Tesla Autopilot, At Times Unsafe For Automatic Lane Changes, Lays Bare Machine-Learning Brittleness

  • 📰 Forbes
  • ⏱ Reading Time:
  • 41 sec. here
  • 2 min. at publisher
  • 📊 Quality Score:
  • News: 20%
  • Publisher: 53%

Technology Technology Headlines News

Technology Technology Latest News,Technology Technology Headlines

Tesla's Autopilot lays bare the brittleness of machine-learning

that seeks to make sense of the images, identifying each pertinent artifact as seen in the visually captured data, and discern which artifacts are important for the driving act and which are not particularly important at the time. A tree that’s standing some twenty feet away from the roadway is probably not as crucial to discern as might be a car that’s in the lane next to you and coming up upon you at a fast rate of speed.

In addition to cameras, there are often also radar sensors involved. These radar units send out radar waves that bounce off objects and return a signal to the radar unit, sometimes, allowing for the AI to gauge the distance to objects, sometimes, depending upon the capabilities of the radar sensor and the kind of objects nearby to the car.for the AI system to process the data being captured about the street scene surrounding the car.

• Or, it could be that the camera is not able to adequately gauge the speed of the car in the other lane, perhaps only noting that the car is back far enough to allow for an opening to make the lane change, but not realizing that the car is fast approaching and will be in the same spot as the lane change once the Tesla maneuvers into the lane (a sensory-related

 

Thank you for your comment. Your comment will be published after being reviewed.
Please try again later.

I hope to own one in future

We have summarized this news so that you can read it quickly. If you are interested in the news, you can read the full text here. Read more:

 /  🏆 394. in TECHNOLOGY

Technology Technology Latest News, Technology Technology Headlines