The technology behind self-driving cars has been racing ahead and as long as they are cruising along familiar streets, seeing familiar sights, they do very well.<\/p>\n
But the University of TorontosFlorian Shkurtisays that when driverless vehicles encounter somethingunexpected, all that progress can come screeching to a halt.<\/p>\n
He offers the exampleof a self-driving car that is following a large truck on a winter road.<\/p>\n
Theres a wind gust and now the snow is coming at you, so you cant see anything, saysShkurti, anassistant professor in the department of mathematical and computational sciences at U of TMississauga who runs the Robot Vision and Learning(RVL) lab.And suppose your LIDAR (light detection and ranging system)misperceives the snow as an array of objects, so it thinks there are a million small objects coming at the car.<\/p>\n
Shkurtis research extends far beyondself-driving cars to autonomous systems in general.How do they learn?How we can make them learn better?How they can successfully navigate complex environments at the service of humans?That includes making sure that robots can handle so-called edge cases, like the snowy truck example cases where the robot comes across a rare scenario, for which theres little or no training data.<\/p>\n
Then you have to either collect more data, or you have to accept that there will be these rare events that your perception system wont recognize, Shkurti says.<\/p>\n
Simulation is an important training tool. Self-driving cars, for example, can be trained on simulated roads and highways before theyre let loose on actual city streets. But scalability remains a challenge. If an autonomous system has to be specially trained for every possible scenario it might encounter, progress will be haltingly slow; there will be no way to take whats been learned from one scenario and scale it up so that the system can handle more general cases.<\/p>\n
In an ideal world, Shkurti says, a robot could learn similar to the way a human would.<\/p>\n
Take, for example,robots that help scientists collect data underwater an effort Shkurti has been involved with for several years.A human diver has to collect data manually, one data point at a time, one location at a time, Shkurti says. Itspainstaking work; its not scalable.<\/p>\n
An autonomous robot, on the other hand, could take over the data collection process if itscapable of maneuvering underwater and equipped with a camera and other sensors. If the robot could understand what its doing if it has a model of what the scientist thinks is important to pay attention toin a particular environment then the robot could collect data on behalf of the scientist.<\/p>\n
Such an approach has many benefits, according to Shkurti: Its much cheaper to deploy additional robots than to train more scientists;and it frees up the scientist to look after higher-level tasks. The scientist can give the robot some hints as to where to collect the data but then the robot can take care of the rest, hesays.<\/p>\n
Shkurti, whodid his undergraduate studies at U of T before earning his PhD in computer science at McGill in computer science, was hired by U of T in 2018.He recently received aConnaught New Researcher Awardfor a project titled Robotics and Machine Learning in the Wild: New Directions in Automated Environmental Monitoring.<\/p>\n
Hey says that while everything about computer science fascinates him, the field of robotics holds special appeal.<\/p>\n
Robotics lets you play in different playgrounds, like control, perception, and machine learning, he says. It allows you to examine these different fields, and I really valued that and I still value it.<\/p>\n
As for the lofty philosophical questions that sometimes crop up when people talk about advanced computer systems such aswhether machines could learn tothink Shkurti prefers to stay focused on the science. Machines can reason, he says, andthey can try to act optimally as they strive to achieve their goals.<\/p>\n
If thats thinking, then theyre doing it, he says. But I dont spend very much time worrying about consciousness. I have enough other things to worry about.<\/p>\n
<\/p>\n
View post:<\/p>\n
Robots in the wild: U of T's Florian Shkurti on overcoming 'edge cases' in machine learning - News@UofT<\/a><\/p>\n","protected":false},"excerpt":{"rendered":" The technology behind self-driving cars has been racing ahead and as long as they are cruising along familiar streets, seeing familiar sights, they do very well. But the University of TorontosFlorian Shkurtisays that when driverless vehicles encounter somethingunexpected, all that progress can come screeching to a halt. Continue reading