Saturday , January 29 2022

The CSIRO researchers are looking for the future of robots


An impression of an artist's ocean, coastline or amphibian river. It would travel with water like water, but to browse and climb the legs. Photo: CSIRO.

Australian CSIRO national science research scientists can see future robots.

In an published paper Nature Machine IntelligenceFuture Science Platform (AIM FSP) of the active integrated matter of CSIRO indicates that the evolution of its engineering questions is constant.

This concept, known as Multi-Level Evolution (MLE), argues that today's robots are struggling with unstructured and complex environments, because they are not sufficiently specialized and have imitated animals of very different adaptations to survive in the environment.

In early January, the author of paper, Dr. David Howard, said that evolution does not matter what.

FIRST: A new UNSW course works by robotics and AI worldwide

"It seeks a much wider design space and creates effective solutions, which would not be immediately for human design.

"An animal like a manta ray or a kangaroo may be unusual, but it is perfectly calibrated for its environment," said Howard.

The paper argues that due to the discovery and characterization of high-speed technologies with less than 20 years of age, in advanced manufacturing and artificial intelligence, robots can design their molecules from one level to another, making them a very challenging circumstance.

Naturally-based algorithms will automatically design robots by combining various materials, components, sensors, and behaviors.

Advanced computer-based modeling was able to quickly test prototypes in simulated "real world" scenarios to determine how well it works.

The final result was simple, small, very integrative, highly specialized and very cost-efficient in the precision, work, environment and engineering of the robot. Self-adapting itself and automatically improving it.

An example would be an environmentally-friendly environmental monitoring robot in extreme environments.

Cross over difficult soils, collect data and eventually degrade it to avoid contaminating the environment.

The proximity of MLE design for robot design depends on land, climate, and other factors.

The robot designed to work in the Sahara Desert should use heat, sand and dust to survive. It could be solar power, slide on the sand dunes and use UV light.

Amazonia's thick and small plant would be a very different challenge.

A robot designed for this area can be crawled over trees and fall records, it works through biomass, such as the plant matter on the jungle floor and degrading it with moisture.

In both cases, MLE will automatically select the appropriate materials and components in a high-design robot design, whenever a robot performs a specific task.

The teams of engineers who demand an infinite process more than designing a single robot today.

Source link