Tuesday, 27 June 2017

Space Robot Technology Helps Self-Driving Cars and Drones on Earth

Support Robots to Navigate Independently
 
The significance of making units of self-driving cars together with grocery delivery through drone could be revealed through an improbable source – autonomous space robots.

An assistant professor of aeronautics and astronautics, Marco Pavone has been creating technologies to assist robots in adjusting to unknown as well as altering environments. Pavone had been working in robotics at Jet Propulsion Laboratory of NASA before coming to Stanford and had maintained relationships with NASA centres together with collaboration with the other departments at Stanford. He views his work in space and Earth technologies as complementary.

 He commented that in a sense, some robotics techniques which tend to have been designed for autonomous cars could be very useful for spacecraft control. Similarly the algorithms which he and his students devised to assist robots make decisions and assessments on their own with a span of a second could help in space exploration as well as they could improve on driving cars and drone from the Earth.

One of the projects of Pavone tends to centre on supporting robots to navigate independently in bringing space debris out of orbit, delivering tools to astronauts and grasp spinning, speeding objects out of the vacuum of space.
 
Gecko-Inspired Adhesives
 
There is no boundary for error while grabbing objects in space. Pavone informed that in space when you approach an object, if you are not very careful in grasping it at the time it is contacted, the object would float away from you. Bumping an object in space would make recovering it very difficult.

Pavone had teamed up with Mark Cutkosky, a professor of mechanical engineering, who had spent the last decade perfecting gecko-inspired adhesives, in order to resolve the grasping issue.

 The gecko grippers support a gentle approach as well as a simple touch in order to `grasp’ an object, enabling easy capture and release of spinning, unwieldy space debris. However the delicate navigations needed for grasping in space is not an easy job. Pavone had stated that one have to operate in close proximity to other objects, spacecraft or debris or any object one might have in space that needs advanced decision making potentials.

 Pavone together with his co-workers developed systems which enabled space robot to independently respond to such flexible situations and competently grab space objects with their gecko-strippers.
 
Perception-Aware Planning
 
The subsequent robot could move as well as grab in real time, updating its decisions at a rate of several thousand times a second. This kind of decision-making technology is said to be beneficial in solving navigation issue with drones that are Earth-bound.

 A graduate student Benoit Landry had stated that for these types of vehicles, navigating at high speed in proximity to buildings, people together with the other flying objects seems difficult to perform. He focused that there seems to be a delicate interplay between making decisions and environmental perception. He added that in this perceptive, several aspects of decision making for independent spacecraft tend to be directly significant to drone control.

Landry together with Pavone have been working on `perception-aware planning’ that enables drones to consider fast routes as well as to `see’ their surroundings besides improved estimate on where they are. The work is presently being extended towards handling of interactions with the humans, a main section to organize autonomous system like the drones and self-driving cars.

 



Reduced Gravity Atmospheres
 
Landry had also mentioned that the background of Pavone at NASA had been a good complement to the academic work. When a robot is said to land on a small solar system body type an asteroid, added challenges tend to come up.

 These atmospheres seem to have total different gravity than the Earth. Pavone had stated that if one were to drop an object from waist-height, the same would take a couple of minute to settle to the ground. Ben Hockman, a graduate student in the lab of Pavone, had worked on a cubic robot known as Hedgehog, in order to deal with low-gravity atmospheres such as asteroids.

 The robot passed through uneven, rugged and low-gravity territories by hopping rather than driving like the traditional rovers. Ultimately, Pavone and Hockman desired Hedgehog to be capable of navigating and carrying out tasks without being obviously told how to perform it by a human located millions of miles away. Hockman had mentioned that the prevailing Hedgehog robot is said to be designed for reduced gravity atmospheres though it could be adjusted for Earth.

It would not hop quite that far since we tend to have more gravity though it could be utilised to cross more rugged territories where wheeled robots are unable to go. Hockman viewed the research that he had been doing with Pavone as core scientific exploration adding that science attempts to answer the difficult questions we don’t know the answers to and exploration seeks to find whole new questions we don’t even know yet how to ask.

No comments:

Post a comment

Note: only a member of this blog may post a comment.