Abstract—This paper describes the design of a robot head, de-veloped in the framework of the RobotCub project. This projectgoals consists on the design and construction of a humanoidrobotic platform, the iCub, for studying human cognition. The nal platform will be approximately 90cm tall, with 23 kg andwith a total number of 53 degrees of freedom.For its size, the iCub is the most complete humanoid robotcurrently being designed, in terms of kinematic complexity. Theeyes can also move, as opposed to similarly sized humanoidplatforms.Speci cations are made based on biological anatomical andbehavioral data, as well as tasks constraints. Different conceptsfor the neck design ( exible, parallel and serial solutions) areanalyzed and compared with respect to the speci cations. The eyestructure and the proprioceptive sensors are presented, togetherwith some discussion of preliminary work on the face design.I. INTRODUCTIONThis paper describes the design and construction of theiCub robotic head/neck system.35858
It is included in the EuropeanProject RobotCub, a large and ambitious project of embodiedcognitive systems [1].The RobotCub project has the twin goals of (1) creatingan open and freely-available humanoid platform, iCub, forresearch in embodied cognition, and (2) advancing our un-derstanding of cognitive systems by exploiting this platformin the study of cognitive development. To achieve this goal weplan to construct an embodied system able to learn: i) how tointeract with the environment by complex manipulation andthrough gesture production & interpretation; and ii) how todevelop its perceptual, motor and communication capabilitiesfor the purpose of performing goal-directed manipulationtasks. iCub will have a physical size and form, similar tothat of a two year-old child, as shown in Figure 1 and willachieve its cognitive capabilities through arti cial ontogeniccodevelopment with its environment.In recent years, the need for new generations of robotshas drifted from industrial automation to human friendlyrobotic systems, able to routinely interact with humans inenvironments such as of ces, homes and hospitals. The studyof humanoid robots is particularly relevant for this type ofinteraction because of anthropomorphism, friendly design,applicability of locomotion, behavior within the human living  environments, small size and so on. In fact, one reason for the(relatively) small size of humanoid type robotic platforms issafety: striding about in homes, a small robot is less likely toharm people by falling on them. Another reason is that shorterlimbs and appendages are easier to move and control.
To meetthese demands, several humanoid robots have been developedin these years.Several small size humanoid robots are already available.Qrio [3] (from Sony), was designed to interact and entertainingpeople by means of motion, speech and vision. Its remarkablemotion capabilities are driven by 38 joints (4 on the neck, xed eyes) each controlled by a separate motor. Qrio sensesits own motion through accelerometers in the torso and feet.It is 58cm height and weighs approximately 6:5Kg.Honda engineers created ASIMO [4] that, with 26 degreesof freedom (2 on the neck,  xed eyes), can walk and performsome tasks much like a human. It is 120cm height and weightsabout 52Kg. It was the  rst robot able to climb stairs and run.The PINO project [5] started with the goals of: developinga platform for research on perception and behaviour, usingmultiple perception channels and many DOFs; investigatingrobot design that are well accepted by the general publicand develop affordable humanoid platform using off-the-shelfcomponents and low-precision materials. The size of the robot is carefully designed to be the size of the 1:5 year-old kids(height 70cm). It has 26 DOFs and 4:5Kg of weight.Another miniature humanoid robot is Fujitsu's HOAP-2 [6].This platform has been programmed to perform movementsfrom the Chinese martial art taiqi, Japanese Sumo wrestlingstances as well as to aid to robotics research. This robot'sweight is less then 7kg and its height is 50cm. It has a totalof 25 DOFs (2 in the neck, the eyes are  xed).The iCub will have head, torso, two arms/hands and twolegs. The legs will be used for crawling but, possibly, notfor biped walking. This will allow the system to explore theenvironment not only by manipulating objects but also throughlocomotion. For this reason, it is particularly important toequip the iCub with enough degrees of freedom to allowtransition between sitting and crawling posture, as well asto look down while manipulating objects lying on the  oor.In the current version of the design, the iCub is about 90cmtall, weighs 23 kg and has a total of 53 degrees of freedomorganized as follows: 7 for each arm, 8 for each hand, 6 forthe head, 3 for the torso/spine and 7 for each leg.The eye-head sub-system will include basic visual process-ing primitives, as well as low-level oculomotor control, visual,inertial and proprioceptive sensors. The iCub will have twoarms with the motor skills and sensory components requiredfor dexterous manipulation. From the control point of view,reaching and grasping primitives will be implemented togetherwith primitives to acquire tactile and proprioceptive informa-tion. It is expected that most of the actuators of the hand willbe located in the forearm. The hands will be underactuated tosave space, power consumption, and cost. This is implementedby means of rigid mechanical couplings, such as single tendonto bend two joints of a  nger alike, or by an elastic couplingof the joints.Most existing humanoid systems have a simpli ed headwith a small number of degrees of freedom. In our case, theinteraction with other robots or inpiduals is very important,justifying the need to include a greater kinematic complexity,while meeting very stringent design constraints, in terms ofweight and size. This paper main contributions are:  Review of anatomical data for neck and eyes  Derivation of the speci cations for a small size robotichead, based on data from humans.  Design of small size serial mechanism with impact pro-tection  Description and comparative analysis of several solutionsfor the (small size) neck design: parallel neck, spring likemechanism based on  exible spine and a serial solution. 机器人iCub头部设计英文文献和中文翻译:http://www.751com.cn/fanyi/lunwen_34027.html