Designers of robot pets are fighting a never-ending battle with consumers to provide entertaining and realistic gadgets that respond to human interaction in ever more nuanced ways, mimicking the behavior of real pet animals or even people. Researchers in Taiwan are now looking at a new design paradigm that could see the development of a robot vision module that might one-day recognize human facial expressions and respond appropriately.
Part of the problem is that robot design takes a long time, while the consumer life cycle of any given product is very short. Moreover, fixed prototypes and repetitive behavior in domestic robots for entertainment is no longer of interest to sophisticated users. Today, they expect their robot pets to be almost as good as the "robots" they see in 3D movies and games.
The researchers, Wei-Po Lee, Tsung-Hsien Yang and Bingchiang Jeng of National Sun Yat-sen University, have now turned to neural networks to help them break the cycle of repetitive behavior in robot toys and to endow them with almost emotional responses to interactions.
The idea of emotional robots seems as pathetic to me as the idea of Japanese men taking virtual robots on vacation. If your abode is too small for a dog at least consider a cat. Or a talking parrot like Alex.
On the bright side, a market for emotional robots will help drive the development of smarter machines so that machine intelligence can become smart enough to some day start a world war between humans and terminators. That war will relieve the ennui of those who are not killed before they get a chance to become underground heroic guerrilla fighters. Most people who need an emotional robot probably stand a chance of feeling more fulfilled by their role in a human-terminator war.
What would be more interesting than emotional robots: dogs genetically engineered for higher intelligence.
What would be far more useful that emotional robots: artificial intelligence that could accelerate the development of rejuvenation therapies.
To test whether residents connected better with Sparky or Aibo, researchers divided a total of 38 nursing home residents into three groups. All were asked questions to assess their level of loneliness. One group saw Sparky once a week for 30 minutes, another group had similar visits with Aibo, and a control group saw neither furry nor mechanical critter.
During visits, Marian Banks, Banks’ wife and co-researcher, brought Sparky or Aibo into a resident’s room and placed the pet companion near the resident. Both pets interacted with residents -- wagging their tails and responding to the people they visited.
After seven weeks, all residents were asked questions about how lonely they felt and how attached they were to Sparky or Aibo.
The residents who received visits from real and artificial pooches felt less lonely and more attached to their canine attention-givers than those who got visits from neither.
There was no statistical difference between whether the real or robotic dog did a better job easing loneliness and fostering attachments.
I would just plain refuse to have my heart warmed up by a visit from Aibo. Get that crude excuse for a silicon-based life form out of here.