A Humanoid may be defined as something that resembles or looks like a human and having characteristics like opposable thumb, ability to walk in upright position, etc.
Fig. 1: Image of Simon Robot developed at Socially Intelligent Machines Lab At Georgia Institute Of Technology That Can Learn And Adapt To Humans
Nowadays the concept of Humanoids is being widely implemented in Robotics and these robots are called Humanoid Robots or may be simply “Humanoids”.
In general Humanoid robots have a torso with a head, two arms and two legs, although some forms of humanoid robots may model only part of the body, for example, from the waist up. Some humanoid robots may also have a 'face', with 'eyes' and 'mouth'. Androids are humanoid robots built to resemble a male human, and Gynoids are humanoid robots built to resemble a human female.
A Humanoid robot is fully automated as it can adapt to its surroundings and continue with its goals.
Fig. 2: Image Of Various Humanoid Robots Wabot-1, P2, HRP-2
Historical Timeline of Humanoids
In 1921, Karel Capek coined the term Robot. After that, during 1941-42, Isaac Asimov proposed Three Laws of Robotics and in 1948, Norbert Wiener formulated the principle of Cybemetics.
The theory of Zero Moment Point (ZMP) was given by Miomir Vukobratovic, in 1969, which is the most fundamental theoretical model of Biped Locomotion.
In 1973, in Waseda University, Tokyo, WABOT-1 was developed, which was capable to communicate with a person in Japanese, could do distance and direction measurements using sensors, artificial ears and eyes and an artificial mouth.
More developments in 20th century include Greenman in 1983, WABOT-2 in 1984, WHL-11 in 1985, the musician robot- WASUBOT in 1985, series of seven biped robots called E0-E6 by Honda in 1986, the full scale anthromorphic robot- MANNY in 1989, with 42 degrees of freedom, Honda’s P1 through P3 in 1993, HADALY and WAIBAN in 1995, SAIKA in 1996 and ASIMO by Honda in 2000.
In 21st Century, further additions were done by Sony’s ORIO in 2001, HOAP and ACTROID in 2003, PERSIA and KHR-1 in 2004, PKD android and WAKAMARU in 2005, TOPIO in 2007, Justin, KT-X and NEXI in 2008, SURALP in 2009, Robonaut-2, SURENA-II and HRP-4C in 2010.
Features of Humanoid Robots
The characteristics features of Humanoid Robots include:
. self-maintenance
. autonomous learning
. avoiding harmful situations to people, property, and itself
. safe interacting with human beings and the environment
Sensors and Actuators
Technical Components: Sensors and Actuators
The two most important components of Humanoid Robots are Sensors and Actuators.
Sensors are the devices which sense something, like environmental parameters (heat, sound, light, temperature, etc), physical and physiological parameters (like movement, orientation, etc.).
Actuators are nothing but the motors which are responsible for the motion and movement of the robots.
Sensors: Proprioceptive and Extroceptive
The position, orientation and the speed of humanoid’s body and joint are sensed by the proprioceptive sensors which consist of accelerometers, tilt sensors, etc.
Extroceptive sensors consist of arrays of tactels (touch receptors) which are used to provide data on what is being touched, forces and torques transferred between robots and other objects.
In humanoid robots, CCD cameras are used for capturing image information and microphones and speakers are used for sound reception and production respectively.
Actuators
Actuators are used to perform like joints and muscles. Mostly humanoid robots use rotator actuators to achieve the effect as human motion. The actuators can be pneumatic, hydraulic, electric or ultrasonic.
Working & Control
Working and Control Mechanism
Concept of Zero Moment Point:
This concept explains the dynamic balance of humanoids during walking which requires information about the contact forces and the current and desired direction of motion.
As per the ZMP Theory, the pressure under supporting foot can be replaced by the appropriate reaction force acting at a certain point of the mechanism’s foot. Since the sum of all moments of active forces with respect to this point is equal to zero, it is termed as the Zero Moment Point (ZMP).
Fig. 3: Figure Illustrates Principle Of ZMP Theory Used In Humanoid Robots
Gi = Gravitation force of the ithlink acting at the mass center Ci
Fi= inertial force of the ith link acting at the mass center Ci
Mi= moment of the inertial force of the ith link for Ci
Ri= the resultant ground reaction force
R= Rv + Rf
M= Mh+Mf
Where subscripts v, h and f denote vertical, horizontal and frictional components respectively.
ZMP is the point on the walking ground surface at which the horizontal components of the resultant moment generated by active forces and moments acting on humanoid links are equal to zero
It is the point on the floor at which the moments around x and y axes generated by reaction force and moment are zero.
