How Do We Teach "Intuition" to Autonomous Machines?

June 18, 2019, Kitchener, Ontario

Posted by: Robert Deutschmann, Personal Injury Lawyer

Currently artificial intelligence and autonomous machines rely on machine learning and programming to ‘learn’ things. The computers then analyze the situation that a machine is in and apply the correct response based on their programming. This could all change if we could give them intuition or that ‘spidey sense’ that humans often rely on to make decisions.

Humans often make what seem to be irrational decisions based on ‘gut feelings’ that direct them to take advantage of an opportunity or to decline it. If we could teach machines this it could actually improve their performance, at least according to Andres Arrieta, assistant PhD of Mechanical Engineering at Purdue university. Purdue researchers have built new sensors that are inspired by living creatures like bats and spiders who have special nerve endings built into their hair and feathers.

Adding this sort of sensor to machines will allow them to collect vastly more data. In nature we don’t collect every piece of data – living creatures filter out the information that they need. This is the skill that needs to be taught to do.

According to the press release from Purdue University:

The nerve endings – mechanosensors – only detect and process information essential to an animal’s survival. They come in the form of hair, cilia or feathers.

“There is already an explosion of data that intelligent systems can collect – and this rate is increasing faster than what conventional computing would be able to process,” said Arrieta, whose lab applies principles of nature to the design of structures, ranging from robots to aircraft wings.

“Nature doesn’t have to collect every piece of data; it filters out what it needs,” he said.

Many biological mechanosensors filter data – the information they receive from an environment – according  to a threshold, such as changes in pressure or temperature.

A spider’s hairy mechanosensors, for example, are located on its legs. When a spider’s web vibrates at a frequency associated with prey or a mate, the mechanosensors detect it, generating a reflex in the spider that then reacts very quickly. The mechanosensors wouldn’t detect a lower frequency, such as that of dust on the web, because it’s unimportant to the spider’s survival.

The idea would be to integrate similar sensors straight into the shell of an autonomous machine, such as an airplane wing or the body of a car. The researchers demonstrated in a paper published in ACS Nano that engineered mechanosensors inspired by the hairs of spiders could be customized to detect predetermined forces. In real life, these forces would be associated with a certain object that an autonomous machine needs to avoid.

But the sensors they developed don’t just sense and filter at a very fast rate – they also compute, and without needing a power supply.

“There’s no distinction between hardware and software in nature; it’s all interconnected,” Arrieta said. “A sensor is meant to interpret data, as well as collect and filter it.”

In nature, once a particular level of force activates the mechanoreceptors associated with the hairy mechanosensor, these mechanoreceptors compute information by switching from one state to another.

Purdue researchers, in collaboration with Nanyang Technology University in Singapore and ETH Zürich, designed their sensors to do the same, and to use these on/off states to interpret signals. An intelligent machine would then react according to what these sensors compute.

These artificial mechanosensors are capable of sensing, filtering and computing very quickly because they are stiff, Arrieta said. The sensor material is designed to rapidly change shape when activated by an external force. Changing shape makes conductive particles within the material move closer to each other, which then allows electricity to flow through the sensor and carry a signal. This signal informs how the autonomous system should respond.

“With the help of machine learning algorithms, we could train these sensors to function autonomously with minimum energy consumption,” Arrieta said. “There are also no barriers to manufacturing these sensors to be in a variety of sizes.”

Posted under Accident Benefit News

View All Posts

About Deutschmann Law

Deutschmann Law serves South-Western Ontario with offices in Kitchener-Waterloo, Cambridge, Woodstock, Brantford, Stratford and Ayr. The law practice of Robert Deutschmann focuses almost exclusively in personal injury and disability insurance matters. For more information, please visit www.deutschmannlaw.com or call us toll-free at 1-866-414-4878.

It is important that you review your accident benefit file with one of our experienced personal injury / car accident lawyers to ensure that you obtain access to all your benefits which include, but are limited to, things like physiotherapy, income replacement benefits, vocational retraining and home modifications.

Practice Areas

  1. Car accidents
  2. Motorcycle accidents
  3. Automobile accident benefits
  4. Catastrophic injury
  5. Brain injury
  6. Paraplegia and Quadriplegia
  7. Spinal cord injury
  8. Drunk driving accidents
  9. Concussion syndrome
  1. Wrongful death
  2. Bicycle accidents
  3. Disability insurance claims
  4. Slip and fall injury
  5. Fractures or broken bone injury
  6. Pedestrian accidents
  7. Chronic pain
  8. Truck accidents
  9. Amputation and disfigurement

Personal Injury Blog

Oct 17, 2019
Safety Is the Priority as Montreal Lowers Downtown Speed Limits
Oct 15, 2019
Car accident deaths are up due to lack of seat belt use - Buckle up
Oct 10, 2019
Good Driving Depends on Good Vision
Oct 08, 2019
Why we see and then forget motorcyclists - CBC Radio
Oct 03, 2019
Spinal Cord Injury is A Leading cause of paralysis in Canadians - Exoskeletons offer hope for mobility
Oct 01, 2019
Revisions to Ontario's Auto Insurance Plan are Problematic

More Personal Injury Articles » 
Review our services

Connect with us

Facebook Twitter Linkedin Youtube Google