Gioel Asuni, Scuola Superiore Sant'Anna
A Bio-Inspired Sensory-Motor Neural Model for a Neuro-Robotic Manipulation Platform
The lecture presents a neural model for visuo-motor coordination of a redundant robotic manipulator in reaching tasks. The model was developed for, and experimentally validated on, a neurobotic platform for manipulation.
The proposed approach is based on a biologically-inspired model, which replicates the human brain capability of creating associations between motor and sensory data, by learning.
The model is implemented here by self-organizing neural maps. During learning, the system creates relations between the motor data associated to endogenous movements performed by the robotic arm and the sensory consequences of such motor actions, i.e. the final position of the end effector.
The learnt relations are stored in the neural map structure and are then used, after learning, for generating motor commands aimed at reaching a given point in 3D space.
The approach proposed here allows to solve the inverse kinematics and joint redundancy problems for different robotic arms, with good accuracy and robustness.
In order to validate this, the same implementation has been tested on a PUMA robot, too.
Experimental trials confirmed the system capability to control the end effector position and also to manage the redundancy of the robotic manipulator in reaching the 3D target point even with additional constraints, such as one or more clamped joints, tools of variable lengths, or no visual feedback, without additional learning phases.
Joint work with:
Gioel Asuni(*), Giancarlo Teti(*), Cecilia Laschi(*), Eugenio Guglielmelli(#) and Paolo Dario(*)
(*)ARTS Lab (Advanced Robotics Technology and System Laboratory)
Scuola Superiore Sant’Anna, Piazza Martiri della Liberta' 33, 56127 Pisa, Italy
(#)Laboratory of Biomedical Robotics & EMC
Universita' Campus Bio-Medico, via Longoni 83, 00155 Rome, Italy
Holger Urbanek, Deutsche Aerospace
Towards a bio inspired grasping system -- Grasping via gating of micro movements
The talk will be about a novel idea for a grasping system for multifingered robotic hands with multiple sensors.
Model based grasp-planners are still not useable in natural environments, as the object to be grasped has to be known exactly - but normally this information is not available.
On the other hand biology proves that manipulation of unknown objects is feasible very well. As it is unlikely that biology uses sophisticated mathematical models, there has to be some other way to solve this problem.
Coming from the quite new biological idea of micromovements and the cerebellum as a big forward-model of our body, an idea for a quite simple, but (hopefully) promising system will be presented.