End-to-End Pixel-Based Deep Active Inference for Body Perception and Action

End-to-End Pixel-Based Deep Active Inference for Body Perception and Action
复制标题

用于身体感知和动作的端到端基于像素的深度主动推理

DOI:
10.1109/icdl-epirob48136.2020.9278105
复制
发表时间:
2019
期刊:
2020 Joint IEEE 10th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob)
影响因子:
--
通讯作者:
Pablo Lanillos
Pablo Lanillos
中科院分区:
--
文献类型:
--
作者:
Cansu Sancaktar;Pablo Lanillos

文献摘要

参考文献

被引文献

相似文献

我们提出了一种基于像素的深度主动推理算法(PixelAI),其灵感来自人体感知和动作。我们的算法将神经科学中的自由能原理(植根于变分推理)与深度卷积解码器相结合,以扩展算法,直接处理原始视觉输入并提供在线自适应推理。我们的方法进行了验证,通过研究身体的感知和行动,在一个模拟的和一个真实的Nao机器人。结果表明,我们的方法允许机器人执行1)动态身体估计其手臂只使用单目摄像机图像和2)自主达到“想象”的手臂姿势在视觉空间。这表明,机器人和人体的感知和行动可以有效地解决看作是一个积极的推理问题的指导下进行感官输入。
We present a pixel-based deep active inference algorithm (PixelAI) inspired by human body perception and action. Our algorithm combines the free energy principle from neuroscience, rooted in variational inference, with deep convolutional decoders to scale the algorithm to directly deal with raw visual input and provide online adaptive inference. Our approach is validated by studying body perception and action in a simulated and a real Nao robot. Results show that our approach allows the robot to perform 1) dynamical body estimation of its arm using only monocular camera images and 2) autonomous reaching to “imagined” arm poses in visual space. This suggests that robot and human body perception and action can be efficiently solved by viewing both as an active inference problem guided by ongoing sensory input.
DOI: 10.1016/j.jmp.2020.102348
发表时间: 2020-06-01
影响因子: 1.8
作者:
Millidge, Beren
通讯作者: Millidge, Beren
DOI: 10.1109/tpami.2016.2567384
发表时间: 2017-04-01
影响因子: 23.6
作者:
Dosovitskiy, Alexey;Springenberg, Jost Tobias;Brox, Thomas
通讯作者: Brox, Thomas