mlstack

mlstack (80)

mlstack

Predictive Processing

Predictive processing is a brain-inspired framework where the nervous system constantly generates predictions about incoming sensory input and then updates those predictions based on the prediction errors it receives. Instead of passively waiting for sensory data, the brain (or robot) actively anticipates what it should…

Continue reading...
mlstack

World Models

World models are internal representations that allow an embodied agent to simulate or predict how the environment will change in response to its own actions. They function like a mental physics engine and planner combined — giving the robot the ability to “imagine” what will…

Continue reading...
mlstack

Embodied Networks

Embodied neural networks are AI models specifically designed or trained with physical interaction and sensorimotor data in mind, rather than relying solely on text, static images, or purely simulated data. Unlike traditional neural networks that process disembodied information, these models integrate perception, planning, and control…

Continue reading...
mlstack

Proprioception

Proprioception is the sense of the body’s own position, movement, and force — often called the “sixth sense.” It tells an agent where its limbs are in space and how much effort is being applied without needing to look. Balance integrates proprioception with vestibular-like information…

Continue reading...
mlstack

Sensors Overview

Sensors provide robots with essential information about the external world and their own internal state, forming the foundation of all embodied intelligence. They include vision sensors (cameras), touch (tactile arrays and force sensors), proprioception (joint position and torque sensors), audio (microphones), inertial measurement units (IMUs…

Continue reading...