Francesca Gandolfo

A Kendama learning robot based on bi-directional theory

A general theory of movement-pattern perception based on bi-directional theory for sensory-motor integration can be used for motion capture and learning by watching in robotics. We demonstrate our methods using the game of Kendama, executed by the SARCOS Dextrous Slave Arm, which has a very similar kinematic structure to the human arm. Three ingredients have to be integrated for the successful execution of this task. The ingredients are (1) to extract via-points from a human movement trajectory using a forward-inverse relaxation model, (2) to treat via-points as a control variable while reconstructing the desired trajectory from all the via-points, and (3) to modify the via-points for successful execution. In order to test the validity of the via-point representation, we utilized a numerical model of the SARCOS arm, and examined the behavior of the system under several conditions. Copyright 1996 Elsevier Science Ltd.

Teaching by Showing in Kendama Based on Optimization Principle

Much progress has been made in the past decade regarding computational understanding of motor learning both in neuroscience and robotics. Our recent interests have been drawn to higher-level task learning rather than simple trajectory following. Learning algorithms such as reinforcement learning and genetic algorithms can be efficiently used for task-level learning if adequate representations of the task are selected. However, Schaal, Atkeson and Botros (1992) pointed out that this selection is the most difficult and critical part of motor learning and if one assumes the pre-existence of proper representations, this amounts to abandoning plans to tackle a major part of the problem. In this paper, based on the dynamic optimization theory for trajectory formation, we propose a general computational theory that derives representations for a wide variety of motor behaviors.

Visual monitoring of robot actions

Present a perspective in relation to the use of vision for the control of robots and for the monitoring of robot actions. In order to explain the approach some experiments have been performed in the field of object manipulation and robot navigation. In particular the monitoring of pushing and tapping actions by means of dynamic measurements performed on image sequences are reported. The experiments show that, not only it seems possible to detect unexpected events, but it is possible to control the arm or vehicle trajectory in such a way to achieve a complex task like the pushing of an unknown object along a predetermined trajectory.<<ETX>>

Vector summation of end-point impedance in kinematically redundant manipulators

One way to control the mechanical interactions between a manipulator and its environment is by specifying the manipulators end-point impedance. According to this view, the function of a controller is to determine the force that the controlled system should apply in response to an externally imposed state, i.e., to determine an output force field. The possibility of combining the output force fields generated by a number of separate controllers in order to approximate an arbitrary output field is considered for a kinematically redundant manipulator. The end-point force fields of a simulated kinematically redundant manipulator operated by a set of spring-link actuators is numerically derived. The results indicate that the linear superposition of end-point fields may provide a quite good approximation of the actual passive behavior. This offers a simple way for solving the ill-posed problems associated with the specification of end-point impedance fields in a kinematically redundant system.

A Kendama learning robot based on a dynamic optimization theory

A general theory of movement pattern perception based on a dynamic optimization theory can be used for motion capture and learning by watching in robotics. We exemplify our methods for the game of Kendama, executed by the SARCOS Dextrous Slave Arm, which has exactly the same kinematic structure as a human arm. Three ingredients have to be integrated for the successful execution of this task. The ingredients were (1) to extract via-points from a human movement trajectory using a forward-inverse relaxation model, (2) to treat via-points as a control variable while reconstructing the desired trajectory from all the via-points, and (3) to modify the via-points for successful execution.