Loading the player ...
- Offer Profile
- The iCub is the humanoid
robot developed at the Istituto Italiano di
Tecnologia as part of the EU project RobotCub and subsequently
adopted by more than 20 laboratories worldwide.
Run by the Robotics, Brain and Cognitive Sciences dept. at the Istituto
Italiano di Tecnologia
Funded by the EU Commision under the Cognitive Systems and Robotics program
- The iCub is the humanoid robot developed at the Istituto Italiano di
part of the EU project RobotCub and subsequently adopted by more than 20
laboratories worldwide. It has 53 motors that move the head, arms & hands,
waist, and legs. It can see and hear, it has the sense of proprioception
(body configuration) and movement (using accelerometers and gyroscopes). We
are working to improve on this in order to give the iCub the sense of touch
and to grade how much force it exerts on the environment.
Scientists at the Cognitive Humanoids Laboratory work at the forefront of
the robotics and neuroscience research implementing models of cognition in
robots of humanoid shape. This heterogeneous group of people aims at
understanding brain functions and realizing robot controllers that can learn
and adapt from their mistakes.
Activities encompass the construction of the hardware, that we call "bodyware",
and software which will make, one day, machines of intelligence comparable
to humans. We call this technology "mindware". On the bodyware side we
developed the iCub, a humanoid robot, shaped as a 4 years old child.
Simultaneously, we are addressing the development of the technologies for
the next generation of robots based on soft and adaptable materials both for
sensing and processing. On the mindware side, the laboratory is involved in
the realization of the cognitive skills of the humanoid robot; that is,
providing the robot with visual, auditory and tactile perception and the
ability to gaze, reach and manipulate objects.
iCub 6-axial force/torque sensor
- iCub mounts four 6-axial force/torque sensors in the
upper arms and legs. They have been designed to be compatible (in size) with
the ATI Mini-45 sensors. The electronics have been miniaturized to fit
inside the sensor which provides digital (CAN) output directly.
CFW-2, 10 port CAN bus PC104 card
- This is a multi-port PC104+ card that hosts 10 CAN bus
ports (managed by two microcontrollers), two Firewire ports and audio
pre-amplifier. A large buffer (2Mbyte) is available for storing CAN messages
and a DMA interface on the PCI bus is also provided.
- The iCub controllers are small microcontroller boards
based on the Freescale 56F807 chip. Each card can connect via CAN bus to the
host CPU (the PC104 card). They are of two different type controlling 4
brushed-dc motors (0.5A each) or 2 brushless-dc motors (48V, 6A continuous,
20A peak). The brushed version is complemented by a small power supply; the
brushless version is made of two parts (one for the logic and the other for
iCub head 1.1
- The iCub head has 6 degrees of freedom and sports two
cameras (Dragonfly2), two microphones (with special pinnae), gyros and
accelerometers (Mtx). It also mounts a PC104 dual core machine with enough
ports to control the entire robot, read data and send them out via a Gbit/s
iCub arm and hand
- The iCub arm and hand are anthropomorphic jointly
sporting 16 degrees of freedom. 7 for the arm (including the wrist) and 9 in
the hand. 8 degrees of freedom are allocated to the thumb, index and middle
fingers thus enabling a fairly large degree of dexterity. The hand can be
integrated with 108 tactile sensors in the fingertips and palm.
iCub legs and torso
- The iCub legs and torso are jointly 15 degrees of
freedom. They were designed mainly for crawling on all fours but tests have
shown that bipedal walking is possible. Force/torque sensors are mounted in
the upper part of each leg.
- New life is being injected into the iCub platform through
the development of force control. This new skill enables safe and gentle
interaction of the robot with human teachers. A short video shows this new
feature used by an experimenter to teach simple actions to a brand new iCub.
- iCub grasping skills have been recently improved as a
result of the Poeticon EU project. Simple actions such as grasping are
combined through the use of spoken language tools developed by the Poeticon
teams. In this example, the iCub pours cereals in a cup.
- New visual skills of the iCub. Independent motion
tracking for the iCub using optical flow. See how we can easily track
objects or people that move independently of the iCub.
- The iCub met the Italian President in 2010 during a visit
to Genoa and IIT. Here the President is asking about the robot after
receiving the latest IIT leaflet from the hand of the iCub. Also in the
picture, Giorgio Metta, Giulio Sandini and Roberto Cingolani.
