Machine learning for autonomous cars - Deepstash

Machine learning for autonomous cars

Currently, there is no widely accepted and agreed basis for ensuring that the machine learning algorithms used in the vehicles are safe.

Autonomous vehicles will use artificial intelligence and machine learning to process the data collected from its sensors to help make decisions about their next actions. The algorithms will help identify the objects detected by the sensors and classify them according to the system's training.

2

STASHED IN:

12

MORE IDEAS FROM Autonomous cars: five reasons they still aren't on our roads

Fully autonomous cars

Elon Musk thinks his company Tesla will have fully autonomous cars available by the end of 2020.

However, there are some fundamental challenges to the safe introduction of these cars before they can be on our roads.

1

STASHED IN:

8

Social acceptability is an issue for those wishing to buy a self-driving car and others who will share the road as numerous automated and autonomous vehicles have been in high-profile accidents.

There is a risk of rejection of this technology if the public is not considered in the decisions about the introduction and adoption of self-driving vehicles. Without collaboration on how to make the car safe and provide evidence of that safety, this project will not gain traction.

STASHED IN:

5

Without recognized regulations and standards, self-driving cars will not be allowed on the road.

Currently, there are no sufficient standards and regulations for a whole autonomous system. The criteria for the safety of existing vehicles assume the presence of a human driver to take over.

STASHED IN:

5

Autonomous cars use several sensors to detect objects such as pedestrians, other vehicles, and road signs.

However, bad weather, heavy traffic, road signs with graffiti on them all interfere with the accuracy of its sensing capability. To enable true autonomous cars, these sensors have to work in all weather conditions anywhere on the planet.

1

STASHED IN:

7

Once an autonomous car is on the road, it will continue to learn. It will detect new objects and be subject to software updates.

The system should continue to be just as safe as its previous version and not forget previously learned behaviors. It should be able to show that any new learning is safe. The industry still has to reach an agreement on these points.

1

STASHED IN:

5

Deepstash helps you become inspired, wiser and productive, through bite-sized ideas from the best articles, books and videos out there.

GET THE APP:

RELATED IDEA

Not ready for self-driving cars

In 2015, the Guardian predicted that self-driving cars would be the norm in 2020. The Business Insider announced in 2016 that there would be 10 million self-driving cars on the road by 2020.

Yet, engineering teams are still struggling to make fully autonomous cars work properly.

3

STASHED IN:

7

STASHED IN:

0 Comments


Artificial eyes look to the future

Bionic eyes have been a mainstay of science fiction for decades, but now real-world research is beginning to catch up with far-sighted storytellers. A raft of technologies is coming to market that restore sight to people with different kinds of vision impairment.

Israeli surgeons implanted the world’s first artificial cornea into a bilaterally blind, 78-year-old man. When his bandages were removed, the patient could read and recognise family members immediately. The implant also fuses naturally to human tissue without the recipient’s body rejecting it.

4

STASHED IN:

28

Sci-Fi’s Influence On Technology

Science fiction, by definition, is science that isn’t real. But looking beyond the fantastical elements there are some very innovative ideas, a lot of which have had a real influence on modern technology. Here are some examples:

  • Star Trek: The motion Picture predicted the Internet as embodied in the movie by the all-enveloping sentient machine ‘V’Ger’.
  • Ready Player One predicted VR technology and virtual workplaces.
  • Total Recall predicted self-driving cars.
  • iRobot predicted self-driving cars that would allow the driver to take over too.
  • Minority Report predicted interactive augmented reality now being commercialized as wearable pair of glasses.
  • Back to the Future predicted the self-tying shoes.

3

STASHED IN:

73