An Incremental Self-Organizing Architecture for Sensorimotor Learning and Prediction
By Luiza Mici, German Ignacio Parisi, Stefan Wermter. 2017
17 citations recorded by the source corpus.
During visuomotor tasks, robots must compensate for temporal delays inherent in their sensorimotor processing systems. Delay compensation becomes crucial in a dynamic environment where the visual input is constantly changing, e.g., during the interaction with a human demonstrator. For this purpose, the robot must be equipped with a prediction mechanism for using the acquired perceptual experience to estimate possible future motor commands. In this paper, we present a novel neural network architecture that learns prototypical visuomotor representations and provides reliable predictions on the basis of the visual input. These predictions are used to compensate for the delayed motor behavior in an online manner. We investigate the performance of our method with a set of experiments comprising a humanoid robot that has to learn and generate visually perceived arm motion trajectories. We evaluate the accuracy in terms of mean prediction error and analyze the response of the network to novel movement demonstrations. Additionally, we report experiments with incomplete data sequences, showing the robustness of the proposed architecture in the case of a noisy visual input.
Subfields: Cognitive Neuroscience, Control and Systems Engineering, Social Psychology, Meta & Continual Learning, Robotics & Control
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Cognitive Neuroscience
In Control and Systems Engineering
In Social Psychology
In Meta & Continual Learning
- Vincenzo Lomonaco
- German Ignacio Parisi
- Davide Maltoni
- Jiangpeng He
- Feng Zhu
- Lorenzo Pellegrini
- Christopher Kanan
- Tyler L. Hayes
In Robotics & Control
Citations in the corpus
- Online Continual Learning on Sequences (2020, background): “Similar GWR-based approaches have been proposed for the incremental learning of body motion patterns [54, 8, 61] and human-object interaction [55].”