- The iCub was in Hannover in April 2010 as part of the
Italian delegation to the international fair and exhibit. Here's interacting
with the German Chancellor Angela Merkel while showing its movement
coordination skills and software reliability. The iCub ran uninterrupted for
three full days attracting thousands of curious visitors
- Lorenzo, the guy in the picture, is checking the new iCub
before running a manipulation experiment. The iCub is being fitted with a
capacitive skin system in the fingertips and palms that enables measurement
of contact and the implementation of safe grasping strategies.
iCub skin and fingertips
- The iCub can mount a capacitive skin system in two forms:
fingertips and generic body skin.
The skin is based on a modular triangular structure and with minimal
Each taxel can be sampled at 50Hz (8bits) and connected via CAN bus to a
- This is a holonomic mobile base for the iCub which mounts
six Swedish wheels, a high perf i7-CPU, it sports wireless connection and
high perf Li-ion batteries. The iCub can stand on top of it and control the
base using a standard interface.
- iCub 2.0 is our experimental (work in progress) new iCub.
The iCub has been upgraded with respect to the torque measurement (joint
level), tension sensors (for tendons), full body skin, fingertips, new head,
additional high-resolution encoder on all brushless motors, many small
improvements in the mechanics and wiring and new foot design (for bipedal
iCub head 2.0
- The new head (tentatively called 2.0) is at an advanced
stage of design and prototypation. ´The neck has been redesigned for more
torque and the eyes will mount zero-backlash harmonic drive gears.
Further small modifications have been implemented especially to support
accurate camera calibration.
- The goal of RobotCub was to study cognition through the
implementation of a humanoid robot the size of a 3.5 year old child: the
iCub. This is a fully open source & hardware project, one of a kind! This is
the project that started the iCub both in hardware and software.
- Xperience will demonstrate that state-of-the-art enactive
systems can be significantly extended by using structural bootstrapping to
generate new knowledge. This process is founded on explorative knowledge
acquisition, and subsequently validated through experience-based
- The Experimental Functional Android Assistant (EFAA)
project will contribute to the development of socially intelligent humanoids
by advancing the state of the art in both single human-like social
capabilities and in their integration in a consistent architecture.
- The MACSi project is a developmental robotics project
based on the iCub humanoid robot. It is funded as an ANR Blanc project from
2010 to 2012.
- The Darwin project aims to develop an "acting, learning
and reasoning" assembler robot that will ultimately be capable of assembling
and disassembling complex objects from its constituent parts
- The ITALK project aims to develop artificial embodied
agents able to acquire complex behavioural, cognitive, and linguistic skills
through individual and social learning. This will be achieved through
experiments with the iCub humanoid robot.
- POETICON is a project that explores the poetics of
everyday life, i.e. the synthesis of sensorimotor representations and
natural language in everyday human interaction. This is related to an old
problem in AI on how meaning emerges.
- The CHRIS project (Cooperative Human Robot Interaction
Systems) addresses the fundamental issues which enable safe Human Robot
Interaction (HRI) from the motoric as well as cognitive point of view.
- The RobotDoC Collegium is a multi-national doctoral
training network for the interdisciplinary training on developmental
cognitive robotics. The RobotDoc Fellows will acquire hands-on experience
through experiments with the open-source humanoid robot iCub.
- RoboSKIN will develop and demonstrate a range of new
robot capabilities based on the tactile feedback provided by a robotic skin
from large areas of the robot body. Up to now, a principled investigation of
these topics has been limited by the lack of tactile sensing technologies.
- Motor skills of humans and animals are still utterly
astonishing when compared to robots. AMARSi aims at a qualitative jump in
robotic motor skills towards biological richness.
- IM-CLeVeR aims to develop a new methodology for designing
robots controllers that can cumulatively learn new efficient skills through
autonomous development based on intrinsic motivations, and reuse such skills
for accomplishing multiple, complex, and externally-assigned tasks.
- The goal of the eMorph project is to design asynchronous
vision sensors with non-uniform morphology, using analog VLSI neuromorphic
circuits, and to develop a supporting data-driven asynchronous computational
paradigm for machine-vision.
- Starting from the assumption that cognition is embodied,
the ROSSI project addresses the question of how the possibility of
communication between agents (e.g. humans and robots) is affected by
differences in sensorimotor capacities.