Today the BBC ran a story “Who will win the race to develop a humanoid robot?” that covers the G1, a humanoid robot built by Chinese firm, Unitree, now featured at the Hannover Messe, one of the world's largest industrial trade shows. And the G1 is just but one of many humanoid robots competing on the market in an array of uses from production to medicine and even social roles.
Today in Japan, there are currently over a quarter million robots being utilised primarily due to declining birth rates and a shrinking workforce. This in addition to the benefits of robotics' economic benefits in covering certain labour sectors and in supporting further industrial mechanization. Aside from these reasons for the steep rise in robotics, Japan has been developing robots specifically to care for older people because there simply are not enough workers available to do rudimentary care work.
However, after over two decades of Japan's experiment with elder care and robotics, it has not proven to be the success that many claim it has been. There are myriad reportsof these robots creating more work for carers than that which they undertake, especially around those suffering from dementia. However, it seems that there may be beneficial uses for humanoid robots as recent studies demonstrate the future potential for these robots working in limited capacities children with autisim spectrum disorders (ASDs).
Robot-assisted therapy (RAT)
Researchers have developed humanoid robots with the hope that they can help children with ASDs in practicing essential social behaviors such as eye contact, facial expression recognition, and turn-taking.
The hope is that robot-assisted therapy (RAT) will improve the treatment of individuals with ASD because of recent advances in artificial intelligence and computer vision that allow these social robots to both read facial expressions and know if someone is frustrated, angry, happy or worried, all without the empathy and true emotional understanding that keeps them from becoming sentient. Developed through the help of artificial intelligence, these robots are often equipped with sensors, cameras, microphones and other technology so they can respond to touch, sounds and visual cues just as humans would.
Ideally, the humanoid technology will develop to a degree that social robots, together with artificial intelligence platforms paired with sensors, cameras, microphones and other technology (eg. computer vision) will be able to interact and engage with humans or other robots. These robots come in many different shapes and sizes and most are not mobile. Some have human-like faces on static pedestals while others are more representative of a small childlike form even if without very good mobility.
Though there are many prototypes in development and being tested, many of these humanoid robots are making the rounds for testing in medical facilities around the planet because there are still severe limitations in what robotics can offer the social care of children with ASD.
Let us explore a few of these models.
FACE (Facial Automaton for Conveying Emotions)
The
The FACE robot has been employed in several clinical trials to demonstrate its effectiveness as a mediator of behavioral interventions on children and on children with ASD. Studies have shown that this device aids in improving imitative skills and shared attention even if the cohort of ASD children enrolled was small. Another study confirms this preliminary evidence, underscoring that all participants have shown an improvement in their imitation abilities and social communication skills after RAT with FACE. Based on these preliminary data, researchers have suggested that treatment with FACE can develop pragmatic emotional responsiveness in children with ASD.
FACE does, however, suffer some limitations. Most obvious is that it is missing motion and mobility. It is also unable to express complex emotions nor can it combine facial emotional expressions with gestures. Lastly, the lack of mobility and motion reduces the variety of human–robot interactions.
Zeno is a humanoid child-size robot with a simple expressive face. It is light with a weight of 6.5 kilos and its dimensions are safe for children with a heigh of 63.5 centimetres. The robot’s face is measured in degrees of freedom, a term widely used to define the motion capabilities of robots, including androids (humanoid robots) through the acronym DoF meaning "degree of freedom."
Roughly speaking, the degree of freedom of a specific joint is the difference of the number of independent constraints from the sum of freedoms of bodies. By that calculation, revolute, helical, and prismatic joints have 1 DoF, cylindrical and universal joints have 2 DoF and spherical joints have 3 DoF.
Zeno's face has 8 DOF, while its neck has 3 DOF, and there is 25 DOF devoted to the body where motors there are actually used for simulating facial expressions. The body is equipped with servomotors for the legs, hips, waist, and shoulders. ZENO is a child-sized and -shaped robot and it has very limited expressive abilities (only six basic emotions). However, this capability combined with motion (it can move its arms and legs) gives it a human-like physical appearance.
While Zeno is more adorable and lifelike than FACE, it actually produces less excellent results clinically. It is known that physical appearance and mobility are two important factors that mediate human–robot interactions which is essential in the context of ASD, given that one of the major impairments in ASD is emotional understanding and recognition. The irony here is that Zeno's main disadvantage is the low number of DoF. The motor system of ZENO is quite limited in its bodily capabilities due to the low number of DoF which in turn drastically decreases the changes to design human-like actions. So, despite Zeno’s being more lifelike in form, ZENO has limited practical use in the context of ASD so far even if it is widely used in autism research. It is expected that this robot, with improvements in emotional recognition, could one day be at the fore oftherapeutic robotics.
Conclusions on the future of androids
As you can see, this technology is far from being implementable in the near future. In fact, some of the models I failed to cover here are utterly frightening and would frighten off most children and adults.
The upside is that advances are continuing towards finding a balance between a humanoid robot that can assist children with ASD in understanding expressions and expressing themselves while facing no danger from the robot. Time will only tell as the field of social care and robotics is exploding and it could be that we see a feasible prototype adopted by learning and healthcare facilities in the coming years.