London, Oct 30: According to an Indiana University neuroscientist, experiments involving real and simulated robots that suggest the relationship between physical movement and sensory inputs, are crucial to the development of more intelligent machines.
Olaf Sporns, a neuroscientist at Indiana University has said that tests involving two real and one simulated robot show that feedback between sensory input and body movement will help better understand the relationship of the robots towards their surrounding world.
Though scientists studying artificial intelligence have traditionally separated physical behaviour and sensory input, “the brain's inputs are not independent".
"For example, motor behaviour has a role to play in what the body senses from the environment," said Sporns.
Sporns said an increasing number of researchers were taking this approach, known as "embodied cognition". For his study, he collaborated with roboticist Max Lungarella from Tokyo University in Japan and created experiments that would test the idea.
The team used a four-legged walking robot, a humanoid torso and a simulated wheeled robot. All three robots had a computer vision system trained to focus on red objects.
They found that the walking and wheeled robots automatically moved towards red blocks in their proximity, while the humanoid robot grasped red objects and moved them closer to its eyes while tilting its head for a better view.
To measure the relationship between movement and vision, the researchers recorded information from the robots' joints and field of vision, and used a mathematical technique to see how much of a causal relationship existed between sensory input and motor activity.
"We saw causation of both kinds. Information flows from sensory events to motor events and also from motor events to sensory events. It is an important experimental demonstration of this aspect of embodied cognition. This work and that of others is now making it more practical and less of a metaphor," Sporns added.
He believes similar experiments will also show the same relationship in animals, as evolution has produced bodies and brains that work together to understand the world.
“Such tests would be much harder to carry out, but researchers are starting to investigate how it might be done,” Sporns said.
“Using similar approaches, it should be possible to produce more efficient cognitive systems, like those in nature, without specialising on a particular task" such as movement or vision,” added Daniel Polani, who researches artificial intelligence at Hertfordshire University in the UK.
Bureau Report