Haptic perception with a robotic hand

We have developed an 8 d.o.f. robot hand which has been tested with three computational models of haptic perception. Two of the models are based on the tensor product of different proprioceptive and sensory signals and a self-organizing map (SOM), and one uses a novel self-organizing neural network, the T-MPSOM. The computational models have been trained and tested with a set of objects consisting of hard spheres, blocks and cylinders. The first computational model, which is based on the tensor product, was able to discriminate between all the test objects. The second and third models could also discriminate between the test objects, but in addition they were capable of shape categorization.