In addition to the concept of ZMP, several other planning and control mechanisms are used for self-collision detection, path planning and obstacle avoidance to allow humanoids to move in complex environments.
Degrees of Freedom (DOF): the number of independent ways in which the robot can exhibit motion.
Development Psychology
Robotics versus Development Psychology
Developmental psychology is most typically employed in robotics research as a source of inspiration. Models from developmental psychology often offer behavioral decomposition and observations about task performance which may provide an outline for a software architecture.
However, a developmental approach to robot construction also provides practical benefits. Human development exploits a gradual increase in both internal complexities to optimize the acquisition of new skills.
Fig. 4: An Image Showing Developmental Psychology In Robotics With Anthromorpic Head For Human Interaction Work
Fig. 5: An Image Showing Developmental Psychology In Robotics With Upper Torso Humanoid For Object Interaction
From an external perspective, the robot’s behavior is quite rudimentary. Given a visual stimulus, typically by waiving an object in front of its cameras, the robot saccades to foveate on the target, and then reaches out its arms toward the target. Early reaches are inaccurate, and often in the wrong direction, but after a few hours of practice, the accuracy improves drastically, - this is the effect of developmental psychology on humanoid.
Fig. 6: Simple Block Diagram Of Devlopmental Psychology In Humanoid Robots
Challenges
Challenges before Humanoid Robotics
The prime problems facing Humanoid research can be given as following four questions:
1. How does the robot know when to imitate?
2. How does the robot know what to imitate?
3. How does the robot map observed actions into behavioral responses?
4. How does the robot evaluate its actions, correct errors, and recognize when it has achieved its goal?
A socially intelligent robot should be able to use imitation for the variety of purposes that humans do. Imitation can be a mechanism for developing social attachments through imitative play and for gaining an understanding of people. Imitation can also be used to explore and expand the range of possible actions. Finally, imitation can be a mechanism for establishing personal identity and discovering distinctions between self and other. A social robot should selectively use imitation to achieve many of these goals.
Faced with an incoming stream of sensory data, the robot must make a number of decisions to determine what actions in the world are appropriate to imitate. The robot must not only be able to distinguish the class of stimuli (including humans and perhaps other robots) which might be a good model but also determine if the current actions of that agent are worthy of imitation. The robot must determine which of these objects and events are necessary to the task at hand , which events and actions are important to the instructional process but not to the task itself , and which are inconsequential. The robot must also determine to what extent each action must be imitated. The robot must also recognize the important aspects of the objects being manipulated so that the learned action will be applied to only appropriate objects of the same class.
ASIMO
Some Humanoids
1. Honda’s ASIMO (Advances Step in Innovative Mobility):
ASIMO can maneuver toward its destination without stopping by comparing any deviation between the input map information and the information obtained about the surrounding area from its floor surface sensor.
Fig. 7: An Image Of Honda’s ASIMO Humanoid
ASIMO is capable of interpreting the postures and gestures of humans and can move independently in response. It can greet approaching people, follow them, move in the direction they indicate, and even recognize their faces and address them by name. Further, utilizing networks such as the Internet, ASIMO can provide information while executing tasks such as reception duties. ASIMO is the world's first humanoid robot to exhibit such a broad range of intelligent capabilities.
MYON
2. MYON
Fig. 8: An Image Of MYON Humanoid Developed By The Neurorobotics Research Laboratory At Humboldt University In Berlin
Myon, a new humanoid robot designed by the industrial design studio, Frackenpohl Poulheim and developed by the Neurorobotics Research Laboratory at Humboldt University in Berlin, Germany, has the special ability to continue to operate even if some of its electronic components are damaged or separated. It’s also the first humanoid robot in the world that can be disassembled into parts (and put back together) while the parts which make it up continue to work steadily and autonomously (its energy source, the central motor and its neural network remain functioning).It has a robust “skin” made out of fiberglass.
Future
A Peep into the Future: Pervasive Robotics
In the near future we will require small, light and cheap robots that exhibit complex behaviors. For example, M2-M4 Macaco project has developed a portable system which is capable of emulating the head of different creatures both aesthetically and functionally. The pervasive robotics will concentrate its research on certain critical aspects like Object Analysis, Social Mechanisms and Navigation.
Conclusion
Though the technology has advanced much in the field of Humanoid Robotics, there are still several problems which need attention. The technological brilliance of the humanoids is required to be sharpened more and the shortcomings in the results must be dealt with properly. The field of Humanoid Robotics is a very promising one and there are a lot of changes needed to be brought in this technology.
