Скачать или смотреть Adaptive robot body learning and estimation through predictive coding

The predictive functions that permit humans to infer their body state by sensorimotor integration are critical to performing safe interaction in complex environments. These functions are adaptive and robust to non-linear actuators and noisy sensory information.

This paper introduces a computational perceptual model based on predictive processing that enables any multisensory robot to learn, infer and update its body configuration when using arbitrary sensors with Gaussian additive noise. The proposed method integrates different sources of information (tactile, visual and proprioceptive) to drive the robot's belief to its current body configuration. The motivation is to provide robots with the embodied perception needed for self-calibration and safe physical human-robot interaction.

This video is for the paper:
Pablo Lanillos and Gordon Cheng: Adaptive robot body learning and estimation through predictive coding. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2018)I, 2018.
DOI: 10.1109/IROS.2018.8593684

More information on: www.selfception.eu