach to motion simulation, based on an anthropomorphic industrial manipulator. This platform offers a larger workspace in comparison to classical hexapod structures (Stewart platforms) while keeping costs limited. The complete motion simulation framework developed for the CyberMotion simulator includes:

1. ad hoc motion cueing algorithm, that translates the motion primitives of a simulated vehicle in a trajectory for the robot;
2. inverse kinematics solution to provide the commands to the robot.

In order to futher extend the motion capabilities of the simulator and increase the immersion in the virtual environment, a novel actuated cabin was developed. This new cabin is endowed with a multiple kinematic behavior, and can produce rotation, translations and a combination of the two.
With the WABS project, the aim is to integrate this framework with human motion perception models, thus producing a perception-based motion simulation.

Planning and control of mobile robots operating with a Human-in-the-Loop

Mobile robots are a major topic of research because their mobility and versatility makes them valuable resources for numerous applications. Despite the signicant research eorts spent to make them fully autonomous, most real-world applications with mobile robots include a human-in-the-loop either because of limitations of the robots or because of regulations. However, the presence of a human introduces several new challenges that need to be addressed, such as i) the need of intuitive control interfaces that allow the action of the operator without requiring intensive training [5], and ii) the need of feedback cues to inform the human of the state of the robot and of the remote site [2].

In the our goal is to solve or mitigate the issues introduced by the human-in-the-loop by developing novel control and planning architectures. These new architectures should enable a single operator to manoeuvre one or multiple mobile vehicles in a remote environment and without the need of training.

Our solution is based on the bilateral shared control paradigm, in which i) the human commands a desired behaviour to the robots by means of a simplied command interface, ii) the robots can correct the human directives or autonomously take care of some aspect of the task, and iii) a feedback loop is closed with the human by combining visual/vestibular cues with a force feedback that is implemented using input devices that are actuated and controlled with a Passive Set-Position Modulation architecture [3] to guarantee passivity.

Within this scenario, we have developed bilateral shared architectures for two scenarios. In the first case, we implemented a framework for the bilateral formation control of a group of unmanned aerial vehicles (UAVs) using only bearing measurements [1] (Fig. 1). In the second case, we have developed a bilateral shared architecture in which the human is moved to the planning level and controls the path followed by the robot rather than the robot itself [4] (Fig. 2). Both approaches have been tested with quadrotor UAVs, and the human operator was able to control the robots easily and without training via two actuated input devices (Fig. 3).

References

1. A. Franchi, C. Masone, V. Grabe, M. Ryll, H. H. Bultho, and P. Robuo Giordano. Modeling and control of UAV bearing-formations with bilateral high-level steering. The International Journal of Robotics Research, Special Issue on 3D Exploration, Mapping, and Surveillance, 31(12):1504{1525, 2012.

2. P. F. Hokayem and M. W. Spong. Bilateral teleoperation: An historical survey. Automatica, 42 (12):2035{2057, 2006.

3. D. J. Lee and K. Huang. Passive-set-position-modulation framework for interactive robotic systems. IEEE Trans. on Robotics, 26(2):354{369, 2010.

4. C. Masone, P. Robuo Giordano, H. H. Bultho, and A. Franchi. Semi-autonomous trajectory generation for mobile robots with integral haptic shared control. Accapted to 2014 IEEE Int. Conf. on Robotics and Automation, 2014.

5. R. Murphy, S. Tadokoro, D. Nardi, A. Jaco, P. Fiorini, H. Choset, and A. Erkmen. Search and rescue robotics. In B. Siciliano and O. Khatib, editors, Springer Handbook of Robotics, pages 1151{1173. Springer, 2008.